M aking The Ethical and Legal Implications of Health Care Informed Consent in the °* Patient-Practitioner Decisions Relationship Volume Two: Appendices Empirical Studies of Informed Consent Freddi; Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research Library of Congress card number 82-600637 For sale by the Superintendent of Documents U.S. Government Printing Office Washington, D.C. 20402 Making | Health: Care ~ Decisions The Ethical and Legal Implications of Informed Consent in the Patient-Practitioner Relationship Volume Two: Appendices Empirical Studies of Informed Consent October 1982 President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research OL D0! P M77] President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research ’ Morris B. Abram, M.A., ].D., LL.D., Chairman, New York, N.Y. H. Thomas Ballantine, M.D., M.S, D.Sc.* Harvard Medical School George R. Dunlop, M.D. University of Massachusetts Mario Garcia-Palmieri, M.D.t University of Puerto Rico Bruce K. Jacobson, M.D. * Southwestern Medical School Albert R. Jonsen, S.T.M., Ph.D.t University of California, San Francisco John J. Moran, B.S.* Houston, Texas Arno G. Motulsky, M.D. University of Washington Daher B. Rahi, D.O. St. Clair Shores, Michigan Anne A. Scitovsky, M.A.t Palo Alto Medical Research Foundation Seymour Siegel, D.H.L. Jewish Theological Seminary of America, New York Lynda Smith, B.S. Colorado Springs, Colorado Kay Toma, M.D.* Bell, California Charles J. Walker, M.D. Nashville, Tennessee Carolyn A. Williams, Ph.D.t University of North Carolina, Chapel Hill *Sworn in August 12, 1982. t Term expired August 12, 1982. Staff Alexander M. Capron, LL.B., Executive Director Deputy Director Barbara Mishkin, M.A, ].D. Assistant Directors Joanne Lynn, M.D. Alan Meisel, J.D. Alan |]. Weisbard, J.D. Professional Staff Mary Ann Baily, Ph.D. Dan Brock, Ph.D. Allen Buchanan, Ph.D. Andrew Burness, M.B.A. Kathryn Kelly, M.S.W. Susan Morgan Marian Osterweis, Ph.D. Renie Schapiro, M.P.H. Daniel Wikler, Ph.D. Research Assistants Michelle Leguay Katherine Locke Ruby J. Sherrod Jeffrey Stryker Special Consultants Caroline Kaufmann, Ph.D. Louis Morris, Ph.D. Administrative Officer Anne Wilburn Editors Sharin Sachs Linda Starke Support Staff Florence Chertok Gretchen Erhardt Ruth Morris Clara Pittman Kevin Powers Nancy Watson President's Commission Commonwealth Fellows and Student Interns Susan Formaker (1982) Kenneth Kim (1982) Eddie Lockard (1982) Stephen Massey (1982) Ruth Oratz (1981) Lisa Rovin (1982) William Thompson (1981) Mindy Werner (1982) This volume consists of materials prepared as background for the Commis- sion’s report on this subject. KF 3%27 I5 M2 | 9422 v. 2 PUBL Table of Contents Volume Two: Appendices (Empirical Studies of Informed Consent) A. A Review of Empirical Studies on Informed Consent and Decisionmaking 1 Information Needs and Desires of Patients 2 Influences on Information Disclosed 9 Patients’ Comprehension of Information 12 Decisionmaking by Patients 13 Conclusion 153 B. Views of Informed Consent and Decisionmaking: Parallel Surveys of Physicians and the Public 17 (Louis Harris and Associates) Executive Summary 17 Index of Tables 27 Introduction 33 Health Providers and Health Consumers: The Context of Informed Consent 53 What Physicians Tell Patients: Patterns of Disclosure 87 When the Patient Is Not Told: Patterns of Withholding 119 Informed Consent 149 Description of Medical Decisionmaking 183 Evaluation of the Medical Decisionmaking Process 201 Decisions to Forego Treatment or Terminate Life 217 Impact of Professional Education and Values on Physicians’ Attitudes and Behavior of Physicians Related to Informed Consent 263 Profiles of the Public and Physician Samples 285 Annotated Questionnaires 293 C. Informed Consent and the Structure of Medical Care (Charles Lidz and Alan Meisel) 8317 Introduction 317 Research Methods 320 The Setting and the Routine 323 Findings 328 Summary of Findings 390 Conclusions 399 D. Treatment Refusal in Medical Hospitals (Paul S. Appelbaum and Loren H. Roth) Methods Results Discussion Conclusion Volume One: Report and Volume Three: Appendices (Studies on the Foundations of Informed Consent) are available from the Superintendent of Documents. 411 414 417 467 476 A Review of Empirical A Studies on Informed Consent and Decisionmaking In order to understand the decisionmaking process and to assess the extent to which patients have a basis for effective participation, the existing empirical literature was reviewed not only on informed consent per se, but also, from the disciplines of sociology and psychology, on communicataion,’ information processing, and decisionmaking in general. This review pro- vided essential background for Volume One of this Report, as well as the impetus for the new empirical studies discussed in this Volume. Answers to the following questions were sought: (1) How much information do doctors disclose to patients and under what circumstances do they withhold information? (2) How much information do patients want? (3) What other factors influence the nature and quantity of fnformation disclosed to or withheld from patients? (4) How do illness and other situational factors affect dis- closure, comprehension, and decisionmaking? (5) To what extent do patients comprehend information about their condition and treatment and how do the method of disclosure and the actual presentation of information affect comprehension? (6) Does information (especially about uncertainties and risks) have negative psychological effects on patients and how does such information affect treatment decisions? October 1982 1 Although health care professionals from many disciplines communi- cate with patients, most empirical studies in this area relate to commu- nication between physicians and patients. In the discussion that fol- lows, therefore, most of the references are to the doctor-patient communication literature. 2 Making Health Care Decisions: Appendix A (7) How do patients make treatment decisions? How and to what extent is information used to make such decisions? The extent to which existing research has answered each of these questions is summarized below. First, however, a few comments need to be made about the nature of the evidence, and the methodological limitations of the studies must be men- tioned. The empirical studies of informed consent are limited both in number and in scope. Generally, rather than asking whether disclosure took place, it is presumed to have occurred. Most studies therefore focus on comprehension of information, which is usually measured in terms of memory and recall rather than understanding. In these studies, as in many of the broader studies of doctor-patient communication, hypothetical situa- tions are often used to elicit physicians’ and patients’ reactions. Thus we have information about what people say they would want or would do under various circumstances, but we know less about what actually occurs when health care professionals and patients meet. Few studies have examined the decision- making process. Little is known, therefore, about how informa- tion is used. The designs of most studies have failed to include control groups. Thus, for example, we do not know whether patients and nonpatients understand or forget differently. Fi- nally, a general shortcoming of existing studies is that they fail to take account of the broad context in which health care is given. Most studies use small groups of patients or nonpatients as subjects and examine one particular treatment. The tremen- dous variety of health care settings, medical problems, the nature of the decisions to be made, and the personal and social characteristics of doctors and patients are virtually ignored. All these methodological shortcomings render the body of existing empirical studies less useful than it might otherwise be. It is difficult to generalize from one idiosyncratic study to a group of people or situations. We are therefore left with impor- tant impressionistic evidence but few definitive answers. Information Needs and Desires of Patients How Much Information Do Physicians Disclose to Patients and Under What Circumstances Do They Withhold Informa- tion? In order for patients to make informed choices about their 2 For two excellent reviews of the shortcomings of existing empirical studies, see Alan Meisel and Loren H. Roth, What We Do and Do Not Know About Informed Consent, 246 J.A.M.A. 2473 (1981); Barbara H. Stanley, Informed Consent and Competency: A Review of Empirical Research, in Natalie Reatig, ed., Competency and Informed Consent: Papers and Other Materials Developed for the Workshop “Empirical Research on Informed Consent with Subjects of Uncertain Compe- tence, National Institutes of Mental Health (Jan. 12-13, 1981). A Review of Empirical Studies 3 health care, there are several kinds of information generally thought to be required, including the nature of the patient's condition and recommended treatment, possible alternative treatments, and the relative risks and benefits of the recom- mended and alternative treatments. Although disclosure of such information is necessary for shared communication and is a key component of the doctrine of informed consent, surpris- ingly little is known about what physicians actually discuss with their patients. It generally appears that physicians dis- close less information than patients want. Four major reasons have been postulated to explain why physicians appear to disclose limited information to their pa- tients. First, it has been suggested that patients do not wish to have such information since if they did want to know more they would ask. Silence on the part of the patients is interpreted, therefore, as lack of interest. Second, it has been argued that patients would not under- stand detailed information about their conditions and about treatments, uncertainties, and risks. The claim here is very simply that laypeople lack the necessary background in order to comprehend fully medical details and that such comprehen- sion may be further compromised by the anxiety surrounding the illness itself. Third, it has been suggested that other situational factors, including particular characteristics of both doctors and pa- tients, time pressures, and multiple priorities, impede full dis- closure. Finally, it has been suggested that physicians withhold information from patients because they are concerned that such information will be distressing or will lead to refusals of medi- cally indicated treatment. Each of these reasons is open to empirical verification. Although there are no definitive data on any one of the explanations offered above, some information about each is discussed below. A number of different methods have been used, including survey techniques, interviews, and observations, to gather both quantitative and qualitative data about information disclosure. Numerous surveys have been conducted in which doctors have been asked what they usually do disclose or would disclose in various hypothetical circumstances. These studies have tended to look narrowly at specific pieces of information that may or may not be disclosed, such as the patient’s diagnosis or prog- nosis, mention of alternative treatments, or risks of treatment. Generally, surveys have found that physicians today are much more open with patients than they were as recently as 20 years ago.’ 3 For example, a survey conducted in 1961 found that 90% of physi- cians preferred not to inform patients of a diagnosis of cancer. See D. Oken, What to Tell Cancer Patients: A Study of Medical Attitudes, 175 4 Making Health Care Decisions: Appendix A In another kind of study, physicians and patients have been interviewed after their discussions in order to discover the content of disclosures. Typically these studies are done in out- patient settings and many have sought to learn what informa- tion was provided concerning drugs that were prescribed. Sub- stantial proportions of these studied claim not to have been informed at all about the nature and purpose of prescribed drugs, and even fewer patients report that they were informed about the side effects of their drugs. Few studies have tried to observe or record actual doctor- patient encounters and those that have may be suspect. The very act of recording can affect the interaction, and people who permit themselves to be observed may be atypical. Many of these studies have also been conducted in outpatient settings and have focused on disclosure of drug information. Their findings generally confirm self-reports by patients that fre- quently little or nothing is said about drugs, their purpose, how to use them, or what to watch for in terms of possible side effects.’ Yet another kind of observational study seeks to examine. communication between health care professionals and patients in a more qualitative way by focusing not on particular pieces of information or on individual encounters, but by examining the communication process in depth in certain special settings. One such recent study of communication with parents on a neonatal intensive care unit provides the kind of subtle data that is impossible to capture in surveys or through examination of any one encounter.’ Here it was found that although staff paid considerable attention to communication with parents—keep- ing them informed regularly and helping them to understand their babies’ conditions by using nontechnical language and J. AM.A. 1120 (1961). By 1977, 97% of physicians said they routinely disclosed cancer diagnoses. D.H. Novak et al., Changes in Physicians’ Attitudes Towards Telling the Cancer Patient, 241 ].A.M.A. 897 (1979). See also Arnold J. Rosoff, INFORMED CONSENT: A GUIDE FOR HEALTH CARE PROVIDERS, Aspen Publications, Rockville, Md. (1981) at 340-76. * See, e.g., H. Nickerson, Patient Education, 1 Hearth Epuc. Mono- GRAPHS 95 (1972); P. Boreham and D. Gibson, The Information Process in Private Medical Consultations: A Preliminary Investigation, 12 Soc. Sci. AND MED. 409 (1978). 5 See, e.g., B.M. Korsch and V.F. Negrette, Doctor-Patient Communica- tion, 227 SCIENTIFIC AMERICAN 66 (1972); Barbara L. Svarstad, Physi- cian-Patient Communication and Patient Conformity With Medical Advice, in David Mechanic, ed., THE GROWTH OF BUREAUCRATIC MEDI- CINE, John Wiley & Sons, New York (1976) at 221-238. ¢ Robert Bogdan, Mary Alice Brown, and Susan B. Foster, Be Honest but Not Cruel: Staff/Parent Communication on a Neonatal Unit, 41 HUMAN ORGANIZATION 6 (1982). A Review of Empirical Studies 5 repeating important information—nonetheless the commu- nication process often left much to be desired. With some exceptions. . .the staff's assessment of parents and the parents’ dispositions toward their babies and the degree to which parents are understanding what is being told to them is often inaccurate. Most assessments of par- ents are based on limited knowledge, derived mainly from short observations, limited conversations, or secondhand reporting of incidents and information. What is known is episodic, not informed by the context of the perinatal expe- rience in the lives of the parents. [In spite of the concern to communicate well and regularly with parents] staff, and physicians especially, spend relatively little time with par- ents. The priority is saving and mending the physical child. in general, staff know very little about what the parents are thinking and about their life outside the hospi- tal.” The authors make several observations about what parents are actually told about their babies. First, staff share the view that “you can never really tell” how a child they are treating will turn out. “In conversations with parents, unless there is very clear evidence of impending death or clear diagnosis of impair- ment, 'you can never really tell’ dominates the tone of commu- nication.” Thus, what is told may be less than candid. Stopping short of complete candor relates to another shared staff view that one should be honest, but not cruel: “While there is varia- tion in the degree to which staff are optimistic rather than pessimistic in what they tell parents about their child’s condi- tion, there is general agreement that you should not take away hope from the parent.” However, the staff share another per- spective, namely “that it is bad to surprise parents with bad news: that it is important to foreshadow possible revelations and if there is a chance of death, there should be some prior warning.” Two other practices relate to the “be honest but not cruel” perspective. One is not to burden parents with too many details about long-term prognosis or problems that will have to be faced later in the childs life. The second is the use of euphemisms to describe the baby’s condition. Although the reason for being euphemistic is to spare parents the anxiety of being direct, this approach can compromise understanding, which eventually may lead to more anxiety than initial straight- forward language would have. Studies such as this one highlight the complexities of com- munication between physicians and patients and serve as a reminder that what is important is not only what is said, but how it is said. The effects of how information is communicated on patients’ comprehension is discussed in a later section. Id. at 11. 6 Making Health Care Decisions: Appendix A How Much Information Do Patients Want? Adequate study of this question is fraught with potential methodological diffi- culties. Asking people hypothetical questions about what they would want to know under various conditions if they were sick is obviously an imperfect measure of what they actually want to know when faced with any particular illness. However, if pa- tients are questioned at the time of illness, the mere fact of asking whether they desire particular information is likely to bias responses. Finally, attempts to determine how much infor- mation patients want by examining their behavior with their physicians are likely to find that most patients ask few ques- tions, thus leading very quickly to the perhaps erroneous con- clusion that patients desire very little information. Due to the many “barriers” to communication it would be unwise to assume that silence on the part of patients indicates a desire not to have information. Patients may not ask questions for reasons that are unrelated to their desire to know about their treatment. For example, patients may be hesitant to question the physician's authority, disclose their ignorance, or be a nui- sance. They may be intimidated. They may not have sufficient background or feel confident in their ability to ask meaningful questions. Furthermore, society has ascribed certain norms and ex- pectations of how patients should act. Traditionally the patient has been seen as submissive to and reliant upon the physician. In addition, illness and hospitalization may make the patient withdrawn and uncommunicative. Physicians may interact with patients during the more acute episodes of illness and therefore not see the person when he or she would be more interested in receiving information. Finally, patients who ask questions or demand information may be perceived by health professionals as uncooperative, unpleasant, or belligerent. Some physicians may be unable to work with assertive patients or may prefer not to.® Despite the difficulties in determining exactly how much information patients want, there is considerable evidence to suggest that patients’ desire for information about their health and medical care has increased in recent years. This change relates to recent shifts in disease patterns and in social atti- tudes. The increase in chronic illness requiring long-term treat- ment and the self-care movement have prompted greater patient ® For an excellent study of the effects of patient questioning on the doctor-patient relationship, see Debra L. Roter, Patient Participation in the Patient Provider Interaction: The Effects of Patient Question Ask- ing on the Quality of Interaction, Satisfaction and Compliance, 5 HearrH Epuc. MONOGRAPHS 281 (1977). In recent years the popular press has also carried stories about this subject. See, e.g., E. Jackson, Coping: The Why Syndrome, WASHINGTON Post, C5 (Sept. 17, 1980). A Review of Empirical Studies 7 autonomy and self-responsibility. The consumer movement has prompted greater demand for information about the risks and benefits of treatments and procedures. These trends have re- sulted in a greater need and desire for patients to know about treatments, a questioning of physician authority, and an ac- ceptance of nontraditional sources of health information. Not- withstanding these trends, the physician is still perceived as the primary and desired source of information.’ Not only have patients’ desires for knowledge about their health care increased, but access to such knowledge has come to be viewed as a right and the physicians’ responsibility to provide it has come to be seen as a moral obligation. For exam- ple, the American Hospital Association's Patients’ Bill of Rights includes the assertion that “the patient has the right to obtain from his physician current information concerning his diag- nosis, treatment, and prognosis in terms the patient can be reasonably expected to understand.” Recent surveys of patients and the general population indi- cate that most people want to know more about their treatment than they are currently told." This desire appears to be spread- ing through all groups in the population. Whereas in the early 1970s it was most prominent among younger and better-edu- cated people, the desire for more information now appears equally across age and educational segments, with perhaps an even greater demand among the elderly and the less well edu- cated members of the population for some kinds of informa- tion." For example, compared to physicians’ judgments about what patients want to know about their drugs, patients them- selves report that they would like to receive detailed informa- * P. Joubert and L. Lasagna, Patient Package Inserts I: Nature, Notions and Needs, 18 CLINICAL PHARMACOLOGY AND THERAPEUTICS 663 (1975). 10 For the complete text of the AHA Patients’ Bill of Rights, see George J. Annas, THE RIGHTS OF HOSPITAL PATIENTS, Avon Books, New York (1975) at 25-27 and the discussion that follows. See also, Joint Com- mission on Accreditation of Hospitals, Accreditation Manual for Hos- pitals, Chicago, 1970. Following the lead of the AHA, the U.S. DHEW adopted a patient bill of rights for skilled nursing facilities participat- ing in Medicare and Medicaid, 39 Fed. Reg. 35774-5 (Oct. 3, 1974). 1 See, e.g., L.A. Morris and S.C. Groft, Patient Package Inserts: A Research Perspective, in K. Melmen, ed., DRUG THERAPEUTICS: CON- CEPTS FOR PHYSICIANS, Elsevier, New York (1981) at 191-217. 12 See, e.g., The Roper Organization, Roper Reports 78-3 and 80-5, New York (1978 and 1980), and David E. Kanouse et al., Informing Patients About Drugs: Summary Report on Alternative Designs for Prescription Drug Leaflets, Rand Corp., Santa Monica, Calif. (1981) R-2800-FDA. 8 Making Health Care Decisions: Appendix A tion about their therapy, with an emphasis on warning and risk information.” The desire for negative information may reflect both lower levels of existing knowledge about this aspect of treatment as well as a desire to know the personal risks being taken in therapy. Although a number of surveys have been conducted in an effort to discover how much information patients wish to have and how much information physicians disclose, the findings are inconclusive and often unreliable due to small samples or low response rates.” The Commission contracted with Louis Harris and Associates to conduct the first parallel national surveys of physicians and the public in order to gather reliable and comparable data regarding attitudes.” A number of studies have indirectly examined the question of how much information patients want by studying reasons for patient satisfaction (or dissatisfaction) with their doctors. Com- munication ranks high on the list of qualities patients view as important to their medical care. There is a growing body of evidence that attests to the value patients place on communica- tion. In a number of studies it has been shown that the quality of the interaction (as measured by such factors as the amount of information provided, time spent with the physician, and the perceived humaneness of the physician) is key to patient satis- faction with medical care.” Promoting patient satisfaction with treatment is often perceived as an important goal of medical treatment since a dissatisfied patient may not return for treat- ment. In addition to cognitive variables, such as providing clear communications, affective and emotional factors also are important determinants of patient satisfaction. The doctor's car- ing, being open and accepting, and taking time and listening are important correlates of the patient’s commitment to the rela- tionship.” Writing a prescription may be perceived as a negative 13 See, e.g., Ruth Faden, et al. Disclosure of Information to Patients in Medical Care, 19 MED. CARE 718 (1981). 14 Rosoff, supra note 3, for example, had only a 24% response rate to his survey of physicians. 15 See Appendix B in this Volume for the complete report of this study. © See, e.g., ].E. Ware and M.K. Snyder, Dimensions of Patient Attitudes Regarding Doctors and Medical Care Services, 13 Mep. CARE 669 (1975); D.J. Kallen and J.J. Stephenson, Perceived Physician Humane- ness, Patient Attitudes and Satisfaction with the Pill as a Contracep- tive, 22 J. HEALTH AND Soc. BEHAV. 256 (1981); and Marian Osterweis and Julianne R. Howell, Administering Patient Satisfaction Question- naires at Diverse Ambulatory Care Sites, 2 J. AMBULATORY CARE MAN- AGEMENT 67 (1979). 17 See, M.R. DiMatteo, L.M. Prince; and A. Tarnata, Patients’ Percep- tions of Physicians’ Behavior: Determinants of Patient Commitment to A Review of Empirical Studies 9 influence on the doctor-patient relationship if it is seen as cutting short or replacing more meaningful communication between the doctor and patient." Despite a general appeal for information, it would be un- wise to assume that all patients want to know everything about their conditions and treatment. There is some evidence to sug- gest that sicker patients place less importance on communica- tion in assessing their satisfaction with care.” Influences on Information Disclosed What Other Factors Influence the Nature and Quantity of the Information Disclosed to or Withheld From Patients? In addition to physicians’ perceptions of what patients want to know, several other important factors influence the amount and nature of the information disclosed to patients. These include uncertainties of several types, personality factors, time pres- sures, social distance between physicians and patients, and fear of negative consequences. There are many different kinds of uncertainty surrounding health care. Some derive from the state of the art of medicine or limitations in a particular physician's knowledge; others are based more on the relationships between providers and pa- tients. At times physicians may withhold information because they fear it may do harm. People turn to physicians for advice and confidence. Yet physicians are frequently faced with un- clear, conflicting, or indefinite information about a patient's disease. Under these circumstances, physicians may withhold information or provide only innocuous news rather than admit their uncertainty. They may fear that patients will misinterpret their lack of knowledge as an indication of a dismal prognosis or that patients will lose confidence because the physician appears to be out of control of the situation.” the Therapeutic Relationship, 4 J. COMMUNITY HEALTH 280 (1979). 8 See, S.A. Wartman et al., Do Prescriptions Adversely Affect Doctor- Patient Interactions?, 71 AMER. J. Pu. HEALTH 1358 (1981). DiMatteo, supra note 17. The significance of medical uncertainty for the doctor-patient rela- tionship was first discussed by Talcott Parsons, THE SOCIAL SYSTEM, Free Press, Glencoe, Ill. (1951) at Chapter Ten and has been elaborated on by Renee C. Fox in many of her writings; see, e.g., Training for Uncertainty, in Renee C. Fox, Essays IN MEDICAL SOCIOLOGY: JOURNEYS INTO THE FIELD, John Wiley and Sons, New York (1979) at 32-48, The Evolution of Medical Uncertainty, 58 MILBANK MEMORIAL QUARTERLY 1, 49 (1980), and, with Judith P. Swazey, THE COURAGE TO FAIL: A SOCIAL VIEW OF ORGAN TRANSPLANTS AND DiaLysis, Univ. of Chicago Press, Chicago (1974) at 40. See also Volume One of this Report at 85-89. 10 Making Health Care Decisions: Appendix A Not knowing how patients will face news about a serious disease or the risk of a certain therapy creates a second type of uncertainty. Physicians may believe that bad news will se- riously upset the patient, make long-term management diffi- cult, or lead to the refusal of treatment the physician believes is necessary. Thus “negative” information may be withheld be- cause of fear that a patients response will exacerbate existing problems or create new ones.” Physicians may underestimate patients’ knowledge and may communicate inadequately because of this mispercep- tion.” Similarly, physicians may anticipate difficulty explain- ing medical concepts to patients, especially those with little education or whose native language differs from the physi- cian’s.” Physicians need to distinguish relevant information from purely technical details, a distinction that may be difficult for some of them to make. Although it is well known that patients often do not understand medical terminology, it is not known how well specific efforts to explain such terms in lay language work. Considerable effort is being expended to de- velop written materials for patients, especially drug informa- tion, in language that can be readily understood, as noted in a later section. Many physicians claim that time pressures are too great to allow for detailed communication.”® Several factors increase these pressures. Other patients waiting to be seen, unexpected emergencies, an inability to find a patient's chart, the absence of a coworker, the ever-present telephone, and the inability to bill for time spent in discussion can all put pressure on the physi- cian to shorten interactions with individual patients. Observa- tional studies seem to indicate, however, that physicians reduce the amount of information given to patients in pressured situa- 21 See Volume One of this Report at 99-100. 22 Robert L. Kane and Kurt Deuschle, Problems in Doctor-Patient Com- munication, 5 MED. CARE 260 (1967). 3 See, e.g., D.A. Pendleton and S. Bochner, The Communication of Medical Information in General Practice Consultations as a Function of Patients’ Social Class, 14A SociaL Sci. AND MED. 669 (1980). See, e.g., C.M. Boyle, Differences Between Patients’ and Doctors’ Interpretation of Some Common Medical Terms, 2 BRITISH MED. J. 286 (1970); J. Samora, L. Saunders, and R. Larson, Medical Vocabulary Knowledge Among Hospital Patients, 2 J. HEALTH AND HUMAN BEHAV. 83 (1961); F.C. Tring and M.C. Hayes-Allen, Understanding and Mis- understanding of Some Medical Terms, 7 Brit. J. Mep. Epuc. 53 (1973); Korsch and Negrette, supra note 5. Although face-to-face time between physician and patient varies depending on the nature of the illness and treatment, national data indicate that the average encounter in ambulatory settings is about 15 minutes. U.S.D.H.E.W., National Center for Health Statistics, The Na- tional Ambulatory Medical Care Survey, 1977 Summary, Series 13, No. 44 (1980), Table D at 6. A Review of Empirical Studies 11 tions only if the patient has had previous experience with the prescribed treatment. For patients with no previous experience, the amount of physician instruction is equal in high- and low- pressure situations.” These few studies indicate an important feature of physi- cian-patient relationships that tends to be overlooked in most studies of informed consent, namely that the relationship often extends beyond a single visit and that physicians are aware of this when they provide information.” For example, it is well known that patients with chronic diseases are often extremely well informed. Their knowledge is unlikely to stem from a single encounter but rather be the result of months or years of communication. Finally, personality factors and stylistic differences on the part of both health care providers and patients should not be overlooked as factors that have profound effects on the commu- nication process. Some people are by nature more talkative or inquisitive than others. Particular combinations of providers and patients may lead to better or worse communication. Styles of interaction may relate to personal characteristics, prior expe- rience with health care, or views about what behavior is appro- priate in a given setting. What is to be hoped for is not unifor- mity in the amount of information that is disclosed, but rather a process of communication in which health care professionals encourage patients to join with them in dialogue so that pa- tients are not inhibited in their questions and providers are not reluctant to answer. How Do Illness and Other Situational Factors Affect Dis- closure, Comprehension, and Decisionmaking? The doctrine of informed consent seems to be based implicitly on a prototypi- cal situation in which a single physician is administering a discrete hospital-based procedure, such as surgery, despite the fact that informed consent is now required in a wide variety of situations that do not conform to this model. Although it seems obvious that the context in which care is given, the nature of the illness and treatment, and the particular people involved will all affect the nature of the disclosure, comprehension, and decisionmaking, there is little empirical evidence of differences in the informed consent process or general nature of the com- munication due to these various factors. It is for this reason the Commission supported an empirical study of how the broad context in which medical care is given affects doctor-patient communications and the decisionmaking process; the results are reported in Appendix C. * Svarstad, supra note 5. 7 Id.; Boreham and Gibson, supra note 4. 12 Making Health Care Decisions: Appendix A Patients’ Comprehension of Information To What Extent Do Patients Comprehend Information About Their Condition and Treatment and How Do the Method of Disclosure and the Actual Presentation of Information Af- fect Comprehension? These issues are discussed at length in Volume One of this Report. Patients’ understanding of health- related information has been shown in several studies to be extremely limited.” But what has generally been studied is either understanding of medical jargon or recall of information given sometime in the past, neither of which is a fair test of patients’ capacity for comprehension. All professions have their own specialized vocabularies that are understood only by the initiated. It is not surprising, therefore, to find that laypeople do poorly on medical vocabulary tests. Ability to remember infor- mation for some period of time may also not be the appropriate thing to examine. The important question is whether patients have the ability to understand important aspects of their condition and treat- ment such that they can use the information to make decisions consistent with their values. Although intelligence and general level of education may be somewhat relevant, the particular circumstances of the illness and treatment are likely to be more important to understanding. Thus, there are really two ques- tions to be answered. First, to what extent can the layperson understand medical information? And second, how and to what extent do the conditions of “patienthood” have an impact on that understanding? Numerous recent studies suggest that through the use of simplified language and written and audiovisual aids as ad- juncts to discussion, laypeople can understand health-related information.” However, no comprehensive analyses have been performed to determine how being ill, frightened, in unfamiliar surroundings, and perhaps in pain—in other words, how being a patient—affect understanding. Although it seems reasonable to suppose that the quality and quantity of understanding will 2 Note 24 supra; and Samuel Gorovitz, DocTORs’ DILEMMAS: MORAL CONFLICT AND MEDICAL CARE, Macmillan Pub. Co., New York (1982) at 38-42. » See, e.g., H.W. Griffith, D. Hayes, and L.P. Hyde, Let Printed Aids Work For You and Your Patients, 21 PATIENT CARE 119 (1973); G.L. Barbour and M.]. Blumenkrantz, Videotape Aids Informed Consent Decision, 240 J.A.M.A. 2741 (1978); E.E. Bartlett, Selection of Educa- tion Strategies, in Lawrence W. Green et al., eds., HEALTH EDUCATION PLANNING: A DIAGNOSTIC APPROACH, Mayfield Pub. Co., Palo Alto, Calif. (1980) at 86-115; Ruth R. Faden, Disclosure and Informed Consent: Does It Matter How We Tell It?, 5 HeaLtH Epuc. MONOGRAPHS 198 (1977). A Review of Empirical Studies 13 diminish as anxiety, symptomatology, and seriousness of ill- ness increase, this has not be adequately studied.” As already mentioned, the use of euphemisms may com- promise patient understanding. Telling a parent, “your baby is very sick” may be the professional's way of notifying parents that their baby is going to die, “but in the context of parents’ lives, it may mean their baby is going to be in the hospital for a few extra days.”* Although health professionals have a wealth of experience to draw on, patients typically have only very limited experience with illness and its treatment. When every- thing is not only new, but also threatening, communications need to be explicit if misunderstanding is to be avoided, and messages may need to be repeated several times before they are really understood. “Parents process information they receive from the staff through the world as they understand it. What staff think they are communicating is often not what parents hear.” Professionals other than physicians may be helpful in the communication process, especially in hospitals where nurses are more likely to have both more time and more contact with patients. On the neonatal unit described above, “good” doctors were those who, among other things, involved nurses in the information process. Increased communication between physi- cians and nurses ensures that the information given to patients is consistent and avoids having patients play one provider off against another to get additional information. Involvement of other staff, if done well, can also facilitate communication. Decisionmaking by Patients Does Information (Especially About Uncertainties and Risks) Have Negative Psychological Effects on Patients and How Does Such Information Affect Treatment Decisions? The answer to this question is discussed at length in Volume One of this Report and will therefore only be summarized here. Al- though physicians have long claimed that some information is so threatening to patients that it will have negative psychologi- cal effects and dissuade people from needed treatment, there are no systematic data to support this claim. Empirical studies seem to show that disclosure of risks and other preparatory information is not only not harmful, but is usually beneficial. “In order to begin to address this very complex question, which relates to the heart of informed consent, the Commission contracted for a study to examine variations in the communication and decision- making process in different medical settings for a variety of illnesses and treatment. The results of that study are reported in Appendix C of this Volume. *! Bogdan, Brown, and Foster, supra note 6 at 13. 2 Id. at 12. 14 Making Health Care Decisions: Appendix A Such information permits patients to know beforehand what to expect, thereby alleviating some of the distress following treat- ment such as surgery.” Several recent studies of the effects of giving patients infor- maion about the risks and side effects of drugs have found that such information does not reduce compliance with drug ther- apy. Furthermore, informed patients report no more side effects than the uninformed but they are more likely to attribute those effects (correctly) to the medicines they are taking.* Because there were really no studies of the effects that information has on treatment refusals, other than for drugs, the Commission contracted for an empirical study to determine the causes and effects of such refusals, with particular attention focused on the role of information. Results of that study are reported in Appendix D. How Do Patients Make Treatment Decisions? How and To What Extent Is Information Used to Make Such Decisions? Most discussions of informed consent assume that information provided by physicians is used to make treatment decisions. However, very few studies have been done to determine just how patients make decisions and what the role of information is in that process. From anecdotal reports and case studies it is clear that many factors come into play in making decisions. These include previous experiences with particular treatments, personal beliefs, and, most notably, the patient’s own assess- ment of what is best in terms of his or her own values and life plans. Such judgments may or may not coincide with physi- cians’ judgments concerning the best medical decision and may therefore be seen as “irrational” by physicians.* Faden and Beauchamps have conducted virtually the only study of the relative influence of a number of factors on treat- ment decisions.” Study subjects who were given information about nonsurgical contraception were questioned afterward about their decisions. Although the information given by the doctor was “useful,” only 12% reported that the information " See Volume One of this Report at 99-102. See also Irving Janis, PSYCHOLOGICAL STRESS: PSYCHOANALYTICAL AND BEHAVIORAL STUDIES OF SURGICAL PATIENTS, John Wiley and Sons, New York (1958); ].M. An- drew, Recovery from Surgery: With or Without Preparatory Instruction for Three Coping Styles, 15 J. PERSONALITY AND SOC. PSYCHOLOGY 223 (1970); LF. Wilson, Behavicral Preparation for Surgery: Benefit or Harm, 4 J. BEHAV. MED. 79 (1981). “ Louis A. Morris and David E. Kanouse, Informing Patients about Side Effects, 5 J. BEHAv. MED. 363 (1982). » C.H. Fellner and J.R. Marshall, Kidney Donors—The Myth of In- formed Consent, 126 AMER. J. PSYCHIATRY 1245 (1970). Ruth R. Faden and Tom L. Beauchamps, Decision-Making and In- formed Consent: A Study of the Impact of Disclosed Information, 7 Soc. INpicATORS RES. 313 (1980). A Review of Empirical Studies 15 was “the most important factor” in determining the decision. Personal feelings and past experience were more often the de- termining factors. Of all the issues in informed consent, decisionmaking has received the least attention. The Commission's study of treat- ment refusals (Appendix D) provides some important informa- tion about how patients actually make decisions. Conclusion Empirical studies provide some insights into the nature of the communication and decisionmaking process in health care, but leave many questions unanswered. It appears that many (if not most) patients do want information, even if they do not ask for it directly, and that the overall quality of communication and time spent with the physician are key factors in patient satisfaction. However, surprisingly little is known about varia- tions in the process that derive from differences in the nature of the illness, proposed treatment, and health care setting, or from differences in the personal and social characteristics of patients and health care professionals. It is not known what level of information patients find most satisfying, nor is much known about variations in the desire for information in the same pa- tient under different circumstances. If the goals articulated in Volume One of this Report— namely, to foster ongoing communication and shared decision- making in the context of a relationship based on mutual respect between patient and provider—are to be realized it is important to be cognizant of the complexities of medical care and to understand the constraints that operate in the real world. The three studies reported on in the remainder of this volume ad- dress some of these complexities and begin to answer some of the difficult questions that have not been dealt with empirically thus far. i E E - Views of Informed Consent and Decisionmaking: Parallel Surveys of Physicians and the Public Louis Harris and Associates* 1 i i L J - EXECUTIVE SUMMARY The President's Commission for the Study of the Ethical Problems in Medicine and Biomedical and Be- havioral Research (PCEMR) was established by an act of Congress to investigate the ethical and legal implications of the requirements of informed consent to undergo medical procedures. After exhaustive re- view of existing research, the Commission determined that the available data on physician and public at- titudes and behaviors regarding informed consent and medical decisionmaking provided an inadequate basis for developing concrete recommendations. Con- sequently, Louis Harris and Associates, Inc., was commissioned by the PCEMR to conduct parallel na- tional surveys of the adult population and the physicians who provide direct patient care to adults, concerning their attitudes about and ex- periences related to disclosure of information and decisionmaking in therapeutic settings. Interviews were completed with a national sample of 805 physi- cians and a national cross section of 1,251 adults. Since the primary goal of this project was to investigate the issue of informed consent, consider- able attention was paid to questions of how much physicians tell their patients about their condition and treatment, as well as to the conditions under which such information is not disclosed. Both phy- sicians and the public agree that full disclosure of case and treatment is desirable. Almost all Amer- icans (94%) say that they want to be told everything * John M. Boyle, Ph.D., Project Director, and Paul J. Brounstein, Ph.D., Senior Analyst. September 1982 18 Making Health Care Decisions: Appendix B about their condition and treatment, even if the news is unfavorable. Physicians appear to recognize this preference, with 86% saying that all or most patients who come to them for treatment want a can- did assessment of their diagnosis and prognosis, even if it is unfavorable. The vast majority of physicians report that they "always" or "usually" discuss most aspects of condi- tion and treatment with their patients. As might be expected, the proportion of doctors who report dis- closing case and treatment information to their pa- tients is greater than the proportion of the public who report that their doctors discuss these matters with them. In most instances, there is a large and reliable difference of approximately 15 to 25 per- centage points between the proportion of doctors who report usually discussing the issue with their pa- tients and the proportion of the public who report that their doctors usually discuss these matters with them. The survey finds that while 98% of doctors say they usually discuss the diagnosis and prognosis with their patients, only 78% of the pub- lic report that their doctor always or usually ex- plains their diagnosis and prognosis to them. Similarly, although 96% of doctors report that they usually discuss with their patients the nature and purpose of the treatment they recommend, only 78% Of the public report that their doctor usually explains this to them. While 84% of physicians report that they usually discuss the pros and cons of the treat- ment they recommend with their patients, only 68% of the public report that their doctor usually explains this to them. The greatest difference between doctor and pa- tient perceptions occurs in the area of side ef- fects. The survey finds that 93% of doctors report that they usually discuss side effects, including inconvenience and pain, that are certain or fairly likely to attend the treatment. However, only 68% of the public report that their doctor usually ex-— plains to them the side effects likely to attend their treatment. This represents a quite substan- tial (25 percentage point) difference in perception. By contrast, doctors and the public are in agree- ment on the frequency with which costs and insurance coverage are disclosed. The survey finds that 47% of doctors say that they usually discuss with their patients the costs associated with the recommended Surveys of Physicians and the Public 19 procedure and whether they would be covered by the patient's health insurance. A virtually identical 45% of the public agree that their doctor usually explains the cost of treatment and their insurance coverage. Although interesting and provocative, these dif- ferences between doctor and patient reports on the frequency of medical information disclosure should not obscure a more important finding. The vast ma- jority of physicians report that they discuss most aspects of condition and treatment with their pa- tients as a matter of course. This is substantiated by the large majority of the public who report that their physician usually discusses these matters with them. Although most physicians usually discuss these matters with their patients, they do not do so with- out reflection. Most doctors (52%) report that at least once a day they find themselves in a situation where they must make a conscious and deliberate evaluation of how much to tell a patient about his or her condition and treatment. This is not to say that doctors routinely withhold information from their patients. Indeed, 46% of physicians say they almost never withhold information about diagnosis or prognosis from a patient, while 57% say that they almost never withhold information about treatment risks and alternatives from a patient. Thus, the conscious and deliberate evaluation appears to focus upon the amount and type of information to be vol- unteered by the physician to the patient, rather than upon whether the patient should be denied in- formation he or she wants to know. Although most of the public wants to be told everything about their condition and treatment, a majority (56%) agree that some patients should be told less about their condition and treatment than other patients. Indeed, fairly substantial majori- ties of the public feel a doctor would be justified in withholding information about condition and treatment from patients if they tell the doctor they don't want to hear bad news (68%) or if the informa- tion might significantly harm the patient's health and well-being (68%). Somewhat smaller majorities or pluralities of the public feel that the doctor is justified in withholding information if it might make the patient anxious or upset (52%) or if the patient's family asks the doctor not to tell the 20 Making Health Care Decisions: Appendix B patient (48%). However, only a minority (38%) feel the doctor is justified in withholding information because it might make the patient unwilling to under- 80 treatment that the doctor thinks is necessary. Doctors' explanations of their most common rea- sons for withholding information are generally con- sistent with public attitudes toward justification of physicians' withholding of information. The pa- tient's ability to cope with the news is the reason for withholding information most commonly given by physicians (34%). The wishes of the patient's fa- mily (21%) and the patient's desire to know (162) are also cited as common reasons. Only 6% of doctors say that the effect of the information on patient compliance is a common reason for withholding information. However, doctors' perceptions of their patients’ ability to understand their condition and treatment is also cited as a primary factor in how much they tell a patient (56%) and as a common reason for with- holding information (28%). This is particularly im- portant since doctors report, on average, that only 79% of their patients are able to understand most as- pects of their treatment and condition, even if rea- sonable time and effort are devoted to explanation. In analyses of physicians’ propensity to disclose information, the proportion of their patients whom they perceive as capable of understanding their con- dition and treatment emerges as the primary determi- nant of actual disclosure. It is noteworthy that the socioeconomic characteristics of the doctors’ patients, particularly education and income, are re- lated to the doctors' perceptions of their patients’ ability to understand their condition and treatment. The public appears generally satisfied with their doctor's disclosure and decisionmaking practice. A majority of the public report that they are very sat- isfied with their doctor's honesty in discussions with them (66%), their doctor's willingness to an- swer questions about condition and illness (61%), their doctor's respect for their treatment prefer- ences (59%), their doctor's concern for their well- being (56%), and the amount of time their doctor spends with them (50%). By contrast, less than 10% of the public report being either very or somewhat dissatisfied with their doctor's honesty or respect for their treatment preferences, and less than 20% are dissatisfied with their doctor's concern for Surveys of Physicians and the Public 21 their well-being, willingness to answer questions, and time spent with them. The survey data, however, suggest two important qualifications of the apparent widespread public sat- isfaction with physician behavior in information dis- closure and medical decisionmaking. First, although 49% of the public are very satisfied with doctors they have used in the past few years, only 35% are very satisfied with doctors in general. This dis- tinction may be explained by the survey finding that 36% of the public have changed doctors because they disagreed with them. Thus, satisfaction with a pa- tient's present physician cannot be equated with gen- eral satisfaction with medical treatment or physician performance. Second, there are some important differences in what the public thinks that patients should be told as a matter of course and what doctors report tell- ing their patients. Generally, the proportion of physicians reporting disclosure as a matter of course equals or exceeds the proportion of the public who expect such disclosure. Substantial differences, however, are found between public expectations and physician practice in the disclosure of treatment costs and treatment risks. While 69% of the public think that physicians should discuss the cost of proposed treatment with the patient as a matter of course, only 38% of physicians report doing so. In the matter of treatment risks, doctors' re- ported disclosure practice (81%) exceeds public expectations (75%) when procedures involve risks of death or serious disability of 1 in 100 cases. How- ever, public expectations exceed physician practice when the procedure involves a risk of death or seri- ous disability of 1 in 1,000 cases (64%; 52%) and when the procedure involves a 1 in 1,000 risk of temporary disability (65%; 43%). In these, and per- haps other areas, public expectations of disclosure of treatment information are not being fully met by current medical practice. The vast majority of the public (94%) want to be told everything about their condition and treatment, even if it is unfavorable. The survey finds that 89% of the public feel that patient rights to infor- mation should be protected by law. More specifi- cally, 72% of the public feel that a doctor should be held legally liable for failing to fully and com~ pletely inform his patient of the risks associated 22 Making Health Care Decisions: Appendix B with a medical procedure. These attitudes may be shaped by the perception of 44% of the public that when a patient doesn't understand his medical treat- ment, it is always or often because the doctor did not explain things well. Physicians report that they usually obtain con- sent from patients for most types of medical proce- dures. Prescriptions and blood tests are the excep- tions, with 53% of physicians saying that they usu- ally do not get consent for prescriptions and 43% for blood tests. The most common form of patient consent obtained for prescriptions, blood tests, mi- nor office surgery, and local anesthesia is oral consent. However, for inpatient surgery, general anesthesia, and radiation therapy, most physicians require written consent from the patients before proceeding. Nearly one-quarter (24%) of the public report that they have been asked to sign a written consent for treatment within the last year. Most of those asked to sign the consent form (71%) say that the doctor explained the consent form to their satisfac- tion. Similarly, 72% report that after reading the consent form, they felt they fully understood the risks of the treatment they were going to undergo. Only 5% report that they refused treatment because of what they learned about the treatment from the consent form. Interestingly, those who feel they did not understand the risks after reading the con- sent forms are more likely to refuse treatment than those who feel they did understand (8% to 47). The parallel surveys of patients and physicians reveal a strong preference for shared decisionmaking in medical treatment. Given alternative ways of mak- ing treatment decisions, most of the public (72%) prefer that the doctor discuss alternatives with the patient and then the two of them decide together how to proceed, compared with approaches that emphasize a unilateral decision by either the patient or the physician. In terms of how patients actually make tredtment decisions, most patients appear to play an active role in these decisions. Among those who have undergone an inpatient surgical operation, 21% say the decision was based upon their own judgment, and another 31% say it was based equally on both their judgment and their doctor's judgment, compared with 40% who say the decision was based on their trust in their doctor's judgment. Similarly, on a Surveys of Physicians and the Public 23 whole range of specific medical decision situations, doctors tend to place the locus of control midway between patient and physician dominance. Regarding shared responsibility for decisions on medical treatment, the survey finds conflict as well as consensus between doctors and patients. Half of the public (51%) report that they have sought a sec- ond opinion from another doctor. One out of five (21%) have refused to undergo treatment recommended by a doctor. one-fifth (20%) report that a doctor has told them to find another doctor if they didn't agree with the doctor's diagnosis and treatment rec- ommendations. And as noted earlier, 36% of the pub- lic have actually changed doctors because they dis- agreed with them. Although doctors are generally supportive of a shared decisionmaking process, most see a profes— sional responsibility to help the patient choose the best available alternative. Three-quarters of phy- sicians (75%) say that it is their responsibility to persuade a patient to accept a medically indicated course of action when a patient disagrees with their recommendation. At the same time, the survey con- firms public perceptions that their doctors are re-~ sponsive to their treatment preferences, particular- ly in extreme circumstances. Using the case of a patient in the end stage of a painful degenerative disease, doctors were asked whether they would re- suscitate her if she suffered a cardiac arrest. Only 5% of doctors who were told she had left instruc- tions that she wished to die naturally say that they would resuscitate her. By contrast, among those doctors who were told that the patient had left in- structions that she wanted everything possible done to prolong her life, 52% say they would resuscitate. Similarly, in cases of terminal diseases that are unresponsive to treatment, most doctors (66%) see the decision between aggressive and palliative ther- apy as a personal rather than a medical decision. The attitudes of physicians concerning the treat- ment of terminally ill patients are complex. On the one hand, a majority of physicians (57%) say that in the absence of guidance from the patient, they would be very likely not to resuscitate a patient dying of a painful, degenerative disease if cardiac arrest should occur. Similarly, 82% of physicians say that it is ethically permissible to (and 847% say they 24 Making Health Care Decisions: Appendix B would be likely to) administer pain relieving drugs to a dying patient in severe distress, even if the size of the required dose might shorten the pa- tient's life. However, only 4% of physicians say that it is ethically permissible to (and only 2% say that they would be likely to) comply with a dying patient's request to have his or her life ended. By contrast, fully 46% of the public say that it would be right for a doctor to comply with a patient's re- quest under these conditions. Within the general pattern of public attitudes described above, some regular and significant dif- ferences are found among certain subgroups in their attitudes toward information disclosure and medical decisionmaking. The older, poorest/low income, and least educated segments of the public are generally the most satisfied with present disclosure practices and least interested in participating in medical de- cisionmaking. These population segments are also the most likely to be ill. Further analysis of the data reveals that the greater satisfaction with cur- rent medical practice among this group is related to their socially dependent status, not their greater experience with medical situations. Better educated people are consistently less satisfied with the present performance of their doctors and the most demanding of greater information disclosure and a broader patient role in decisionmaking. Physicians are far more homogeneous than the public in both their reported attitudes and prac- tices regarding information disclosure and medical decisionmaking. However, even among physicians, some differences emerge. The nature of the physi- cians' patient characteristics appears to affect at- titudes toward disclosure and decisionmaking. Spe- cifically, the proportion of patients whom the doc- tors think are able to understand treatment issues is directly related to the physicians' disclosure Practices and attitudes toward informed consent. Medical education -- types of school and types of courses -- seems to play only a minor role in Physicians' disclosure or treatment behavior. This is consistent with physicians’ ranking of medical training as least influential on the way they practice medicine, compared with their personal values and their experiences in practice. In terms of the present legal requirements for informed consent, physicians seem relatively well Surveys of Physicians and the Public 25 aware of the meaning and intent of informed con- gent. However, three-quarters of physicians (76%) frankly admit that they do not know what legal stan- dard for informed consent is applicable in the state in which they practice. Indeed, many of those who think they know the applicable legal standard cite the wrong standard for their state (32%). The public's understanding of the term "informed consent" is far more rudimentary than physicians’ understanding. Indeed, 21% of the public admit they don't know what the term "informed consent" means. Among those who do recognize the concept, few iden- tify the term with patients’ understanding of treat- ment risks and alternatives as compared with pa- tients' being informed of the nature of condition and recommended treatment. Perhaps as a result of this perception, 86% of the public agree that a pa- tient's signature on a written consent form estab- lishes that the patient has given consent. Both the public and physicians are generally supportive of formal written consent for medical treatment. Majorities of both the public (65%) and doctors (64%) agree that written consent forms are helpful to doctor-patient communications. And when asked about the effects of written consent forms, 63% of doctors cite positive effects of the forms, compared with 29% who cite negative effects. However, despite the fact that majorities of both physicians (76%) and the public (89%) agree that patients' rights to information about treatment risks and alternatives should be protected by law, the survey findings cast some doubt about whether the present legal requirements are successful. Only a minority of physicians (32%) agree that the legal requirements for obtaining informed consent are clear and explicit, while a majority (58%) dis- agree. The inability of all but a tiny fraction of doctors to correctly identify the present legal standard in their state for informed consent tends to support this view. Further, the survey findings raise questions about whether present legal standards are appropri- ate. Today, most states recognize either an average, reasonable practitioner standard or an average, reasonable patient standard for disclosure. A plurality of both the public (46%) and physicians (42%) choose a third standard for disclosure over 26 Making Health Care Decisions: Appendix B these other two. Both physicians and the public think the best definition of a physician's legal obligation to disclose treatment information should be: "the physician should disclose the information that the articular patient being treated would consider relevant to his decision whether or not to accept proposed treatment." Although this might be a more stringent standard than either of the present legal standards, it conforms to both public and Physician perceptions that the needs, desires, and abilities of patients to deal with treatment issues and information are highly individual. Surveys of Physicians and the Public 27 Index of Tables Chapter 1: Introduction 1-1 Specialty Group Participation by Experimental Group, October 12, 1979 «.qccinaciminmirms som sm 45 1-2 Sample Disposition: Physicians.................... 47 1-3 Final Sample Disposition: General Public........... 48 1-4 Interview Completion by Calls and Incentive: Physicians ..........ouiuiiiiiiiiiiii iii 49 1-5 Expected Sampling Error (Plus or Minus) at 95% Confidence Level (Simple Random Sample). ........ 51 1-6 Pooled Sampling Error Expressed as Percentages for Given Sample Sizes (Assuming p=q) .............. 52 Chapter 2: Health Providers and Health Consumers: The Context of Informed Consent 2-1 Current Health Status: General Public .............. 67 2-2 Employment: General Public ...................... 68 2-3 Disability and Hospitalization: General Public ...... 69 2-4 Chronic and Life-Threatening Illness: General Public 69 2-5 Type of Life-Threatening Illness: General Public... .. 70 2-6 Usual Source of Care: General Public............... 71 2-7 Usual Source of Care: General Public............... 72 2-8 Number of Times Doctor Seen in Last Year: General PUDME .. 0 cv ses is 3 os 5 ibn 008 30000 F900 8 30.008 0 4 8 0608 B04 73 2-9 Hospitalization, Visits to Doctor, and Written Consent: General Public. .......................... 74 2-10 Surgical Operation Requiring Overnight Hospitalization: General Public .................... 75 2-11 Physician Characteristics. ......................... 76 2-12 Practice Location: Physicians ...................... 77 2-13 Sox: PhySICIans. . oss sus sna wes sss mm ames omen mm wm 78 2-14 Recency of Training: Physicians. ................... 79 2-15 Percentage of Patients 65 and Over: Physicians... ... 79 2-16 Percentage of Adult Patients Who Are College Educated: Physicians uses asmsrmmssnnivmormss spe 80 2-17 Percentage of Adult Patients Who Are Poor: PHYSICIANS «vv vvvvvmn smn nals ain ss mds 508 85 Wik 830 8 900 5 80 2-18 Mean Percentage of Patients With Serious Illness and Ability To Understand Treatment: Physicians ....... 81 2-19 Mean Number of Hours Spent on Direct Patient Care: Physicians ...........ooiiiiiii 82 2-20 Mean Number of Hospital Inpatients: Physicians. . .. 83 2-21 Mean Number of Hospital Outpatients: Physicians .. 84 28 Making Health Care Decisions: Appendix B 2-22 Mean Number of Patients Seen Each Hour of Direct Patient Care: Physicians : «sss s wie s soi saws srs asi ve 85 2-23 Satisfaction With Physician: General Public. ........ 86 Chapter 3: What Physicians Tell Patients: Patterns of Disclosure 3-1 3-2 3-3 3-4 3-5 3-6 3-7 3-8 3-9 How Often Physicians Discuss Treatment Issues With Patients: Physicians. .............................. 105 Whether Physicians Initiate Discussion of Treatment Issues With Their Patients: Physicians.............. 106 Disclosure of Various Information Items: Physicians (Rosoff, 1981) .........oiiii i 107 Frequency of Discussions With Patients: General Public, PHYSICIANS « urs ums 0500 2 500 4 555 5.518 45 9808 28 8 0.500 4 108 Public Expectations vs. Physician Practice: General Public, PhysiCIans . «iu. vu. coms on es ems sies sss ams smo s 109 Predictors of Physician Disclosure ................. 110 Predictors of Patient Expectations About Disclosure . 111 How Much Physicians Disclose to Patients About Routine Care: General Public ...................... 112 How Much Physicians Disclose to Patients About Serious Illness: General Public. .................... fe Extent to Which Patients Are Kept Fully Informed: General Public ................................... 114 Whether Patients Would be Willing to Pay More for More Explanation of Routine Care Issues: General PUDNG overs ony wins wi 0 0 5 i 90000 0 3 100 2 3000 0 50 6 115 Who is Responsible for Keeping Patients Fully Informed: General Public. ......................... 116 Who is Responsible for Keeping Patients Fully Informed: Physicians ............................. 117 Chapter 4: When The Patient is Not Told: Patterns of Withholding 4-1 4-2 4-3 4-4 4-5 4-6 4-7 4-8 How Often Physicians Tell Patients of Condition: Physicians ............ cco, 134 Factors Influencing How Much to Tell: Physicians... 135 Desire for Condition Information: General Public.... 136 Patient Desire for Candid Diagnosis and Prognosis: PIYSICIANS «viv 0 500 5 wi 5 505 4 9.00008 000 £3 000 1 um 4 0m # sm mi 137 Requested Withholding of Bad News: General Public 138 Appropriateness of Differential Disclosures: General PUBIC: & 55:5 5.516 8 i 1 0 mw mm om ems nn 00 9 139 Justification for Physician Withholding Information: General Public ................................... 140 Best Judge of Information to be Disclosed: General PUBME . vou vies aos wis wn 3 500 5 wi 4 8 1s 5 5005 5 5s 50805 5 806 50 141 Surveys of Physicians and the Public 29 4-9 Justification for Physician Withholding Information: General Publi . cuss: mms sho in sums suis sms s mas ven 142 4-10 How Often Information About Treatment Risks is Withheld: Physicians uz: sw remiss bm svmes sm sons wns 143 4-11 Information About Diagnosis or Prognosis is Withheld: Physicians ................... 0000 144 4-12 Most Common Reason for Withholding Information: Physicians ........... cc. i iii 145 4-13 Patients Requesting Nondisclosure of Bad News: Physicians ...........coiiiiiiiiiiiiiiiiiiiiin 146 4-14 Information Withheld From Patient Discussed With Family: Physicians. «oc «sacs wm smn suns piss mons smvs wor» 147 Chapter 5: Informed Consent 5-1 Physician Liability for Failure to Disclose Risks: General PUbliC .civvisvssvmismisnmsamer impose rows 164 5-2 Physician Responsibility for Patients’ Understanding of Treatment: General Public. ...................... 165 5-3 Informed Consent: General Public.................. 166 5-4 Informed Consent: Physicians ..................... 167 5-5 Types of Consent: Physicians ...................... 168 5-6 Written Consent: General Public ................... 169 5-7 Physician Explanation of Consent Form: General PUIBTAT cr wo sos cn es i 5 sis he hs # hse dp J Joh $0 170 5-8 Understanding of Risks Described in Consent Form: General. PUDME « , om voor sao eo mns sum mom nmin «mime vias v0 171 5-9 Refusal of Treatment Because of Consent Form: General PUblE ...ovivvssnvonmsrnmsmrsisms ims emmenn 172 5-10 Effect of Written Consent Forms: Physicians ........ 173 5-11 Best Disclosure Standard: General Public........... 174 5-12 Best Disclosure Standard: Physicians. .............. 175 5-13 Applicable Disclosure Standard: Physicians. ........ 176 5-14 Knowledge of State Law: Physicians. ............... 177 5-15 Attitudes About Informed Consent: General Public, PRYSICIANG «cov vo mis wis 4.8 5.50008 R30 6 50000 5 200% 5 000» 0005 3 010 178 5-16 Predictors of Public Attitudes About Whether Patient Rights to Information Should be Protected by Law... 179 5-17 Predictors of Physicians’ Preferences Between Time Spent in Patient Consultation and Patient Care. ..... 180 5-18 Predictors of Public's Preferences Between Time Spent in Patient Consultation and Patient Care. ..... 181 5-19 Predictors of Public’s Attitudes Toward Consent Forms: Doctor-Patient Communications. ............ 182 30 Making Health Care Decisions: Appendix B Chapter 6: Description of Medical Decisionmaking 6-1 6-2 6-3 6-4 6-5 6-6 6-7 Basis for Patient's Assent to a Surgical Operation: General Public i. onsssssossnmssmmsomss ems pmes ness Refused Treatment: General Public ................. Changed Physician Because of Disagreement: General Public ....ou:iaiinssmnismesnmss ais bmpsmuss Physician Suggested Change of Doctor: General PUDBHIG . vo «iv 3 vi ii 55 65 9560 7 8 0 3 90 5 3 GRATE R95 58 8 Changed Doctors Because of Disagreement: General PUDNC .... vcs vin vnvnnmssnmss nis sine soins nid 5s s Sought Second Opinion: General Public............ Locus of Decisionmaking Mean Scores: Physicians . . Physician's Responsibility to Persuade Patient: Physicians .............ooiiiiiiiiiiiiii Chapter 7: Evaluation of the Medical Decisionmaking Process 7-1 7-2 7-3 7-4 7-5 7-6 Treatment Decisions: General Public. ....... Fm ev np Doctor's Respect for Patient's Treatment Preferences: General Public ................................... Effect of Increased Patient Role in Decisionmaking: General Public ................. ool Effect of Increased Patient Role in Decisionmaking: Physicians ..............oooiiiiiiiii Reasons for Effects of Increased Patient Role in Decisionmaking: Physicians .......vvssuiensonys ans Reasons for Effects of Increased Patient Role in Decisionmaking: General Public ................... Chapter 8: Decisions to Forego Treatment or Terminate Life 8-1 8-2 8-3 8-4 8-5 8-6 8-7 8-8 8-9 Responsibility for Final Choices if Seriously Ill: General Public Treatment Decisions When Patient is Too Sick to Decide: General Public............................ Provided Treatment Instructions: General Public... .. Left Written Treatment Instructions: General Public. . Should Law Allow Physicians to Honor Patients’ Written Instructions: General Public................ Desire to be Informed of Diagnosis of Cancer: General Public i cisussnissassniivmssmesnmsrmssnes Provide Estimate of Life Expectancy to Dying Patients: ‘General PUBIC... «ow ois wis ssi sions s si 4 5 aes Provide Estimate of Life Expectancy to Dying Patient: Physicians ............... oc iii iii Decision Between Aggressive Therapy and Supportive Therapy: General Public................ Surveys of Physicians and the Public 31 8-10 Who Should Make Treatment Decision: Physicians .. 248 8-11 Resuscitation of a Dying Patient: Physicians ........ 249 8-12 Initiation of Discussion with Dying Patient About Resuscitation: Physicians. ........... cocoa 250 8-13 Resuscitation of Dying Patient Who Requested Life Maintenance: Physicians... 251 8-14 Resuscitation of Dying Patient Who Provided Written Instructions: Physicians. .............ooooiiinnn 252 8-15 Likelihood of Administering Narcotics to Reduce Pain: Physicians. ....cocovivsvvmnrssisneassscnsen 253 8-16 Permissibility of Administering Narcotics to Reduce Pain: Physicians. .........coovnvniiiiiiiiinnnnns 254 8-17 Permissibility of Administering Palliative Drugs: Ceneral Public .. cov vsiissimsromismassmessmrmme ems 255 8-18 Termination of Life as Ethically Permissible: Physicians ........covoveeiiiiiiiiiii 256 8-19 Likelihood of Terminating Life: Physicians. ......... 257 8-20 Permissibility of Physician Complying With a Request for Termination: General Public............ 258 8-21 Should Law Allow Administration of Potentially Life-Abbreviating Drugs to Reduce Pain: Physicians . 259 8-22 Should Law Allow Administration of Potentially Life-Abbreviating Drugs: General Public ............ 260 8-23 Should Law Allow Physicians to Comply With Dying Patient's Request to End Life: Physicians ..... 261 8-24 Should Law Allow Physicians to Comply With Dying Patient's Request to End Life: General Public.. 262 Chapter 9: Impact of Professional Education and Values on Physicians’ Attitudes and Behavior of Physicians Related to Informed Consent 9-1 Profile of Physicians by Sex, Father's Education, Religious Background, and Ability to Speak English. 271 9-2 Ability to Speak English: Physicians ............... 272 9-3 Medical School Attended: Physicians .............. 273 9-4 Formal Training in Medical School: Physicians. ..... 274 9-5 Formal Training in Medical School: Physicians. ..... 275 9-6 Greatest Influence on Practice: Physicians .......... 276 9-7 Greatest Infuence on Practice: Physicians ........... 277 9-8 Responsibility for Patients’ Knowledge of Condition: PhysICIBNG .....vvvevrmnsiiosvevussnsnrmesanaenss 278 9-9 Initiate Discussions of Risks With Patients: PhySiCIans .......coovvvrrrrreesrnavnrrnsseesunnss 279 9-10 Frequency Doctors Withhold Information: Physicians 280 32 Making Health Care Decisions: Appendix B 9-11 Procedures for Which No Consent is Obtained: Physicians ...... BWR LD § EE SE BN 4 on 281 9-12 Increasing Patient's Role in Decisionmaking: Physicians ....................................... 282 9-13 Likelihood of Not Ordering Resuscitation: Physicians 283 9-14 Initiation of Discussion With Patient About Resuscitation: Physicians. ......................... 284 . Appendix A: Profiles of the Public and Physician Samples A-1 Profile of the Public Sample ....................... 286 A-2 Profile of the Physician Sample.................... 289 Surveys of Physicians and the Public 33 CHAPTER 1: INTRODUCTION Background The decade of the 1970s witnessed significant changes in traditional doctor-patient relation- ships. Public opinion surveys charted a steep de- cline in public confidence in the medical profes- sion. Public consciousness concerning patients’ rights increased dramatically. The courts redefined physician obligations in achieving informed consent from their patients prior to medical and surgical procedures. Issues of termination of care were raised in courts and in the media. Medical malprac- tice cases accelerated in their incidence and severity. The legal and ethical issues generated from the changing public attitudes and law regarding patient information and decisionmaking led to the es- tablishment of the President's Commission for the Study of the Ethical Problems in Medicine and Bio- medical and Behavioral Research (PCEMR). The PCEMR was established in 1978 by an act of Congress (Pub- lic Law 95-622) to investigate these problems. Spe- cifically, the act required the PCEMR to investigate "the ethical and legal implications for . . . the requirements of informed consent to . . . undergo medical procedures" and to report its findings (in- cluding any recommendations for legislation or ad- ministrative action) to the President, Congress, and the heads of relevant administrative agencies. The PCEMR determined that existing empirical knowledge of physician and public attitudes and be- haviors with regard to the underlying principles of informed consent provided an inadequate base for recommendations in this area. Moreover, in view of the fact that legislative enactments and admin- istrative regulations dealing with informed consent implicitly operate on a single prototypical model -- despite the fact that informed consent is required in a wide variety of situations that do not conform to this model -- there is reason to be concerned that the standards governing the obtaining of in- formed consent may be unrealistic since they fail to take into account the fact that medical research and practice is a widely diverse, not monolithic, en- terprise. 34 Making Health Care Decisions: Appendix B For these reasons the PCEMR decided to contract for several empirical studies, including national surveys of physicians and the general public. Par- allel surveys of physicians and the public on cur- rent practices with regard to decisionmaking and the disclosure of information under various circum- stances, perceptions about what should occur, and understanding of the law on these subjects were thought to be particularly useful. Louis Harris and Associates was commissioned as an independent contractor, and not as an agent of the government, by the PCEMR to conduct a national survey of doctor-patient communication and medical decisionmaking as part of the PCEMR's larger study of the ethical and legal implications of informed consent. The PCEMR was also able to use this survey effort to obtain needed information for some of their concurrent studies, including decisions about whether or not to forego life-sustaining therapy. Because the PCEMR felt that the greatest contri- bution to relevant knowledge would be made by plac- ing the assessment of the central issues of interest firmly within the context of the patient-provider relationship, parallel surveys of nationally repre- sentative cross sections of the public and physi- cians were effected. These parallel cross-sectional assessments allow for the investigation of the de- gree of convergence in perspectives of the key ac- tors involved in the information exchange process as well as for the investigation of points of and underlying reasons for divergences in these perspectives. Specifically, the commissioned survey sought to ascertain valid and reliable information from the public and practitioners concerning: Information Disclosure sn-ormation Jisclosure -- The amount and type of information dis- closed regularly about medical condition and possible treatment; -- Respective doctor and patient roles in the information disclosure process; -- Physicians' motivations and/or reasons for disclosure or nondisclosure of information; and Surveys of Physicians and the Public 35 -- The general public's perception of phy- sician nondisclosure as acceptable as well as the specific conditions under which the public views nondisclosure as acceptable. Informed Consent -— Perceived definition and scope; —- Preferred standard defining a physician's legal obligations; -- Type of consent (oral, written, both, or none) obtained by physicians for a number of procedures; -- Perception of written consent forms as beneficial or harmful to meaningful doctor-patient interchange, information disclosure, therapeutic process and outcome; -- Attitudes toward the general doctrine of informed consent; and -- Perceptions of legal liability under the informed consent doctrine. Therapeutic Decisionmaking —- Perceived roles of patient and physician in the medical decisionmaking process; —- Conditions that moderate these perceived roles, including type and severity of ill- ness, prognosis, and family intervention; and -- Perceived impacts of increasing the pa- tient's role in medical decisionmaking. Decisions to Forego Treatment -- Respective doctor and patient roles; -- Decisions about nonresuscitation; —- Views concerning the ethics of non- resuscitation; 36 Making Health Care Decisions: Appendix B -=- Physician and patient views on the ef- fects of patient input into this de- cisionmaking process; -- Physician perceptions of ethical be- havior and reports of actual behavior; —-- Abbreviating and terminating life; -- Physician perceptions of ethical be-~ havior and reports of actual behavior; -= Public views concerning the ethics of such behaviors; -= Physician and patient views on the ef- fects of patient input into the de- cisionmaking process; and -= Physician and patient views on the role the law should play in mediating life- abbreviating or life-terminating pro- cedures. In order to accomplish these research objectives, Louis Harris and Associates conducted parallel sur- veys of a national sample of physicians engaged in primary care of adults (N=805) and a national sample of the noninstitutionalized adult population of the United States (N=1,251). This project is the first parallel national survey of doctors and patients re- garding information disclosure and decisionmaking in the course of medical treatment. Sampling Methodology: Public Survey The goal of the public survey was to generate population estimates of the attitudes and experi- ences of the adult, noninstitutionalized population of the United States. Since the survey was conduct- ed by telephone, all nontelephone households in the United States were effectively excluded from the sample. According to the most recent estimate of the Bureau of the Census, the total noncoverage of the survey would be expected to be 8%. Although this noncoverage rate is quite low, it is distribut- ed unequally among regions, communities, and demo- graphic groups. Low telephone incidence rates are most typical of certain rural and semirural areas in the southern United States and in multifamily dwell- Surveys of Physicians and the Public 37 ing units in inner cities, particularly in low income areas. In some inner city communities, the telephone incidence rate may be as low as 70%. The problem of noncoverage of certain population areas and groups in telephone surveys has led Louis Harris and Associates to develop an area-probability- based sampling frame within which a random digit dialing (RDD) sampling procedure is used. The Harris organization employs a first-stage population-based sampling frame, rather than a telephone-exchange- based sampling frame, in order to draw inferences about the total household population of the United States, rather than about the total population with telephones. We treat the absence of telephones in some part of the population as a bias in our sample, rather than eliminate that bias in the sample by specifying a more limited universe (population with telephones). If this fundamental assumption is ac- cepted, then the rationale for the area-probability framework for random digit dialing becomes clear. The likelihood for selection into the sample should depend upon population incidence, rather than upon telephone incidence. Given geographic differences in telephone incidence, this says implicitly that a person with a telephone from the same community as a person without a telephone will provide a better measure of nontelephone individuals' opinions and experiences than a telephone individual in another community. Thus, we develop a geographically stra- tified sample of the total household population of the United States as the basis for apportioning the second-stage telephone numbers in national telephone surveys. The initial stage of sample construction requires the development of a national-area-probability sample based upon the distribution of the adult population of the United States. First, the adult household population of the country is stratified by region and type of place. For regional stratification the Uni- ted States is divided into four regions as follows: 1. East: Maine, New Hampshire, Vermont, Massa- chusetts, Rhode Island, Connecticut, New York, New Jersey, Pennsylvania, Delaware, Maryland, District of Columbia, and West Virginia. 2. South: Virginia, North Carolina, South Caro- lina, Florida, Georgia, Kentucky, Tennessee, Alabama, Mississippi, Arkansas, Louisiana, Texas, and Oklahoma. 38 Making Health Care Decisions: Appendix B 3. Midwest: Ohio, Michigan, Indiana, Illinois, Wisconsin, Minnesota, Iowa, Missouri, Kansas, Nebraska, South Dakota, and North Dakota. 4, West: Montana, Wyoming, Colorado, New Mex- ico, Arizona, Utah, Idaho, Nevada, Cali- fornia, Oregon, and Washington. We also use four Bureau of Census definitions for type of community: 1. Central City: Every place defined as a cen- tral city by the Bureau of the Census. 2. Suburb: Every place that is not a central ————————— . . . . city but is within an urbanized area as de- fined by the Bureau of the Census. 3. Town: Every city, town, or village that is not an urbanized area but is within the town boundaries. 4. Rural: Every place that is not included in any of the other three categories. The total sample for the study is allocated to each of these sixteen strata in proportion to the total adult household population in each stratum. Within each stratum, counties are selected as the primary sampling units. These primary sampling units are selected in proportion to the distribution of the population within the stratum. Operational- ly, a listing is constructed of the latest estimates of the adult population of every county within each region in rank order -- P;;(Ai80/Pigp); then a running cumulative total of gross sums is produced. Next, a random number |, which is less than t/n, where t is the adult population of the stratum, is selected. The sample points (n) are then assigned according to where the numbers V, (U + t/n), (b+ 2t/n), (V+ 3t/n). . . .(V + (n-1)t/n) fall on the running cumulative total of the adult population within that stratum. This same procedure is then applied to each county within each stratum to form primary sampling units (PSUs). At the next stage of selection, one telephone number for each PSU is randomly selected from the updated Louis Harris and Associates library of tele- phone directories. The selected numbers are then Surveys of Physicians and the Public 39 altered by dropping the last two digits (in central city and suburban areas) or the last three digits (in towns and rural areas) and replacing them with randomly selected numbers. As many two- Or three- digit randomly selected numbers as needed are appended until a working residential number is reached or until an interview is completed. Tech- nically, this method of sampling produces an epsem sample of all published telephone banks, where the sampling fraction f = n/N for all elements in all strata. Each eight-digit telephone number (area code and the first five digits) is generated and recorded on a sample card. Interviewers receive a group of sam- ple cards plus another card with five two-digit ran- dom numbers to be added to the existing partial tel- ephone numbers. The interviewers add one set of random digits to the five-digit number on the sample card. RANDOM DIGITS SAMPLE CARD 10 32 47 (516) 964-82 59 64 For example, the first number called in this case would be (516) 964-8210. If the call results in a completion, the interviewer moves to the next sample card. Only one completed interview for each sample card is permitted. However, if the outcome of the call is a refusal, screenout, noneligible, termi- nate, or disconnect, the interviewer retains the same index card but moves on to the next random dig- it ending: (516) 964-8232. If the number dialed results in a busy signal or a ringing but unanswered phone, the interviewer places the card to the side and goes to the next sample card. If all of the random digit numbers are attempted and none leads to a completion, the interviewer hands the card back to the supervisor. This second stage sampling technique is known as random digit dialing (RDD). The use of RDD sampling 40 Making Health Care Decisions: Appendix B eliminates the otherwise serious problem of unlisted telephone numbers. Nationwide, approximately 20% of all phone subscribers have unlisted phones. More- over, significant variation occurs among demographic groups, with the number of unlisted phones reaching a high of 26% in the West, 29% in large metropolitan areas, 25% among those earning $5,000 to $10,000, and 32% among nonwhites. Thus, as directories grow out of date, noninclusion rates in cities like New York and Chicago exceed 40% among some demographic groups. For these reasons, using published phone listings as the universe is inadequate for telephone surveys and inferior to using random digit dialing. The last stage in the sample selection is the identification of the designated respondent in each household. In each household contacted, the inter- viewer first determines whether there are any males aged eighteen or over in the household. If there is more than one adult male in the household, the youngest adult male is the designated respondent. If there are no adult males in the household, then the youngest adult female in the household becomes the designated respondent. These selection proce- dures adjust the sampling procedures for the likeli- hood of reaching women more than men and older peo- ple more than younger people in random household contacts by phone. These sampling procedures yield a highly representative sample of the adult popu- lation of the United States. Sampling Methodology: Physicians Sample The most complete enumeration of physicians in the United States can be found in the Fisher-Stevens listings. These listings include both AMA and non- AMA physicians and are updated every two weeks by the AMA. They include key identifiers (such as pri- mary specialty and type of practice) that may be im- portant elements in sampling. One of the few limi- tations of these listings is that they do not in- clude telephone numbers. The universe of physicians for this survey was limited to physicians who engage in direct care of adult patients. This universe includes the special- ties of internal medicine, general practice, family practice, obstetrics and gynecology, and surgery. The universe excludes specialties that are ancillary to the administration of medical treatment, e.g., radiology and anesthesiology; specialties that pro- Surveys of Physicians and the Public 41 vide noninvasive care, e.g., psychiatry; and spe- cialties concerned with treatment of children, e.g., pediatrics. The universe also excludes all physi- cians who do not engage in direct care of patients, regardless of their specialty. Both office-based and hospital-based doctors were included in the study population. Since a complete enumeration of the study popu- lation of this survey was available through the AMA data base, we were able to conduct a simple random sampling from this universe. Simple random sampling is a method of selecting n units out of the total N such that every one of the NC, distinct elements has an equal chance of being drawn. A simple random sample is drawn sequentially in practice. At any point in the draw there must be an equal chance of selection for any element in the population not al- ready drawn. Under most circumstances, simple random sampling has the distinct advantage of reducing the variance of sample estimates. Statistical formulas for spec- ifying the sampling precision associated with par- ticular sample sizes are based upon the assumption of simple random sampling. Operationally, systematic sampling usually is employed when a simple random sample is desired. This sampling is done by systematically selecting every "ith" person in the sample universe. In this case the "ith" refers to a constant interval, which is determined by the following formula: i = N/n where N is the number of elements in the population and n is the desired number of elements in the sam- ple. The elements in the sampling universe are listed in a random order. A computer-generated ran- dom number is used to select an initial number be- tween 1 and i to establish a random start. The con- stant interval i is sequentially accumulated until all sampled elements have been designated. This procedure can be demonstrated to be statistically identical to the method whereby individual elements are selected at random without replacement from the population. The specifications for the sampling of the phy- sician population were developed by Louis Harris and 42 Making Health Care Decisions: Appendix B Associates. The actual conduct of the sampling from the AMA data base was performed by the Survey and Data Resources Division of the American Medical As- sociation. In order to allow for ineligibility, nonlocation, and nonresponse among the sample, the drawn sample was three times as large (2,400) as the achieved sample size desired (800). The procedures should yield a pure sample of the population of physicians in the United States who are engaged in the direct care of adults on more than a part-time basis. Data Collection Procedures Both the public and the physician surveys were conducted by telephone. The length of both in- terviews was dictated by the maximum effective length of telephone interviews with the study pop- ulations. The public questionnaire was developed for a thirty-minute interview -- the maximum dura- tion of a telephone interview among any population before serious problems of termination and refusal are encountered. The interviews with physicians required a shorter duration, because they were con- ducted during office hours. Consequently, a twenty- minute questionnaire was developed for the physician sample. A traditional telephone survey was inappropriate for the physician sample. The objective of the sur- vey of physicians was to elicit professional judg- ments concerning the diagnosis and treatment of pa- tients. Physicians are accustomed to making judg- ments on the merits and circumstances of particular cases. Therefore, a series of hypothetical case studies was developed to allow physicians to make more realistic estimates. Case studies require reading and consideration, rather than snap judgments. Therefore, the physi- cians survey was developed as a mail-assisted tel- ephone survey. This methodology involves mailing an advance copy of the questionnaire to the physicians, along with a letter directing the doctors' attention to the particular questions that the doctors might wish to reflect upon prior to the interview. One week after the mailout of the questionnaires, ex- ecutive interviewers from Louis Harris and As- sociates contacted the sampled physicians to arrange Surveys of Physicians and the Public 43 an appointment for the telephone interview. If a physician had not received the questionnaire, an- other survey instrument was mailed prior to the in- terview. The mail-assisted telephone methodology was employed only in the physicians sample. The public sample received no advance copy of the ques- tionnaire prior to their telephone interviews. The survey instruments were developed on the ba- sis of extensive discussions among the Louis Harris and Associates project staff and PCEMR staff and consultants. Focus groups were held with physicians and the public in order to better identify the issue structure among the study groups. Draft question- naires were developed by the Harris staff, reviewed by the PCEMR staff, and pretested among the study groups. Three separate pretests were conducted among the physicians until a survey instrument emerged that appeared to satisfy both the re- searchers' and the participants' needs. For the physicians sample, some of the questions of interest took alternate forms. Since repeated measures of the same respondent on each version of a question were impractical, several split ballot items were developed. For example, half of the sam- ple was asked whether they discussed certain issues with their patients -- always, usually, sometimes, rarely, or never, while the other half of the sample was asked whether they initiated discussion of the item with their patients as a matter of course or only if asked. In order to accommodate these split ballot items, four versions of the physicians ques-— tionnaire were developed. Similarly, two versions of the public questionnaire were developed to ac- commodate more limited use of the split ballot tech- nique. Physician Incentives Since physicians were interviewed during their working hours, it was considered appropriate to of- fer them an honorarium for their time. Previous re- search among physicians had demonstrated that fi- nancial incentives were crucial to attain high res- ponse rates among this population. The published data from the 1979 Center for Disease Control im- munization survey is quite revealing. For a twenty- to thirty-minute telephone interview on a non- commercial topic for a respected government health 44 Making Health Care Decisions: Appendix B agency, response rates were reported at the end of a five-and-a-half-week field period, as shown in Table 1-1. These findings clearly indicated that some fi- nancial incentive would be required, and they strongly suggested that a $50 incentive would be necessary to attain the desired 70% response rate among the general and family practitioners and internists who comprise the bulk of the sample. However, in conjunction with the Office of Manage- ment and Budget, a test of the differential effects of the two levels of incentive -- $25 and $50 -- was arranged. Although physicians were promised a small honorarium in the cover letter, the amount was not specified. When interviewers contacted a physician, they systematically alternated the amount of the incentive, i.e., the first doctor contacted by an interviewer was offered $25, the second was offered $50, the third was offered $25, and so on. Careful records of sample disposition by type of response rate were kept so that the effects of the incentive could be determined. Participation Rates The surveys of physicians and the general public were conducted betwzen April 15 and June 3, 1982. The relatively long field period of seven weeks was necessitated by the number of callbacks required to complete interviews among the sample of physicians. In some cases, more than ten callbacks had to be made over a period of weeks before an interview was successfully completed. A total of 805 interviews were completed with physicians. The completed sample was evenly bal- anced between the four interview forms. Interviews were completed with 68% of the eligible physicians who were reached during the field period -- this represents a remarkably good response rate for this population. For a detailed breakdown of the re- sponse rate for the physician sample, see Table 1-2. A total of 1,251 interviews were completed with the national cross section of the noninstitution- alized adult population of the United States. In- terviews were completed with 70% of the eligible households that were reached during the field peri- od. This response rate is fairly representative for TABLE 1-1 SPECIALTY GROUP PARTICIPATION BY EXPERIMENTAL GROUP, OCTOBER 12, 1979 NO INCENTIVE $25 $50 RESPONSE RESPONSE RESPONSE N RATE N RATE N RATE ALL PHYSICIANS 137 58% 147 697% 145 77% GENERAL AND FAMILY PRACTITIONERS 38 37% 35 54% 39 64% INTERNISTS 41 39% 40 487% 32 78% PEDIATRICIANS 34 79% 39 87% 40 90% INDUSTRIAL PHYSICIANS 24 92% 33 88% 34 76% NOTE: FROM "PHYSICIAN RESPONSE TO A TELEPHONE SURVEY, EFFECTS OF MONETARY INCENTIVE LEVEL." GUNN, W.J. AND RHODES, I.N., PUBLIC OPINION QUARTERLY, VOL.45, NO.1, 1981, pp. 109-115. o1[qn 9} PU SUBRIDISAY{ JO SAdAING Sh 46 Making Health Care Decisions: Appendix B telephone interviews of this length. A substantial portion of the public cannot be induced to spend half an hour on the telephone. For this reason thirty minutes is considered to be the upper limit for telephone interviewing. A detailed breakdown of the response rate for the public sample is presented in Table 1-3. The difference in incentives appeared to have no appreciable effect on the willingness of doctors to participate in the survey. Among those doctors who were offered a $25 incentive, a participation rate of 86% was obtained. This is statistically in- distinguishable from the 88% participation rate ob- tained from the doctors who were offered a $50 honorarium (Table 1-2).1 The hypothesis that the larger incentive might have promoted quick response to the survey was explored, even if it did not af- fect the ultimate outcome. Scrutiny of survey com- pletion rates by the number of callbacks, however, reflects no difference between the two incentive levels (Table 1-4). lyote that the refusals obtained before an offer of an incentive was made are excluded from this calcu- lation. This accounts for the difference between participation rates and the overall response rate. Surveys of Physicians and the Public 47 TABLE 1-2 SAMPLE DISPOSITION: PHYSICIANS x = NOT ASKED. PHYSICIANS INCENTIVE TOTAL EXPERIMENT $25 $50 % % TOTAL SAMPLE DRAWN 2106 NOT LOCATED 418 x x NO TELEPHONE LISTING (326) UNLISTED NUMBER (33) NO ANSWER (59) INELIGIBLE 104 x x NOT AVAILABLE DURING SURVEY PERIOD 70 x x OUTSTANDING AT END OF SURVEY PERIOD 302 x x TOTAL ELIGIBLE CONTACTS 1183 434 ° 490 REFUSED PRIOR TO INCENTIVE OFFER 259 x x REFUSED AFTER INCENTIVE OFFER 118 59 59 TERMINATED 1 1 - COMPLETED INTERVIEW 805 374 431 RESPONSE RATE 68.0% 86.2% 88.0% 48 Making Health Care Decisions: Appendix B TABLE 1-3 FINAL SAMPLE DISPOSITION: GENERAL PUBLIC GENERAL PUBLIC TOTAL SAMPLE DRAWN 3385 NOT IN SERVICE 794 NO ANSWER/BUSY 318 QUOTA MET 62 CALLBACKS IN FIELD AT END OF INTERVIEW PERIOD 139 OTHER NONELIGIBLE 272 TOTAL ELIGIBLE CONTACTS 1800 REFUSALS 532 TERMINATED 17 COMPLETED 1251 RESPONSE RATE 69.5% Surveys of Physicians and the Public 49 TABLE 1-4 INTERVIEW COMPLETION BY CALLS AND INCENTIVE: PHYSICIANS TOTAL $25 $50 CUMULATIVE N % N % N % 1ST CALL 145 18 63 17 82 19 2ND CALL 354 44 169 45 185 43 3RD CALL 501 62 233 62 268 62 4TH CALL 610 76 294 79 316 73 STH CALL 667 83 311 83 356 83 6TH CALL 713 89 335 90 378 88 7TH CALL 739 92 350 94 389 90 8TH CALL 765 95 358 96 407 94 9TH CALL 780 97 364 97 416 97 10 OR MORE 805 100 374 100 431 100 Expected Sampling Error The objective of the sampling procedures and field procedures outlined in the previous sections was to produce an unbiased sample of the study pop- ulations. An unbiased sample will share the same properties and characteristics of the total pop- ulation from which it is drawn, subject to a certain level of sampling error. The maximum expected sampling error for a simple random probability sam- ple of population elements may be derived from the following formula: o p{q) Var (x) = zg /2]n-1 where: var (X) = the expected sampling error of the mean of some variable, expressed as a proportion; 50 Making Health Care Decisions: Appendix B z = the standard normal score for some con- fidence limit; P = some proportion of the sample displaying a certain characteristic or giving a certain answer on some variable or ques- tion; q = 1.0 - p; and n = the size of the sample. Assuming a desired confidence level of 95%, the for- mula can be solved for the maximum expected sampling error for various sample sizes and various sample splits on study characteristics. The maximum expected sampling error for the full sample of 805 physicians would be + 3.5 percentage points. In other words, if 50% of the sample of physicians state that they make a conscious and de- liberate evaluation of how much to tell a patient at least once a day, we can be 95% confident that if every doctor in the entire study population was in- terviewed, the results would vary no more than be- tween 46.5% and 53.5%. The larger sample size of the public sample in- creases the sampling precision somewhat. The max- imum expected sampling error for a sample of 1,250 would be + 2.8% points at the 95% confidence level. Table 1-5 gives the expected size of sampling error for subgroups of various sizes, and at different re- sponse distributions on the dependent variable. The reader may use the table to estimate the expected limits of sampling error for various findings re- ported in the body of the text. Estimating Statistical Significance To test whether or not a difference between two sample proportions is statistically significant, some simple calculations are necessary. Call the total sampling error (i.e., 1.96[ BD of the first sample S; and the total sampling error of the sec- ond sample response Sj. Then, the sam ing error of the difference, Sq ={S1 ¢ + Sp 2, Any difference between observed proportions that exceeds Sq is a significant difference. Note that this technique is mathematically equivalent to generating Surveys of Physicians and the Public 51 standardized tests of the difference between proportions. TABLE 1-5 EXPECTED SAMPLING ERROR (PLUS OR MINUS) AT 95% CONFIDENCE LEVEL (SIMPLE RANDOM SAMPLE) PERCENTAGE OF THE SAMPLE OR SUBSAMPLE GIVING SIZE OF A CERTAIN RESPONSE OR DISPLAYING A CERTAIN SAMPLE OR CHARACTERISTIC FOR PERCENTAGES NEAR: SUBSAMPLE 10 OR 90 20 OR 80 30 OR 70 40 OR 60 50 1,300 1.6 2.2 2.5 2.7 2.7 1,200 1.7 2.3 2.6 2.8 2.8 1,100 1.8 2.4 2.7 2.9 3.0 1,000 1.9 2.5 2.8 3.0 3.1 900 2.0 2.6 3.0 3.2 3.3 800 2.1 2.8 3.2 3.4 3:5 700 2.2 3.0 3.4 3.6 3.7 600 2.4 3.2 3.7 3.9 4.0 500 2.6 3.5 4.0 4.3 4.4 400 2.9 3.9 4.5 4.8 4.9 300 3.4 4.5 5.2 5.6 5.7 200 4.2 5.6 6.4 6.8 6.9 150 4.8 6.4 7.4 7.9 8.0 100 S.9 7.9 9.0 9.7 9.8 75 6.8 9.1 10.4 11.2 11.4 50 8.4 11.2 12.8 13.7 14.0 NOTE: ENTRIES ARE EXPRESSED AS PERCENTAGE POINTS (+ or -). For example, if 44% of females in the sample (n = 651) answered "yes" to a particular question, compared to 50% of the males in the sample (n = 600), this six percentage-point difference would be sig- nificant since S; = 3 Ja-Lhecfaaaiesl, So = 4.0 (for males), and Sq =43.74 + 4.04 = 5.4. On the other hand, if 447% of the people from the East (n = 326), answered "yes" to a question, compared to 52 Making Health Care Decisions: Appendix B 50% of the people from the Midwest (n = 326), then the six percentage-point difference between these smaller groups would not be significant since S1 = 5.7 (for _the East Sp = 5.7 (for the Mid- west) and Sq ={5.72 + 5.72 = 8.1. The sampling error for response distributions of groups of different sizes is given in the Table 1-6. The actual sizes of the groups and the res- ponse patterns are reported in the data tables throughout this report. Thus, with a limited amount of interpolation and calculation, the statistical significance of differences observed in the tables can be established easily. However, note that es- timates of pooled sampling error were calculated based upon a binomial distribution where p = 0.5. As the observed distribution becomes more disparate from p = 0.5 (e.g., p = 0.3) the magnitude of dif- ferences between observed proportions required to yield statistically significant contrasts de- creases. Thus, the values of pooled sampling error presented in Table 1-6 should be viewed as indicat- ing the maximum difference values required for yielding statistically significant contrasts. TABLE 1-6 POOLED SAMPLING ERROR EXPRESSED AS PERCENTAGES FOR GIVEN SAMPLE SIZES (ASSUMING p=q) SAMPLE SIZE 1100 10.2 7.5 1000 10.3 7.6 6.5 900 10.3 7.6 6.6 5.9 800 10.4 2:7 6.7 6.0 5.6 700 10.5 7.8 6.8 6.1 5.8 5.4 600 10.6 8.0 7.0 6.3 5.9 5.7 500 10.7 8.2 7.2 6.6 6.2 400 11.0 8.5 7.5 6.9 300 11.3 9.0 8.1 200 12.0 9.8 100 13.9 100 200 300 400 500 600 SAMPLE SIZE Surveys of Physicians and the Public 53 CHAPTER 2: HEALTH PROVIDERS AND HEALTH CONSUMERS: THE CONTEXT OF INFORMED CONSENT Introduction Before considering the extent and pattern of in- formation disclosure and decisionmaking in the med- ical treatment situation, it is important to under- stand the context in which these interactions occur. The characteristics and attitudes of both health care providers and health care consumers may significantly influence these interactions. More specifically, the physicians' experience, specialty, patient load, and practice characteristics may have a profound effect on the way they view the require- ments, outcomes, and process of securing patients’ informed consent to undergo medical procedures. So too, might patients' health status, general satis- faction with physicians, and experience with medical procedures be related to their views of the informed consent process. This chapter attempts to put into perspective later discussions of informed consent -- the core issue of the current research -- by describing the results of the two parallel surveys of the general public (patients) and physicians, and how the re- sults pertain to the medical practice environment, including the distribution of professional services by specialty, location, patient load, and health care service utilization patterns. The Health Status of the General Public The study focuses on the attitudes and pref- erences of the general public concerning medical in- formation disclosure and decisionmaking in the med- ical treatment process. A central question in any attitudinal research project is to what extent are or are not attitudes grounded in experience. That question is particularly relevant in this study of public attitudes toward informed consent. There- fore, we begin this report by investigating the cur- rent health status and health experience of the pub- lic sample. The noninstitutionalized adult population of the United States considers itself to be quite healthy (Table 2-1). When asked to rate their own health 54 Making Health Care Decisions: Appendix B compared to that of others their age, 39% of the public respondents describe their health as ex- cellent, while another 43% say that their health is good. A less than positive assessment of current health is offered by 11% of the public, who describe their health as fair. Only 6% of the public say that their health is poor compared to others their own age. Poor health is related to age and income. When the percent of persons comprising a specific cat- egory of self-health rating is calculated, one can see that while those aged 65 and over comprise 16% of the total American public, they are overrepre- sented among those who rate their health as fair (32%) or poor (38%). The relationship between age and health is an expected part of the biological process. The relationship between income and health is equally striking and well known. The proportion of the public in excellent health rises from 20% in households with incomes of $7,500 or less, to 31% in households making $7,501 to $15,000, to 40% in households with incomes of $15,001 to $25,000, to 46% in households with incomes of $25,001 to $35,000, and to 52% of households with incomes over $35,000. By contrast, less than 7% of individuals with incomes over $35,000 describe themselves as be- ing in fair or poor health, compared to 35% of those with incomes of $7,500 or less. Some of the observed relationships between in- come and health may be spurious. More than one- third (34%) of those in poor health are retired (Table 2-2). Since the retired population is also the older population, age rather than income is the likely source of poor health. Another 16% of those in poor health are disabled or too ill to work. Thus, half of those in poor health are also limited in income by either their condition or their ad- vanced age. Not all of the relationships between income and health, however, are products of spurious cor- relations. Even when the data is controlled for age, a relationship between income and health per- sists. This relationship has been repeatedly ob- served in health surveys for decades. The relation- ship between race and health is one illustration of this effect (Table 2-]). While 15% of the white Surveys of Physicians and the Public 55 population report fair or poor health, we find that almost twice as many minority respondents (28%) re- port fair or poor health. Indeed, the true dif- ference in health status between white and minority populations is even greater when one considers that the minority population is considerably younger than the white population. Observation: The relationships among health status, income, age, race, and related sociodemographic vari- ables is well known in the literature of med- ical sociology. We have spent some time out- lining the relationships in this data set be- cause it is important in the later study of at- titudes toward informed consent and medical de- cisionmaking. Because of the strong relation- ship between health status, income, age, and ed- ucation, it is generally to be expected that if another variable is correlated with one of these factors, it will co-vary with the rest of them. This makes it more difficult to infer causal re- lationships from simple two-way and even three- way cross tabulations. Two other indicators of health condition provide additional measures of current health status. The survey finds that 17% of the adult population report that their activities are limited because of a health condition or disability (Table 2-3). The measure of current activity limitation is highly correlated with the self-rating of health. While only 6% of those in excellent health report any ac- tivity limitation, 74% of those in poor health re- port activity limitations. Interestingly, the pro- portion that reports activity limitations appears to double between each level of the personal health ra- ting in a reasonably uniform manner. The survey also finds that 15% of the adult pop- ulation reports being hospitalized overnight on at least one occasion during the past year. It is in- teresting to note that this percentage is virtually identical to the 16% of the public who reported be- ing hospitalized overnight during the past year in the Louis Harris and Associates 1978 survey of Hos- pital Care in America. The incidence of recent hospitalization is re- lated to current health status, but much less so 56 Making Health Care Decisions: Appendix B than are activity limitations. Among those in ex- cellent health, 10% have been hospitalized overnight in the past year. The incidence of hospitalization increases to 15% for those in good health and 20% for those in fair health. More than a third (35%) of those in poor health have been hospitalized over- night in the past year. For those who were hos- pitalized overnight in the past year, the average number of nights spent in the hospital increases from 6.0 among those in excellent health to 24.6 for those in poor health. Experience with Chronic and Life-Threatening Illness The issues of life-threatening illness are not limited to the small subset of Americans in poor health. Three out of ten Americans (31%) report that they or another member of their immediate fam- ily have had a life-threatening illness or medical condition (Table 2-4). Although life-threatening illnesses are more common among those in poor health, we find that 27% of those in excellent health and 32% of those in good health report a life-threatening illness in their family. Indeed, 6% of those in excellent health and 8% of those in good health say that they have had a life- threatening condition. The life-threatening conditions reported by the sample reflect the leading causes of death in Amer- ica. Among those who report a life-threatening con- dition in the family, 23% mention malignant neo- plasms (cancer), 21% report heart attacks, 16% re- port other cardiovascular diseases, 7% mention de- generative diseases, 6% report trauma, 57% report stroke, and 2% report hypertension. It is a note- worthy reflection on the changes affected by medical science since 1900 that in 1982 among those who re- port experience with a life-threatening disease in the family, only 1% mention infections or infectious diseases (Table 2-5). It is not surprising, then, that many Americans have had experience with chronic conditions. When they were asked if there were anyone in their im- mediate family who was chronically sick or who need- ed medical treatment and hospitalization fairly of- ten, the interviewers found that 16% of American households faced such a situation (Table 2-4). Surveys of Physicians and the Public 37 Utilization of Health Care Facilities The personal and family health conditions of Americans represent one aspect of their health ex- perience. Another major aspect of this experience is their health care behavior in seeking treatment for these medical conditions. Most Americans (81%) report that they have a regular source of care that they go to when they are sick or need advice about their health (Table 2-6). The 19% who have no usual source of medical care are not the old, poor, and sick. Quite the contrary, those who have no regular source of medical care tend to be the younger and healthier members of the public. They have no regular source of care because they don't need regular care. The majority of Americans (51%) say that they go to a doctor's office as their usual source of med- ical care (Table 2-7). However, 16% of the general public use clinics as their usual source of medical care; 9% use health centers. Surprisingly, neither income nor health insurance coverage seems to have a orderly effect on the usual source of medical care. It was noted earlier that 15% of the public had spent at least one night in the hospital during the past year. Exclusive of this hospitalization ex- perience, nearly three-quarters (74%) of the Amer- ican public went to see a doctor about their health at least once during the past year (Table 2-8). The major exception was the young and healthy population with no usual source of medical care. Forty-seven percent of individuals with no usual source of care saw a doctor about their health in the previous year. On average, individuals went to see a doctor 3.1 times in the past year, not counting hospitaliza- tions (Table 2-8). As might be expected, there is an inverse relationship between the number of doctor visits during the past year and the current health status of the individual. Those in excellent health saw a physician fewer than 2 times (X = 1.8), those in good health saw a doctor about 3 times (Xx = 2.6), those in fair health saw a doctor 5 times (X = 5.3), and those in poor health saw a doctor 12 times (x = 12.0), on average. Thus, as demonstrated repeatedly in the literature of health care utilization, there is a very strong relationship between health care need and health-seeking behavior, including health 58 Making Health Care Decisions: Appendix B care utilization. This is certainly not to say that no unmet needs exist. Rather, one can simply ac- knowledge that patterns of health-seeking behavior among different sociodemographic groups can only be examined if the need for health care is controlled between groups. This treatment experience leads inevitably to the question of informed consent. A substantial portion of the public has had recent experience with disclosure procedures related to cause and treatment information. Almost one-quarter of the public (24%) report that they have been asked to sign a written consent for treatment within the past year (Table 2-9). However, this report may underestimate the actual incidence of consent situations. The survey finds that the reported incidence of written consent rises significantly with education -- from 19% of those without high school degrees, to 26% of those with college degrees. This is a surprising finding since all other indicators of frequency of treatment tend to decline with education. This sug- gests one of two possible conclusions. Either doc- tors are more likely to ask for written consent from college educated patients than from patients with less education, or the underreporting of written consent procedures tends to decline with education. The latter seems the more plausible explanation. If so, the actual incidence of recent experience with informed consent would be substantially higher then the 24% who report having experienced written con- sent procedures in the past year. Although the frequency with which physicians ob- tain written or oral consent varies with the treat- ment procedure, virtually every doctor reports ob- taining consent from patients for inpatient surgical procedures. Thus, almost everyone who has undergone inpatient surgery during the past twenty years would have been exposed to the consent process. The survey finds that 6 out of 10 Americans (61%) report that they have undergone a surgical op- eration that required an overnight stay in the hos- pital (Table 2-10). This finding provides further evidence of the pervasiveness among the public of the type of treatment experience that is governed by the laws of informed consent. The survey demon- strates that at least one-quarter of the public was recently involved in a treatment experience that re- quired formal consent, while more than half of the Surveys of Physicians and the Public 59 public was involved in a previous treatment ex- perience that required formal consent. Health Providers Although a variety of professions (physician, nurse, medical technician) and institutions (hos- pitals, clinics, HMOs) play important roles in the delivery of health care to the American public, the physician remains the main health care provider. Not all physicians, however, are engaged in direct medical care of adult patients -- the research area of primary interest in this study. Consequently, the universe of health providers examined in this study was restricted to the medical specialties of general practice, family practice, internal med- icine, obstetrics and gynecology, and surgery. Since the purpose of the study was to examine doctor-patient communication in the context of medical decisionmaking, specialties that treat only children (pediatricians), those that have little direct communication with the patient (anesthesi- ologist, radiologist), and those that are not engaged in invasive treatment (psychiatrist) were excluded from the survey. The most common specialty among the physicians surveyed is internal medicine (29%) (Table 2-11). Together, the general practice and family practice specialties rival the number of internists (29%). More than 2 out of 5 sampled physicians (22%) are surgeons,2 involved in either general surgery or other surgical specialties. Doctors specializing in obstetrics and/or gynecology represent 13% of sam- pled physicians. Finally, 6% of physicians who de- fine their primary specialty as one of the five cat- egories above in the AMA listings, defined them- selves somewhat differently in the interview situa- tion (e.g., geriatrics, industrial physician, etc.). Only physicians who reported spending at least twenty hours a week on direct patient care were included in the survey. 2 . Bod : We will treat the term "physician" throughout this report as the equivalent of doctor of medicine, thus including surgeons. 60 Making Health Care Decisions: Appendix B Most physicians in this survey are office based (66%). By comparison, 29% of physicians have a hospital-based main practice. Most of the remaining physicians report that their main practice is equal- ly divided between the office and the hospital (3%). As would be expected, medical specialties of general practice/family practice (87%) and obstet- rics/gynecology (74%) are primarily office based (Table 2-12). To a lesser extent, internal medicine (62%) is also office based. By contrast, most sur- geons (51%) report that their main practice is hos- pital based. The vast majority of physicians (94%) are men (Table 2-13). The sex distribution of doctors be- tween medical specialties ranges from only 2% of surgeons who are women to 13% of Ob/Gyn practi- tioners who are women. The proportion of physicians who are women rises from 3% of those who received their medical degrees prior to 1967, to 7% of those who received their degrees during the years from 1967 and 1972, to 10% of those who received their degrees after 1973. A substantial proportion of physicians are rel- atively young. Nearly one-quarter of primary care physicians (24%) received their degrees in 1976 (Table 2-14). Another 23% received their degrees between 1966 and 1975. Only 15% of primary care physicians received their medical degrees prior to 1945; therefore, they would be approaching retire- ment age. It is important to note that there is a strong relationship between physicians' age (or medical school cohort) and their practice location. More than half (53%) of hospital-based physicians re- ceived their medical degrees in 1976. By contrast, only 12% of office-based physicians graduated from medical school in the past five years. Medical Practice: Types of Patients The medical care practice of a physician may be best described by the patient mix that the doctor treats, The nature of the patient load and the types of patients seen are bound to affect the ex- periences of the physicians in the area of informa- tion disclosure and treatment decisionmaking. Fur- Surveys of Physicians and the Public 61 thermore, differences in practice characteristics may affect attitudes toward informed consent and decisionmaking. For these reasons, physicians were asked to estimate the percentage of their adult pa- tients who are: -- Aged 65 or over; -- College educated; di Poor; ee Suffering from a serious illness or health condition; -- Able to understand most aspects of their treatment and condition. It should be noted that the patient mix reported by the physicians may not match the population distri- bution of the public sample. As we noted in the previous section, certain segments of the public are significantly less healthy (e.g., old, poor, mi- nority respondents) than are other segments of so- ciety. Since doctors on average tend to see more patients who are ill than well patients, it is likely that they will see a somewhat biased sample of the total public. The aging population =-- those who are 65 and over -- represent 16% of the adult population of the United States. However, physicians report that per- sons over the age of 65 comprise more than one-third (X = 35.3%) of the adult patient load, on average (Table 2-15). Hospital-based doctors are more like- ly to see more patients over 65 (X = 40.4%) than office-based doctors (X = 32.8%). The incidence of older patients also varies with the physicians' spe- cialty, with older patients ranging from a mean of 9.8% of the caseload for Ob/Gyn practitioners to a mean of 50.6% of the caseload for internists. Physicians report that about one-quarter (X = 25.3%) of their patients, on average, are col- lege educated (Table 2-16). This estimate is ap- proximately equal to the incidence of the college- educated individuals in the general adult population (23%). There is relatively little variation in the distribution of college-educated patient load across the various medical specialties, practice locations, or ages of the doctors. Poor patients represent approximately one- quarter (X = 24.6%) of the patient load, on average, according to physicians' reports (Table 2-17). 62 Making Health Care Decisions: Appendix B Since there is no equivalent definition of poverty for the public sample, it is not possible to esti- mate whether the poor are overrepresented as pa- tients. Hospital-based doctors report a substan- tially greater proportion of poor patients (X = 26.9%) than do office-based doctors (¥ = 19.3%). There is little difference by specialty in the pro- portion of poor patients. Physicians report that more than one-third (X = 36.3%) of their patients, on average, are suffering from serious illnesses or health con- ditions (Table 2-18). The incidence of serious ill- ness is related to the incidence of poverty among the patient population. Among medical practices where 25% or less of the patients are poor, physi- cians report that approximately 34% of their case load involves seriously ill patients. The case load of seriously ill patients increases to about 44% in medical practices where over half of the patient population is poor. Physicians also report that approximately 79% of their patients, on average, are able to understand most aspects of their treatment and condition, if reasonable time and effort are devoted to explana- tion. The corollary is that physicians estimate that about 21% of their patients cannot understand most aspects of their treatment and conditions. There 1s a relationship between the proportion of patients who are poor and the proportion of patients who can understand treatment. In practices where poor patients represent 25% or less of the patient load, doctors report that an average of 82% of their patients can understand treatment issues. However, in practices where poor patients represent more than half of the patient load, doctors report that only 70% of their patients can understand treatment is- sues. Since the issue of patient understanding goes to the heart of informed consent and patient participa- tion in medical decisionmaking, practice character- istics may have a substantial effect on doctors’ attitudes and perceptions in these areas. Physicians' Work Load Physicians report spending an average of 53.2 hours a week on patient care activities in 1982 (Table 2-19). This reported weekly estimate is Surveys of Physicians and the Public 63 slightly higher than the 1980 average of 49.6 hours practiced per week reported by the AMA Survey of Physicians.3 The number of hours practiced per week is relat- ed to the location of the physicians' office, when the physicians began practice, and the reported com- position of the physicians' clients. Specifically, hospital-based physicians report spending more time on direct patient care than do office-based physi- cians (X = 61.8 vs. X = 49.7 hours). The time spent on direct patient care per week also varies with the physicians' specialty. Surgeons report spending the most time in patient care (X = 57.1 hours) while general and family practitioners report spending the least time in patient care duties (X = 49.5 hours). The survey also assessed the numbers of inpa- tients and outpatients seen by doctors in an average week. On average, physicians see approximately 17 hospital inpatients each week (Table 2-20). As one would expect, the number of inpatients seen is re- lated to specialty, with surgeons seeing the great- est number of inpatients (X = 22.3) and general and family practitioners and obstetricians seeing the fewest (X = 13.3 and 13.7 respectively). Also, practice location (X hospital = 27.7, X¥ office = 12.6), and the percentage of seriously ill patients a doctor treats (X more than 50% = 24.9, X 26% to 50% = 19.1, ¥ 0% to 25% = 12.8), functioned much as would be expected. The number of outpatients seen in an average week was also calculated (Table 2-21). On average, physicians see 72 outpatients each week. The pat- tern of these data presents a mirror image of the data reported for inpatients. Here, surgeons see the fewest number of outpatients (X = 48.1) while general and family practitioners, and obstetricians and gynecologists see the greatest number of out- patients (X = 102.2 and 88.0 respectively). office-based physicians see more outpatients (x = 88.0) than do hospital-based physicians (X = 34.9). Physicians among whose patients there is a high prevalence of serious illness see far few- Fourteenth Periodic Survey of Physicians, March 1980. Center for Health Services Research and Development, American Medical Association. 64 Making Health Care Decisions: Appendix B er outpatients than do physicians whose patients are not generally seriously ill (X = 44.9 vs. 71.4 and 85.4 respectively) .4 Because the number of patients seen and the num- ber of hours spent in direct patient care comprise a physician's work load, a standardized index was cre- ated to better investigate the factors actually re- lated to work load and to better describe the medical practice as observed in this survey. Quite simply, the total number of patients seen in an average week in direct patient care activities was calculated by adding together the number of in- patients and outpatients seen in an average week. This sum was divided by the number of hours worked each week. The resultant quotient is expressed as the average number of patients a physician sees per practice hour. Note, however, that this is not necessarily the same as the number of patients per hour that a doctor will see in his office or the average time he spends with each patient. On average, the physicians sampled see almost two patients per hour of practice (X = 1.8) (Table 2-22). General and family practitioners and Ob/Gyn specialists see the greatest number of people each hour (X = 2.4 and 2.0 respectively), while surgeons see the fewest number of patients during each hour of practice (X = 1.3). Thus, as could be expected, surgeons seem to work the longest hours, both in ab- solute terms and relative to the number of patients they see. Similarly, hospital-based physicians, in gen- eral, see far fewer patients (Xx = 1.1) than do office-based physicians (¥X = 2.1). Patient-Provider Relationships: Public Attitudes Toward Physicians The particular issues of informed consent and medical decisionmaking arise in a broader setting of doctor-patient interactions and relations. The ——————— The average number of outpatients seen per week re- ported by physicians in this 1982 survey (72) is almost identical to the average number of office visits per week (78.7) reported by the physicians in the 1980 AMA survey. Surveys of Physicians and the Public 65 health needs and behaviors of the public have an im- portant influence on the disclosure and decision- making process. The abilities of the public to ac- cept and understand treatment information and the physician's perception of those abilities shape the context of treatment decisions. In addition to these conditions and behaviors, medical decision- making takes place in a particular context of doctor-patient relations. The degree of patient satisfaction with medical providers, particularly the patient's trust in the doctor, may have consi- derable impact on the disclosure and decisionmaking process. In order to evaluate public attitudes toward doctors, particularly in the context of doctor- patient communications and decisionmaking, a number of questions were asked of the general public sam- ple. At the most general level, the survey finds that three-quarters of the public (76%) say that they are at least somewhat satisfied with doctors, in general, while 35% say they are very satisfied. But when queried about the doctors they and their families have used in the last few years, the pro- portion of the population at least somewhat satis-— fied increases to 84%. More importantly, almost half of the public (49%) say that they are "very satisfied" with the doctors they have used in the last few years. That represents a 14 percentage point improvement in the approval rating given to doctors in general. Observation: The public respondents give their doctors, in particular, a better approval rating than they do doctors in general. One possible explanation is that the negative attention given to the medical profession over the past ten years has somewhat eroded public trust in the profession. Public opinion polls have recorded a loss of confidence in medicine over the past decade, along with declining confidence in many other institutions. However, the particular doctor who treats the public is still given a very pos- itive rating by his patients. Generally speaking, the vast majority of the American public seem to be satisfied with the crit- ical elements in doctor-patient relationships 66 Making Health Care Decisions: Appendix B (Table 2-23). Two-thirds of the public (66%) report that they are ''very satisfied" with their doctor's honesty in his or her discussions with them. Another 23% are at least somewhat satisfied with the honesty of their doctor's discussions with them. Only 8% of the public say that they are either somewhat dissatisfied (6%) or very dissatisfied (2%) with the honesty of their doctor. A majority of the public (56%) are very satis- fied with their doctor's concern for their well- being. Another 30% of the public are at least some- what satisfied with their doctor's concern for their well-being. This perception has critical impact on doctor-patient relationships because it means that patients assume that their doctor will attempt to act in the patients' best interests. Most Americans are also quite satisfied with their doctor's respect for their treatment preferen- ces, as well as their doctor's honesty and willing- ness to discuss their condition with them. The sur- vey finds that 59% of the public are "very satisfied" with their doctor's respect for their treatment preferences, and an additional 29% are at least '"somewhat satisfied." Only 7% of the public say that they are 'somewhat dissatisfied" (5%) or "very dissatisfied" (2%) with their doctor's respect for their treatment preferences. These survey find- ings suggest that the relationship between medical service providers and health care consumers still resembles the traditional health care model more than newer models of a medical marketplace. The public respondents generally are quite satisfied that their doctor is concerned for their well-being, that their doctor is honest in discussions with them, and that their doctor will respect patients’ treatment wishes. TABLE 2-1 CURRENT HEALTH STATUS: GENERAL PUBLIC Q.: I would like to begin with some background questions about your own health. First compared to others your own age, would you say that your health is excellent, good, fair, or poor? BASE EXCELLENT GOOD FAIR POOR *LESS THAN 0.5%. GENERAL PUBLIC AGE INCOME 65 $7,501 $15,001 $25,001 MORE 18-34 35-50 51-64 AND $7,500 TO TO TO THAN RACE TOTAL YEARS YEARS YEARS OVER OR LESS $15,000 $25,000 $35,000 $35,000 WHITE MINORITY 1251 488 334 226 195 164 219 341 229 214 1027 201 2 z % x x z 2 x 2 % 2 x 39 42 50 29 24 20 31 40 46 52 40 36 43 48 38 45 36 44 45 48 42 40 45 36 11 9 6 14 24 19 14 10 7 7 10 17 6 1 5 12 15 16 10 3 4 * 5 11 o1[qnd 9Y} Pue SUBIDISAYJ JO SABAING A 68 Making Health Care Decisions: Appendix B G-F3 TABLE 2-2 EMPLOYMENT: GENERAL PUBLIC Q.: Are you presently employed full time, part time, unemployed, retired, a student, a housewife, or what? GENERAL PUBLIC AGE REPORTED HEALTH STATUS 18-34 35-50 51-64 65 AND TOTAL EXCELLENT GOOD FAIR POOR YEARS YEARS YEARS OVER BASE 1251 488 334 226 195 482 550 144 73 2 2 z z x z x 2 2 EMPLOYED FULL TIME 54 63 57 34 23 64 72 52 6 EMPLOYED PART TIME 8 8 8 11 4 9 7 10 5 UNEMPLOYED 5 4 6 8 10 8 4 6 1 RETIRED 14 9 12 27 34 - 1 13 Nn STUDENT 4 5 4 1 - 9 * - 1 HOUSEWIFE 12 1n 12 16 12 11 12 1 15 DISABLED, TOO ILL TO WORK (VOL.) 2 - 1 3 16 - 2 5 1 ON DUTY IN ARMED FORCES (VOL.) * »* - - * * - OTHER * - - 1 - - * - - *LESS THAN 0.5%. Surveys of Physicians and the Public 69 G-4,5,6 TABLE 2-3 DISABILITY AND HOSPITALIZATION: GENERAL PUBLIC Q.: Are your activities limited in any way because of a health condition or disability? Q.: Have you been a patient overnight in a hospital during the past year? Q.: How many nights have you spent in the hospital during the past year? GENERAL PUBLIC REPORTED HEALTH STATUS TOTAL EXCELLENT GOOD FAIR POOR BASE 1251 482 550 144 73 2 x Y 2 x ACTIVITY LIMITATION YES 17 6 13 35 74 NO 83 9% 87 63 26 NOT SURE 1 * 1 2 - HOSPITALIZATION DURING THE PAST YEAR YES 15 10 15 20 35 NO 85 90 85 80 65 BASE 180 48 76 28 28 MEAN NUMBER OF NIGHTS IN HOSPITAL AMONG THOSE HOSPITALIZED IN PAST YEAR 9.8 6.0 6.8 10.2 24.6 *LESS THAN 0.51. G-8A,9A TABLE 2-4 CHRONIC AND LIFE-THREATENING ILLNESS: GENERAL PUBLIC Q.: Including yourself, is there anybody in your immediate family who is chronically sick or who needs medical treatment or hospitalization fairly often? Q.: Have you or a member of your immediate family ever had a life-threatening illness or medical condition? Who was that? GENERAL PUBLIC REPORTED HEALTH STATUS TOTAL EXCELLENT GOOD FAIR POOR BASE 1251 482 550 144 73 2 2 2 2 x LIFE-THREATENING ILLNESS IN THE FAMILY YES 31 27 32 36 50 NO 68 73 68 64 49 LIFE-THREATENING ILLNESS (SELF) YES 10 6 8 18 30 NO 90 94 92 82 70 CHRONIC ILLNESS IN THE FAMILY YES 16 14 14 26 31 NO 84 86 86 74 69 70 Making Health Care Decisions: Appendix B G-9C TABLE 2-5 TYPE OF LIFE-THREATENING ILLNESS: GENERAL PUBLIC BASE: HAD A LIFE-THREATENING ILLNESS OR MEDICAL CONDITION IN FAMILY Q.: What was the life-threatening illness or medical condition you or a member of your immediate family had? TOTAL GENERAL PUBLIC BASE 385 x ALL MALIGNANT NEOPLASMS 23 HEART ATTACKS AND OTHER RELATED CARDIOVASCULAR DISEASES 21 OTHER CARDIOVASCULAR DISEASES 16 CHRONIC DEGENERATIVE DISEASES 7 TRAUMA, ACCIDENTS, SEVERE INJURIES 6 INTESTINAL DISORDERS 5 STROKE 5 BENIGN GYNECOLOGIC-OBSTETRIC CONDITIONS 3 ASTHMA AND ALLERGIES 2 HYPERTENSION, HIGH BLOOD PRESSURE 2 INFECTIONS 1 ALL OTHER MENTIONS 13 NOT SURE/REFUSED 1 TABLE 2-% USUAL SOURCE OF CARE: GENERAL PUBLIC Q.: Is there a particular clinic, health center, doctor's office, or other place that you usually go to when you are sick or need advice about your health? Q.: What kind of place is that -- a clinic, a health center, a hospital, a doctor's office, or some other place? GENERAL PUBLIC DOCTOR'S TOTAL NONE OFFICE OTHER BASE 1251 236 639 376 z 2 z z AGE 18-34 YEARS 41 53 36 43 35-50 YEARS 24 23 25 24 51-64 YEARS 18 14 19 19 65 AND OVER 16 10 19 14 HEALTH STATUS EXCELLENT 39 47 39 33 GOOD 43 41 44 43 FAIR 11 6 il 15 POOR 6 6 5 8 o1[qNd 9Y} PUB SURIDISAY JO SA8AIng 14 TABLE 2-7 USUAL SOURCE OF CARE: GENERAL PUBLIC Q.: Is there a particular clinic, health center, doctor's office, or other place that you usually go to when you are sick or need advice about your health? Q.: What kind of place is that -- a clinic, a health center, a hospital, a doctor's office, or some other place? GENERAL PUBLIC INCOME $7,501 $15,001 $25,001 MORE HEALTH COVERAGE $7,500 TO TO TO THAN MEDI- TOTAL OR LESS $15,000 $25,000 $35,000 $35,000 HMO CARE OTHER NONE BASE 1251 = 164 219 341 229 214 212 "222 "306 "70 2 z z % % % x z % 2 USUAL SOURCE OF CARE NONE 19 18 22 16 20 18 18 10 18 36 DOCTOR'S OFFICE 51 48 38 54 55 56 43 60 54 31 OTHER 30 34 40 30 26 26 40 29 28 33 TL g xrpuaddy :suoistoa(] aie) yjeay Sune Surveys of Physicians and the Public 73 G-7 TABLE 2-8 NUMBER OF TIMES DOCTOR SEEN IN LAST YEAR: GENERAL PUBLIC Q.: How many times have you seen a doctor about your health during the past year, not counting any hospitalization? GENERAL PUBLIC USUAL SOURCE OF CARE a REPORTED HEALTH STATUS DOCTOR'S TOTAL EXCELLENT GOOD FAIR POOR NONE OFFICE OTHER BASE 1251 482 S50 144 73 236 639 “376 2 2 ¥ x 2 x x 2 1 TIME 25 32 24 7 18 29 2 2-3 TIMES 25 20 30 29 13 16 26 28 4-6 TIMES le 8 15 21 23 7 14 16 7-10 TIMES 3 2 4 8 5 3 3 5 10 TIMES OR MORE 7 3 4 12 43 2 7 9 NONE 26 36 23 16 9 53 21 19 NOT SURE - - - - - - - - REFUSED - - - - - - - - MEAN NUMBER OF TIMES 3.1 1.8 2.6 5.3 12.0 1.5 2.8 3.4 74 Making Health Care Decisions: Appendix B G-4,5,7,52 TABLE 2-9 HOSPITALIZATION, VISITS TO DOCTOR, AND WRITTEN CONSENT: GENERAL PUBLIC Q.: Have you been a patient overnight in a hospital during the past year? Q.: How many nights have you spent in the hospital during the past year? Q.: How many times have you seen a doctor about your health during the past year, not counting any hospitalization? Q.: Written consent forms are now standard practice in many hospitals for surgery and certain other medical procedures. Within the last year, have you been asked to sign a written consent for treatment? GENERAL PUBLIC LESS THAN HIGH SCHOOL HIGH SCHOOL COLLEGE TOTAL GRADUATE GRADUATE GRADUATE BASE 1251 242 726 280 HAVE YOU BEEN A PATIENT IN A HOSPITAL IN THE LAST YEAR ES 152 172 152 132 NO 852 83% 85% 87% MEAN NUMBER OF NIGHTS SPENT IN THE HOSPITAL LAST YEAR FOR THOSE HOSPITALIZED (N = 180) 9.8 12.3 10.2 5.5 MEAN NUMBER OF VISITS TO THE DOCTOR IN LAST YEAR 3.1 4.1 3.0 2.7 HAVE YOU BEEN ASKED TO SIGN A WRITTEN CONSENT FOR TREATMENT YES 242 192 24% 262 NO 762 812 762 746% G-10A TABLE 2-10 SURGICAL OPERATION REQUIRING OVERNIGHT HOSPITALIZATION: GENERAL PUBLIC Q.: Have you ever undergone a surgical operation that required an overnight stay in the hospital? GENERAL PUBLIC AGE 18-34 35-50 51-64 65 AND TOTAL YEARS YEARS YEARS OVER BASE 1251 488 334 226 195 % % % % % YES, HAVE UNDERGONE 61 46 63 72 83 NO, HAVEN'T UNDERGONE 39 54 37 28 17 o1[qnd ay} pue sueisAyd jo sfoaing Gl 76 Making Health Care Decisions: Appendix B TABLE 2-11 PHYSICIAN CHARACTERISTICS TOTAL PHYSICIANS BASE 805 z SPECIALTY INTERNAL MEDICINE 29 GENERAL/FAMILY PRACTICE 29 OB/GYN 13 SURGERY 22 ALL OTHERS 6 PRACTICE LOCATION OFFICE 66 HOSPITAL 29 BOTH 3 OBTAINED MEDICAL DEGREE 1966 AND BEFORE 55 1967 TO 1972 12 1973 AND AFTER 33 SEX MALE 94 FEMALE 6 TABLE 2-12 PRACTICE LOCATION: PHYSICIANS Q.: Is your main practice, that is, where you have most of your contact with patients, office-based or hospital-based? PHYSICIANS OBTAINED SPECIALTY MEDICAL DEGREE GENERAL/ 1966 1967 1973 INTERNAL FAMILY ALL AND TO AND TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS BEFORE 1972 AFTER BASE 205 233 231 108 177 45 441 97 266 % Z % % % % % % % PRACTICE LOCATION OFFICE 66 62 87 74 42 49 80 72 42 HOSPITAL 29 31 10 21 51 44 14 23 56 o1[qnd 9Y} pue SUBIISAYJ JO shang LL Q.: Respondent's Sex: BASE MALE FEMALE TABLE 2-13 SEX: PHYSICIANS PHYSICIANS SPECIALTY GENERAL INTERNAL FAMILY ALL TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS 805 233 231 108 177 45 7% % % 7% 7 % 94 94 94 87 98 100 6 6 6 13 2 = 8. gd xtpuaddy :suorsioa(] aren) yiesy Sure Surveys of Physicians and the Public 79 P-F2 TABLE 2-14 RECENCY OF TRAINING: PHYSICIANS Q.: In what year did you graduate from medical school? PHYSICIANS PRACTICE LOCATION TOTAL OFFICE HOSPITAL BASE 805 ~~ 533 237 z z 2 OBTAINED MEDICAL DEGREE BEFORE 1945 15 18 6 1946-1965 39 46 19 1966-1970 9 9 ? 1971-1975 14 14 14 1976 AND AFTER 24 12 53 NOT SURE - - ~ NO ANSWER/REFUSED * * - TESS THAN 0.5%. P-6A TABLE 2-15 PERCENTAGE OF PATIENTS 65 AND OVER: PHYSICIANS Q.: Now I would like to ask you a few questions about the kinds of patients that you see. Approximately what percentage of your adult patients would you say are aged 65 and over? PHYSICIANS SPECIALTY GENERAL/ INTERNAL FAMILY ALL PRACTICE LOCATION TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS OFFICE HOSPITAL BASE 805 233 231 108 177 45 533 232 % x x z x 2 z % 0% to 25% 42 11 48 97 39 49 47 31 26% to 50% 38 47 40 3 45 33 36 41 51% to 75% 17 36 10 - 14 9 14 23 76% AND OVER 4 7 2 - 2 9 2 6 NOT SURE/REFUSED - - - - - = - - MEAN PERCENTAGE 35.3 50.6 31.8 9.8 35.8 32.8 32.8 40.4 STANDARD ERROR 0.8 1.8 1.8 0.9 1.4 3.1 0.9 1.5 80 Making Health Care Decisions: Appendix B P-6B TABLE 2-16 PERCENTAGE OF ADULT PATIENTS WHO ARE COLLEGE EDUCATED: PHYSICIANS Q.: Approximately what percentage of your adult patients would you say are college educated? TOTAL PHYSICIANS BASE 805 2 0X to 25% 62 26% to 50% 30 51% to 75% 6 76% AND OVER 2 NOT SURE /REFUSED MEAN PERCENTAGE 25.3 STANDARD ERROR 0.7 P-6D TABLE 2-17 PERCENTAGE OF ADULT PATIENTS WHO ARE POOR: PHYSICIANS Q.: Approximately what percentage of your adult patients would you say are poor? PHYSICIANS PRACTICE LOCATION TOTAL OFFICE HOSPITAL BASE 805 533 232 Z z 2 0% to 25% 69 78 45 26% to 50% 18 14 28 51% to 752 7 4 15 76% AND OVER 6 4 12 NOT SURE/REFUSED - = = MEAN PERCENTAGE 24.6 19.3 26.9 STANDARD ERROR 0.8 0.9 1.8 Surveys of Physicians and the Public TABLE 2-18 MEAN PERCENTAGE OF PATIENTS WITH AND ABILITY TO UNDERSTAND TREATM SERIOUS ILLNESS ENT: PHYSICIANS 81 Q.: What percentage of your total patients would you characterize as suffering from serious illness or health conditions? Q.: What percentage of your patients would you say are able to understand most aspects of their treatment and condition, if reasonable time and effort are devoted to explanation? TOTAL BASE 805 2 MEAN PERCENTAGE OF: PATIENTS SUFFERING FROM SER10US ILLNESS 36.3 STANDARD ERROR 1.0 PATIENTS ABLE TO UNDERSTAND ASPECTS OF THEIR TREATMENT AND CONDITION 79.4 STANDARD ERROR 0.7 PHYSICIANS POOR PATIENT. 252 51% OR 26X- OR 1 LESS _50X MORE T3500 “143 "107 z x 2 P-1 TABLE 2-19 MEAN NUMBER OF HOURS SPENT ON DIRECT PATIENT CARE: PHYSICIANS Q.: Let me begin by asking you a few general questions about your practice. Including both inpatient and outpatient care, approximately how many hours do you spend on direct patient care activities in an average week? PHYSICIANS OBTAINED PATIENTS WITH PATIENTS SPECIALTY MEDICAL DEGREE SERIOUS ILLNESS UNDERSTAND ING POOR PATIENTS CENERAL/ PRACTICE 1966 1967 1973 MORE TREATMENT 25% SIT INTERNAL FAMILY ALL LOCATION AND TO AND OX 261 THAN ~ 0% 702 90% OR 262 OR TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS OFFICE HOSPITAL BEFORE 1972 AFTER TO 251 TO 50% 502 TO 69% TO 89% TO 100% LESS TO S0X MORE BASE 805 233 231 108 177 45 533 232 441 97 266 367 229 "193 ~ 150 270 _ 364 500 143 107 MEAN NUMBER OF HOURS 53.2 54.5 49.5 52.6 57.1 54.8 49.7 61.8 48.1 52.6 62.2 50.7 53.7 58.0 56.3 53.5 51.8 51.6 52.9 62.3 STANDARD ERROR 0.6 1.2 1.0 1.7 1.6 2.6 0.6 1.5 0.6 1.6 1.3 0.8 1.2 1.5 1.7 1: 0.8 0.7 1.6 2.3 z8 dg xipuaddy :suoistoa(] are) yj[eay Sure TABLE 2-20 MEAN NUMBER OF HOSPITAL INPATIENTS: PHYSICIANS Q.: Approximately how many different hospital inpatients would you estimate that you see in an average week? PHYSICIANS PATIENTS WITH SPECIALTY SERIOUS ILLNESS GENERAL/ PRACTICE MORE INTERNAL FAMILY ALL LOCATION 0% 26% THAN TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS OFFICE HOSPITAL TO 25% TO 50% 50% BASE 805 233 231 108 177 45 533 232 367 229 193 MEAN NUMBER OF INPATIENTS 17.2 19.2 13.3 3.7 22.3 20.1 12.6 27.7 12.8 19.1 24.9 STANDARD ERROR 0.6 1.2 1.2 1-1 1.4 2.5 0.6 1.3 0.7 1.2 1.5 o1iqng dy} pue suerdisAyd jo sAoAINg P-5 TABLE 2-21 MEAN NUMBER OF HOSPITAL OUTPATIENTS: PHYSICIANS Q.: Approximately how many different patients would you estimate that you see on an outpatient basis during an average week? PHYSICIANS PATIENTS WITH SPECIALTY SERIOUS ILLNESS GENERAL/ PRACTICE MORE INTERNAL FAMILY ALL LOCATION 0% 26% THAN TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS OFFICE HOSPITAL TO 25% TO 50% 50% BASE 805 233 231 108 177 45 533 232 367 229 193 MEAN NUMBER OF OUTPATIENTS 71.9 53.4 102.2 88.0 48.1 65.9 88.0 34.9 85.4 71.4 44.9 STANDARD ERROR 1.9 2.6 4.4 4.4 2.6 3.3 2.3 2.2 2.8 3.9 2.7 v8 dg x1puaddy :suorstoa( aren yyeay Sunyey P-1,4,5 TABLE 2-22 MEAN NUMBER OF PATIENTS SEEN EACH HOUR OF DIRECT PATIENT CARE: PHYSICIANS Q.: Let me begin by asking you a few general questions about your practice. Including both inpatient and outpatient care, approximately how many hours do you spend on direct patient care activities in an average week? Q.: Approximately how many different hospital inpatients would you estimate that you see in an average week? Q.: Approximately how many different patients would you estimate that you see on an outpatient basis during an average week? PHYSICIANS PRACTICE SPECIALTY LOCATION GENERAL/ INTERNAL FAMILY TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS OFFICE HOSPITAL BASE 805 233 231 108 177 45 533 232 MEAN NUMBER OF PATIENTS SEEN EACH HOUR 1.8 1.5 2.4 2.0 1.3 1.7 2.1 1.1 STANDARD ERROR .04 .05 .09 .08 .05 .11 .05 .04 o1qngd 9Y} pue SUBIOISAYJ Jo sAdAING G8 TABLE 2-23 SATISFACTION WITH PHYSICIAN: GENERAL PUBLIC Q.: Now, I'd like to ask you some more general questions about health care. In light of your experience, how satisfied are you with (READ EACH ITEM)? Would you say you are very satisfied, somewhat satisfied, somewhat dissatisfied, or very dissatisfied? TOTAL GENERAL PUBLIC YOUR DOCTOR'S DOCTORS YOUR WILLINGNESS YOUR THE YOU AND DOCTOR'S TO ANSWER DOCTOR'S YOUR AMOUNT OF YOUR FAMILY RESPECT QUESTIONS HONESTY DOCTOR'S TIME YOUR DOCTORS HAVE USED FOR YOUR ABOUT YOUR IN HIS CONCERN DOCTOR IN IN THE LAST TREATMENT CONDITION DISCUSSIONS FOR YOUR SPENDS BASE: 1251 GENERAL FEW YEARS REFERENCES AND ILLNESS WITH YOU WELL-BEING WITH YOU 1 x I 2 z z z VERY SATISFIED 35 49 59 61 66 55 50 SOMEWHAT SATISFIED 41 35 29 25 23 30 29 SOMEWHAT DISSATISFIED 16 n 5 8 6 9 12 VERY DISSATISFIED 6 3 2 4 2 3 6 NOT SURE 2 2 5 2 2 2 3 98 SUOISTOa(] a1e) Yj[eay Suryey g x1ipuaddy Surveys of Physicians and the Public 87 CHAPTER 3: WHAT PHYSICIANS TELL PATIENTS: PATTERNS OF DISCLOSURE Introduction The first question about informed consent con- cerns how much doctors tell their patients about the patients' condition and treatment. The law requires a good deal more than disclosure of fact. The fun- damental issue in informed consent is whether or not the patient understood the risks of and alternatives to the medical procedure(s) before agreeing to them. Nonetheless, the information disclosure is the sine qua non for understanding. Little is known about what doctors tell their patients. One major research effort, conducted in 1976-77 by Arnold Rosoff, attempted to explore the frequency of physician disclosure of some key case and treatment information elements. However, the limitations of the sample and the response rate make any generalization from the Rosoff survey extremely tenuous. Perhaps even less is known about whether pa- tients find current disclosure patterns adequate or inadequate. This is a crucial issue since the law is relatively vague on the particular specifications for adequate disclosure. For instance, doctors are generally held accountable by the courts for not disclosing the risks of treatment to a patient be- fore undertaking a medical procedure. However, there is no general standard for the probability level at which a risk must be disclosed. If doctors are not responsible for disclosing remote risks of treatment, then the issue must be resolved as to what constitutes a remote risk -- 1 in 100, 1 in 1,000, 1 in 10,000 or 1 in 1,000,000? The following chapter provides a systematic as- sessment of the current extent and pattern of physi- cian disclosure of case and treatment information to patients. The survey of physicians provides the bulk of this information but certain key disclosures are also measured in the public survey. The public survey provides an assessment of the perceived ade- quacy of current disclosure practices, as well as a measurement of public norms concerning what physi- cians should be telling their patients. 88 Making Health Care Decisions: Appendix B What the Patient is Told: The Physician's View The extent of physician disclosure of case and treatment information to patients has never been analyzed on the basis of a nationwide study. The closest parallel is the 1976-77 survey of physician practices conducted by Arnold kosoff.? The sample design for the Rosoff study, however, included only certain medical specialties (internal medicine and surgery) and certain purposefully selected states (Pennsylvania, New Jersey, Arizona, Delaware, Kan- sas, California, Rhode Island, and the District of Columbia). Moreover, the Rosoff survey experienced a response rate of 247%. Hence, Rosoff readily ad- mits that "no claim is made that the sample thus represents the 'typical American physician.'" The Rosoff study, nonetheless, provides a good starting point for the analysis of physician disclo- sure patterns. The 1976-77 survey asked physicians about the frequency with which they discussed twenty-one areas of treatment information with their patients. Some of these areas are regarded by law as mandatory subjects for disclosure. The results of the Rosoff survey indicated that ‘physicians in all categories report a high level of disclosure on information items that the law would regard as man- datory subjects for disclosure." This approach -- with some significant varia- tions -- was replicated in the 1982 Louis Harris and Associates survey of physicians. The first impor- tant variation from the 1976-77 study was a split- format investigation of the nature of physicians’ discussions with patients. Half of the sample of doctors were asked the question in a manner very similar to the Rosoff question: '"In your practice, how often do you discuss each of the following things with the patient -- always, usually, some- times, rarely, or never?" The other half of the physician sample was asked: "In your practice, do you initiate discussion about each of the following things with the patient as a matter of course, or discuss it only if the patient asks about it?" A second important variation from the Rosoff approach was providing a set of response categories 5 tniorned Consent, A Guide for Health Care Provid- ers, Arnold J. Rosoff, Rockville, Maryland; Aspen Systems Corporation, 1981. Surveys of Physicians and the Public 89 that do not force respondents toward one or the oth- er end of the response scale. The Rosoff question required doctors to describe the discussions as -- always, usually, rarely, or never. The current study provides the categories always, usually, some- times, rarely, or never. Finally, although there is substantial overlap between the issue areas being probed in the two surveys, there is also substantial divergence. Some of the Rosoff categories were merged or dropped. Some new categories were added. Despite the differences between the methodology of the 1976 and 1982 surveys, we find a common con-— clusion. A very high level of disclosure is report- ed by physicians on most items that the law would regard as mandatory subjects of disclosure (Table 3-1). Following Rosoff's convention of treating items that are "always" or '"usually" disclosed as "commonly disclosed," we find that 98% of physicians commonly disclose diagnosis and prognosis of a pa- tient's condition (98% in Rosoff); 96% of physicians commonly disclose’ the nature or purpose of the treatment option they recommend (99% in Rosoff); and 93% of physicians commonly disclose side effects that are certain or fairly likely to attend the treatment (87% in Rosoff). The Rosoff study found that the vast majority of doctors (91%) commonly disclosed the risks generally recognized as attending the recommended treatment. In the 1982 survey, we asked the physician sample about the frequency of disclosure of three types of treatment risks, which varied by probability and seriousness of outcome. The 1982 survey finds that 82% of physicians commonly disclose risks of death or serious disability that are about 1 in 100; 59% of physicians commonly disclose risks of death or serious disability that are about 1 in 1,000; and 57% of physicians commonly disclose risks of tem= porary disability that are about 1 in 1,000. When the question is phrased in terms of initiating dis- cussion, 81% of physicians respond that as a matter of course they initiate discussion with their pa- tients about risks of death or serious disability that are about 1 in 100 (Table 3-2); 52% initiate discussion as a matter of course about risks of death or serious disability that are about 1 in 1,000; and 43% initiate discussion as a matter of course about risks of temporary disability that are about 1 in 1,000. 90 Making Health Care Decisions: Appendix B The survey findings clearly indicate that most doctors would usually inform a patient of a treat- ment risk that involves death or serious disability in 1 out of 100 cases; at the same time, the survey also finds that 1 out of 5 doctors would not. When the risk of death or serious disability from treat- ment declines to 1 out of 1,000 cases, little more than half of Practicing physicians would initiate discussions with their patient about it as a matter of course (Table 3-2). Less than half of all doc- tors would initiate a discussion with their patient about a treatment risk of temporary disability that was 1 in 1,000. Since the courts have steadfastly avoided setting a probability level that requires disclosure of risk, it is impossible to conclude whether or not these findings mean that doctors are generally in compliance with risk disclosure stan- dards or not. However, it is notable that a court decision in one state found that any risk of death, no matter how slight, must be disclosed. In another court decision, a physician was held accountable for not disclosing a risk of injury that occurred in less than 1 out of 1,000 cases. Outside of the areas of information disclosure mandated by law, we find no real decline in the fre- quency with which physicians commonly disclose treatment information to their patients or initiate discussions with their patients about the matter. Indeed, although there is clear variation by infor- mation area in the probability that a physician will usually discuss it with patients as a matter of course, we find that a majority of physicians report commonly disclosing thirteen of the fourteen issue areas investigated. The one issue about which most physicians do not initiate discussions with their patients as a matter of course concerns the costs associated with recom- mended procedure and whether these costs would be covered by the patient's health insurance. The sur- vey finds that 38% of doctors say that they initiate discussion of the costs of the treatment with the patient as a matter of course, while 56% say that they discuss it only if the patient asks about it. Similarly, 47% of physicians say that they always (15%) or usually (32%) discuss costs of treatment with their patients, while 34% say they discuss it sometimes, 15% discuss it rarely, and 4% never dis- cuss it (Table 3-1). (It is noteworthy that the 47% of physicians who report commonly initiating dis- Surveys of Physicians and the Public 91 cussion of costs of treatment (Table 3-1) is indistinguishable from the 46% found in the Rosoff study (Table 3-3)). In other areas, we find that the frequency with which physicians discuss treatment and case issues with their patients is surprisingly high (Table 3-1). The survey finds that 84% of doctors commonly disclose the pros and cons of the treatment they recommend compared to other alternatives. Three- quarters or more (77%) commonly discuss any uncer- tainties about whether the recommended treatment will be successful and what the different views in the profession are about the best course of treat- ment (75%). Sixty-two percent of doctors commonly discuss uncertainties in their own mind about the best course of treatment with the patient. One of the most interesting findings is the high level of physician disclosure of the probable impact of the condition and treatment on the patient's life (Table 3-1). The probable impact of the condition and treatment on the patient's job role, where there is a likely impact, is commonly discussed by 86% of doctors. The probable impact of the condition and treatment on the patient's family life is commonly disclosed by 78% of doctors. It is noteworthy that these survey findings are much higher than the Rosoff study, which found disclosure rates of 697% and 47%, respectively (Table 3-3). The survey findings describe a medical pro- fession where the common pattern is physician initiation of discussion with the patient of medical issues that are related to the patient's condition and treatment (Table 3-2). Less than 5% of physi- cians do not usually disclose diagnosis, prognosis, nature and purpose of recommended treatment, and side effects likely to attend treatment. The vast majority of physicians discuss with the patient their uncertainties about the patient's condition or the recommended treatment. The only real question mark concerning physician disclosure has to do with risk. It is clear that physicians' disclosure of risks of treatment are related to the seriousness of the consequences and the probability of those con- sequences occurring. Physicians do not necessar- ily disclose all risks of death or serious dis- ability from treatment, particularly if they are not more than 0.1%. 92 Making Health Care Decisions: Appendix B What the Patient is Told: The Patient's View The physicians sample does not provide an un- biased method for measuring the frequency of dis- closure of case and treatment information to pa- tients. The norm and legal principle of full dis- closure have been widely disseminated. Doctors have a self-interest in reporting high levels of informa- tion disclosure to patients. Therefore, our national sample of the adult pop- ulation of the United States was asked about the frequency with which their doctors discuss certain information with them. The patients' report of the frequency of disclosure provides an important coun- terweight to the doctors’ report. It should be not- ed, however, that the patients' report is no more intrinsically valid than is the doctors" report. Countless observational studies have demonstrated that doctors think that they have disclosed informa- tion that was never discussed, while patients claim not to have been informed of matters that observers have recorded. Perceptual biases work both ways. So, the frequency of disclosure from the patients’ report provides another perspective on the flow of doctor-patient communications, but not necessarily a better one. Generally speaking, the proportion of doctors who report disclosing case and treatment information to their patients is greater than the proportion of the general public who report that their doctors discuss these matters with them (Table 3-4). In most instances, there is a large and reliable dif- ference of appproximately 15 to 25 percentage points between the proportion of doctors who report com- monly discussing the issue with their patients and the proportion of the general public who report that their doctor commonly discusses these matters with them. The survey finds that while 98% of doctors say they commonly discuss the diagnosis and prog- nosis with their patients, only 78% of the general public report that their doctor commonly explains their diagnosis and prognosis to them. Similarly, although 96% of doctors report commonly discussing the nature and purpose of the treatment they recom- mend with their patients, only 78% of patients re- port that their doctor commonly explains this to them. While 84% of physicians report that they com- monly discuss the pros and cons of the treatment they recommend with their patients, only 67% of the pub- Surveys of Physicians and the Public 93 lic report that their doctor commonly explains this to them. The greatest difference between doctor and pa~ tient perceptions occurs in the area of side ef- fects. The survey finds that 93% of doctors report that they commonly discuss side effects, including inconvenience and pain, that are certain or fairly likely to attend the treatment. However, only 68% of the public report that their doctor commonly ex— plains to them the side effects likely to attend their treatment. This represents a 25 percentage point difference in perception. By contrast, doctors and the general public are in agreement on the frequency with which costs and insurance coverage are disclosed. The survey finds that 47% of doctors say that they commonly discuss with their patients the costs associated with the recommended procedure and whether those costs would be covered by the patient's health insurance. A virtually identical 45% of the public agrees that their doctor commonly explains the cost of treatment and their insurance coverage. The striking agreement between doctors and the general public on the disclosure of cost informa- tion, compared to the striking disagreement in other areas of medical disclosure, should stimulate some interesting speculation. Certainly professional norms and legal requirements are least likely to inflate physician reports in this area, since there is no obligation to routinely discuss these matters with patients. There is certainly no reason for the public to overstate the frequency with which their doctor told them how much their medical treatment would cost them. However the differences are explained for these various items, it seems reasonable to accept the patients' estimate of the frequency of disclosure as the lower bounds for disclosure and the physician estimate as the upper bounds. Despite the magnitude of the difference in the upper and lower bounds of the estimated frequency of disclosure of important case and treatment information, one vital fact stands out. Almost 8 out of 10 patients report that their doctor usually explains the diagnosis and prognosis of their condition to them, along with the nature and purpose of the treatment the doctor rec— ommends. Nearly 7 out of 10 patients report that 94 Making Health Care Decisions: Appendix B their doctor explains the side effects likely to attend the recommended treatment, as well as the pros and cons of alternative treatments. This strongly suggests that one of the opera- tional goals of informed consent is being achieved -- the vast majority of doctors are explaining the major issues of condition and treatment to the vast majority of their patients. The patients who report that their doctor does not usually discuss these matters with them and the doctors who report that they do not usually discuss these matters with their patients are exceptions. Of course, this does not mean that no problem exists. The survey findings, however, clearly establish the limits of the problem. Patient Preferences and Physician Practice of Disclosure Shab eite of Jisc Osure Whereas the sampled physicians were asked wheth- er or not they initiated discussions with their pa- tients on selected issues of case and treatment, the public respondents were asked whether or not they thought doctors should tell their patients about each of these things as a matter of course, or only if the patient asked. A comparison of the propor- tions of the public who think that the doctor should disclose particular types of information with the proportion of doctors who report disclosing that information serves two important purposes. At min- imum, for those who believe that physician reports of their behavior tend to reflect physician norms more than physician practice, the comparison pro- vides an opportunity to compare public and physician norms of disclosure. At maximum, for those who ac- cept physician reports of their disclosure behavior at face value, it provides an opportunity to assess how well physician practices live up to public ex- pectations. The comparison of what the public says that it wants the doctor to discuss and what doctors say that they do discuss with their patients yields a substantial amount of agreement (Table 3-5). The proportion of doctors who say that they usually dis- cuss as a matter of course diagnosis and prognosis of condition, nature and purpose of recommended treatment, side effects likely to attend treatment, and pros and cons of alternative treatments exceed the proportion of the public sample that think that Surveys of Physicians and the Public 95 the doctor should be expected to initiate dis- cussions of these issues. Thus, doctors' norms and possible behavior seem to meet adequately public expectations in these areas. The issue of risk disclosure for different prob- abilities of risk yields an interesting comparison. For risks of death and serious disability that occur in 1 of 100 cases, the proportion of doctors who report initiating discussions with their patients (81%) is greater than the proportion of the public who think that the doctor should initiate such a discussion (75%). However, on more remote risks of treatment, patient expectations exceed physician behavior. For risks of death or serious disability that occur in 1 of 1,000 cases, 64% of the general public think that the doctor should initiate dis- cussion with the patient while only 52% of physi- cians report initiating discussions under those cir- cumstances. When there is a risk of temporary dis- ability of 1 in 1,000, 65% of the public think that their doctor should initiate disclosure; only 43% of doctors say they do initiate a discussion under these circumstances. Two things are notable in these comparisons of preferences and practices related to treatment risks. First, fairly substantial differences exist between doctors and the public on the disclosure of remote risks. In essence, it would appear that a larger segment of the general public feel that they should be informed of relatively remote risks, with- out having to ask than may currently be receiving such information (based on the proportion of physi- cians reporting disclosure). Second, the source of this imbalance between public expectation and physi- cian behavior appears to lie in different attitudes toward risk in the two populations. The likelihood of physician disclosure of risk increases with the probability of occurrence and seriousness of con- sequences: 43% of doctors initiate discussion with their patients about a 1 in 1,000 risk of temporary disability; 52% initiate discussion about a 1 in 1,000 risk of death or serious disability; and 81% initiate discussion about a 1 in 100 risk of death or serious disability. By contrast, there is rel- atively little difference in the proportion of the public who think that they should be told about a 1 in 100 risk of death or serious disability (75%) and a 1 in 1,000 risk of temporary disability (65%). There is no difference in the proportion of the pub- 96 Making Health Care Decisions: Appendix B lic who want to be told about a risk of death or se- rious disability (64%) or temporary disability (65%) that is 1 in 1,000. Observation: These findings help to explain why courts have found against doctors for failure to disclose fairly remote risks. Doctors discriminate be- tween treatment risks on the basis of their probability and their consequences. Patients, on the other hand, are much less likely to make such discriminations. A fairly substantial maj- ority of the public (65%) have said that they think doctors should initiate discussions with their patients about treatment risks of tem- porary disability, even if the probability is 1 in 1,000. In other areas of information disclosure, the current practice of disclosure by doctors tends to meet or exceed public expectations in these areas. One exception, however, is the cost of the procedure and the patient's insurance coverage. Only 38% of physicians say that they initiate discussion of treatment costs with their patients as a matter of course, while 69% of the public think their doctors should initiate such discussions. Predictors of Disclosure The pattern of information disclosure in the therapeutic milieu was explored further to sta- tistically determine the best predictors of physi- cians' reported disclosure practices, as well as those predictors that best distinguish among the general public regarding their perceptions of how much doctors should tell. Initially, a simple additive disclosure index was created for each of the two samples surveyed. Here, items with regard to disclosure of information about condition and treatment, treatment risk, phy- sician uncertainty, and possible treatment impacts were summed (survey items 13, 14, 15 -- physicians, and 25, 26, 27 -- public). Indices of different versions of the questions (i.e., frequency of dis- cussion vs. initiate discussion in the physician sample) were calculated separately. In the public sample this index was created for only one-half of Surveys of Physicians and the Public 97 the respondents -- those asked how frequently physi- cians should initiate relevant discussions. Re- sponse scales on different versions of the physician questionnaire were made compatible. Subscale inter- correlations revealed a very high consistency of responses regarding the perceptions of disclosure. That is, physicians who say they initiate discussion or always or usually tell patients about treatment risk also say they initiate discussion or always or usually disclose information regarding the potential impact of treatment. Thus, a single additive index, weighted as a function of the number of the individ- ual's responses to the items, was calculated. Principal component analyses were performed to determine the factor structure of sample descrip- tors/predictors. Among physicians, four factors emerged. Two factors were most heavily weighted by patient load characteristics. The first was com posed primarily of the percent of the old and ser- iously ill, while the second was weighted by the percent of college educated and poor patients a phy- sician treats. The third factor weighted character- istics of physicians' training experiences, includ- ing coursework in medical ethics, law, and doctor- patient communication. The final factor involved the location of physicians' practice (hospital or private office) and the recency of training. Among the public, six factors emerged. Three were composed primarily of single variables -- race, age, and experience with chronic illness. The other three factors were most heavily weighted by self- rated health status and frequency of physician vis- its (health status); education and income (socio- economic status); and respondent sex and location of care. Constructed factors and other potential explana- tors were then employed in stepwise regressions on each of the two forms of the created disclosure in- dices for the physician sample and the singular in- dex in the public sample. Because the overlap of the results of the regression analyses was substan- tial in the physician sample, further multiple re- gression analyses employed the overall additive mea- sure of disclosure. Three significant predictors (p < .05) of physi- cians' reported disclosure emerge. The primary de- terminant of how much a physician disclosed is the 98 Making Health Care Decisions: Appendix B percent of patients he or she perceived as being capable of understanding their condition or treat- ment when reasonable time and effort are devoted to explanation (Table 3-6). As the proportion of pa- tients able to understand case and treatment issues increased, so too did the incidence of physician disclosure. This finding is consistent with biva- riate analyses reported earlier. The second predictor to enter the equation is more surprising -- the educational attainment of the physician's father. The greater the educational attainment of the father, the greater is the pre- disposition of the physician to disclose all rel- evant treatment information. It is interesting to note that even within the restricted range of socio- economic characteristics of physicians, background factors still seem to exert some influence over at- titudes and behavior. The last significant predictor to enter the re- gression equation is one of the constructed practice characteristic factors. The analysis reveals that as the proportion of college educated patients in- creases and the proportion of poor patients de- creases in a practice, physicians become more apt to indicate that they disclose more information. This again highlights that the primary determinant of how much a physician discusses depends on the perceived capability of the patient to understand. Education may be thought of as an immediately transferable proxy of perceived ability to understand. Thus, high education (college graduate) should be posi- tively related to perceived ability to understand, while low education (poverty as a surrogate) should be negatively related. Although these three variables are statistically significant predictors of physician disclosure, they explain only a limited amount of this behavior. Only 6% of the variance between doctors in dis- closure is explained by the final regression equa- tion. Among the public, ill health (i.e., poor self- health ratings and increasingly high number of phy- sician visits), increasing amounts of time hospital- ized, those viewing the physician as the best judge of how much information should be disclosed, those not having first-hand experience with a life- threatening illness, and minority group status are Surveys of Physicians and the Public 99 each significantly related to reports that doctors should tell patients less about their condition and treatment (Table 3-7). Again, we see that indica- tors of chronic dependence on medical services -- ill health and time hospitalized are significantly related to the belief that doctors should disclose less information. Alternately, the experience of a life-threatening illness -- a situation where need may be acute -- is related to the belief that physi- cians should disclose more. Note that while actual disclosure by physicians certainly seems to be a function of education/understanding, among the pub- lic, the perceptions of how much a physician should disclose is not dramatically affected by education (i.e., the constructed socioeconomic status factor). The patterns of patient characteristics that emerge from this multivariate analysis as a pre- dictor of disclosure are consistent with the find- ings of earlier analyses. However, it should be noted that the final regression equation explains only 11% of the variance in patient expectations about whether physicians should disclose treatment information to their patients. Public Satisfaction with Medical Disclosure A majority of the American public are satisfied with how much doctors tell their patients, both about matters of routine care and in cases of ser- ious illness. In matters of routine care, 56% of the public say that doctors generally tell their pa- tients about the right amount (Table 3-8). In cases of serious illness, 53% of the public feel that doc- tors generally tell their patients about the right amount (Table 3-9). Despite majority satisfaction with the extent of physician disclosure concerning medical condition and treatment, a full one-third of the public (33%) say that they think doctors generally tell their pa- tients too little in cases of serious illness. In matters of routine care, the proportion of the pub- lic who think that doctors tell their patients too little increases to 38%. In almost no cases (2%) does the public feel that doctors generally tell their patients too much. Neither current health status nor past ex-~ perience with a life-threatening illness seems to have a major effect on perceptions of the adequacy 100 Making Health Care Decisions: Appendix B of physician disclosures. Those who have experienced a life-threatening illness in the family are slight- ly more likely to feel that doctors tend to disclose too little in cases of serious illness. By con- trast, those who are currently in poor health are most likely to think doctors generally tell too much, particularly in matters of routine care. However, there is no strong or consistent relationship be- tween health experience and attitudes about the adequacy of the amount of information doctors gen-— erally disclose to their patients. Age and education, on the other hand, have a strong relationship with public perceptions of the adequacy of present disclosure by doctors. The per- ception that doctors generally tell their patients too little in cases of serious illness rises from 27% among those without a high school degree to 38% among college graduates. More dramatically, the proportion of the sample who think doctors generally tell their patients too little in matters of routine care rises from 28% of those without a high school diploma to 45% among college graduates. Observation: Critical attitudes toward the extent of physi- cian disclosure of medical information to their patients rises directly with education. This may represent a greater and less readily sat- isfied demand for disclosure of and participa- tion in medical treatment among the better edu- cated. Or it may represent a more trusting and dependent relationship between the doctor and the less educated patient. It was noted earlier that the public tends to be more critical of doctors, in general, than they are critical of their own doctors. This appears to be the case in the area of the disclosure of medical information as well. The national sample of Amer- icans was asked: "How would you rate your doctor on keeping you fully informed about questions and decisions relating to your medical care that you consider important?" More than two-thirds of the public say that their doctor keeps them informed on all (37%) or most (30%) of the issues related to their medical care that they consider important (Table 3-10). By comparison, 18% of the public say that their doctor keeps them informed on some issues Surveys of Physicians and the Public 101 that they consider important, 6% say that they are kept informed on few issues, and 3% say that their doctor keeps them informed on none of the important issues related to their medical care. If adequate disclosure is defined as being kept informed on most of the important issues related to patient care, then the public is somewhat more satisfied with the amount of information disclosed by their own doctor (67%) than by doctors in general (53%). It should be noted, however, that full dis- closure, according to the average reasonable patient standard or particular patient standard, would pre- sumably require physicians to disclose all medical information considered important to the patient. By this standard, less than 4 out of 10 Americans feel they are receiving adequate medical disclosures. Unfortunately, the issue of what the public con- siders to be adequate disclosure -- all or most im- portant issues -- is not investigated in this study. The relationship between education and satisfac- tion with the extent of physician disclosure is found in attitudes toward the particular physician, as well as physicians in general. Nearly one-half of those without a high school degree (48%) say that their doctor keeps them informed about all important medical issues. By contrast, less than one-third of college graduates (33%) say that their doctor keeps them informed about all important medical issues. More than one-quarter of college graduates (26%) say that they are informed of only some of the medical issues they consider important, compared to 11% of those without a high school degree. How dissatisfied is the public with present dis- closure by doctors? How important is increased dis- closure to the public? One way of measuring the personal importance of a particular behavior or con- dition to the public is to ask whether or not re- spondents would actually pay to change the situa- tion. In this case, respondents were asked: "If your doctor spent more time explaining things to you about your routine care, would you be willing to pay more?" Nearly 4 out of 10 (38%) of those surveyed re- port that they would be willing to pay more if their doctor spent more time explaining things to them about their routine care (Table 3-11). These are not the individuals with the largest discretionary 102 Making Health Care Decisions: Appendix B budgets. Quite the contrary -- 44% of those with a family income of $7,500 or less say that they would be willing to pay more for more explanations -- a higher proportion than in any other income group. Presumably, those who are willing to pay more for more time with the physician are also those who are most dissatisfied with the amount of time the physician spends with them. However, the most fre- quent critics of the adequacy of physician dis- closure -- particularly with regard to routine care —= are also the least willing to pay for more time and information. Among college graduates, fully 602 say that they would not be willing to pay for great- er disclosure by the physician, compared with the 47% of those without a high school diploma who say that they are not willing to pay. This may suggest that the real concern about the adequacy of physi- cian disclosure may be greater among the least edu- cated and the poorer segments of the population, even though criticism of physician disclosure is less widespread among these groups than among the more educated and the wealthier segments of the population. Responsibility for Full Disclosure The doctrine of informed consent establishes the right of patients to be "adequately" informed by their doctors before agreeing to medical proce- dures. The boundaries of the legal right to full disclosure have been expanded in the last two dec- ades. It has been argued that the legal changes have been responsive to changing public attitudes toward patient rights. It is certainly clear that the belief that the patient has a right to full disclosure is virtually universal among the adult public of the United States in 1982. The national sample was asked: "Do you think that patients should have the right to all available information about their condition and treatment if they wish to have it?" Fully 97% of the American public report that the patient should have the right to full disclosure. Less than 2% of the adult population feel that the patient should not have that right. Although the public clearly believes that pa- tients have a right to be fully informed about their Surveys of Physicians and the Public 103 condition and treatment if they so wish, the respon~ sibility for such disclosure is not established by this question alone. Hence, the public sample was asked whose responsibility it is to make sure the patients are fully informed about their condition and treatment -- the doctors' or the patients’. Two- thirds of the public (66%) report that they feel it is primarily the doctors’ responsibility to make sure that the patients are fully informed (Table 3-12). By contrast, only 20% feel that it is primarily the patients' responsibility to make sure that they are fully informed about their condition and treatment. Another 12% of the adult population volunteer the opinion that doctors and patients should be equally responsible for ensuring that the patients are fully informed. In other words, 78% of the public feel that the doctors' responsibility is at least as great as the patients' responsibility for ensuring full disclosure, while 66% feel that it is primarily the doctors’ responsibility. It is important to note that when doctors are asked the same question, the attitude that it is the doctors' responsibility to ensure that the patients are fully informed is found to be more widespread among physicians than it is among the public. More than three-quarters of physicians (77%) say that they feel it is primarily the doctors' responsibil- ity to ensure that the patients are fully informed about their condition and treatment (Table 3-13). Another 19% of doctors volunteer that the doctors and patients are equally responsible for ensuring that the patients are fully informed. Less than 3% of practicing physicians in the United States think that it is primarily the patients’ responsibility to see that they are fully informed about their condi- tion and treatment. The perception of physicians’ responsibility for ensuring that patients are fully informed is most widespread among the medical specialties with the greatest number of seriously ill patients -- inter- nal medicine (79%) and surgery (80%) -- while the perception is least widespread in Ob/Gyn (73%), which also has the fewest seriously ill patients. However, the perception of physician responsibility is not directly tied to the incidence of seriously ill patients. The perception that ensuring that patients are fully informed is primarily the doctors' respon 104 Making Health Care Decisions: Appendix B sibility increases with the proportion of patients that the doctors feel can understand most aspects of their treatment and condition. Among physicians who report that less than 70% of their patients are able to understand most aspects of their treatment and condition if reasonable time and effort are devoted to explanation, 71% feel that it is primarily the doctors' responsibility to make sure patients are fully informed about their treatment. By contrast, among physicians who report that 90% to 100% of their patients are able to understand their condi- tion and treatment, 81% feel that making sure pa- tients are fully informed is primarily the doctors’ responsibility. In other words, doctors are more likely to accept responsibility for making sure that their patients are fully informed if they feel pa- tients are able to understand their condition and treatment, given reasonable efforts of explanation. Among the public, the perception that being ful- ly informed is primarily the responsibility of the patients is inversely related to education. Those without a high school degree are more likely to see full disclosure as the patients’ responsibility (26%) than are high school graduates (21%) or col- lege graduates (15%). Most dramatic, however, is the difference between whites (18%) and minorities (31%) in the perception that patients are primarily responsible for being fully informed about their condition and treatment. Surveys of Physicians and the Public 105 P-13,14,15 TABLE 3-1 HOW OFTEN PHYSICIANS DISCUSS TREATMENT ISSUES WITH PATIENTS: PHYSICIANS Q.: There are a variety of issues that doctors may or may not discuss with their patients regarding the patient's condition and possible treatment. In your practice, how often do you discuss each of the following things with the patient -- always, usually, sometimes, rarely, or never? Q.: Another set of issues that may or may not be discussed with patients involves the risks of adverse consequences associated with treatment. In your practice, how often do you discuss each of the following things with the patient -- always, usually, sometimes, rarely, or never? Q.: Often there are significant uncertainties associated with medical decisions. When such uncertainties exist, how often do you discuss (READ EACH ITEM) with the patient -- always, usually, sometimes, rarely, or never? PHYSICIANS SOME - NOT BASE: 404 ALWAYS USUALLY TIMES RARELY NEVER SURE NATURE AND PURPOSE OF TREATMENT OPTION YOU RECOMMEND 62 34 4 - - - DIAGNOSIS AND PROGNOS1S OF PATIENT'S CONDITION 2 56 42 2 * - - SIDE EFFECTS THAT ARE LIKELY TO ATTEND TREATMENT 1 53 40 6 1 - * RISKS OF DEATH OR SERIOUS DISABILITY (1 IN 100) 2 52 30 9 6 1 2 PROS AND CONS OF THE TREATMENT YOU RECOMMEND COMPARED TO ALTERNATIVES 2 38 46 13 3 “ 1 UNCERTAINTIES OR RESERVATIONS ABOUT DIAGNOSIS 2 37 47 13 3 - PROBABLE IMPACT OF CONDITION AND TREATMENT ON PATIENT'S JOB ROLE 2 36 50 11 2 1 » UNCERTAINTIES ABOUT WHETHER THE RECOMMENDED TREATMENT WILL BE SUCCESSFUL 2 34 43 16 4 2 - PROBABLE IMPACT OF CONDITION AND TREATMENT ON PATIENT'S FAMILY LIFE 2 32 46 18 4 » 1 RISKS OF DEATH OR SERIOUS DISABILITY (1 IN 1,000) 30 29 22 13 4 2 RISKS OF TEMPORARY DISABILITY (1 IN 1,000) 25 32 23 15 2 2 DIFFERENT VIEWS WITHIN THE PROFESSION ABOUT THE BEST COURSE OF TREATMENT 2 23 52 19 5 1 1 UNCERTAINTIES ABOUT THE BEST COURSE OF TREATMENT 1 22 40 26 9 3 1 COSTS ASSOCIATED WITH THE RECOMMENDED PROCEDURE 1 15 32 34 15 4 1 *LESS THAN 0.51. NOTE: SPLIT-BALLOT ITEM, ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. 106 Making Health Care Decisions: Appendix B P-13,14,15 TABLE 3-2 WHETHER PHYSICIANS INITIATE DISCUSSION OF TREATMENT ISSUES WITH THEIR PATIENTS: PHYSICIANS Q.: There are a variety of issues that doctors may or may not discuss with their patients regarding the patient's condition and possible treatment. In your practice, do you initiate discussion about each of the following things with the patients as a matter of course, or discuss it only if the patient asks about it? Q.: Another set of issues that may or may not be discussed with patients involves the risks of adverse consequences associated with treatment. In your practice, do you initiate discussion about each of the following things with the patient as a matter of course, or discuss it only if the patient asks about it? Q.: Often there are significant uncertainties associated with medical decisions. When such uncertainties exist, do you initiate discussion about it with the patient as a matter of course, or discuss it only if the patient asks about it? PHYSICIANS DISCUSS INITIATE ONLY NEVER BASE: 404 DISCUSSION IF ASKED DISCUSS NOT SURE NATURE AND PURPOSE OF TREATMENT OPTION YOU RECOMMEND 2 96 3 - hd DIAGNOSIS AND PROGNOSIS OF PATIENT'S CONDITION 2 99 1 - * SIDE EFFECTS THAT ARE LIKELY TO ATTEND TREATMENT 2 95 4 1 1 RISKS OF DEATH OR SERIOUS DISABILITY (1 IN 100) 2 81 16 1 3 PROS AND CONS OF THE TREATMENT YOU RECOMMEND COMPARED TO ALTERNATIVES: 2 84 14 1 1 UNCERTAINTIES OR RESERVATIONS ABOUT DIAGNOSIS z 90 8 1 1 PROBABLE IMPACT OF CONDITION AND TREATMENT ON PATIENT'S JOB ROLE 2 79 20 1 1 UNCERTAINTIES ABOUT WHETHER THE RECOMMENDED TREATMENT WILL BE SUCCESSFUL z 80 16 2 2 PROBABLE IMPACT OF CONDITION AND TREATMENT ON PATIENT'S FAMILY LIFE z 74 23 1 2 RISKS OF DEATH OR SERIOUS DISABILITY (1 IN 1,000) 2 52 40 S$ 4 RISKS OF TEMPORARY DISABILITY (1 IN 1,000) 2 43 48 4 4 DIFFERENT VIEWS WITHIN THE PROFESSION ABOUT THE BEST COURSE OF TREATMENT 2 75 22 2 1 UNCERTAINTIES ABOUT THE BEST COURSE OF TREATMENT 2 66 27 4 4 COSTS ASSOCIATED WITH THE RECOMMENDED PROCEDURE 2 38 56 4 2 *LESS THAN 0.52. NOTE: SPLIT-BALLOT ITEM, ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. Surveys of Physicians and the Public 107 TABLE 3-3 DISCLOSURE OF VARIOUS INFORMATION ITEMS: PHYSICIANS (ROSOFF, 1981) PHYSICIANS ALL ALL OLD Rif Nim RCLE COMMONLY PHYSICIANS INTERNISTS INTERNISTS INTERNISTS § DISI.OSEY _ N=BOO Ne321 N=120 N=198 HEE pee A. DIAGNOSIS® YES 98 100 100 100 100 100 100 NO 2 - - - - - - B. UNCERTAINTIES REDIAGNOSIS YES 91 9% 99 9% 93 92 9- no 9 “ 2 1 7 8 s C. NATURE OF TREATMENT® YES 99 99 101 99 99 95 9: no 1 - - 1 1 s 2 D. PURPOSE CF TREATMENT® YES 99 99 100 99 100 100 100 wo 1 - - 1 - - - E. RISKS OF TREATYENT® YES 91 93 93 9 92 93 91 no 8 6 9 6 7 1 9 F. NEGATIVE CONSEQUENCES® YES 87 9 9 90 8s 8 8 nN 12 8 6 n 13 17 1 G. BINEFITS EXPECTED® YES 98 99 99 9 98 99 98 NO 2 1 1 2 1 1 2 H. PROBABILITY OF SUCCESS® YES 90 0 89 8 92 9: 94 no 9 n 1 n s 6 s 1. PERSONAL CLINICAL YES 66 66 $7 2 64 6 6s EXPERIENCE No 33 3 “ 28 3 3s 33 J. COSTS OF TREATMENT YES “tb bb 38 “8 “1 el 3) no sa $5 63 50 6 58 5? K. WHAT HEALTH YES » 2 26 36 3- 28 «2 INSURANCE COVERS No 62 67 % $3 63 7”? $7 L. ALTERNATIVE YES 8 a 8 8s 85 87 90 TREATMENTS POSSIBLE® wo 15 15 1 15 1n 13 9 MW. RISKS OF ALTERNATIVE YES 81 8 81 0 8 9 87 MODES * no 17 17 20 16 16 2 1 N. BENZFITS EXPECTED FROM YES 79 80 79 [4 79 7% 82 ALTERRATIVES® wo 20 20 21 20 20 2% 1 0. PROBABILITY OF SUCCESS YES 73 74 Te 7% 7 Te 7 OF ALTERKATIVES® "0 25 25 28 33 28 25 25 P. PROS AND CONS OF ALTERNATIVES AND YES 8 [BD os 8 79 9 8 PROPOSED TREATMENT »o 16 1“ 15 le 19 20 18 Q. COSTS OF ALTERNATIVES YES 29 3 2% 3» 20 20 2 "0 70 3 7% 62 7 7 7” R. WHAT HEALTH INSURANCE YES 21 1 15 25 le 12 16 COVERS OF ALTERNATIVES wo ” 10 8 7 ~“ 87 80 §. PROBABLE IMPACT ON YES “1 $3 “9 4 “0 «0 «0 FAMILY wo 51 “tb $0 a 8 59 6 T. PROBABLE IMPACT ON JOB Yes 69 7 70 3 73 7% 70 "0 30 28 30 26 27 26 26 U. PROGNOSIS IF NO TREATMENT® YES 92 9 90 92 95 9% 93 0 . . 10 ? 4 3 6 TTY TTT SMANDATORY DISCLOSURE UNDER LAW. 108 Making Health Care Decisions: Appendix B G-13,14,25 P-13,14,25 TABLE 3-4 FREQUENCY OF DISCUSSIONS WITH PATIENTS: GENERAL PUBLIC, PHYSICIANS Physicians Q.: There are a variety of issues that doctors may or may not discuss with their patients regarding the patient's condition and possible treatment. In your practice, how often do you discuss each of the following things vith the patient -- always, usually, sometimes, rarely, or never? Physicians Q.: Another set of issues that may or may not be discussed with patients involves the risks of adverse consequences associated with treatment. In your practice, how often do you discuss each of the following things with the patient -- always, usually, sometimes, rarely, or never? General Public Q.: Do you usually come avay from your doctor feeling that you have understood the important issues relating to your treatment -- fully, more than adequately, or less than adequately? General Public Q.: When a patient doesn't understand his medical treatment, how often is it because the doctor did not explain things well? Would you say always, often, sometimes, rarely, or never? Physicians Q.: In your practice, how often do you discuss each of the following things with the patient -- always, usually, sometimes, rarely, or oever? General Public Q.: There are a variety of issues that doctors may or may not discuss with their patients regarding the patient's condition and possible treatment. Do you feel that doctors should initiate discussion about each of the following things with the patient as a matter of course or discuss it only if the patient asks about it? SOME - BASE: 626 GENERAL PUBLIC; ALWAYS USUALLY TIMES RARELY NEVER NOT SURE 404 PHYSICIANS. DIAGNOSIS AND PROGNOSIS GENERAL PUBLIC 2 52 26 13 5 2 3 PHYSICIANS 2 56 42 2 » - - COSTS AND INSURANCE GENERAL PUBLIC 2 25 20 14 13 24 4 PHYSICIANS 2 15 32 34 15 4 1 NATURE AND PURPOSE OF RECOMMENDED TREATMENT N 1 1 52 26 13 4 3 3 PHYSICIANS 1 62 34 4 - - - PROS AND CONS OF RECOMMENDED TREATMENT GENERAL PUBLIC 1 43 24 17 6 4 5 PHYSICIANS 2 38 46 13 3 - 1 SIDE EFFECTS LIKELY GENERAL PUBLIC 2 39 29 15 2? 6 4 PHYSICIANS 2 53 40 6 1 - - “LESS THAN 0.31. NOTE: SPLIT-BALLOT ITEM, ASKED OF ONE-HALF OF GENERAL PUBLIC SAMPLE AND PHYSICIAN SAMPLE. Surveys of Physicians and the Public 109 G-25,26,27 P-13,14,15 TABLE 3-5 PUBLIC EXPECTATIONS VS. PHYSICIAN PRACTICE: GENERAL PUBLIC, PHYSICIANS Q.: There are a variety of issues (patient's condition, riswvs of adverse consequences sociated with treatment, uncertainties sociated vith medical decisions) that doctors may or may not discuss with their patients regarding the patient's condition and possible trestment. Do you feel tra: the doctors should tell their patients sbout/In your practice, do you iritiste discussion about each of the following things with the patient as a matter of course, or discuss it only if the patient asks about it? DISCUSS BASE: 1251 GENERAL PUBLIC; INITIATE ONLY NEVER 404 PHYSICIANS. DISCUSSION IF ASKED DISTUSS MCT SURE DIAGNOSIS AND PROGNOSIS OF CONDITION GENERAL PUBL IC--SHOULD 2 75 21 - . PHYS1CIANS--DO 2 99 1 - . COSTS OF PROCEDURE AND INSURANCE COVERAGE GENERAL PUBL 1C--SHOULD 2 69 27 . “ PHYSICIANS-=DO 2 38 56 4 2 NATURE AND PURPOSE OF RECOMMENDED TREATMENT GENERAL PUBLIC--SHOULD x 85 13 * 2 PHYS1CLANS==-DO 1 96 3 - . PROS AND CONS OF RECOMMENDED TREATMENT GENERAL PUBLIC--SHOULD 2 81 16 - 3 PHYS1CIANS--DO t 84 14 1 1 SIDE EFFECTS THAT ARE LIKELY TO OCCUR GENERAL PUBLIC--SHOULD 2 88 10 - 2 PHYSICIANS--DO 2 95 4 1 1 1-100 RISKS OF DEATH OR SERIOUS DISABILITY GENERAL PUBLIC--SHOULD x 75 22 . 2 PHYSICLANS--DO 1 81 16 1 3 1-1,000 RISKS OF DEATH OR SERIOUS DISABILITY ENERAL PUBLIC--SHOULD 2 64 33 1 3 PHYSICIANS--DO x 52 40 5 4 1-1,000 RISKS OF TEMPORARY DISABILITY GENERAL PUBLIC--SHOULD 2 65 32 . 3 PHYS1CLANS--DO x 43 48 4 4 DOCTOR'S UNCERTAINTIES ABOUT DIAGNOSIS ' GENERAL PU == 1 75 23 . 2 PHYSICIANS--DO 2 90 8 1 1 DOCTOR'S UNCERTAINTIES ABOUT BEST COURSE OF TREATMENT GENERAL PUBL IC--SHOULD 1 80 18 1 2 PHYSICIANS--DO 2 66 27 4 4 DIFFERENT VIEWS ABOUT BEST COURSE OF TREATMENT GENERAL PUBL I1C--SHOULD 2 76 21 . 3 PHYSICIANS--DO z 75 22 2 1 DOCTOR'S UNCERTAINTIES ABOUT WHETHER TREATMENT WILL BE SUCCESSFUL GENERAL PUBL 1C--SHOULD 2 83 1% . 2 PHYS1CLANS--DO 1 80- 16 2 2 PROBABLE IMPACT ON FAMILY LIFE GENERAL PUBLIC--SHOULD 1 8 18 . “ PHYSICIANS--DO 1 Va- 23 1 2 PROBABLE IMPACT ON JOB RATE GENERAL PUBLIC--SHOULD 1 79 18 - 3 PHYSICIANS--DO 1 79 20 1 1 “LESS TRAN 0.5%. DEPENDENT VARIABLE: MULTIPLE R R SQUARE ADJUSTED R SQUARE STANDARD ERROR TABLE 3-6 PREDICTORS OF PHYSICIAN DISCLOSURE SUMMARY MEASURE OF DOCTORS' REPORTED FREQUENCY OF INFORMATION DISCLOSURE (P.13-15) 0.26090 0.06807 0.06390 0.10327 —= VARIABLES IN THE EQUATION ANALYSIS OF VARIANCE REGRESSION RESIDUAL F = 16.31265 DF SUM OF SQUARES 3 0.52189 670 7.14514 SIGNIF F = 0.0000 VARIABLE B SE B BETA T SIG T PATIENT'S ABILITY TO UNDERSTAND 0.00115 0.21200-03 0.20988 5.435 0.0000 FATHER'S EDUCATION 0.00411 0.00196 0.07870 2.102 0.0359 PATIENT'S EDUCATION -0.01418 0.00675 -0.08124 -2.097 0.0363 (CONSTANT) 0.76397 0.01961 38.962 0.0000 FOR BLOCK NUMBER 1 PIN = 0.050 LIMITS REACHED. MEAN SQUARE 0.17396 0.01066 Ort fd x1puaddy :suoisia( aren) yjeay Sunyepy TABLE 3-7 PREDICTORS OF PATIENT EXPECTATIONS ABOUT DISCLOSURE DEPENDENT VARIABLE: SUMMARY MEASURE OF ATTITUDES ABOUT WHETHER PHYSICIANS SHOULD INITIATE DISCLOSURE (G-25 VERSION E1) MULTIPLE R 0.34052 ANALYSIS OF VARIANCE R SQUARE 0.11596 DF SUM OF SQUARES MEAN SQUARE ADJUSTED R SQUARE 0.10653 REGRES SION 0.64158 0.12832 STANDARD ERROR 0.10212 RESIDUAL 469 4.89137 0.01043 F = 12.30331 SIGNIF F = 0.0000 sm et et tt VARIABLES IN THE EQUATION VARIABLE B SE B BETA T SIG T HEALTH STATUS -0.02674 0.00724 -0.16784 -3.695 0.0002 RACE -0.04291 0.01282 -0.14701 -3.346 0.0009 BEST JUDGE OF DISCLOSURE -0.03308 0.00940 -0.15315 -3.520 0.0005 EXPERIENCE OF LIFE-THREATENING ILLNESS 0.02239 0.00723 0.13846 3.098 0.0021 HOSPITAL DAYS -0.00146 0.56300-03 -0.11563 -2.592 0.0098 (CONSTANT) 0.97277 0.02415 40.288 0.0000 o1[qnd 8Y} pue suelIsAYd jo sAoaing 11 G-16 TABLE 3-8 HOW MUCH PHYSICIANS DISCLOSE TO PATIENTS ABOUT ROUTINE CARE: GENERAL PUBLIC Q.: For routine care, do you think that doctors generally tell their patients too much, too little, or about the right amount? GENERAL PUBLIC EDUCATION AGE LESS 65 THAN HIGH 18-34 35-50 51-64 AND HIGH SCHOOL COLLEGE TOTAL YEARS YEARS YEARS OVER SCHOOL GRADUATE GRADUATE BASE 1251 488 334 226 195 242 726 280 % % 7% % % % % % TOO MUCH 2 2 1 2 4 4 2 1 TOO LITTLE 38 43 40 35 27 28 39 45 ABOUT THE RIGHT AMOUNT 56 55 55 58 59 59 59 31 NOT SURE 4 1 4 5 10 9 2 3 cll f x1puaddy :suorsioa( aren yireay Sunyey TABLE 3-9 HOW MUCH PHYSICIANS DICLOSE TO PATIENTS ABOUT SERIOUS ILLNESS: GENERAL PUBLIC Q.: In cases of serious illness, do you think that doctors generally tell their patients too much, too little, or about the right amount? GENERAL PUBLIC EDUCATION AGE LESS 65 THAN HIGH 18-34 35-50 51-64 AND HIGH SCHOOL COLLEGE TOTAL YEARS YEARS YEARS OVER SCHOOL GRADUATE GRADUATE BASE 1251 488 334 226 195 242 726 280 % % 7% 7% % % % % TOO MUCH 2 2 2 2 4 4 2 2 TOO LITTLE 33 38 34 28 24 27 33 38 ABOUT THE RIGHT AMOUNT 53 52 54 53 56 51 56 45 NOT SURE 11 7 10 17 16 12 9 13 o1[qng ay) pue suerIsAyd jo sfoaing gil TABLE 3-10 EXTENT TO WHICH PATIENTS ARE KEPT FULLY INFORMED: GENERAL PUBLIC Q.: How would you rate your doctors on keeping you fully informed about questions and decisions relating to your medical care that you consider important? Would you say that they have kept you fully informed on all important issues, most important issues, only some important issues, few important issues, or none of the issues you consider important? GENERAL PUBLIC EDUCATION AGE LESS © 65 THAN HIGH 18-34 35-50 51-64 AND HIGH SCHOOL COLLEGE TOTAL YEARS YEARS YEARS OVER SCHOOL GRADUATE GRADUATE BASE 1251 488 334 226 195 242 726 280 Z Z Z Z Z Zz Z Z ALL 3; 31 36 44 47 48 36 33 MOST 30 33 30 28 23 25 x 3] an SOME 18 22 19 14 12 11 18 26 FEW 6 7 7 5 6 5 8 5 NONE 3 3 3 3 3 4 3 3 NOT SURE 5 4 4 6 9 7 5 3 vit dg x1puaddy :suoisa(] aren yjeaH Sune TABLE 3-11 WHETHER PATIENTS WOULD BE WILLING TO PAY MORE FOR MORE EXPLANATION OF ROUTINE CARE ISSUES: GENERAL PUBLIC Q.: If your doctor spent more time explaining things to you about your routine care, would you be willing to pay more? BASE YES, WOULD BE WILLING NO, WOULD NOT BE WILLING DEPENDS NOT SURE GENERAL PUBLIC EDUCATION INCOME LESS $7,500 $7,501 $15,001 $25,001 MORE THAN HIGH OR TO TO TO THAN HIGH SCHOOL COLLEGE TOTAL LESS $15,000 $25,000 $35,000 $35,000 SCHOOL GRADUATE GRADUATE 1251 164 219 341 229 214 242 726 280 x z z z Z z x Z x 38 44 36 33 40 41 40 38 33 54 48 55 59 51 53 47 54 60 4 bi 3 4 3 2 5 3 3 5 3 6 4 6 4 8 S 4 o1[qng dy) pue SueIOISAYd JO sfaAIng SIT TABLE 3-12 WHO IS RESPONSIBLE FOR KEEP ING PATIENTS FULLY INFORMED: Q.: Do you think that it is patient is fully informed abo responsibility to ask the doc BASE PRIMARILY THE DOCTOR'S RESPONSIBILITY PRIMARILY THE PATIENT'S RESPONSIBILITY EQUALLY RESPONSIBLE (VOL. ) NOT SURE GENERAL PUBLIC primarily the doctor's responsibility to make sure that the ut his condition and treatment, or the patient's tor about these things? GENERAL PUBLIC EDUCATION LESS THAN HIGH RACE HIGH SCHOOL COLLEGE TOTAL WHITE MINORITY SCHOOL GRADUATE GRADUATE 1251 1027 201 242 726 280 Z Z Z Z Z z 66 68 58 60 68 68 20 18 31 26 21 15 12 13 10 13 10 16 1 1 1 1 1 1 911 fq x1puaddy :suorsiae(] aren yjeey Sunyeiy P-24 TABLE 3-13 WHO IS RESPONSIBLE FOR KEEPING PATIENTS FULLY INFORMED: PHYSICIANS Q.: Do you think that it is primarily the doctor's responsibility to make sure that the patient is fully informed about his condition and treatment, Or the patient's responsibility to ask the doctor about these things? PHYSICIANS PATIENTS UNDERSTANDING TREATMENT TOTAL 0% TO 69% 70% TO 89% 90% TO 100% BASE 805 150 270 384 % 7% % z TOTAL PRIMARILY THE DOCTOR'S RESPONSIBILITY 77 71 76 81 PRIMARILY THE PATIENT'S RESPONSIBILITY 2 3 5 1 EQUALLY RESPONSIBLE (VOL.) 19 24 19 17 NOT SURE 1 2 * 1 *LESS THAN 0.5%. o1[qnd 8Y) pue SueIISAY JO sAdAING i11 Surveys of Physicians and the Public 119 CHAPTER 4: WHEN THE PATIENT IS NOT TOLD: PATTERNS OF WITHHOLDING Introduction There appears to be widespread agreement between doctors and the general public that the doctor should initiate discussions with the patient on im- portant issues concerning treatment. Moreover, doc- tors and patients generally agree that in most mat- ters of concern to patients with regard to their condition and treatment, doctors usually initiate discussion with their patients as a matter of course. However, there are situations that arise in which doctors do not disclose all information on all matters relevant to their patients' condition, prog- nosis, or treatment. Note that relatively few doc- tors say that they always tell all their patients everything they might wish to know. Both the public and the medical profession recognize reasons and circumstances under which information should and will be withheld from patients. The purpose of this chapter is to explore the magnitude, pattern, and reasons for physician nondisclosure. How Much to Tell the Patient? We noted in the previous chapter that physicians recognize that most patients want to be given a can- did assessment of their diagnosis and prognosis, even if it is unfavorable. And the survey finds that most physicians report initiating discussions with their patients as a matter of course on most important areas of treatment and condition. At the same time, we find that doctors do not necessarily tell all patients everything about their condition and treatment. Decisions are continually being made by physicians about how much to tell patients about their condition and treatment. The national sample of physicians was asked: ... How often do you find yourself in a situation where you must make a conscious and deliberate eval- uation of how much to tell a patient about his or her condition or treatment -- would you say several times a day, daily, weekly, rarely, or never?" Ac- cording to the physicians, 27% make a serious eval- uation of how much to tell a patient several times a 120 Making Health Care Decisions: Appendix B day (Table 4-1). Another 25% of the physician sam- ple report that they make an evaluation of how much to tell a patient daily. Thus, questions regarding disclosure occur at least daily for more than half of all physicians engaged in the care of adult pa- tients. Another 20% of physicians say that this type of situation occurs weekly for them. Only 25% of physicians say that they rarely have to make a serious evaluation of how much to tell patients, and 3% say it never happens. The frequency with which physicians must make decisions about how much to tell patients about their condition or treatment is an important find- ing. However, it should be clearly noted that the question does not address the frequency with which a physician decides to withhold certain information from a patient. The physician may have to choose daily about whether or not to divulge certain in- formation to a patient, but this does not mean that physicians withhold information on a daily basis. This latter issue is investigated later in this chapter. The incidence of disclosure decisions is not the same for all doctors. The nature of a physician's practice has a lot to do with how often he or she will have to make a serious evaluation of how much to tell patients about their condition and treat- ment. For instance, one-third of office-based doc- tors (33%) rarely or never have to decide how much to tell a patient. By contrast, only 14% of hospital-based doctors say that they rarely or never have to make a conscious decision about how much to tell a patient. The distinction between office-based and hospital-based practices reflects differences in the types of patients seen. Nearly four-fifths (79%) of doctors with less than one-quarter of their patients with serious illnesses are office based, while more than one-half (52%) of doctors with more than 50% of their patients with serious illnesses are hospital based. It is not surprising then that more than one-third (36%) of doctors with relatively few ser- iously ill patients rarely or never have to make disclosure decisions, compared with 18% of those most of whose patients are seriously ill. It follows that physicians who specialize in Ob/Gyn are more likely to rarely or never have to Surveys of Physicians and the Public 121 make disclosure decisions (47%) than surgeons (25%) or internists (21%). However, it is notable that even in Ob/Gyn, 40% report that cases arise at least daily that require them to decide how much to tell the patient about her condition or treatment. Daily decisions on how much to tell a patient are reported by 56% of physicians in internal medicine, 54% of surgeons, 50% of physicians in general practice or family practice, and 53% of all others. What Affects Disclosure? If doctors do not tell all their patients every- thing about their condition and treatment, then the question arises as to what determines disclosure. To answer this question, the national sample of phy- sicians was asked: "What are the primary factors that influence how much you tell a patient about his condition or treatment?" The patient's perceived ability to understand the treatment or condition issues emerges as the factor most often cited by physicians as influencing the extent of this disclosure (Table 4-2). More than one-half of the physicians (56%) interviewed volunteer patient understanding as a primary factor in influencing how much they tell patients about their condition or treatment. This factor is most salient among doctors who report a substantial case load of patients who cannot understand most aspects of treatment, even if reasonable time and effort are devoted to explanation. In practices where physi- cians report that less than 70% of their patients are able to understand most aspects of treatment, the patient's ability to understand treatment issues is volunteered by 70% of doctors as a primary factor in determining the extent of disclosure. In prac- tices where 90% or more of patients are able to un- derstand most aspects of their treatment, the survey finds 48% of physicians volunteering the patient's ability to understand treatment issues as a primary factor in disclosure decisions. Observation: Physicians engaged in the primary care of adults report that approximately 21% of their patients, on average, cannot understand most aspects of their condition and treatment, even if reason-— able time and effort are devoted to explana- 122 Making Health Care Decisions: Appendix B tion. This inability to understand treatment issues raises its own set of informed consent issues. The patient's ability to understand the issues -- as perceived by the doctor -- is the most common factor in physicians' decisions about how much to tell a patient. A number of other concerns are also cited by physicians as primary factors in influencing dis- closure. The seriousness of the condition is cited by 30% of doctors as a primary factor in influencing how much to tell the patient, while 31% cite the patient's ability to cope with the news. As we shall see later in this report, both doctors and the public feel that the effect of the information on the patient should be a primary factor in decisions on disclosure. One-quarter of the physician sample (25%) cite how much the patient wants to know as a primary fac- tor in influencing how much the doctor tells. A much smaller 13% of physicians cite the wishes of the patient's family as a primary factor in in- fluencing disclosure. And only a miniscule 2% of physicians volunteer legal responsibilities as a primary factor in influencing how much they tell their patients. Less than 1 in 10 doctors (9%) volunteer that they tell their patients everything. The distribu- tion of this reported behavior among physicians var- ies somewhat with the type of patients seen in the practice. The proportion of physicians who say that they tell their patients everything declines as the proportion of seriously ill patients in the practice increases (11%-6%) and increases as the proportion of patients who can understand most aspects of their treatment increases (5%-12%). Neither specialty nor location of practice, however, has a strong or con- sistent effect on this behavior. Do Patients Want to be Told Everything? One-quarter of physicians cite patient interest in their condition and treatment as a primary factor in determining the extent and nature of disclosure. Do the public respondents want to know everything about their condition and treatment, or only the positive aspects? The survey finds near unanimity Surveys of Physicians and the Public 123 among the public on this issue. Ninety-four percent want their doctor to tell them everything about their condition, even if it is unfavorable (Table 4-3). Only 4% of the public say that they don't want their doctor toc tell them everything, while 1% are uncertain. Physicians recognize this preference among their patients. Doctors were asked: "How many patients who come to you for treatment want you to give them a candid assessment of their diagnosis and prog- nosis, even if it is unfavorable?" Eighty-six per- cent of doctors say that all (17%) or most (69%) of the patients who come to them want to know the truth, even if it is unfavorable (Table 4-4). Observation: The perceived preference of the patient toward the extent of medical disclosure is an enormous- ly important issue in how much a doctor will tell a patient. Doctors admit that they make judgments about how much a patient wishes to know without explicit statements from the pa- tient. Hence, the degree to which doctors and patients correctly perceive the others' pref- erences is a critical issue in doctor-patient communications. It is very reassuring to find that almost all doctors recognize that most pa- tients want full disclosure, even if it involves bad news. The rather small difference between the 94% of the public who want to know everything about their condition and the 86% of physicians who feel that most of their patients want a candid assessment of their condition may be partially explained by the difference between a national sample of the general public and a sample of physicians reporting about patients. Fully 82% of the public sample describe their health as good or excellent. By contrast, more than one-half of the physicians interviewed report that one-quarter or more of their patients are seriously ill. The attitudes toward full dis- closure of medical condition may be somewhat dif- ferent in a healthy population than in a sick pop- ulation. This is supported by the finding that only 3% of those who describe their health as excellent or good do not want their doctor to tell them every- 124 Making Health Care Decisions: Appendix B thing about their condition. By contrast, 8% of those who describe their health as fair or poor do not want their doctor to tell them everything about their condition. It seems reasonable to infer from these findings that the incidence of individuals who say that they do not want to hear bad news from their doctor may be somewhat greater among the patient population, where bad news is a more realistic expectation and a more pressing concern, than among a healthy pop- ulation. However, even among those in poor health and at high risk of serious illness, the vast ma- jority want their doctor to tell them the truth about their condition and treatment. It is reassur- ing that the vast majority of doctors recognize this. Even among physicians who report that more than one- half of their patients are seriously ill, we find that 85% believe that all or most of their patients want a candid assessment of their condition. Who Doesn't Want the Doctor to Tell Everything? There is relatively little difference by spe- cialty in doctors' perceptions that most of their patients want a candid assessment of their condi- tion. For instance, 87% of internists, who tend to deal with patients who are ill, think that most of" their patients want to know the truth, compared with 92% of practitioners who tend to deal with healthy patients. Similarly, there is a relatively small distinction between the 88% of office-based physi- cians and the 82% of hospital-based physicians who think that most of their patients want a candid as- sessment. In both of these areas, it is presumably the greater proportion of seriously ill patients that accounts for the limited differences in physi- cian perception of patient preferences. A much more substantial difference exists be- tween physicians whose practices are distinguished by other patient characteristics. In practices in which poor patients represent 50% or less of all patients, 89% of doctors think that all or most of their patients want a candid assessment of their condition. However, in practices where more than half the patients are poor, we find that only 737% of physicians think that most of their patients want the truth. Surveys of Physicians and the Public 125 Are the poor less likely to want a candid as- sessment of their condition, even if it is unfavor- able? The evidence from the survey suggests that a slightly higher proportion of members of less priv- ileged segments of society may not want to hear bad news from their doctor. However, the difference is quite small and overshadowed by the fact that most, indeed nearly all patients want to be told every- thing. The incidence of persons not wanting their doc- tor to tell them everything, even if it is unfavor- able, declines from 6% of those with an income of $7,500 or less, to 5% of those with an income of from $7,501 to $15,000, to 4% of those with an in- come of over $15,000. More convincingly, although 4% of whites don't want their doctor to tell them everything, this increases to 8% among minorities. A striking set of differences emerges from con- sideration of the population distribution of those respondents who report that they have asked a doctor not to tell them bad news. The proportion of in- dividuals from minority groups (5%) is three times the proportion of whites (2%) who have asked a doc- tor not to tell them bad news (Table 4-5). Four times as many individuals from households with an income of $7,500 or less (4%), compared with in- dividuals from households earning more than $35,000 (1%), have asked doctors not to tell them bad news. Similarly, five times as many individuals without a high school diploma (5%) as college graduates £12) have asked not to be told bad news about their con- dition. Although the differences between groups are quite small, the pattern is consistent and con- vincing. The probability is somewhat greater that an individual of lower socioeconomic status will ask his or her doctor not to convey bad news. Although this difference is quite small, it would appear that the difference is recognized by physicians. Observation: The slightly higher incidence of patients asking not to be told bad news among low income and minority individuals appears to have a dis- proportionate impact on physicians' perception of how many patients want to know everything about their condition. This is further com= pounded by the relationship between physicians’ 126 Making Health Care Decisions: Appendix B perceptions of patients' ability to understand treatment issues and physicians' perceptions of how many patients want a candid assessment of their condition. Low income patient populations are more likely to be seen by physicians as hav- ing more difficulty in understanding treatment issues. Hence, these two factors tend to create the perception that many of their patients do not want full disclosure. Justification for Nondisclosure Virtually the entire adult population of the United States subscribes to the belief that patients have a right to all available information about their condition and treatment. Moreover, they want to know everything about their own condition and treatment, even if it is bad news. At the same time, this does not represent acceptance of a com- plete and uniform disclosure standard by the pub- lic. The public sample was asked: "Do you think that some patients should be told less about their condition and treatment than other patients, or should doctors provide the same amount of informa- tion to all patients?" A majority of 56% of the public respond that some patients should be told less about their condition and treatment than other patients (Table 4-6). Only 36% feel that doctors should provide the same amount of information for all patients. Interestingly, the acceptance of differences in the degree of disclosure among the general public is more strongly rooted among those with a college de- gree and older members of the public. Only 47% of the 18-34 age group feel that some patients should be told less, compared with 63% of those aged 65 and over. Similarly, while 52% of those without a high school diploma think some patients should be told less than others, this attitude is held by 64% of college graduates. These patterns are particularly noteworthy since age is inversely related to educa- tion. Thus, both age and education must have an important independent effect on public approval of differences in the degree of disclosure. Under what conditions can patients be told less than the whole truth about their condition or treat- ment? The public sample was asked whether they felt that a doctor would be justified in withholding in- Surveys of Physicians and the Public 127 formation about condition and treatment from pa- tients -- under five specific conditions. More than two-thirds of the public (68%) report that they feel a doctor would be justified in withholding informa- tion if the patient has told the doctor that he doesn't want to hear bad news, while 27% of the pub- lic feel that a doctor would not be justified in withholding information under these conditions (Table 4-7). Similarly, a substantial majority of 68% of the public feel that a doctor would be jus- tified in withholding information if the information might significantly harm the patient's health and well-being, while 27% feel that the doctor would not be justified in withholding information. The public overwhelmingly accepts the doctor's right to withhold information about condition and treatment from a patient who has expressed the wish not to be told, or under conditions where the infor- mation might significantly harm the patient's health. The acceptability of withholding informa- tion from patients declines rather precipitously when the request is made by the family rather than the patient or where the harm to the patient from disclosure is not significant. Although 68% of the public feel that the doctor would be justified in withholding information if the patient has asked not to be told bad news, only 487% feel that the doctor would be justified in withholding information if the patient's family has asked the doctor not to tell the patient. Similarly, while 68% of the public feel that the doctor would be justified in with- holding information that might significantly ham the patient, only 52% consider the doctor justified in withholding information that might make the pa- tient anxious or upset. Although the extent of public acceptance of these four justifications for withholding informa- tion from the patient varies considerably, at least a plurality of the American public consider the doc- tor justified in withholding information under these four conditions. However, a majority of the public (55%) say that a doctor would not be justified in withholding information from the patient on the grounds that the information might make the patient unwilling to undergo treatment that the doctor thinks is necessary. Only a minority (38%) consider patient compliance with prescribed medical therapy as sufficient grounds for nondisclosure by doctors. 128 Making Health Care Decisions: Appendix B Observation: These findings suggest that the public permits the doctor substantial leeway in deciding what information should be disclosed to the patient. The doctor is seen as justified in withholding information that he or she believes will harm the patient or that the patient does not want to hear. However, a line is drawn at the point of medical decisionmaking. Most of the public feel that the doctor is not justified in withholding information that might lead the patient to choose a course of treatment other than the course prescribed by the doctor. In other words, it appears that the public approves of nondisclosure when it prevents direct harm, but not for the purpose of limiting patient choice, even of potentially harmful courses of action. Even if the information about the patient's con- dition and treatment is not inherently harmful, questions remain as to how much information doctors should disclose to their patients -- how much they are obligated to disclose. Too much information may be confusing to patients. It also requires a sub- stantial amount of the time and effort of the doc- tor. Too little information violates the perceived right of patients to all information that they wish to know about their condition and treatment. It also circumscribes the patients! right to informed choice of medical treatment. Hence, it is important to know who should decide how much information should be disclosed to patients. The public sample was asked: "Who do you think is the best judge of the amount of information that should be disclosed to the patient?" The survey finds that the general public splits on this ques- tion -- with 45% saying that the patient is the best judge of how much should be disclosed and 44% saying that the doctor is the best judge (Table 4-8). An- other 8% say that it depends, and 3% are not sure. The level of education of the respondent has a substantial effect on the perception of who is the best judge of how much should be disclosed to the patient. Among those without a high school diploma, a majority (51%) say that the doctor is the best judge, while a minority (35%) say that the patient is the best judge. Among those who are high school graduates, there is an almost even division between Surveys of Physicians and the Public 129 those who think that the patient is the best judge (46%) and those who think that the doctor is the best judge (44%). However, among college graduates, a majority (53%) think that the patient is the best judge of how much should be disclosed, while only a minority (36%) think that the doctor is the best judge. The more widespread acceptance among the least educated of the doctor's predominant role in de- cisions about disclosure is paralleled to a certain extent in the justifications for nondisclosure. While 57% of those without a high school diploma feel that the doctor would be justified in withhold- ing information that might make the patient anxious or upset, only 41% of college graduates consider this adequate justification (Table 4-9). Similarly, 53% of those without a high school diploma feel that the doctor would be justified in withholding infor- mation if the patient's family has asked him not to tell the patient, while only 45% of the college gra- duates consider this adequate justification. Final- ly, 41% of those with less than a high school diplo- ma, compared with 30% of college graduates, feel that doctors would be justified in withholding in- formation if the information might make the patient unwilling to undergo treatment that the doctor thinks is necessary. Clearly, the public perception of the patient's role in the medical information disclosure process tends to increase with the level of education of the public. Those with less than a high school diploma tend to see the doctor as the best judge of how much to tell patients about their condition and treat- ment. The college graduate, on the other hand, not only tends to see the patient as the best fudge of how much information should be disclosed, but also tends to restrict the range of conditions under which doctors are justified in withholding informa- tion from patients. In two areas of disclosure, however, the college graduates depart from this general pattern. Educa- tion level has no effect on the widespread opinion that a doctor would be justified in withholding in- formation from patients who tell the doctor that they don't want bad news. Moreover, if the informa- tion might significantly harm the patient's health and well-being, the proportion of the public who see the doctor as justified in withholding the informa- 130 Making Health Care Decisions: Appendix B tion rises from 65% of those without a high school diploma, to 67% of those with a high school diploma, to 73% of college graduates. Frequency of Physician Nondisclosure A key question in the study of doctor-patient communication is whether or not physicians actually withhold information from their patients concerning their condition and treatment. As we noted earlier, over one-half of the physicians engaged in the care of adults (52%) report that at least one case will arise each day in which they must make a deliberate and conscious evaluation of how much to tell a patient about his or her condition and treatment. How often, then, do doctors decide that a patient should not be told something about his or her con- dition and treatment? The sample of physicians was asked how often they withhold information about condition and treat- ment from their patients. Doctors were first asked about information related to condition: "Sometimes a physician feels it is advisable to withhold in- formation about diagnosis or prognosis from a pa- tient. In your practice would you say this occurs once a day, once a week, once a month, a few times a year, or almost never?" Then they were asked about withholding information related to the patient's treatment: "Sometimes a physician feels it is ad- visable to withhold information about treatment risks and alternatives from a patient. In your practice would you sav this occurs once a day, once a week, once a month, a few times a year, or almost never?" The survey demonstrates that, at least according to physicians, withholding information from patients about their condition and treatment is a fairly un- common occurrence. Although more than one-half of physicians (52%) report that situations arise daily that require a conscious and deliberate evaluation of how much to tell patients about their condition and treatment, only 2% of physicians report actually withholding information concerning diagnosis and prognosis on a daily basis, and only 4% report with- holding information about treatment risks and alter- natives at least once a day (Tables 4-10 and 4-11). Most typically, physicians report almost never withholding information about condition or treatment Surveys of Physicians and the Public 131 from patients. Nearly one-half of physicians (46%) report that they almost never withhold information about diagnosis and prognosis from a patient. A clear majority of physicians (57%) report that they almost never withhold information about treatment risks and alternatives from a patient. To the ex- tent that physicians do withhold information from their patients, this type of situation is most like- ly to occur only a few times a year for both issues of diagnosis/prognosis (32%) and treatment risks/al- ternatives (24%). When doctors are asked the most common reasons for withholding information about condition or treatment from a patient, the answers are quite similar to the primary factors cited by doctors in evaluating how much to tell a patient. The most common reason for withholding information from the patient is the patient's ability to cope with the news (34%), while 17% cite the related concept of the seriousness of the patient's condition (Table 4-12). Once again, the patient's ability to understand his or her condition or treatment is a leading reason given by doctors (28%) for with- holding information, while another 17% cite the phy= sical or mental incapacity of the patient. Finally, 16% of physicians indicate that how much the patient wants to know is a common reason for withholding in- formation, while a somewhat larger 21% cite the wishes of the patient's family as a common reason for withholding information. An analysis of the distribution of withholding by doctors' practice characteristics gives a good deal of credence to these explanations. In with- holding information about condition (diagnosis and prognosis), the proportion of seriously ill patients in the practice has an effect on the frequency of withholding. Only 39% of physicians with practices composed primarily of seriously ill patients report that they almost never withhold information about diagnosis and prognosis. By contrast, 52% of phy- sicians with practices in which a relatively limited number of the patients are seriously ill report that they almost never withhold information about con- dition. The proportion of patients in the practice who can understand most aspects of their condition and treatment also has a clear effect on the fre- quency of withholding information. The frequency of withholding information related to treatment risks and alternatives is quite strong- 132 Making Health Care Decisions: Appendix B ly affected by the proportion of patients in the practice who can understand most aspects of their condition and treatment. Among physicians who re- port that less than 70% of their patients can under- stand their condition and treatment, the survey finds 427% saying that they almost never withhold in- formation about treatment risks and alternatives. By contrast, 637 of physicians who report that 90% or more of their patients are able to understand their condition and treatment say that they almost never withhold treatment information. In the case of treatment information, there is little or no re- lationship between the likelihood of withholding information and the incidence of serious illness among patients. Observation: The issues of patient competence and ability to cope with bad news appear to be differentially related to issues of disclosure of case and treatment information. The data suggest that disclosure of treatment information is fairly directly related to patients' ability to under- stand the information. Doctors seem to be say- ing that they are willing to disclose treatment information to the extent that they perceive patients are able to understand the informa- tion. However, the disclosure of information about diagnosis and prognosis is much more in- timately bound to the seriousness of the condi- tion and the perceived ability of the patient to cope with the information. From both the legal and ethical standpoints, it is particularly important to know the extent to which physicians withhold information at the express wish of their patients. A substantial number of physicians (16%) report that one of the most common reasons for withholding information from a patient is the extent to which the patient wants to be in- formed. However, it is unclear whether this is a wish expressed by the patient or an inference drawn by the physician about the patient's wishes. Because of the importance of this issue, doctors were asked directly: "For those patients to whom you do not disclose bad news, how often is it be- cause they tell you directly they don't want to know it?" Only one-quarter of physicians who report ever Surveys of Physicians and the Public 133 withholding bad news from their patients say that it is always (12%) or often (11%) because the patients have told them directly that they don't want to know it (Table 4-13). Most physicians who withhold bad news from patients say that it is rarely (45%) or never (16%) because they have been asked by patients not to tell them. Most doctors who withhold information about treatnent from their patients, however, inform the patient's family. Four out of five doctors (80%) report that in situations where they withhold in- formation about treatment from a patient, they usu- ally discuss the information with another family member (Table 4-14). Only 7% of physicians say that when they withhold information from a patient, they rarely or never discuss it with the patient's fam ily. Internists and surgeons are most likely to discuss withheld information with another family member, while obstetricians and gynecologists are least likely to do so. P=9 TABLE 4-1 HOW OFTEN PHYSICIANS TELL PATIENTS OF CONDITION: PHYSICIANS Q.: How often do you find yourself in a situation where you must make a conscious and deliberate evaluation of how much to tell a patient about his or her condition or Treatment -- would you say several times a day, daily, weekly, rarely, or never? PHYSICIANS PATIENTS WITH SERIOUS SPECIALTY ILLNESS GENERAL/ PRACTICE MORE INTERNAL FAMILY ALL LOCATION 0Z TO 262% THAN TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS OFFICE HOSPITAL 25% TO 50% 50% 805 233 231 108 177 45 533 232 367 229 193 2 z z Z z Z 2 % z z 2 SEVERAL TIMES A DAY 27 28 28 22 27 26 26 26 23 29 31 DAILY 25 28 22 18 27 27 22 34 20 30 28 WEEKLY 20 22 23 13 20 9 18 25 21 16 22 RARELY 25 19 24 44 23 24 30 13 33 21 16 NEVER 3 2 2 3 2 11 3 1 3 3 2 NOT SURE * * * - 2 2 1 = =~ * 1 aap. *LESS THAN 0.5%. bet SUOISIOa(] aIe) [eal Suryey g xrpuaddy Surveys of Physicians and the Public 135 P-10 TABLE 4-2 FACTORS INFLUENCING HOW MUCH TO TELL: PHYSICIANS Q.: What are the primary factors that influence how much you tell a patient about his condition or treatment? What else? PHYSICIANS PATIENTS WITH PATIENTS SERIOUS ILLNESS UNDERSTANDING MORE TREATMENT ren 0x TO 26% THAN OX TO 70% 903 TO TOTAL 25% TO 50% 50% 69% TC E33 100% BASE 805 367 229 193 150 ie 38% 2 x 2 z 2 2 2 PATIENT'S ABILITY TO UNDERSTAND HIS CONDITION OR TREATMENT 56 53 60 57 70 57 48 PATIENT'S STABILITY/PERSONAL STRENGTH/EMOT IONAL MATURITY/ ANTICIPATED REACTION TO NEWS/ ABILITY TO COPE WITH NEWS 3 31 31 32 28 33 32 SERIOUSNESS OF PATIENT'S CONDITION/DIAGNOSIS/PROGNOSIS 30 30 32 30 36 31 28 HOW MUCH THE PATIENT WANTS TO KNOW/ASKS, PATIENT'S CURIOSITY 25 22 29 28 28 25 24 THE WISHES OF THE PATIENT'S FAMILY 13 33 14 15 13 15 12 PATIENT'S PHYSICAL/ MENTAL INCAPACITY 12 12 11 10 8 13 12 1 TELL THEM EVERYTHING 9 11 9 6 5 2? 12 EFFECT OF INFORMATION ON PATIENT'S COMPLIANCE WITH TREATMENT 8 9 7 8 6 9 8 UNCERTAINTY/CERTAINTY OF DIAGNOS1S/PROGNOSIS 5 6 4 5 5 6 4 SERIOUSNESS/RISKS OF TREATMENT 5 3 6 4 7 6 3 PATIENT'S POSITION IN FAMILY, EFFECT OF CONDITION/ TREATMENT ON FAMILY FUTURE/LIFESTYLE 4 3 7 4 2 6 4 EFFECT OF INFORMATION ON PATIENT'S HEALTH AND WELL BEING “6 4 3 3 6 3 FAMILY /OUTSIDE SUPPORT AVAILABILITY “4 3 4 3 3 4 DOCTOR-PATIENT RELATIONSHIP, RAPPORT WITH PATIENT 33 3 33 4 2 MEDICAL/LEGAL RESPONSIBILITY 2 2 1 4 3 3 2 TIME AVAILABILITY, IF I HAVE THE TIME 2 2 3 2 4 2 2 ALL OTHER MENTIONS 1 1 1 1 1 1 NOT SURE/REFUSED 1 1 * 1 * 1 "LESS THAN 0.5%. c-21 TABLE 4-3 DESIRE FOR CONDITION INFORMATION: GENERAL PUBLIC Q.: Do you want your doctor to tell you everything about your medical condition, even if it is unfavorable? YES, WANT NO, DO NOT WANT NOT SURE GENERAL PUBLIC REPORTED HEALTH STATUS TOTAL EXCELLENT GOOD FAIR POOR 1251 482 550 144 73 % 7% Z % % 94 96 99 90 90 4 3 3 8 8 1 1 1 2 2 981 d xipuaddy :suoistoaq are) yiesy Supe TABLE 4-4 PATIENT DESIRE FOR CANDID DIAGNOSIS AND PROGNOSIS: PHYSICIANS Q.: How many patients who come to you for treatment want you to give them a candid assessment of their diagnosis and prognosis, even if it is unfavorable? Would you say all, most, some, few, or none? BASE ALL MOST SOME FEW NONE NOT SURE PHYSICIANS SPECIALTY POOR PATIENTS GENERAL/ PRACTICE 25% 51% INTERNAL FAMILY ALL LOCAT ION OR 26% TO OR TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS OFFICE HOSPITAL LESS _ 50% MORE 805 233 231 108 177 45 533 232 500 143 107 % % % % % % 3 % z Z z 17 11 18 21 18 29 18 14 16 18 18 69 76 68 71 63 67 70 68 73 nn 55 9 11 6 6 14 2 9 11 9 8 14 4 1 8 2 4 - 3 5 2 4 11 * - - - 1 2 - 1 * - 2 * 1 - - 1 - * 1 * - - I — *LESS THAN 0.5%. o1qng ay} pue sueroIsAyd jo shearing LET TABLE 4-5 REQUESTED WITHHOLDING OF BAD NEWS: GENERAL PUBLIC Q.: Have you ever asked a doctor not to tell you bad news? GENERAL PUBLIC EDUCATION INCOME LESS $7,500 $7,501 $15,001 $25,001 MORE THAN HIGH HIGH OR TO TO TO THAN RACE SCHOOL SCHOOL COLLEGE TOTAL LESS $15,000 $25,000 $35,000 $35,000 WHITE MINORITY GRADUATE = GRADUATE GRADUATE BASE 1251 164 219 341 229 214 1027 201 242 726 280 x Zz Z xX Zz x Z Z Zz Z 2 YES, HAVE ASKED 2 4 2 2 2 1 2 5 5 2 1 NO, HAVE NOT ASKED 97 96 97 98 97 99 98 95 95 98 99 NOT SURE * - * - * - * - * * - *LESS THAN 0.5%. 8€T fg xrpuaddy :suorstoa(y aren) yjesy Sunyepy G-20 TABLE 4-6 APPROPRIATENESS OF DIFFERENTIAL DISCLOSURES: GENERAL PUBLIC Q.: Do you think that some patients should be told less about their condition and treatment than other patients, or should doctors provide the same amount of information to all patients? GENERAL PUBLIC EDUCATION AGE LESS 65 THAN HIGH HIGH 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE TOTAL YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE BASE 1251 488 334 226 195 242 726 280 % % % % % % % % SOME TOLD LESS 56 47 64 64 63 52 55 64 SAME FOR ALL 36 48 29 27 25 35 38 29 NOT SURE 8 5 7 10 12 13 7 6 o1[qnd 9Y) pue sueISAY Jo sAoaIng 6€1 140 Making Health Care Decisions: Appendix B TABLE 4-7 JUSTIFICATION FOR PHYSICIAN WITHHOLDING INFORMATION: GENERAL PUBLIC Q.: Sometimes physicians feel that it is advisable to withhold information about diagnosis, prognosis, or treatment risks and alternatives from patients. Under each of the following conditions, do you feel that a doctor would be justified in withholding information about condition and treatment from a patient, or not? TOTAL SENERAL PUBLIC NOT BASZ: 1251 JUSTIFIED JUSTIFIED NOT SURE IF THE PATIENT TELLS vs DOITOR THAT HE DOESN'T - TO HEAR BAD NEWS 2 68 27 4 IF TARE INFORMATION MIGHT FICANTLY HARM PATIENT'S HEALTH AND WELL-BEING 2 68 27 5 IF THE INFORMATION MIGHT MAKE THE PATIENT ANXIOUS OR UPSET 2 52 41 7 IF THE PATIENT'S FAMILY ASKS THE DOCTOR NOT TO TELL THE PATIENT x 48 42 9 IF THE INFORMATION MIGHT MAKE THE PATIENT UNWILLING TO UNDIRGO TREATMENT THE DOCTOR THINKS 1S NECESSARY 2 38 55 7 G-19 TABLE 4-8 BEST JUDGE OF INFORMATION TO BE DISCLOSED: GENERAL PUBLIC Q.: Who do you think is the best judge of the amount of information that should be disclosed to the patient -- the patient or the doctor? GENERAL PUBLIC EDUCATION AGE LESS 65 THAN HIGH HIGH 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE TOTAL YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE BASE 1251 488 334 226 195 242 726 280 2 % % % % Xz % PATIENT 45 55 49 37 27 35 46 53 DOCTOR bt 38 40 48 58 51 44 36 DEPENDS (VOL) 8 4 9 10 10 8 6 9 NOT SURE 3 3 2 5 5 6 3 2 o1[qng ay) pue suerdIsAyd jo sAoaIng vl 142 Making Health Care Decisions: Appendix B TABLE 4-9 JUSTIFICATION FOR PHYSICIAN WITHHOLDING INFORMATION: GENERAL PUBLIC Q.: Sometimes physicians feel that it is advisable to withhold information about diagnosis, prognosis, or treatment risks and alternatives from patients. Under each of the following conditions, do you feel that a doctor would be justified in withholding information about condition and treatment from a patient or not? GENERAL PUBLIC BASE IF THE PATIENT TELLS THE DOCTOR HE DOESN'T WANT TO HEAR BAD NEWS IF THE INFORMATION MIGHT SIGNIFICANTLY HARM PATIENT'S HEALTH AND WELL-BEING IF THE INFORMATION MIGHT MAKE THE PATIENT ANXIOUS OR UPSET IF THE PATIENT'S FAMILY ASKS THE DOCTOR NOT TO TELL THE PATIENT IF THE INFORMATION MIGHT MAKE THE PATIENT UNWILLING TO UNDERGO TREATMENT DOCTOR THINKS NECESSARY TOTAL YEARS YEARS YEARS OVER GRADUATE EDUCATION LESS 65 THAN HIGH HICH 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE GRADUATE GRADUATE 726 280 1251 488 334 226 19% 242 2 x x x x x 1 1 68 64 172 22 70 68 69 67 68 65 77 63 66 65 67 7 52 43 56 64 58 57 55 41 48 41 52 56 su 53 48 45 18 33 39 39 46 “1 40 30 Surveys of Physicians and the Public 143 P-19 TABLE 4-10 HOW OFTEN INFORMATION ABOUT TREATMENT RISKS IS WITHHELD: PHYSICIANS Q.: Sometimes a physician feels it is advisable to withhold information about treatment risks and alternatives from a patient. In your practice would you say this occurs once a day, once a week, once a month, a few times & year, or almost never? BASE ONCE A DAY ONCE A WEEK ONCE A MONTH A FEW TIMES A YEAR ALMOST NEVER NOT SURE *LESS THAN 0.5%. PHYSICIANS PATIENTS WITH PATIENTS SERIOUS ILLNESS UNDERSTANDING MORE TREATMENT 0% TO 26% THAN 0% TO 70% 90% TO TOTAL 25% TO 50% 50% _69% TO 89% 100% 805 367 229 193 150 270 384 % 2 x 2 2 x 2 4 3 5 3 4 4 3 6 S 9 6 8 8 4 9 7 11 9 13 9 6 24 26 24 21 31 21 23 57 59 51 60 42 57 63 1 * * 1 2 * * 144 Making Health Care Decisions: Appendix B TABLE 4-11 INFORMATION ABOUT DIAGNOSIS OR PROGNOSIS IS WITHHELD: PHYSICIANS Q.: Sometimes a physician feels it is advisable to withhold information about diagnosis or prognosis from a patient. In your practice would you say this occurs once a day, once a week, once a month, a few times a year, or almost never? BASE ONCE A DAY ONCE A WEEK ONCE A MONTH A FEW TIMES A YEAR ALMOST NEVER NOT SURE *LESS THAN 0.5X. PHYSICIANS PATIENTS WITH PATIENTS SERIOUS ILLNESS UNDERSTAND ING MORE TREATMENT 0X TO 26% THAN 0% TO 70% 90% TO TOTAL _25% TO 50% 50% 69% TO 89% 100% 805 367 229 193 150 270 384 z z Z z z Zz z 2 1 4 2 4 2 2 7 5 7 10 9 8 S 12 10 13 16 11 14 11 32 31 36 2 36 33 30 46 52 40 39 38 43 52 1 * * 2 1 1 1 Surveys of Physicians and the Public 145 Qu TABLE 4-12 MOST COMMON REASON FOR WITHHOLDING INFORMATION: PHYSICIANS condition or treatment? What others? BASE PATIENT'S STABILITY/PERSONAL STRENGTH EMOTIONAL MATURITY/ANTICIPATED REACTION TO NEWS/ABILITY TO COPE WITH NEWS PATIENT'S ABILITY TO UNDERSTAND HIS CONDITION OR TREATMENT THE WISHES OF THE PATIENT'S FAMILY PATIENT'S PHYSICAL /MENTAL INCAPACITY SERIOUSNESS OF PATIENT'S CONDITION/ DIAGNOS1S/PROGNOS1S HOW MUCH THE PATIENT WANTS TO KNOW/ ASKS, PATIENT'S CURIOSITY EFFECT OF INFORMATION ON PATIENT'S ‘HEALTH AND WELL-BEING 1 TELL THEM EVERYTHING UNCERTAINTY/CERTAINTY OF DIAGNOS1S/ PROGNOSIS PATIENT IS TOO YOUNG/TOO OLD EFFECT OF INFORMATION ON PATIENT'S COMPLIANCE WITH TREATMENT SERIOUSNESS/RISKS OF TREATMENT TIME AVAILABILITY ALL OTHERS NOT SURE/REFUSED TOTAL PHYSICIANS — 805 2 34 28 21 17 XY? 16 What are the most common reasons for you to withhold information about TABLE 4-13 PATIENTS REQUESTING NONDISCLOSURE OF BAD NEWS: PHYSICIANS BASE: DON'T GIVE CANDID ASSESSMENT OF DIAGNOSIS TO ALL PATIENTS Q.: For those patients to whom you do not disclose bad news, how often is this because they tell you directly they don't want to know it? Would you say always, often, sometimes, rarely, or never? BASE ALWAYS OFTEN SOMETIMES RARELY NEVER NOT SURE PHYSICIANS PATIENTS WITH SERIOUS ILLNESS MORE 0% TO 26% THAN TOTAL 25% TO 50% 50% 668 "296 198 160 % 2 % 2 12 11 11 17 11 12 12 n 13 12 11 16 45 47 46 43 16 16 19 12 Ivl g x1puaddy :suoisa( aren) yiesy Sunyepy TABLE 4-14 INFORMATION WITHHELD FROM PATIENT DISCUSSED WITH FAMILY: PHYSICIANS Q.: In those situations in which you withhold information about treatment from a patient, do you usually, sometimes, rarely, or never discuss the information with another family member? PHYSICIANS SPECIALTY GENERAL/ INTERNAL FAMILY ALL TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS BASE 805 233 231 108 177 45 % % 2 2 % % USUALLY 80 84 75 68 88 73 SOMETIMES 10 10 14 6 7 9 RARELY 4 3 6 6 1 2 NEVER 3 2 3 10 1 2 NOT SURE 3 1 2 8 3 11 o1iqng ay} pue suedsAyd jo sAoaing LY1 Surveys of Physicians and the Public 149 CHAPTER 5: INFORMED CONSENT Introduction The physician's responsibilities regarding in- formation disclosure are governed by the doctrine of informed consent. However, this doctrine is not clear-cut; neither the requirements for disclosure placed upon the physician or the patient, nor the amount of disclosure required so that informed con- sent obtained is clearly articulated. It seems largely impossible to specify a single standard that would identify the level of disclosure required to ensure that a patient is adequately informed. This lack of specificity is clearly reflected in the clear and significant variation between states in the legal standards and case law generated around the issues of information disclosure and informed consent. This chapter presents data relevant to physicians' and the general public's understanding of informed consent and related issues, including perceived locus of responsibility for disclosure in the hope of stepping outside of the legal arena to examine the thoughts and preferences of those di- rectly involved in the consent process -- the health care providers and service consumers. Disclosure and the Law We have noted in the previous chapters that the vast majority of Americans expect their doctors to keep them fully informed about their condition and treatment. Virtually all Americans (97%) feel that patients have a right to be told everything about their condition and treatment. And almost all want their doctor to tell them everything about their medical condition, even if the news is unfavorable (94%). What if doctors do not tell their patients ev- erything that patients want to know? What if doc- tors do not give their patients a candid assessment of their condition and treatment? An overwhelming majority of the public (72%) think that doctors should be held legally liable for failing to fully and completely inform their patients of the risks associated with their medical procedures (Table 5-1). 150 Making Health Care Decisions: Appendix B The concept of informed consent is enormously attractive as the basis of patient autonomy in med- ical decisionmaking. The general public blames inadequate explanations from doctors for much of patient ignorance of treatment issues and concerns. The public sample was asked: "When a patient doesn't understand his medical treatment, how often is it because the doctor did not explain things well?" The survey finds that 44% of the public think that inadequate explanations by doctors are often responsible for lack of patient understanding of treatment, while only 11% think that this is rarely or never the case (Table 5-2). The doctrine of informed consent generally at- tempts to place physicians as responsible for en- suring such understanding prior to treatment. How- ever, the problems of such an approach are myriad. As noted earlier, physicians report that approxi- mately 21% of their patients are unable to under- stand most treatment issues, even if reasonable ef- fort is expended. Physicians withhold information from some patients, and the public agrees that some patients should be told less than others. Against this backdrop of mixed attitudes and behaviors concerning disclosure and its legal im- plications, we have investigated public and profes- sional attitudes toward, and experience with, con- sent to treatment. The Meaning of Informed Consent The term "informed consent" is familiar to most of the public and virtually all physicians. Both doctors and the general public were asked: '"What does the term 'informed consent' mean to you?" Only 21% of a national cross section of adults say that they aren't sure or don't know what informed consent means (Table 5-3). Only 3% of doctors say that they don't know what it means (Table 5-4). Both the public (43%) and doctors (58%) most commonly describe informed consent as informing pa- tients about their condition and recommended treat- ment. The public closely associates the term with permission or consent to treatment (43%), while the association with agreement to treatment is somewhat less common among physicians (26%). Only a handful of doctors (4%) or the public (7%) identify informed Surveys of Physicians and the Public 151 consent with the written consent form that patients are required to fill out prior to treatment. The disclosure of treatment risks and alterna- tives is mandatory under informed consent. Most doctors' explanations of informed consent reflect this understanding. Nearly one-half of the physi- cians (47%) describe informed consent in terms of explaining treatment risks to the patient. In ad- dition, 14% of physicians mention the explanation of treatment alternatives in their description of in- formed consent. By sharp contrast, only 8% of the public mention the disclosure of treatment risks as part of informed consent, and only 5% mention the disclosure of treatment alternatives. The crux of informed consent, however, is pa- tient understanding, not simply disclosure by the doctor. Many doctors seem to recognize this funda- mental concept. One-third of the physician sample (34%) describe informed consent in terms of patients fully understanding the nature of their condition and the method of treatment. Nearly one-quarter (23%) of the doctors interviewed identify informed consent as patient understanding of treatment risks. Patient understanding of treatment alterna- tives is mentioned by 8% of physicians. Although a substantial number of physicians rec- ognize patient understanding as the central dimen- sion of informed consent, most of the public do not make this connection. Only 9% of the public de- scribe informed consent in terms of patients under- standing of the nature of his or her condition and method of treatment. Further, only 2% of the public associate informed consent with patient wunder- standing of treatment risks, and 1% of the public associate it with understanding of treatment alternatives. Among the public, the most accurate appraisals of the meaning of informed consent are found among the college educated. Patient understanding of the nature of his or her condition and treatment is vol- unteered by 5% of those without a high school di- ploma, compared with 16% of those with a college degree. Patients being informed about treatment risks is associated with informed consent by 3% of those without a high school diploma, compared with 17% of college graduates. The most educated are also more likely to describe informed consent in 152 Making Health Care Decisions: Appendix B terms of patient understanding of treatment risks and treatment alternatives. Experience with life-threatening illnesses is also a factor in the understanding of informed con- sent. Only 16% of those who have faced a 1life- threatening illness don't know what informed consent means, compared with 23% of those who have never faced such an illness. However, the survey finds that those in poor health are least knowledgeable about informed consent. Fully 40% of those in poor health do not know what informed consent means, com- pared with 14% of those in excellent health. As might be expected from these data, respon- dents with the lowest household incomes are far less knowledgeable than those with a higher income about the meaning of informed consent. Respondents with the lowest household incomes are far more likely (39%) than respondents with the highest household incomes (13%) to report that they do not know what the term "informed consent" means. Similarly, those aged 65 and over (36%) are far less likely to know the meaning of informed consent than their younger counterparts (22%). Consent for Medical Procedures Do physicians obtain their patients' consent for the full range of medical treatments that may be covered by the law? The national sample of physi- cians was asked whether in their practice they usu- ally get written consent, oral consent, or no con- sent from patients for each of nine general proce- dures. The survey findings demonstrate that the type of consent that doctors obtain prior to treat- ment, if any, varies dramatically with the type of medical procedure. Two of the most widespread medical procedures are frequently conducted without obtaining any pa- tient consent. A majority of physicians (53%) re- port that they do not usually get consent from their patients for presciptions -- the most common of med- ical treatments (Table 5-5). Similarly, 43% of doc- tors report that they do not usually get patient consent for blood tests, a routine but clearly in- vasive procedure. Most doctors claim that they usually get patient consent for minor office surgery and the use of lo- Surveys of Physicians and the Public 153 cal anesthesia. However, oral rather than written consent is usual in these procedures. For minor office surgery, 49% of doctors report that they usu- ally obtain oral consent, compared with the 22% who obtain written consent. Similarly, 51% obtain oral consent for the use of local anesthesia, compared with the 19% who usually obtain written consent. Setting bones and diagnostic X-rays involving injections are rarely practiced without patient con- sent. In setting bones, the proportion of doctors who rely exclusively on oral comsent (19%) is about the same as the proportion who require written con- sent (17%). For diagnostic X-rays, the proportion of doctors who require written consent (30%) exceeds the proportion who rely exclusively on oral consent (23%). Procedures involving inpatient surgery and gen- eral anesthesia are almost invariably accompanied by written consent from patients. Not all physicians engage in surgical procedures that require the use of general anesthesia or inpatient surgery. Among those who do engage in these medical procedures, approximately 96% obtain written consent from their patients prior to treatment. Observation: The frequency of obtaining written consent to treatment rises directly with the risk of death or serious disability as a result of the medical procedure. Since these procedures are under- taken in hospitals, the institutional role in the specification of the type of consent obtained prior to treatment should not be dis- counted. Risks and side effects are relatively common with prescription medication, yet patient consent to prescription is rare. In this case, treatment occurs largely outside of an institu- tional setting. Patient Experience with Written Consent The legal requirements of informed consent do not specify written consent to treatment. Nonethe- less, it appears that the vast majority of doctors who engage in medical procedures involving the greatest risk to patients usually require written 154 Making Health Care Decisions: Appendix B consent. Whether written consent is a reasonable analogue to informed consent remains to be determined. Most of the public has had some experience with written consent forms. Indeed, nearly one-quarter of the public (24%) report having been asked to sign a written consent form for treatment within the past year (Table 5-6). The recent experience of these individuals with the medical consent process is quite illuminating. Most of those who have been asked to sign a written consent form in the past year have been sat- isfied with the doctor's explanation of the form. More than 7 out of 10 (71%) say that the doctor ex- plained the consent form to their satisfaction (Table 5-7). In looking at who was not satisfied with the doctor's explanation, individuals with no usual source of care (38%) stand out in sharp con- trast to those with private doctors (29%) or those usually accessing other sources of care (23%). The ability of the written consent form to help ensure informed consent receives mixed reviews. Seventy-two percent of those who have been asked to sign a consent form say that they felt they fully understood the risks of treatment they were about to undergo after reading the consent form (Table 5-8). However, one-quarter (25%) of those asked to sign a consent form report that they did not feel they ful- ly understood the risks of treatment after reading the form. As expected, educational attainment has a substantial impact on understanding. Reported un- derstanding of the written consent form increases from 57% of those without a high school diploma to 78% of those with a college degree. Those with no usual source of care are less likely to understand the consent form (63%) than those with a private doctor (73%) or another usual source of care (75%). Observation: A primary issue in informed consent is the gen- eral ability of the patient to understand infor- mation related to his or her condition and treatment. The less educated report greater difficulty with understanding treatment risks from either a written form or a doctor's explanation. Surveys of Physicians and the Public 155 Education, however, is not the only factor in- fluencing patient understanding of treatment. Individuals without a usual source of care -- who tend to be relatively well educated -- also have considerable difficulty. This suggests that the efforts of an interested doctor may be quite important. The survey finds that only 5% of those asked to sign a written consent form have ever refused treat- ment as a result of what was learned about treatment from the written consent form (Table 5-9). Those who refused treatment as a result of the information disclosed in a consent form are disproportionately drawn from those in fair (9%) and poor (9%) health, compared with those in excellent (4%) or good (4%) health. This may reflect differences in the ser- iousness of the medical procedures being proposed for these different groups. Physician Evaluation of Written Consent Because of the widespread use of written consent forms, the national sample of physicians was asked to evaluate the effect of written consent forms. Among the physicians surveyed, 22% feel that the written consent forms have had no effect (Table 5-10). Another 8% of doctors say that they are not sure what effects the written consent forms have had. The majority of physicians (63%) report some positive effects for the written consent forms, while those who report negative effects are in a distinct minority (29%). Interestingly, the pos- itive and negative effects attributed to the consent forms are virtually mirror images of each other. While 2% of doctors say written consent forms have eroded doctor-patient relationships, 16% say that they have improved doctor-patient communications. Although 6% of doctors say that the consent forms have increased lawsuits, 18% say that they have pro- vided legal protection for doctors and hospitals. Finally, while 14% of doctors say that written con- sent forms have increased patient fears and anxie- ties, 28% say that they have improved patient aware- ness concerning medical procedures and have given patients more realistic expectations. Written consent forms are clearly an instrument that cut both ways. The real question is whether, 156 Making Health Care Decisions: Appendix B on balance, they have had a more positive or more negative effect on the patient and the doctor. By a two-to-one margin, doctors appear to feel that the effect of written consent forms has been more pos-— itive than negative. The one negative criticism that has no positive counterpart is the observation by 6% of physicians that written consent forms may cause patients to avoid needed treatment. Indeed, we know that 5% of patients who are asked to sign a written consent form refuse treatment. Remember, however, that phy- sicians' belief in the necessity of treatment is one possible justification for withholding information that is not accepted by the public. From the stand- point of the public, the ability to refuse treatment -- even when the physician thinks the treatment is necessary —- is an effect of disclosure. Legal Standards for Disclosure The public believes that patients have a right to be told everything that they wish to know about their condition or treatment. When patients don't understand their treatment, most of the public be- lieves that it is at least sometimes the physician's fault. Moreover, nearly three-quarters of the pub- lic think that doctors should be held legally liable for failing to fully and completely inform their patients of the risks associated with their medical procedures. This begs the question, 'What standard of dis- closure must physicians satisly to fulfill their legal obligation to fully and completely inform their patients?" At the present time, most states have adopted (by law or court decision) one of two legal standards for informed consent. The oldest standard is the average reasonable practitioner standard, which states that the physi- cian should disclose the information that the av- erage, reasonable practitioner would disclose under the same or similar circumstances. In some states this has been superseded by the "new rule" of aver- age reasonable patient. This standard requires the physician to disclose the information that the av- erage, reasonable patient would consider relevant to his or her decision whether or not to accept the proposed treatment. Surveys of Physicians and the Public 157 A third standard, which has not yet been for- mally adopted as law in any state, would require the physician to disclose the information that the par- ticular patient being treated would consider rele- vant to the decision whether or not to accept the proposed treatment. Although this standard has yet to be adopted, the case law in one state {i.e., Or- lahoma) supports its application to litigation cen- tering on issues related to informed consent. Both the public and the physicians were asked which of these three standards would be the best standard to define physicians' legal obligation to disclose treatment information to their patients. As expected, physicians (26%) are somewhat more likely than the public (18%) to choose the average reasonable practitioner standard (Tables 5-12 and 5-13). The corollary is that the public (28%) is more likely than physicians (21%) to prefer the av- erage reasonable patient model. The differences are minor, however, compared with the agreement of sub- stantial numbers of physicians and the public on what they consider the best standard of disclosure. Strong pluralities among both the public (46%) and physicians (42%) consider the particular patient standard as the best standard for disclosure. Observation: The particular patient standard is probably the most popular standard with doctors and patients because it reflects medical norms. It would not be surprising if they were asked about their own behavior, if most doctors would probably de- scribe their professional responsibility in this way. It is doubtful that either doctors or the public recognize that this standard is poten- tially the most stringent and difficult to fulfill. The survey yields some evidence that doctors’ preferences for disclosure standards are affected by their age cohort. Doctors who have received their medical degree after 1973 are more likely than older doctors to prefer the average reasonable patient standard. Doctors who received their medical degree from 1967 to 1972 are more likely than either young- er or older doctors to prefer the average reasonable practitioner standard. Finally, doctors who re- ceived their medical degree prior to 1967 are the most likely to prefer the particular patient stan- 158 Making Health Care Decisions: Appendix B dard. It is noteworthy that these preferences are consistent with the norms or legal standards present during the formative years of the cohort's profes- sional education. Despite these differences, how ever, we should not neglect the main finding that the particular patient standard is the most popular disclosure standard among these groups. Do Physicians Know the Law? We noted earlier that the vast majority of phy- sicians are familiar with informed consent, and most seem to understand it. However, the particular re- quirements for conforming to the law with regard to informed consent will depend upon the standard for disclosure in the particular state. Since phy- sicians stand at legal risk for all medical proce- dures under the doctrine of informed consent, it seems likely that physicians would be quite aware of the legal standard pertaining to their practice. Thus, doctors were asked: 'Do you know which stan- dard is applicable in the state(s) in which you practice?" The survey finds that fewer than one-quarter of primary care physicians (23%) claim to know their legal obligation regarding information disclosure (Table 5-13). In examining differences between doc- tors, an even more striking finding emerges from the survey. Claiming knowledge of legal standards is least common among recent medical school graduates (17%), compared with those who graduated before 1967 (27%). And although surgeons report most frequently that they are knowledgeable about state standards of informed consent (30%), internists, who are equally at risk, least often report knowing their state's law (19%). The proportion of doctors who actually know the standard of disclosure in their state is consider- ably less than the proportion who claim to know it. Those who report that they know the applicable stan- dard in their state were asked to name that stan- dard. Respondent answers were then compared with their state's standard.® A bare majority of phy- ® classification of state by disclosure standard was conducted by the staff of the PCEMR. A state- by-state analysis of informed consent laws appears in Appendix L, Volume 3 of the PCEMR's report on Medical Decisionmaking. Surveys of Physicians and the Public 159 sicians (54%) (Table 5-14) who claim to know their state's standard can correctly identify the legal standard that applies in their state. Thus, only about 9% of practicing physicians in the United States can be said to be correctly aware of the par- ticular standards of informed consent under which they are legally responsible. Attitudes Toward Issues of Informed Consent There is very close agreement between physicians and patients on the value of written consent forms for doctor-patient communications. The survey finds that 65% of the public and 64% of physicians agree that written consent forms are helpful to doctor- patient communications (Table 5-15). The perception that consent forms have a communications value is somewhat more widespread among doctors whose case- load is primarily poor (72%) than among doctors with the fewest poor patients (61%). This pattern is paralleled among the public, where college educated individuals are more likely to doubt the communica- tions value of consent forms (40%), compared with those without a high school diploma (26%). The survey finds a much broader consensus among doctors, compared with the public, on the value of doctor-patient communications. A majority of the public (53%) feel that the time doctors spend dis- cussing diagnosis, prognosis, and treatment with patients would not be better spent in patient care. This attitude is shared by more than three quarters (77%) of doctors. By contrast, the attitude that the patient's rights to information should be protected by law is somewhat more widespread among the public (89%) than among doctors (76%). However, it receives over- whelming agreement among both groups. There is rel- atively little variation by income or education in the public's attitude toward patients' rights. How- ever, respondent age seems to affect the level of agreement expressed by the public regarding the leg- al protection of patients' rights for information. The survey finds 93% of those aged 18-34 in favor of legal rights, compared with 79% of those aged 65 and over. Although consent forms may be helpful in doctor- patient communications, this effect is not perceived 160 Making Health Care Decisions: Appendix B as their primary goal. More than three quarters of the public (79%) feel that the primary purpose of consent forms is to protect physicians from law- suits. Although a substantial proportion of physi- cians disagree (43%), nonetheless a majority of doc- tors (54%) say that the primary purpose of consent forms is to protect the physician. Even though doctors and patients agree that con- sent forms exist to protect the doctor, there is substantial disagreement about the legal standard itself. Only a minority of doctors (32%) feel that the legal requirements for informed consent are clear and explicit. By contrast, a majority of the public (53%) feel that the legal requirements are clear and explicit. From the standpoint of the public, a patient's signature on a written consent form establishes that the patient has given consent. Nearly the entire public (86%) agrees that a signed consent form es- tablishes consent, and 57% agree strongly with this notion. Although a majority of doctors (62%) agree, only 30% agree strongly. It is important to note that there is no ap- preciable difference in the attitude expressed as a function of having experienced a life-threatening illness in the family. This suggests that those who have signed consent forms have viewed them as of- ficial. Three of the seven survey items above were se- lected and employed as dependent measures in multi- ple regression analyses because of their importance to fully understanding attitudes toward informed consent. Two items -- "patient's rights to informa- tion about treatment risks and alternatives should be protected by law" and "the time spent discussing diagnosis, prognosis, and treatment with patients could be better spent in taking care of patients" -- are key to understanding attitudes towards informed consent. The third item -- "written consent forms are helpful to doctor-patient communications" -- bridges concerns about the impact of a specific as- pect of informed consent procedures and the broader, patient-provider relationship. Physicians' attitudes toward the legal protec- tion of patients' rights to treatment information were not significantly related to any of the eight Surveys of Physicians and the Public 161 predictors (i.e., practice descriptors and selected demographic characteristics) employed in the regres- sion. This finding, though somewhat surprising, may reflect the clear ambivalence that physicians appear to feel regarding the role of legal institutions in the practice of medicine. For example, within the context of medical decisionmaking, there is a de- crease in agreement noted when physicians are asked whether the law should sanction the administration of pain-relieving, potentially life-abbreviating drugs to a dying patient in severe distress who has asked for the drug and who understands the potential consequences of the treatment (53% say the law should allow this), compared with when physicians are asked if "they would be likely to" (84% say they would be likely to) or "is it ethically permissible to" (82% say it is ethically permissible to) ad- minister narcotics under these circumstances. When the same set of questions is asked concerning termination, the percentage of physicians endorsing legal sanctioning of physician behavior increases dramatically (26%), compared with those saying they would be likely to (2%), or feel it ethically per- missible to (4%) end a dying patient's life. These points will be elaborated upon in Chapter 8 of this report. Unlike the physician sample, public respondents’ attitudes concerning the legal sanctioning of their rights to information vary in predictable ways. Two factors emerge as significant predictors of at- titudes. Both respondents' hospitalization history (i.e., length of hospitalization) and age are in- versely related to the degree to which they endorse legal protection of patients' information rights (Table 5-16). Again, these data provide further evidence that those most dependent on medical ser-— vices are also least interested in playing an active role, and are most satisfied with the current prac- tices. The primary predictor of physicians' attitudes regarding the trade-offs between disclosure and treatment is their reported workload. As the number of patients treated per hour increases so too does physicians' belief that time spent talking could be better spent in treatment (Table 5-17). Similarly, as the percentage of old or ill patients in physi- cians' caseload increases, physicians' agreement with this position increases. 162 Making Health Care Decisions: Appendix B Two other predictors enter the final regression equation. The first is a factor weighted by loca- tion of the physician's practice (private office or hospital) and recency of training. The valence of this factor indicated that both recency and a hos- pital location are related to the attitude that the time spent in discussing treatment information is an essential part of treatment and is not better spent in physically taking care of patients. Similarly, as the percentage of persons perceived to be capable of understanding treatment information comprise a greater proportion of the physician's caseload, phy- sicians are more apt to feel that the time spent discussing treatment is valuable and not easily traded off against increased time spent in direct treatment. Clearly, these data again demonstrate that char- acteristics of a physician's patient population have a strong effect on attitudes towards disclosure. However, the importance of the physical constraint of time itself (i.e., workload) and other practice characteristics (i.e., office location and recency of training) are also found to be important pre- dictors of physicians' attitudes toward information disclosure. Regression analysis of public data yield four significant predictors of attitudes concerning whether the time spent discussing treatment with patients could be better spent in patient care. Socioeconomic status and minority group membership emerge as the two primary predictors of the extent of agreement with the statement that time spent in discussion is better spent in treatment (Table 5-18). Among those with the lowest socio- economic status and among minority groups, the pref- erence for care as opposed to discussion is high- est. Again, we see that those with the least educa- tion and with the lowest level of household income (socioeconomic status), as well as minority group respondents, voice a much more diminished need or desire for information than do wealthier, more edu- cated, or white respondents. These findings mirror the results of similar analyses performed on physi- cian data discussed immediately above. Among the public, females and those who think patients are the best judge of the amount of infor- mation that should be disclosed are far more likely than males or those who think the physician is the Surveys of Physicians and the Public 163 best judge to value physician time spent in discussion. The final item analyzed by multiple regression techniques focused on the impact of written consent forms on doctor-patient communication. The only significant predictor of physicians' attitudes on this measure is that physicians treating relatively high concentrations of college educated patients and low concentrations of poor patients feel most strongly that written consent forms are helpful to physician-patient communication. However, unlike many of the survey items dis- cussed in this report, the data pattern observed among the public is directly at odds with that ob- served among physicians. Public respondents with a high socioeconomic status (high education and high income) disagree most vehemently with the notion that written consent forms are helpful to doctor- patient communication (Table 5-19). Recall that physicians treating high percentages of highly educated patients and low concentrations of poor patients are most optimistic about the impact of written consent forms. In addition to respondents' socioeconomic stat- us, experience with a life- ~-threatening illness and usual source of care enter the regression equation as significant predictors. As personal experience with a life-threatening illness increases, disagree- ment with the statement, ''written consent forms are helpful to doctor-patient communications," increases sharply. Thus, persons most likely to have first- hand experience with written consent forms are least likely to feel consent forms facilitate communica- tion with the physician. On the other hand, those obtaining their routine care in a physician's office feel most strongly that written consent forms are helpful in establishing better doctor-patient com- munications, compared with those usually receiving care at other medical facilities or those having no usual source of care. Again, these are persons who most probably have established a communicative rela- tionship with their physician and who therefore probably have received a substantial amount of verbal explanation of the form itself. In this light, the written form may be perceived as a supplemental tool to be used in obtaining information. TABLE 5-1 PHYSICIAN LIABILITY FOR FAILURE TO DISCLOSE RISKS: GENERAL PUBLIC Q.: Should a doctor be held legally liable for failing to fully and completely inform his patient of the risks associated with a medical procedure? BASE YES, SHOULD BE LIABLE NO, SHOULD NOT BE LIABLE NOT SURE GENERAL PUBLIC EDUCATION LESS THAN HIGH HIGH SCHOOL SCHOOL COLLEGE TOTAL GRADUATE GRADUATE GRADUATE 1251 242 726 280 % % A % 72 70 72 73 21 21 22 18 7 9 6 8 P91 dg x1puaddy :suorsioe(q are) yjeay Sune TABLE 5-2 PHYSICIAN RESPONSIBILITY FOR PATIENTS' UNDERSTANDING OF TREATMENT: GENERAL PUBLIC Q:: When a patient doesn't understand his medical treatment, how often is it because the doctor did not explain things well? Would you say always, often, sometimes, rarely, or never? GENERAL PUBLIC EDUCATION LESS - THAN HIGH HIGH SCHOOL SCHOOL COLLEGE TOTAL GRADUATE GRADUATE GRADUATE BASE 1251 242 726 280 x z Zz % ALWAYS 12 17 12 8 OFTEN 32 14 34 42 SOMETIMES 37 41 37 35 RARELY 8 10 8 7 NEVER 3 4 3 2 NOT SURE 7 14 6 5 o1[qnd 9Y) pue sueIsAY{ Jo sfeaing e917 166 Making Health Care Decisions: Appendix B TABLE 5-) INFORMED COWSENT: GEWERAL PUBLIC v. 1 vould like to turn vour attention to o final set of questions sbout informed consent. Wnat does the ters informed consent” wean to you! What else’ LIC Al THAN HIGH 18-34 33-50 Si-6e AND $CHOOL SCHOOL COLLEGE _RIPORTED WEALTH STATUS TAL YEARS YLAKS YEARS 3 CRAD.ATE GRAD ATE GRAZ. ATE DX.7 NT GC: Falk Fok BASE 1251 “8h 3a 2 ial 28 0 350 Te. 73 1 3 3 3 3 3 3 1 1 1 «) ~“ “ “4 3 n ““ 2 n “ 2 22 2 El « “a Ld . 26 n “ “7 » 3 1 2 26 2 DEZISIONS, DOCTOR NLEDS INPUT, PATIENT -5 DOITOR WOW TO TREAT 19 20 20 20 13 1 1 2» 12 12 nn 7 1 ic PATIENT UNDER- STANDS NATURE OF COKDITION AND METHOD OF TRLATNENT 9 . 12 . 7 s . 1 3 . “. ? + “. PATIENT INFORMED/ TOD ABOUT TREAT- MENT RISKS . ’ 10 . 3 3 . ” . 3 2 1 . . WRITTEN FORM PATIENT FILLS OUT, WRITTEN CONSENT 7 ’ . 0 . 3 1 . . “ 1 “ . CONSENT TO TERMINATION WONRESUSCITATION, PORIQDINC LIFE SUPPORT OR OTHER TREAT NT . . . . . . . 7 . 3 3 . “ 3 PATIENT INFORMED/ TOLL ABOUT TREAT- MENT ALTERNATIVES CINILUDING : MO TREAD NT) s s . . 3 1 . 0 3 3 3 - 2 3 CONSENT TO SURGERY 2 3 2 2 1 2 3 1 1 2 1 1 1 2 PATIENT UNDERSTANDS TREATMENT RISKS 2 2 . . od 1 1 . 2 1 - 2 1 1 MEGAL DOCUMENT 1 3 1 3 . . 2 . 1 1 . - 1 . PATIENT DNDERSTANDS TRLATHENT ALTERMA- TIVES INCLUDING FOREGOING TREATWNENT 1 1 1 = - - 1 2 1 1 . * 1 . ALL OTHERS 1 2 . 1 1 1 2 1 1 1 1 - 1 i WOT SURE/RLPUSED 1 2 12 10 » “0 1 10 . nn ’ 1s Surveys of Physicians and the Public 167 P-32 TABLE 5-4 INFORMED CONSENT: PHYSICIANS Q.: Now I would like to turn your attention to 8 final set of questions about informed consent. What does the term ‘informed consent ' mesn to you? What else? PHYSICIANS PATIENTS PATIENTS WITH UNDERSTANDING PRACTICE SERIOUS ILLNESS TREATMENT LOCATION 0% 26% MORE 0% 70% 90% T0- THAN TO TO TO TOTAL OFFICE HOSPITAL 25% 502 S02 69% 8921 100 2 id BASE “5% $513 37 38 13 13 TO 1 2 1 : 1 2 “3 “H ALL OTHER INFORMING DISCLOSING, INCLUDING INFORMING PATIENT ABOUT NATURE OF HIS CONDITION AND RECOMMENDED TREATMENT se 59 56 s8 57 60 $7 59 59 DISCLOSING’ EXPLAINING TREATMENT RISKS TO PATIENT &7 46 47 so 38 50 45 47 48 ALL OTHER PATIENT UNDERSTANDING, INCLUDING GENERAL "PATIENT FULLY UNDERSTANDS NATURE OF HIS CONDITION AND METHOD OF TREATMENT" 34 32 41 33 36 34 36 33 34 PATIENT GIVES PERMISSION/ AGREES TO TREATMENT 26 26 28 23 32 25 27 26 26 PATIENT UNDERSTANDS TREATMENT RISKS 23 21 28 23 26 21 23 2 22 INFORMING /DISCLOSING/ EXPLAINING TREATMENT ALTERNATIVES, INCLUDING WO TREATMENT, TO PATIENT 16 16 14 13 15 1 13 16 13 PATIENT 1S ABLE TO DECIDE/ IS ALLOWED TO CHOOSE HIS TREATMENT 9 8 8 7 10 10 11 11 6 PATIENT UNDERSTANDS TREATMENT ALTERNATIVES, INCLUDING FOREGOING TREATMENT 8 8 10 8 10 7 7 9 8 FORM/DOCUMENT PATIENTS FILL OUT 4 [3 3 3 7 2 3 4 4 LEGAL DOCTRINE, LAN 3 3 2 2 3 4 3 4 2 ALL OTHERS 1 1 2 1 2 1 1 1 1 NOT SURE/NOTHING/ ITS MEANINGLESS 3 o 2 bo 2 4 4 4 3 168 Making Health Care Decisions: Appendix B TABLE 5-5 TYPES OF CONSENT: PHYSICIANS Q.: In your practice, do you usually get written consent, oral consent, or no consent from patients before each of the following procedures? TOTAL PHYSICIANS NOT APPLI- WRITTEN . ORAL BOTH NO CABLE NOT BASE: 805 CONSENT CONSENT (VOL.) CONSENT (VOL. ) SURE INPATIENT SURGERY 2 62 3 11 * 24 * MINOR OFFICE SURGERY 2 22 49 8 6 14 * SETTING BONES 2 17 19 4 4 55 1 GENERAL ANESTHESIA x 48 2 ? * 42 % LOCAL ANESTHESIA 2 19 51 6 13 10 1 DIAGNOSTIC X-RAYS INVOLVING INJECTIONS 2 30 23 5 7 34 1 BLOOD TESTS 2 2 49 * 43 6 1 PRESCRIPTIONS 2 1 42 2 53 2 1 RADIAT ION THERAPY 2 28 8 5 2 55 2 *LESS THAN 0.5%. TABLE 5-6 WRITTEN CONSENT: GENERAL PUBLIC Q.: Written consent forms are now standard practice in many hospitals for surgery and certain other medical procedures. Within the last year, have you been asked to sign a written consent for treatment? GENERAL PUBLIC EDUCATION LESS THAN HIGH HIGH REPORTED HEALTH STATUS SCHOOL SCHOOL COLLEGE TOTAL EXCELLENT GOOD FAIR POOR GRADUATE GRADUATE GRADUATE BASE 1251 482 S50 144 73 242 726 280 7 2 2 2 % % % 2 YES, HAVE BEEN ASKED 24 22 23 28 35 19 24 26 NO, HAVE NOT BEEN ASKED 76 78 77 72 65 81 76 74 orqng ay) pue suerIsAyd jo searing 691 G-53A TABLE 5-7 PHYSICIAN EXPLANATION OF CONSENT FORM: GENERAL PUBLIC BASE: HAVE BEEN ASKED TO SIGN WRITTEN CONSENT Q.: Did the doctor explain the consent form to your satisfaction? GENERAL PUBLIC EDUCATION USUAL SOURCE LESS THAN OF CARE HIGH HIGH DOCTOR SCHOOL SCHOOL COLLEGE TOTAL NONE OFFICE OTHER GRADUATE GRADUATE GRADUATE BASE 288 48 147 93 242 726 280 % Z % % Z % % YES, HE DID 71 62 72 76 59 77 65 NO, HE DID NOT 29 38 29 23 41 23 35 041 dg xtpuaddy :suorsioa(] aren yjesy Sune G-53B TABLE 5-8 UNDERSTANDING OF RISKS DESCRIBED IN CONSENT FORM: GENERAL PUBLIC BASE: HAVE BEEN ASKED TO SIGN WRITTEN CONSENT Q.: After reading the consent form, did you feel that you fully understood the risks of the treatment you were going to undergo? GENERAL PUBLIC EDUCATION USUAL SOURCE LESS THAN OF CARE HIGH HIGH DOCTOR SCHOOL SCHOOL COLLEGE TOTAL NONE OFFICE OTHER GRADUATE GRADUATE GRADUATE BASE 288 48 147 93 48 175 65 % % 7% % % % % YES, FULLY UNDERSTOOD 72 63 73 75 57 74 78 NO, DID NOT 25 35 26 20 37 24 21 NOT SURE 3 2 1 5 6 3 1 o1[qng ay) pue sueIsAyd jo sheaing 141 G-53C TABLE 5-9 REFUSAL OF TREATMENT BECAUSE OF CONSENT FORM: GENERAL PUBLIC BASE: HAVE BEEN ASKED TO SIGN WRITTEN CONSENT Q.: Have you ever refused treatment because of what you learned about the treatment from the written consent form? GENERAL PUBLIC EDUCATION LESS THAN HIGH HIGH REPORTED HEALTH STATUS SCHOOL SCHOOL COLLEGE TOTAL EXCELLENT GOOD FAIR POOR GRADUATE GRADUATE GRADUATE BASE 288 99 122 39 28 48 175 65 % Z % % Z z % x YES, REFUSED 5 4 4 9 9 6 6 2 NO, DID NOT 95 96 96 91 87 94 94 98 NOT SURE * - - = 4 - 1 5 *LESS THAN 0.5%. ZL dg x1puaddy :suoistoa(] aren yj[eay Suryepy Surveys of Physicians and the Public 173 Q.: Written consent forms are now standard practice in many hospitals for and certain other medical procedures. surgery TABLE 5-10 EFFECT OF WRITTEN CONSENT FORMS: PHYSICIANS been the effect of these forms? What else? BASE POSITIVE IMPROVED PATIENT AWARENESS ABOUT MEDICAL PROCEDURES, MORE REALISTIC PATIENT EXPECTATIONS ABOUT RISKS AND CONSEQUENCES OF TREATMENT, PATIENTS KNOW MORE ABOUT THEIR TREATMENT LEGAL PROTECTION FOR DOCTORS AND HOSPITALS IMPROVED DOCTOR-PATIENT COMMUNICATIONS, PATIENTS ASK MORE /BETTER QUESTIONS, BETTER DOCTOR-PATIENT RELATIONSHIPS ALL OTHER POSITIVE COMMENTS NEGATIVE INCREASED PATIENT FEARS/ANXIETY/ WORRIES ABOUT THEIR CONDITION AND TREATMENT CAUSES PATIENTS TO AVOID NECESSARY TREATMENT, REDUCES PATIENT COMPLIANCE WITH THERAPY DOESN'T PROVIDE DOCTORS OR HOSPITALS WITH LEGAL PROTECTION, INCREASES LAWSUITS REDUCES / IMPAIRS /ERODES DOCTOR-PATIENT RELATIONSHIP, MAKES PATIENTS DISTRUST THEIR DOCTORS ALL OTHER NEGATIVE COMMENTS NO EFFECT ALL OTHER MISCELLANEOUS COMMENTS NOT SURE In your experience, what has TOTAL PHYSICIANS 805 x 63 28 16 29 174 Making Health Care Decisions: Appendix B G-51 TABLE 5-11 BEST DISCLOSURE STANDARD: GENERAL PUBLIC Q.: Three general standards have been proposed to define a physician's legal obligation to disclose treatment information to his patients. I would like to know which of these standards do you feel would be the best disclosure standard? GENERAL PUBLIC EDUCATION LESS THAN HIGH HIGH SCHOOL ~~ SCHOOL COLLEGE TOTAL GRADUATE GRADUATE GRADUATE BASE 1251 2462 728 280 2 2 2 2 PHYSICIAN SHOULD DISCLOSE THE INFORMATION THAT THE AVERAGE REASONABLE DOCTOR WOULD ISCLOSE UNDER THE SAME OR SIMILAR CIRCUMSTANCES 18 16 17 20 PHYSICIAN SHOULD DISCLOSE THE INFORMATION THAT THE AVERAGE REASONABLE PATIENT WOULD CONSIDER LEVANT TO HIS DECISION WHETHER OR NOT TO ACCEPT THE PROPOSED TREATMENT 28 24 30 28 PHYSICIAN SHOULD DISCLOSE THE INFORMATION THAT THE PARTICULAR PATIENT BEING TREATED WOULD CONSIDER RELEVANT TO HIS DECISION WHETHER OR NOT TO ACCEPT PROPOSED TREATMENT 46 43 47 45 NO GENERAL STANDARD OF DISCLOSURE 1S DESIRABLE (VOL.) 1 2 1 3 NOT SURE 7 15 6 4 Surveys of Physicians and the Public 175 P-34 TABLE 5-12 BEST DISCLOSURE STANDARD: PHYSICIANS Q.: Three general standards have been proposed to define a physician's legal obligation to disclose treatment information to his patients. I will read you these standards and would like to know which one you feel would be the best disclosure standard. PHYSICIANS OBTA MEDICAL DEGREE 1966 1967 1973 AND TO AND TOTAL BEFORE 1972 OVER BASE 805 441 97 266 2 2 z 2 PHYSICIAN SHOULD DISCLOSE THE INFORMATION THAT THE AVERAGE, REASONABLE PRACTITIONER WOULD DISCLOSE UNDER THE SAME OR SIMILAR CIRCUMSTANCES 26 25 33 23 PHYSICIAN SHOULD DISCLOSE THE INFORMATION THAT THE AVERAGE, REASONABLE PATIENT WOULD CON- SIDER RELEVANT TO HIS DECISION WHETHER OR NOT TO ACCEPT THE PROPOSED TREATMENT 21 18 25 27 PHYSICIAN SHOULD DISCLOSE THE INFORMATION THAT THE PARTICULAR PATIENT BEING TREATED WOULD CONSIDER RELEVANT TO HIS DECISION WHETHER OR NOT TO ACCEPT PROPOSED TREATMENT 42 45 35 39 NO GENERAL STANDARD OF DISCLOSURE 1S DESIRABLE (VOL.) 8 9 6 6 NOT SURE b 3 3 5 P-35 TABLE 5-13 APPLICABLE DISCLOSURE STANDARD: PHYSICIANS Q.: Do you know which standard is applicable in the state(s) in which you practice? PHYSICIANS OBTAINED SPECIALTY MEDICAL DEGREE GENERAL/ 1966 1967 1973 INTERNAL FAMILY ALL AND TO AND TOTAL MEDICINE PRACTICE OB/GYN SURGERY OTHERS BEFORE 1972 OVER BASE 805 233 231 108 177 45 441 97 266 % % % % % % % % % YES, KNOW 23 19 23 26 30 13 27 24 17 NO, DOES NOT KNOW 76 81 77 73 70 87 73 76 82 9.1 fd x1puaddy :suoisioa(] aren Uj[eay Sunyey Surveys of Physicians and the Public 177 P-35 TABLE 5-14 KNOWLEDGE OF STATE LAW: PHYSICIANS Q.: Do vou know which standard is applicable in the state(s) in which you practice’ NO 76.5 (616) YES 23.4 (188) PERCENT OF TOTAL PHYSICIAN SAMPLE NOT SURE N=13 LX INCORRECT N = 60 7.42 NOT ASSIGNABLE N= 53 6.6% CORRECT N= 72 8.9% PERCENT CORRECT OF ALL PHYSICIANS 72/805 = 8.94% PERCENT CORRECT OF PHYSICIANS SAYING THEY KNOW WHICH STANDARD IS APPLICABLE IN STATES THAT HAVE ADOPTED A CLEARLY IDENTIFIABLE STANDARD 72/132 = 54.5% WHETHER PHYSICIAN CAN CORRECTLY NAME STANDARD OF DISCLOSURE IN ANOTHER STATE PHYSICIANS WHO SAY THEY DO KNOW TOTAL WHICH LAW 1S PHYSICIAN APPLICABLE SAMPLE BASE 188 805 2 1 CORRECT 38 9 INCORRECT 32 7 NOT ASSIGNABLE 28 7 NOT SURE 2 » *LESS THAN 0.52. 178 Making Health Care Decisions: Appendix B G-54 P-38 TABLE 5-15 ATTITUDES ABOUT INFORMED CONSENT: GENERAL PUBLIC, PHYSICIANS Q.: Now, I am going to read you a series of statements that have been made about informed consent, and I would like to know whether you tend to agree strongly, agree somewhat, disagree somewhat, or disagree strongly. BASE: 1251 GENERAL PUBLIC; AGREE AGREE ~~ DISAGREE DISAGREE NOT 805 PHYSICIANS. STRONGLY SOMEWHAT SOMEWHAT STRONGLY SURE PATIENTS' RIGHTS TO INFORMATION ABOUT TREATMENT RISKS AND ALTERNATIVES SHOULD BE PROTECTED BY LAW GENERAL PUBLIC 63 26 5 4 PHYSICIANS 38 38 10 10 » sw TIME SPENT DISCUSSING DIAGNOSIS, PROGNOSIS, AND TREATMENT WITH PATIENTS COULD BE BETTER SPENT IN TAKING CARE OF PATIENTS GENERAL PUBLIC PHYSICIANS 22 THE LEGAL REQUIREMENTS FOR OBTAINING INFORMED CONSENT ARE CLEAR AND EXPLICIT GENERAL PUBLIC PHYSICIANS 22 31 21 11 16 13 19 28 30 10 2 THE PRIMARY PURPOSE OF CONSENT FORMS IS TO PROTECT PHYSICIANS FROM LAWSUITS GENERAL PUBLIC PHYSICIANS 48 31 12 5 20 34 24 19 0 ws WRITTEN CONSENT FORMS ARE HELPFUL TO DOCTOR-PATIENT COMMUNICATIONS GENERAL PUBLIC PHYSICIANS 29 36 19 11 26 38 18 15 0 (“XV THE REQUIREMENTS OF INFORMED CONSENT PUT TOO MUCH EMPHASIS ON DISCLOSURE OF REMOTE RISKS ENERAL PUBLIC PHYSICIANS 13 31 30 13 12 31 42 16 8 4 A PATIENT'S SIGNATURE ON A WRITTEN CONSENT FORM ESTABLISHES THAT THE PATIENT HAS GIVEN CONSENT GENERAL PUBL PHYSICIANS 57 29 9 3 2 30 32 20 14 4 LR] TABLE 5-16 PREDICTORS OF PUBLIC ATTITUDES ABOUT WHETHER PATIENT RIGHTS TO INFORMATION SHOULD BE PROTECTED BY LAW DEPENDENT VARIABLE: G-54A MULTIPLE R 0.14709 ANALYSIS OF VARIANCE R SQUARE 0.02164 DF SUM OF SQUARES MEAN SQUARE ADJUSTED R SQUARE 0.01983 REGRESSION 2 13.41605 6.70803 STANDARD ERROR 0.74913 RESIDUAL 1081 606.64587 0.56119 F = 11.95323 SIGNIF F = 0.0000 VARIABLE B SE B BETA i) SIG T HOSPITAL DAYS 0.00991 0.00267 0.11212 3.717 0.0002 AGE 0.00377 0.00130 0.08741 2.898 0.0038 (CONSTANT) 1.30563 0.06056 21.560 0.0000 FOR BLOCK NUMBER 1 PIN = 0.050 LIMITS REACHED. o1[qng 8Yy} pue SueIISAYd jo shoring 6L1 DEPENDENT VARIABLE: MULTIPLE R R SQUARE ADJUSTED R SQUARE STANDARD ERROR TABLE 5-17 PREDICTORS OF PHYSICIANS' PREFERENCES BETWEEN TIME SPENT IN PATIENT CONSULTATION AND PATIENT CARE P-38B 0.24098 ANALYSIS OF VARIANCE 0.05807 DF 0.05278 REGRESSION 4 0.91848 RESIDUAL n2 F = 10.97403 SUM OF SQUARES 37.03098 600.64660 SIGNIF F = 0.0000 == VARIABLES IN THE EQUATION -=====mmm—mmmm meme VARIABLE B SE B BETA I SIG T WORKLOAD 0.17053 0.03313 0.19676 5.147 0.0000 OLD/ILL PATIENTS 0.15785 0.05709 0.10565 2.765 0.0058 PRACTICE LOCATION 0.06382 0.02290 -0.10287 -2.787 0.0055 PATIENTS UNDER- STAND TREATMENT -0.00396 0.00179 -0.08163 -2.209 0.0275 (CONSTANT) 1.76128 0.15606 11.286 0.0000 FOR BLOCK NUMBER 1 PIN = 0.050 LIMITS REACHED. MEAN SQUARE 9.25774 0.84360 081 fd x1puaddy :suorsa(] aren) yj[eay Sunyey TABLE 5-18 PREDICTORS OF PUBLIC'S PREFERENCES BETWEEN TIME SPENT IN PATIENT CONSULTATION AND PATIENT CARE DEPENDENT VARIABLE: G-54B MULTIPLE R 0.27649 R SQUARE 0.07644 ADJUSTED R SQUARE 0.07299 STANDARD ERROR 1.07924 ANALYSIS OF VARIANCE REGRESSION RESIDUAL F = 22.12074 SE B BETA DF SUM OF SQUARES 4 103.06190 1069 1245.13442 SIGNIF F = 0.0000 T SIG T VARIABLE B SOCIOECONOMIC STATUS -0.19036 RACE 0.42065 BEST JUDGE OF DISCLOSURE 0.26941 SEX -0.16816 (CONSTANT) 1.69143 0.03083 -0.18476 0.08955 0.13890 0.06597 0.12022 0.06671 -0.07501 0.17808 FOR BLOCK NUMBER 1 PIN = 0.050 LIMITS REACHED. -6.174 0.0000 4.698 0.0000 4.084 0.0000 -2.521 0.0119 9.498 0.0000 MEAN SQUARE 25.76547 1.16477 o1[qnd oY} pue sueroisAyd jo sfeaing 181 TABLE 5-19 PREDICTORS OF PUBLIC'S ATTITUDES TOWARD CONSENT FORMS: DEPENDENT VARIABLE: G-54G DOCTOR-PATIENT COMMUNICATIONS MULTIPLE R 0.14678 ANALYSIS OF VARIANCE R SQUARE 0.02155 DF SUM OF SQUARES ADJUSTED R SQUARE 0.01877 REGRESSION 3 21.81824 STANDARD ERROR 0.96819 RESIDUAL 1057 990.82529 F = 7.75848 SIGNIF F = 0.0000 mmm VARIABLES IN THE EQUATION —==—====--cmmmmmoo__ VARIABLE B SE B BETA 1 SIG T SOCIOECONOMIC STATUS -0.09227 0.02736 0.10270 3.372 0.0008 LIFE THREATENING ILLNESS 0.13297 0.04457 0.09089 2.983 0.0029 LOCATION OF CARE -0.07847 0.03858 -0.06202 -2.034 0.0422 (CONSTANT) 2.13198 0.11122 19.169 0.0000 FOR BLOCK NUMBER 1 PIN = 0.050 LIMITS REACHED. MEAN SQUARE 7.27275 0.93739 z81 dg x1puaddy :suorsioa( aren yieey Sunyepy Surveys of Physicians and the Public 183 CHAPTER 6: DESCRIPTION OF MEDICAL DECISIONMAKING Introduction The dynamic qualities of the doctor-patient re- lationship are perhaps nowhere better portrayed than when viewed in the context of medical decision- making. Moreover, it is within this context that the actual impact of the doctrine of informed con- sent must be judged. One of the primary goals of the informed consent process is to educate the con- sumers of medical services so that they can parti- cipate freely and responsibly in the therapeutic decisionmaking process. This study attempts to assess the public's role in the medical decisionmaking process by measuring a series of self-reported behaviors related to infor- mation seeking and patient autonomy. At the same time, physicians' perceptions regarding the locus of treatment decisionmaking responsibility are assessed. Locus of Decisionmaking: The Public's View The public reports substantial input into the medical decisionmaking process, both in terms of seeking information and in persevering to see that their wishes regarding treatment are honored. This is consistent with their views expressed on the top- ic of information disclosure (See Chapters 3 and 4), including their stated desire for full disclosure of medical information regarding their condition and treatment. One clear indicator of patient input into the decisionmaking process lies in the public's report of the locus of responsibility for their decision to undergo a surgical procedure. Persons undergoing a surgical operation certainly have experience with informed consent procedures. In this instance, when an invasive procedure is proposed and a patient's life may be at substantial risk, the perceived locus of decisionmaking responsibility is especially il- lustrative of patients' general desire to maintain some semblance of control over the medical decision- making process regarding their treatment. When asked about the locus of decisionmaking regarding undergoing a recent surgical procedure, the majority (52%) of qualified respondents note that their own 184 Making Health Care Decisions: Appendix B judgment was equally as important as the doctor's (31%), or was predominant (21%), in the final decision to have the procedure performed (Table 6-1). Still, 2 of 5 respondents (40%) -- a very sizable minority -- note that it was the physician's judgment and not their own that served as the pri- mary motivation for undergoing the surgical procedure. Other indicators regarding patient input into the decisionmaking process also portray the patient as active. Surprisingly, more than 1 in 5 (21%) of the respondents in the general public sample report that they have refused to undergo treatment recom- mended by a doctor (Table 6-2). Although the ma- jority of the public (79%) note that they have never rejected a physician's prescribed treatment, the sizable minority of people actively rejecting treat- ment portrays the physician-patient relationship as dynamic and potentially volatile. In a similar vein, the public was asked if they had ever changed physicians because they disagreed with them. More than one-third (36%) of the re- spondents surveyed note that they have changed doc- tors because they disagreed with them (Table 6-3). Clearly this movement of the public among providers reveals that people feel free to seek alternative sources of treatment. Underlying this freedom must be consumers' perceptions that adequate treatment can be obtained from a number of alternative sources. This finding is quite divergent from what might be expected if the public viewed physicians as extraordinary individuals and not as a group of sim- ilarly qualified and capable people. In point of fact, when more than one-third of the general public report feeling free enough to move from one physi- cian to another -- that is, to shop around a bit -- there exists a substantial hint that the context within which medical practice exists is becoming more firmly rooted in the reality of the market- place, and the public's view of physicians as ser- vice providers is becoming more deeply entrenched. Looking at the issue of changing physicians from another angle, we find that 1 in 5 (20%) of the pub- lic sample report that physicians have suggested that they find another doctor because of disagree- ments regarding diagnosis or recommended treatment (Table 6-4). Although we cannot describe the intent of this recommendation on the doctors’ part -- Surveys of Physicians and the Public 185 whether the physician makes his or her recommenda- tion solely to benefit the patient (i.e., to make sure that the patient becomes aware of a consensus of medical opinion and obtains treatment), or be- cause he or she feels that this patient cannot be worked with further, or both -- we can use these data to better interpret the impetus of the patient's decision to change physicians. Of the general public reporting that they have changed physicians because of disagreements, more than one fourth (27%) note that a treating physician had at least once recommended that they seek medical advice and help elsewhere (Table 6-5). Again, these data do not enable strict causal inferences to be drawn, but they do illuminate the probable causal sequence in many instances. It is not infeasible that in roughly 1 of 4 patients' decisions to change physi- cians it was the physican who initiated the move, while in the remaining 73% of such instances, the patients initiated the move. These data again high- light the dynamic give and take interaction between physician and patient observed in this survey. These data, taken together, help explain pre- viously cited differences in the reported levels of satisfaction with physicians in general and with respondents’ specific physicians. Recall that 76% of the public sample report that they are satisfied with physicians in general (35% are very satisfied; 41% are somewhat satisfied). When asked about their own physician, 84% of respondents report that they are satisfied (49% are very satisfied; 35% are some- what satisfied). The substantial increase in re- ported satisfaction with respondents’ own physi- cians, compared with physicians in general, may in part be the result of their seeking out physicians whom they like or trust better. The movement of more than one-third of the respondent sample (36%) from one physician to another because of a disagree- ment could readily explain registered differences between satisfaction with physicians in general and satisfaction with physicians whom they are now seeing. A final indicator of the public's level of ac- tivity or participation in the medical decision- making process focuses on information seeking. The public was asked whether they have ever sought a second opinion from another doctor. We must note at the outset of this discussion that seeking second opinions does not necessarily mirror or serve as a 186 Making Health Care Decisions: Appendix B conceptually clean substitute for voluntary exposure to information. Although many people may seek sec- ond opinions because they are uncertain and need to obtain information, many more will do so because their physician has directed them to do so or be- cause they simply cannot, or do not, want to believe the diagnosis of their condition or the necessity of the recommended treatment, and are seeking discon- firmation from another credible source. However, the measure can and does yield illustrative find- ings, and though we cannot be certain that the mag- nitude of the responses reflects purely information- seeking behaviors, we can be relatively sure that the pattern of the data does meaningfully reflect this issue of interest, and thereby describes more fully the patient's role in the medical decision- making process. The majority of persons queried report that they have actively sought out a physician other than their administering physician to obtain a second opinion (51%) (Table 6-6). Still, a large minority report that they have never sought a second opinion (49%) regarding a physical condition, complaint, or treatment. Voluntary exposure to multiple sources of information has been demonstrated to reflect a desire to obtain certainty on an issue. Generally, if the person had no intention of translating this certainty into action, there would be no reason to initiate the information search in the first place. Thus, we can view this exposure to alternative sources of information as a means used by the con- sumer of medical services to become educated so that he or she can better participate in decisions re- garding his or her own case. Again, though we may not know exactly from these data the proportion of respondents voluntarily seeking information in order to gain certainty regarding a medical condition or treatment, we can view these data as indicating that the public is at least somewhat active in this regard. Differences Among the Public in Decisionmaking The previous data clearly present the public as active in the medical decisionmaking process. How- ever, not all identifiable subgroups comprising the total sample report being equally active. On the issue of decisionmaking, respondent age, health sta- tus, education, and household income similarly af- Surveys of Physicians and the Public 187 fect obeserved response patterns. Here we see the variables of age, household income, education, and health status working jointly to play an important moderating role. Those most socially dependent -- the elderly, those in poor health, the poor, and the least educated -- report taking a less active role in the decisionmaking process. Specifically, those aged 65 and over (54%), those in poor health (56%), and those without a high school diploma (41%) are more likely to have relied on a physician's judgment as the primary determinant to undergo a surgical procedure than those who are younger (34%), healthier (34%), or better educated (36%). Similarly, the old (13%), the poorly educat- ed (13%), and the poor (19%) are far less likely to have refused treatment than those who are younger (23%), better educated (30%), or wealthier (24%). The consistency of these data patterns extends to each and every measure regarding the patient's input into the medical decisionmaking process dis- cussed. Examination of the survey data clearly dem- onstrates that the aged, the poor, those attaining lower educational levels, and those in poor health are less likely to have ever sought second opinions, changed physicians because they disagreed with them, or been told by a physician to seek treatment else- where than are those who are younger, wealthier, more educated, or healthier. Clearly, because of their greater use of medical services, such individuals have, in absolute terms, the greatest opportunity to affect their treatment. However, because of the nature of their ills or be- cause of their expectations regarding the doctor- patient relationship or possible health outcomes, these persons who are dependent on medical services are more apt to move through treatment without tak- ing as active a part in the decisionmaking process as are those who are younger, healthier, or better educated. It is also interesting to note that this pattern of dependency or limited expectations mir- rors that reported earlier concerning information disclosure. The oldest, the least well educated, the poor, and those in poor health see physicians as the best judge of how much information to disclose, are least likely to want to know everything about their treatment, and are generally most satisfied with their physicians and their information dis- closure practices. These are the same people who 188 Making Health Care Decisions: Appendix B seem least well informed about medical disclosure standards in that fully 36% of the elderly, 40% of the least educated, 39% of those with an income of $7,500 or less a year, and 40% of those in poor health have no idea about the meaning of the term "informed consent." Locus of Decisionmaking: The Physicians' View The results of the current survey indicate that patients report substantial input into the medical decisionmaking process, including making the deci- sion to undergo an operation (52%), refusing recom- mended treatment (21%), changing doctors because of disagreements about their treatment (36%), and shop- ping for second opinions (51%). This general pic- ture of the patient as an actor and not merely a passive recipient of health care services is bol- stered by physicians' responses to several of the survey items. On a broad range of treatment issues, doctors regularly note that the locus of decisionmaking rests dually on the administering physician and on the patient. Many split ballot items were employed in this survey, both to maximize information gain and to obtain convergence of perspectives on a num- ber of issues. One series of six items required physicians to indicate the appropriate locus of med- ical decisionmaking on a bipolar, seven-point scale anchored at the endpoints by "1," patient's deci- sion, and "7," doctors's decision. A response of "4" therefore indicates equal input into the deci- sionmaking process. The decision items were de- liberately selected to represent the breadth of pos- sible decision situations. For example, the choice of antibiotics for the treatment of strep throat or amputating a gangrenous limb were thought to fall almost completely within the physician's decision- making domain. Amniocentesis for pregnant women over thirty-five and the decision to hospitalize a patient with a mild case of pneumonia were thought to be located more in the domain of patient's de- cision. Items concerning beginning dialysis for end-stage renal disease and initiating estrogen treatment for menopausal problems were thought to fall at an intermediate position on this continuum. In general, physicians' ratings coincided with a priori assumptions concerning where each item was Surveys of Physicians and the Public 189 located on the continuum. More specifically, those items thought to represent situations in which both physician and patient have relatively equal inputs (i.e., dialysis for end-stage renal disease and in- itiating estrogen treatment for menopausal problems) fall on or quite near the midpoint of the decision scale (X = 4.2 and ¥ = 3.8, respectively) (Table 6-7). One of the items for which the patient's in- put into the decisionmaking process was thought to predominate (i.e., amniocentesis) falls clearly within the patient's decisionmaking domain (X = 3.3), while the other item (hospitalizing a mild case of pneumonia) falls just to the physician's side of equal decisionmaking (X = 4.4). Lastly, the two items felt to require more significant input from the physician than the patient (i.e., choice of antibiotics and amputation of a gangrenous limb) are rated clearly as primarily physician decisions (Xx = 6.4 and X = 4.9, respectively). Though each item provides useful information, it is perhaps most useful to think of these six items as comprising a balanced scale and to observe the overall results as ‘indicative of actual physician decisionmaking preferences. To accomplish this, individual item means and overall "scale" means were tabulated. Examination of these data indicates little ab- solute variation. However, the observed patterns of the mean scores, both for the individual items and for the overall total, fit with those already re- ported. Overall, physicians seem to indicate a strong preference for joint, almost equal, decision- making -- note that the scale mean for the physician sample is 4.5. Even on treatment issues logically falling completely within the physician's decision- making domain (e.g., choice of antibiotics) we see some willingness to include the patient's views in making the final decision. On the other five treat- ment issues, all mean scores fall between 3 and 5, indicating a slight preference for decisionmaking falling to either the patient or the physician, but with a clear emphasis on the preference for incor- porating both views. This, in point of fact, fits precisely with the stated views and inputs into the decisionmaking process viewed by the general public as most appropriate. It is clear that doctors, in general, say that doctors should elicit patient input in the formula- 190 Making Health Care Decisions: Appendix B tion of treatment decisions. As such, this pre- scription represents a professional, operational standard. However, there are occasions when patient willingness to make input is either absent or inap- propriate (e.g., unconscious patient, area outside patient expertise or comprehension). Also, patients sometimes disagree with physicians, even when the medical treatment recommended is strongly indicated. In these situations, physicians cannot adhere to their operational standard of including patient in- put into the decision unless they do not intend to initiate therapy. Instead, they may attempt to per- suade the patient and impose their judgment of what treatment is required, thereby attempting to ensure that a medical remedy to the patient's problem is attempted. Doctors were asked about their perceived re- sponsibility regarding persuading a patient to accept the recommended, medically indicated course of treatment when the patient disagrees with the recommendation. Seventy-five percent of the phy- sicians surveyed report that it is their respon- sibility to persuade the patient (Table 6-8). How- ever, 6% say they have no such responsibility to the patient, while an additional 18% mention that their responsibility is situationally specified. Although we cannot move directly from these data to the patients' reports that 20% have had doctors recommend that they seek help elsewhere because of personal or procedural disagreements or that 36% have changed doctors because of such disagreements, it is clear that doctors and patients disagree rel- atively frequently and that the response of the phy- sician can take as many forms as that of the patient. Differences Among Physicians in Decisionmaking Physicians, like the public, do not respond as a single group with a unified vision or perception regarding their own practices and/or preferences regarding medical decisionmaking. Interestingly, many of the characteristics that seem to distinguish those physicians who most strongly support the pref- erence or practice of joint decisionmaking more and unilateral decisionmaking less mirror those charac- teristics that tended to discriminate among identi- fiable subgroups of public respondents. Specifical- ly, concentration of the aged, the seriously ill, Surveys of Physicians and the Public 191 and those unable to understand treatment in a doc- tor's practice reliably differentiate among the phy- sician supporters of joint medical decisionmaking. Also, because of the interrelationship between re- cency of physician training and location of practice with patient characteristics, these classificatory variables also discriminate well between those stat- ing the greatest and least preference for equal patient-physician input into the medical decision— making process. Examination of the overall means derived from the six loci of treatment decision items indicates that physicans trained most recently (4.3), compared with physicians who received their degree before 1967 (4.6), feel that medical decisions should be made jointly by doctor and patient. Moreover, phy- sicians with high percentages of seriously ill pa- tients (4.6), patients unable to understand treat- ment (4.9), or poor patients (4.6) are more likely to say that they feel that treatment decisions should be made more by the treating physician than are physicians treating few seriously ill patients (4.5), patients unable to understand the treatment (4.5) or poor patients (4.5). Although these dif- ferences appear small, the standard errors of the estimates are smaller yet. In general, reported differences are at least marginally significant (p< 0.1). These data indicate that although training vari- ables (i.e. recency of graduation) may predispose physicians to differ in their views on the impor- tance of patient input in medical decisionmaking, the characteristics of the patients themselves may require the physician to exert more control over the process. Recall that the poor, the less educated, those in poor health, and the aged are the public respondents most likely to be passive in the medical decisionmaking process. Here we see that physicians treating high proportions of low income, old, and poorly educated patients view physician input as more essential than do other doctors for determining medical treatment. Perhaps this is a reaction to their patients' decreased level of activity in the medical decisionmaking process. In like fashion, an examination of the survey data indicates that several patient composition var- iables are, in fact, related to physicians who re- port that they feel responsible for persuading pa- 192 Making Health Care Decisions: Appendix B tients to accept medically indicated treatment. More specifically, physicians most recently begin- ning their practice (79%), physicians with a high percentage of seriously ill patients (76%), or whose patients are, on average, poor (82%), are more like- ly to report being inclined to persuade disagreeing patients to accept recommended treatment than are physicians who received their degree before 1967 (75%) or physicians who generally treat few serious- ly ill (74%) or poor patients (75%). If there is an implicit professional ethic or normative standard of behavior regarding the physi- cians' role in medical decisionmaking, little vari- ability in physicians' responses on relevant items should have been manifested. Yet, systematic variation in physicians' responses to these items, including measures directly related to patient au- tonomy, are evidenced. From the survey results re- ported above, it seems that physicians’ perceptions of their role vis-3d-vis medical decisionmaking are multiply-determined by the interplay and dynamics of the physicians' training, personal beliefs, and interaction style, as well as patient character- istics and interaction style. G-10B BASE: TABLE 6-1 BASIS FOR PATIENT'S ASSENT TO A SURGICAL OPERATION: HAVE UNDERGONE A SURGICAL OPERATION REQUIRING AN OVERNIGHT HOSPITAL STAY GENERAL PUBLIC Q.: When you agreed to the surgical operation, did you base your decision primarily on your trust in or was it based primarily on your own understanding and judgment of the your doctor's judgment, situation? BASE DOCTOR'S JUDGMENT OWN JUDGMENT EQUALLY (VOL.) NOT SURE REPORTED HEALTH STATUS GENERAL PUBLIC AGE 65 18-34 35-50 51-64 AND EDUCATION LESS THAN HIGH SCHOOL HIGH GRAD- SCHOOL COLLEGE TOTAL EXCELLENT GOOD FAIR POOR YEARS YEARS YEARS OVER UATE GRADUATE GRADUATE 774 273 342 98 60 230 216 162 161 165 436 171 x x % x % Z Z z Z x z Z 40 34 41 45 56 34 35 40 54 41 41 36 21 21 24 18 10 30 21 18 14 21 21 22 31 36 26 34 35 22 36 40 29 31 n 31 7 8 9 4 - 15 7 2 2 6 6 10 o11qnd dy} pue suedIsAyd jo sLaaIng €61 G-29 TABLE 6-2 REFUSED TREATMENT: GENERAL PUBLIC Q.: Have you ever refused to undergo treatment recommended by a doctor? GENERAL PUBLIC EDUCATION LESS THAN AGE HIGH INCOME 65 SCHOOL HIGH $7,501 $15,001 $25,001 MORE 18-34 35-50 51-64 AND GRAD- SCHOOL COLLEGE $7,500 TO TO TO THAN TOTAL YEARS YEARS YEARS OVER UATE GRADUATE GRADUATE OR LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 488 334 226 195 242 726 280 164 219 341 229 214 Z Z Z % Z Z x Zz z Z Z 2 Z YES, REFUSED 21 23 22 21 13 13 20 30 19 15 20 24 24 NO, HAVE NOT REFUSED 3 77 78 78 87 87 80 70 81 85 80 76 76 v61 dg xrpuaddy :suorstoa(y aren yjesy Sune TABLE 6-3 CHANGED PHYSICIAN BECAUSE OF DISAGREEMENT: GENERAL PUBLIC Q.: Have you ever changed doctors because you disagreed with them? GENERAL PUBLIC EDUCATION LESS THAN AGE HIGH INCOME 65 SCHOOL HIGH $7,501 $15,001 $25,001 MORE 18-34 35-50 51-64 AND GRAD- SCHOOL COLLEGE $7,500 TO TO TO THAN TOTAL YEARS YEARS YEARS OVER UATE GRADUATE GRADUATE OR LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 488 334 226 195 242 726 280 164 219 341 229 214 % % % % % % % 3 % % % z % YES, HAVE CHANGED 36 38 40 33 31 29 35 46 30 34 36 36 44 NO, HAVE NOT CHANGED 64 62 60 66 69 71 65 54 70 66 64 64 56 o1qng 9Y} pue suerdIsAyd jo sfeaIng S61 TABLE 6-4 PHYSICIAN SUGGESTED CHANGE OF DOCTOR: GENERAL PUBLIC Q.: Has a doctor ever suggested that you find another doctor if you don't agree with his diagnosis and treatment recommendation? GENERAL PUBLIC EDUCATION LESS THAN AGE HIGH INCOME 65 SCHOOL HIGH $7,501 $15,001 $25,001 MORE 18-34 35-50 51-64 AND GRAD- SCHOOL COLLEGE $7,500 TO TO TO THAN TOTAL YEARS YEARS YEARS OVER UATE GRADUATE GRADUATE OR LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 488 334 226 195 242 726 280 164 219 341 229 214 z z % z % % 2 % z % z 2 2 YES, HAS SUGGESTED 20 24 19 19 12 16 21 21 17 19 17 17 26 NO, HAS NOT SUGGESTED 80 76 81 81 88 84 79 79 83 81 83 83 74 961 dg x1puaddy :suorsioa(] aren) yjeay Sune G-31,32 TABLE 6-5 CHANGED DOCTORS BECAUSE OF DISAGREEMENT: GENERAL PUBLIC Q.: Have you ever changed doctors because you disagreed with them? Q.: Has a doctor ever suggested that you find another doctor if you don't agree with his diagnoses and treatment recommendations? HAS A DOCTOR EVER SUGGESTED THAT YOU FIND ANOTHER ROW Z DOCTOR IF YOU DON'T AGREE WITH HIS DIAGNOSIS AND TREATMENT RECOMMENDATIONS? TOTAL 2 YES NO TOTALS HAVE YOU EVER CHANGED DOCTORS YES 27 73 36 BECAUSE YOU DISAGREED 10 26 WITH THEM? NO 16 84 64 10 54 TOTAL 20 80 100 o1qngd 9Y} pue sueroIsAyd jo sAeaing L61 TABLE 6-6 SOUGHT SECOND OPINION: GENERAL PUBLIC Q.: On your own, have you ever sought a second opinion from another doctor? GENERAL PUBLIC EDUCATION LESS THAN AGE HIGH INCOME 65 SCHOOL HIGH $7,501 $15,001 $25,001 MORE 18-34 35-50 51-64 AND GRAD- SCHOOL COLLEGE $7,500 TO TO TO THAN TOTAL YEARS YEARS YEARS OVER UATE GRADUATE GRADUATE OR LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 488 334 226 195 242 726 280 164 219 341 229 114 2 > z z 2 x ” x 2 % x 2 2 YES, HAVE SOUGHT 51 55 49 53 42 39 53 57 49 44 54 52 54 NO HAVE NOT SOUGHT 49 45 51 46 58 61 47 43 51 56 46 48 46 861 dg xtpuaddy :suorsioa(] aren yjesy Sune BASE AVERAGE MEAN SCORE HOSPITALIZE A MILD CASE OF PNEUMONIA AMNIOCENTESIS FOR PREGNANT FEMALES OVER 35 CHOICE OF ANTIBIOTICS FOR STREP THROAT INITIATE DIALYSIS FOR END-STAGE RENAL DISEASE AMPUTATE A GANGRENOUS LIMB ESTROGEN TREATMENT FOR MENOPAUSAL PROBLEM TABLE 6-7 LOCUS OF DECISIONMAKING MEAN SCORES: PHYSICIANS PHYSICIANS PATIENTS OBTAINED PATIENTS WITH UNDERSTANDING POOR MEDICAL DEGREE SERIOUS ILLNESS TREATMENT PATIENTS 1966 1967 1973 0 6X MORE AND TO AND TO TO THAN TO TO TO OR TO OR TOTAL BEFORE 1972 LATER 251 50% 50% 691 89% 100% LESS 501 MORE 207 $3 "143 172 122 103 72 13 198 255 71 54 6.6 4.6 bb 6.3 4.5 bk 4.6 4.9 4.5 4.5 6.5 4.4 6.6 G6 6.5 4.5 bl bb 63 4.5 5.3 All Ak bb 6.2 6D 3.3 3.6 3.3 3.1 3.2 3.4 3.3 3.1 3.1 3.4 3.3 3.2 3.6 6.4 6.4 6.4 6.2 6.3 6.3 6.5 6.8 6.4 6.3 6.4 6.3 6.3 4.2 4.4 4.1 6.0 4.1 4.1 43 6.1 6.2 4.2 4.0 4b 4.6 4.9 4.9 4.9 4.8 5.0 4.8 4.9 5.4 5.1 4.8 4.9 4.8 4.9 3.8 4.1 3.5 3.6 4.0 3.7 3.8 4.9 4.0 3.8 3.8 3.8 4.1 Sp — rep NOTE: SPLIT BALLOT ITEM, ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. o1[qng ay} pue surIOISAYJ JO sAeaing 661 TABLE 6-8 PHYSICIAN'S RESPONSIBILITY TO PERSUADE PATIENT: PHYSICIANS Q.: When a patient disagrees with your recommendation for a particular procedure or course of treatment that you think is strongly medically indicated, do you feel that it is your responsibility to try to persuade the patient to accept the medically indicated course of action, or do you feel that you have no responsibility to do this? PHYSICIANS OBTAINED PATIENTS WITH POOR MEDICAL DEGREE SERIOUS ILLNESS PATIENTS 1966 1967 1973 0% 26% MORE 25% 26% 51% AND TO AND TO TO THAN OR TO OR TOTAL BEFORE 1972 LATER 25% 50% 50% LESS 50% MORE BASE 805 441 97 266 367 229 193 500 143 107 7% % % 7% % Z % % % Z RESPONSIBILITY TO PERSUADE 75 75 62 79 74 76 76 75 73 82 NO RESPONSIBILITY 6 6 6 5 5 6 7 5 6 8 DEPENDS (VOL. ) 18 18 31 15 20 17 15 19 20 9 002 fd x1puaddy :suorsa(y aren) yieay Sunyepy Surveys of Physicians and the Public 201 CHAPTER 7: EVALUATION OF THE MEDICAL DECISIONMAKING PROCESS Introduction The previous chapter presented the medical de- cisionmaking process as interactive and dynamic. Physicians describe the medical decisionmaking pro- cess as a joint process involving both doctors and patients. Further, the public notes that they have exercised their judgment and made significant input into medical decisions, including undergoing sur- gical procedures, changing doctors, and refusing recommended treatment. The present chapter explores the public's stated preferences regarding medical decisionmaking in general and their satisfaction with current modes of medical decisionmaking, and reports on evaluative responses elicited from physi- cians and the public alike on hypothetical changes to the therapeutic decisionmaking process. In gen- eral, the public reports being quite satisfied with the medical decisionmaking process. They, like the physicians, however, report that medical decision-— making can still be improved by encouraging patients to take a more active role in participating in the decisionmaking process. These issues are explored in this chapter. The Public's Preferences Regarding Medical Decisionmaking It is clear that for a number of treatment sit- vations, physicians report that medical decision- making is patterned on a model that requires sub- stantial input from both physicians and patients. As was pointed out in the previous chapter, physi- cians report that mutual medical decisionmaking characterizes the selection of treatment strategies that apply to a number of situations, including the hospitalization of a mild case of pneumonia, amnio- centesis for pregnant women over 35, initiation of dialysis for end-stage renal disease, and estrogen treatment for menopause problems. Even in decisions thought to fall heavily in the physicians' decision- making domain (i.e., amputate a gangrenous limb or choice of antibiotics for strep throat), doctors' responses indicate that meaningful contributions to the formulation of the decision are made by patients on a regular basis. In this regard, an important 202 Making Health Care Decisions: Appendix B question that arises revolves around the public's reaction toward or preference for such joint decisionmaking. Respondents have made clear that they generally feel they should participate in decisions related to their medical treatment. When asked directly which of five decisionmaking approaches they prefer, al- most 3 out of 4 respondents (72%) report that a dec- isionmaking process in which "the doctor discusses alternatives with the patient and the two of them decide together how to proceed" is the one they fav- or (Table 7-1). An additional 19% of the public respondents report favoring unilateral decision- making by the patient either after the doctor pres- ents all available alternatives (7%) or after the doctor presents a recommendation for patient ap- proval or rejection (12%). In total, more than 9 of 10 public respondents (91%) report that the patient should be at least an equal partner in the medical decisionmaking process, and 1 of 5 respondents note that they believe that with sufficient information from the doctor, patients would be completely cap- able of and should make the treatment decision for themselves. Only 1 of 14 (7%) of the respondents polled prefer physicians making unilateral treatment decisions. Public Satisfaction with the Medical Decisionmaking Process The consistency of the data regarding the actual process of medical decisionmaking and the public's preferences for specific decisionmaking arrangements indicate clearly that the most common and popular form of treatment decisionmaking involves both pa- tient and physician in an interactive information exchange. It follows that if, in fact, such a feed- back loop actually exists and patient preferences regarding medical decisionmaking are being met, pa- tients should be at least somewhat satisfied with their role in achieving the treatment they desire. The public sample was asked to report their experienced-based evaluations of their doctor's respect for their treatment preferences. Almost 6 of 10 (59%) of the respondents in the general public report being very satisfied with their physician's respect for their preferences regarding treatment (Table 7-2). An additional 29% of the respondents Surveys of Physicians and the Public 203 report being somewhat satisfied. Thus, almost 9 of 10 respondents (88%) report being at least somewhat satisfied with their physician's respect for their treatment preferences. Only about 1 out of 14 (7%) respondents note any dissatisfaction (5% somewhat dissatisfied, 2% very dissatisfied) with their phy- sician's respect for their treatment preferences. The Effect of Respondent Characteristics on Prefe- rences and Satisfaction with Medical Decisionmaking Consistent reports of the public having input into treatment decisions, and doctors having respect for patient treatment preferences in the decision- making process, along with the fact that respondents among the public sample clearly indicate that they prefer joint decisionmaking converge and suggest strongly that such a decisionmaking process is cur- rently in effect and that the large majority of both physicians and respondents are generally satisfied with the roles carved out for each. Recall that, when asked, physicians also note a strong preference for joint decisionmaking across a broad selection of treatment situations. Although there seems to be general agreement regarding both the preference for and imputed ex- istence of joint medical decisionmaking, and high levels of satisfaction with the extant medical dec- isionmaking process are expressed, these perceptions are not universal. Again, respondents' membership in identifiable population subgroups affects, in a predictable way, the level of satisfaction expressed and the stated preferences for decisionmaking strat- egies. For example, greatest levels of satisfaction with physicians' respect for their treatment pref- erences are reported by the poorest respondents (65%), those having attained only minimal formal education (68%), and those reporting that their health status is poor (76%). Again, these patient characteristics (including age) similarly affect several variable measures of interest and together represent the segment of the public that is at once both most dependent and most passive within the con- text of the medical decisionmaking milieu. In ad- dition, this segment of the public is generally the most satisfied, not only with the amount of informa- tion disclosed to them, but, as we see here, with the perceived importance of the input they have made into the medical treatment decisionmaking process. 204 Making Health Care Decisions: Appendix B Similarly, respondent standing on these clas- sificatory variables predicts adequately preferences regarding the approach taken to medical decision- making. Recall that when asked directly which of five decisionmaking approaches they prefer, almost 3 out of 4 respondents (72%) report that a joint de- cisionmaking process is the one they favor. A neg- ative relationship is observed between income and endorsement of this decisionmaking approach. As respondents’ household income decreases, the percent favoring joint decisionmaking drops, on average, 17%, from about 74% (mean percent of the four highest income categories) to 57% (household income of $7,500 or less). As decreases in endorsing mutual decisionmaking occur in these groups, increases in endorsing extremes of decisionmaking are observed. Specifically, the poorest respondents are far more likely than those with a higher income to endorse a treatment decisionmaking strategy in which the doc- tor makes unilateral decisions (12% vs. 5% -- arith- metic mean of the other four income groups), or in which patients, after learning of all alternatives, decide for themselves (12% vs. 7% -- arithmetic mean of the other four income groups). These same data patterns describe adequately preferences observed as a function of increasing respondent age. Positive relationships are observed to exist between reported health status and education with endorsement of joint decisionmaking. As reported health status and education level increase, the per- centage of respondents favoring mutual decision-mak- ing increases (poor health, 52%; excellent health, 79%; no high school diploma, 62%; college degree, 77%). Those respondents in poor health are far more likely than respondents in better health to endorse patients selecting a treatment alternative from a list of alternatives presented by the physician (14%), accepting or rejecting a physician's recommended treatment (22%), or to a lesser extent, simply allowing physicians to make unilateral treat- ment decisions (10%). The distribution of pref- erences among those with the least education mirrors that of the poorest respondents -- unilateral de- cisions made by either the doctor (15%) or the pa- tient (9%) are more likely to be favored by those without a high school diploma than by those with a college degree (4% and 6%, respectively). Observation: Ubservation Clearly, the predictable variation in response patterns as a function of income, education, Surveys of Physicians and the Public 205 health status, and a set of associated variables leads to the conclusion that persons we could characterize as most dependent and passive with- in the context of therapeutic decisionmaking are most satisfied with the physician's respect for their treatment preferences, and somewhat more extreme in their preferred approach to decision- making, feeling unilateral decisions made either by the physician or patient are preferable -- compared with other respondents, who more clear- ly favor joint decisionmaking. It is also in- teresting in this light to recall from previous chapters that these individuals generally under- stand least and are least active in the actual formulation of medical decisions regarding their treatment. Perceived Impact of Modification of Decisionmaking Though it is clear that both the public and phy- sicians surveyed report being satisfied with the medical decisionmaking process as it exists, neither group may be optimally satisfied. It is possible that the process could be made better. To tap one aspect of change and avenue of possible improvement, both physicians and the general public were asked about the potential impact of increasing the pa- tient's role in the medical decisionmaking process. In general, 64% of the public sample and 56% of the physician sample feel that increasing the pa- tient's role in medical decisionmaking would improve the quality of medical care (Tables 7-3 and 7-4). From these data we can feel secure that although patients are currently participating in the de- cisionmaking process, there is definitely room for their role to expand. Similarly, virtually equal proportions of the physician and public samples view increasing the patient's role in medical decision- making as having no effect on the quality of medical care (22% and 22%, respectively). In contrast to the relatively close agreement observed between the segments of the physician and public samples regarding their views that the rela- tionship between increasing the patient's role in medical decisionmaking and the quality of medical care is positive or null, the relative proportions of these samples asserting that the relationship is negative -- that increasing the patient's role will 206 Making Health Care Decisions: Appendix B have deleterious effects on the quality of medical care -- are more widely disparate. More spe- cifically, while only about 1 in 12 (8%) of the gen- eral public feel that increasing the patient's role in the medical decisionmaking process would reduce the quality of health care, more than 1 in 7 (16%) of the physicians surveyed report such a belief. In a followup to this item, the survey attempted to probe the reasons behind each respondent's belief about the effect of increasing the patient's role in medical decisionmaking and the quality of medical care. The most popular rationale expressed by ser- vice providers is that increasing patient participa- tion would lead to increased patient compliance with medical treatment, thereby increasing the quality of medical care. This response is endorsed by almost one-third (32%) of the responding physicians. Howe- ver, it is cited by only 6% of the public. On the other hand, almost one-fourth (24%) of the public believe that increasing the patient's role would improve medical care through the vehicle of improved doctor-patient relationships (Table 7-6). Although this response is similar to doctors’ reports that increased patient participation would yield increas- ed compliance, it is at least one step more remote. The public's initial concern is that they are heard and respected. This sets up a meaningful inter- change, mutual decisionmaking, and better results —- better suited to the public's needs. Doctors may view this same interchange as getting patients to take responsibility for themselves, adhere to mu- tually decided upon regimens, and thereby achieve a better medical result. It is interesting to note that although the public, in large numbers, per- ceives improved doctor-patient relationships as the vehicle by which improved medical outcomes would be obtained, the only comparable response from physi- cians is that it improves patients' confidence in their doctor (8%). Another example of divergent perspectives is noted in the responses provided by physicians and patients regarding the negative impact on medical care resulting from increasing the patient's role. Almost one-third of physician responses to this open-ended item portray negative aspects of increas- ing the patient's role. Fully 20% note that pa- tients are not knowledgeable enough to understand medical issues. Another 9% note that patients can- not make objective decisions about themselves. Al- Surveys of Physicians and the Public 207 so, 3% note that increasing the patient's role would lead to increased noncompliance with treatment. Patients rarely report reasons for their belief that increasing the patient's role would decrease the quality of medical care. Only 8% of the public report that such a relationship would obtain. Though differential perspectives do emerge om a few response items, there seems to be a relatively good correspondence between the reports of the pub- lic and physicians on the bulk of response items. The most popular response among the public is that increasing the patient's role would improve medical care by educating the patient (28%). This response is the second most popular response among physicians (31%). Differential Perspectives Regarding the Effects of Increasing the Patient's Role in Medical Decisionmaking Reports that increasing the patient's role in medical decisionmaking would improve the quality of medical care are predictably affected by the cluster of dependency/passivity related variables among the public sample and by physician and practice charac- teristics among the physician sample. More specifically, within the public sample, as health status, household income, and educational attainment decrease, the percent of respondents bel- ieving that increasing the patient's role would im- prove the quality of health care also decreases. Those in excellent health (63%), those with a col- lege degree (67%), or those with a household income of over $35,000 (66%) are more likely to see a pos= itive link between increasing the patient's role in medical decisionmaking and an increase in the qual- ity of health care than those in poor health (59%), those without a high school diploma (52%), or those with a household income of $7,500 or less (56%). Respondents who report that their health status is poor or household income low are more likely to en-— dorse the postulate that increasing patient input would negatively affect the quality of health care, whereas low educational status is more directly linked to respondents' uncertainty about the prob- able effects of increasing the patient's role in medical decisionmaking. 208 Making Health Care Decisions: Appendix B Analysis of the physician data indicates that physicians beginning practice more recently are much more likely to believe that increases in the pa- tient's role would result in improvements in the quality of medical care (64% for those who obtained their degree from 1967 to 1972, 63% for those who obtained their degree after 1972), compared with their colleagues practicing for a longer period of time (52%). It is important to recall that younger physicians are the ones most likely to have patients who are older, less educated, and in poor health. These physicians are also the least likely to report that all or most patients would prefer to have med- ical decisions made for them. Also recall that the aged, the least educated, the poor, and those in poor health describe themselves as relatively pas- sive in the medical decisionmaking situation. It has been observed that younger doctors are on the one hand as likely or more likely to report bel- ieving that improvement in medical care is a func- tion of getting patients more involved and on the other hand, are generally less likely to voice the opinion that increased patient participation would affect negatively the quality of medical care; this may be taken to represent the effects of some under- lying characteristic of more recent medical school training, or some secular change in our social en- vironment, or perhaps some aspect of the patients’ characteristics. In this vein, it is also interesting to note that the characteristics of physicians' patients are related to their perception of the relationship of patient participation with the quality of medical care. As the percent of clients with serious ill- nesses rises and the percent of clients perceived as able to understand treatment falls, the percent of doctors reporting that they feel increasing the pa- tient role would have deleterious effects on the quality of medical care increases. TABLE 7-1 TREATMENT DECISIONS: GENERAL PUBLIC Q.: Now, 1'd like to ask you a few questions about treatment decisions. There are many ways in which decisions may be made about medical treatment. Which of the following approaches would you prefer? GENERAL. PUBLIC EDUCATION LESS THAN INCOME HIGH HIGH $7,501 $15,001 $25,001 MORE JRTED HEALTH STATUS SCMOUL SCHL COLLEGE $7,500 TO TO TO THAN TOTAL EXCELLENT GOOD FATR POOR CRADUATE GRADUATE CRADUATE OR LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 482 550 144 13 72 126 280 164 219 31 229 214 1 2 1 x 2 1 1 1 2 1 z 2 1 THE DOCTOR PRESENTS ALL OF THE AVAILABLE ALTERNATIVES AND ALLOWS THE PATIENT TO DECIDE 7 6 8 7 14 9 7 6 12 8 8 8 3 THE DOCTOR DISCUSSES ALTERNATIVES WITH THE PATIENT AND THE TWO OF THEM DECIDE TOGETHER HOW TU PROCEED 12 9 n 65 52 62 74 17 57 7n 78 mn 72 THE DOCTOR PRESENTS HIS RECOMMENDATION TO THE PATIENT FOR THE PATIENT [0 ACCEPT OR REJECT 12 1 12 12 22 10 0 1 15 13 10 n 15 THE DOCTOR DECIDES WHAT SHOULD BE DONE AND DOES IT 7 & 6 13 10 15 5 b 12 6 3 3 9 DEPENDS (VOL.) * Ci 1 - - * * . 1 - * * * NOT SURE 1 . 2 2 1 4 . 2 3 2 1 * 1 —_— LESS THAN 0.5%. o1[qnd 9) pue sueIdISAY Jo sfaAIng 60¢ TABLE 7-2 DOCTOR'S RESPECT FOR PATIENT'S TREATMENT PREFERENCES: Q.: Now, I'd like to ask you some more general questions about health care. are you with your doctor's respect for your treatment preferences? GENERAL PUBLIC GENERAL PUBLIC In light of your experience, how satisfied EDUCATION LESS THAN INCOME HIGH HIGH $7,501 $15,001 $25,001 MORE REPORTED HEALTH STATUS SCHOOL ~~ SCHOOL COLLEGE $7,500 TO TO TO THAN TOTAL EXCELLENT GOOD FAIR POOR GRADUATE GRADUATE GRADUATE OR LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 482 SS0 144 73 242 726 280 164 719 341 229 214 z x x 2 % 2 % z 2 % z % % VERY SATISFIED 59 61 56 56 76 68 58 55 65 62 55 58 57 SOMEWHAT SATISFIED 29 25 32 3% 17 20 30 32 27 22 31 28 34 SOMEWHAT DISSATISFIED 5 6 6 4 1 4 5 7 3 4 7 8 6 VERY DISSATISFIED 2 2 2 1 3 1 2 2 2 3 2 3 1 NOT SURE 5 5 4 5 4 7 4 4 3 8 5 2 3 ote fd xtpuaddy :suorsa(] aren yiesH Sunyepy G-35 TABLE 7-3 EFFECT OF INCREASED PATIENT ROLE IN DECISIONMAKING: GENERAL PUBLIC Q.: Do you believe that increasing the patient's role in medical decisionmaking is likely to improve the quality of medical care, reduce the quality of medical care, or have no effect on it? GENERAL PUBLIC EDUCATION LESS THAN INCOME HIGH HIGH $7,501 $15,001 $25,001 MORE REPORTED HEALTH STATUS SCHOOL SCHOOL COLLEGE $7,500 TO TO TO THAN TOTAL EXCELLENT GOOD FAIR POOR GRADUATE GRADUATE GRADUATE OR LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 482 550 144 13 242 726 280 164 219 341 229 214 z x z % z % % % z 2 % % z IMPROVE THE QUALITY 64 63 65 62 59 52 66 67 56 61 66 68 66 REDUCE THE QUALITY 8 9 6 6 11 8 8 6 10 6 7 10 5 HAVE NO EFFECT 22 23 21 23 19 23 21 23 21 24 21 19 26 NOT SURE 7 6 8 8 9 15 6 4 12 9 6 3 2 o1iqng 9Y} pue sueIsAyd jo sLaaing 11e TABLE 7-4 EFFECT OF INCREASED PATIENT ROLE IN DECISIONMAKING: PHYSICIANS Q.: Do you believe that increasing the patient's role in medical decisionmaking is likely to improve the quality of medical care, reduce the quality of medical care, or have no effect on it? PHYSICIANS PATIENTS PATIENTS WITH UNDERSTANDING POOR OBTAINED MEDICAL DEGREE SERIOUS ILLNESS TREATMENT PATIENTS 1966 1967 1973 0% 26% MORE 0% 70% 90% 25% 26% 51% AND TO AND TO TO THAN TO TO TO OR TO OR TOTAL BEFORE 1972 AFTER 25% 50% 50% 69% 89% 100% LESS 50% MORE BASE 401 234 44 123 195 107 90 78 136 186 245 72 53 2 2 2 % ro 0% % r 2 2 2 IMPROVE THE QUALITY 56 52 64 63 58 54 56 58 60 53 56 60 55 REDUCE THE QUALITY 16 20 7 11 14 18 18 23 13 14 16 15 11 HAVE NO EFFECT 22 20 25 24 21 25 20 14 22 25 21 24 26 NOT SURE 6 8 4 2 7 3 7 5 5s 8 7 1 8 NOTE: SPLIT BALLOT ITEM, ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. 212 fg x1puaddy :suorsioa(] aren yjeey Sunyey Surveys of Physicians and the Public 213 »-22,23 TABLE 7-8 REASONS POR EFFECTS OF INCAZASED PATIENT ROLE IN DECISIOMMAKING: PEYSICLANS Q.: Do you believe that inc of medical care, reduce the ing the patient's role in medical decisionmaking is likely to improve the quality ty of medical care, or have no effect on it? Q.: Why do you say that? PEYSICLANS PATIENTS PATIENTS WITH UNDERSTANDING POOR OBTAINED MEDICAL DEGREE SERIOUS ILLNESS TREATMENT mii 1966 1967 1 oF 26% womk AND T x TO or 1 ¥ ~ = 3 ~38 AND TOTAL BEFORE 1972 AFTER BASE of Jo ~ 123 1 1 3 3 » HH $3 -g52 wgzd ~Hg3 IT IMPROVES PATIENT COMPLIANCE WITH TREATMENT, INCREASES COMPLIANCE WITH MEDICAL JUDGMENTS, INCREASES THEIR COOPERATION IN THEIR TREATMENT, MAKES THEM MORE WILLING TO ACCEPT THE DOCTOR'S JUDGMENT 2 n 3% 36 32 Nn n 3» 29 46 28 IT INCREASES PATIENT KNOWLEDGE AND AMARENESS OF TREATMENT /BEST TREATMEWT/ ALTERNATIVE TREATMENTS, GIVES THEM MORE INFORMATION UNDERSTANDING OF TREATMENT, ELIMINATES QUACK CURES 3 26 36 3» 33 1 32 30 3 an 2 32 PATIENTS ARE NOT KNOWLEDGEABLE ENOUGH TO MAKE PROPER DECISIONS, PATIENTS DOR'T KNOW ENOUGH/DON'T NAVE ENOUGH INFORMATION/HAVE POOR OR INCOMPLETE INFORMATION/DOR'T UNDERSTAND TMZ MEDICAL 1SSUES 20 23 14 12 17 Ww 22 n 19 10 n 1 19 PATIENTS CANNOT MAKE OBJRC- TIVE DECISIONS ABOUT TMEIR TREATMENT, PATIENTS ARE TOO EMOT1ONAL/TOO ARX100S/TOO FRIGHTENED TO CHOOSE TME BEST TREATWENT ’ n ’ S ’ ’ . 13 5 10 . . 13 IT IMPROVES PATIENT COWNFI- DENCE IN THEIR TREATMENT/ DOCTOR/MEDICAL CARE . . 2 10 10 . 3 1 1n . 10 3 4 IT IMPROVES PHYSICIAN PERFOR- MANCE, PHYSICIAN MUST BR BETTER INFORMED AND MORE MONEST WITH PATIENT, MAKES DOCTOR DISCUSS THE ALTER- NATIVES AND RISKS, ALLOWS DOCTOR TO CONSIDER PATIENT PREFERENCES, DECREASES DOCTORS' INFLATED SERSE OF SELF . [3 n 10 10 s . . 10 . un 3 . WOTE: SPLIT BALLOT ITEM, ASKED OF ONI-NALY OF PHYSICIAN SAMPLE. (CONT LWUED) 214 Making Health Care Decisions: Appendix B r-22,23 TABLE 7-3 (CONTINUED) SEASONS FOS EPFECTS OF LECASASED PATISNT BOLE IN DECISIOMMAKING: PEYSICLARS C PATIENTS WITM WEARING PooR OBTA GREE SER 1 TREATMENT Mi Ae Mga 8 we LLNESS re AND T° AND ™ 10 T™ T™ TO 7. 9! nas hone 3 5 ~ 11] 3] -8 ok 891 1002 NORE THAN 7 mi ne 1 3 3 “3 4 “Sg of Ld BOCTOR MOT PATIENT DETEANINGS QUALITY OF MEDICAL CAME, QUALITY OF PUYSICIAN AND MOT PATIENT BOLE DETERNIMES QUALITY OF CARE, PATIENT'S ROLE 1S TO RELATIONSHIP BETWEEN PATIENT ROLE AND QUALITY OF MEDICAL CARE, PATIENT'S ROLE IS TO TAKE CARE OF SELF . . ’ 6. 7 6 10 S ? ’ 7 . ’ PATIENTS MAKE THE SAE DECISIONS REGAADLESS OF WOW MUCH THEY PARTICIPATE, USUALLY ONLY OME ACCEPTABLE JOURSE OF ACTION, WO REAL CHOICE, MOST PATIENTS LEAVE THE DECLS LOW TO THE DOCTOR WMEM ALL 18 SAID AND DOME . . . 10 . 5 7 . 3 . 5S 10 ‘. IT VILL INTERFERE VITM PHYSICIAN'S TREATMENT, PATISNT MAY DEVIATE FROM TREA' AND CAUSE SELF MARM, PATIENTS WON'T DO WMAT TMEIR DOCTORS TELL THEN TO DO, IT WILL REDUCE PATIENT COMY LDENCE 18 ALL OTNEAS 7 . ’ . 4 ¢ nu 13 3 8 S . 135 — OTE: SPLIT BALLOT ITEM, ASKED OF OME-NALY OF PEYSICLAN SMPLE. Surveys of Physicians and the Public 215 6-33.36 Q.: Do you believe that care, reduce the quality Q.: Way do you say that? IT INCARASES PATIENT LHOVLEDGE ASD AMAREWESS OF TRAATMENT/ BEST TREATWENT/ ALTERNATIVE TREATMENTS, GIVES PATIENTS ORE INFORMAT LON/ BETTER UNDERSTANDING OF TLATINT, ELININATES QUA umes IMPROVES DOCTOR- PATIENT KALATLOWSM| INCREASES (NFORMAT LOW 7L0wW/DUCTOR-PAT LENT COMMUNI CAT LOW IMPROVES MEDICAL DEC IS 10MAK IHG, UNF IDENLE /RASE IN MEDICAL DEL IS LOMNAK LUG PATIENTS ARE DON'T KM ENOULH/ HAVE EHOUCH INFORMAT IUN/ ARE WOT EMUWLEUCLABLE HOCH TO MAKE PEOPLE DEC IS LONS, PATIEM(S WAVE POOK OR INCOMPLETE INFURMAT LOM, DON'T UMILKSTAND Tie MLUICAL I330LS IMPROVES PHYSICIAN IT IMPROVES PAT LENT CONFIDENCE |W THEIR TREATMENT /DOCTOR /MEDICAL CARE, IMCAZASLS LASE OF ADNINISTERING TREATIENT, INCREASES act cant DECRLASES UNNECESSARY TRAATMANT/SIDE EFFECTS PATIENT RIGHTS, DOCTOR RESPONSIBILITIES INVOLVERENT IN TREATMENT /CENERAL COMLLAN FUR WEALTH DOCTOR, WOT PATIENT DLTRAMINLS QUALITY OF MEDICAL CARE PATIENTS WOULD MAKE THE SAME DACISIONS AS DOCTORS DU IF THEY NAD THe MIGHT [NFORRATION OTMER/MO EFFECT /NO BALAT LOMSH LP BLASONS FOR EFFECTS OF INCASASED PATIENT SOLS LN BECISIONMALLEG: @ENEAAL Lic TABLE 1-4 incrassing the patient's role ia medical decisiommaking is likely te improve the quality of medicel of medical care, © have me effect oa it! —_— _ cemaweuc Tuas 125 “82 $50 1 3 1 20 til 20 2% i un n 1" nu n 1" nu 12 1 n ’ "w . . 10 . . . . . 3 . 3 . . 1 3 1 , , . 0) nv n GRADUATE amar prone) wis Seige \T LOW wicH Jaa SCHOOL m2 be | pl 1 3 1" 3» 1" El , n , 1 . 12 . w . , 3 . 3 . 3 s 1 1 1 . n n » 1» HAS 125,000 3s " Surveys of Physicians and the Public 217 CHAPTER 8: DECISIONS TO FOREGO TREATMENT OR TERMINATE LIFE Introduction There is no area of medical decisionmaking more visible or ethically complex than that of deciding to forego treatment or to terminate life. Clearly, the issues involved are controversial. The rel- atively recent highly publicized medical cases such as that of Karen Anne Quinlan have brought these issues to the forefront of public concern. This is perhaps best reflected in the dramatic shifts over time in public attitudes regarding terminally ill patients' right to die or to "death with dignity." Almost a decade ago (Harris, 1973), 62% of the general public felt that terminally ill patients had the right to forego life-sustaining therapy if that was their desire. Five years ago (Harris, 1978), the percentage of the public endorsing this state- ment increased to 71%. Last year (Harris, 1981), 78% of the public indicated that such a patient should be allowed to die. Similar shifts in at- titude have been noted in endorsing terminally ill patients' right to have their life ended. Here the proportion of Americans endorsing the statement in- creased almost 20%, from 37% to 56% in 1981. The current research provided a vehicle to study, in some detail, the intricate issues involved in making decisions to forego treatment, to ter- minate treatment, and to terminate life, from the perspective of both the administering physician and the general public; in the final analysis, the pub- lic is the object of such administrating and must bear the consequences of such decisions. The survey included several items that attempted to tap re-~ spondents' views on the rights of patients; it also included items on the responsibilties of physicians to the law, to some implicit or explicit profes- sional ethic, and to their patients. The results of these study findings are presented below. Final Choices About Medical Treatment The general public respondents were asked if they would prefer to be responsible for final de- 218 Making Health Care Decisions: Appendix B cisions about their medical treatment if they became seriously ill or whether they would prefer that their physician make the decision (Table 8-1). Almost one-half (43%) of the public surveyed felt that patients and not doctors should be responsible for final choices about medical treatment. Fourteen percent of the public felt that such decisions should be made jointly by both physician and pa- tient. However, a very large minority (38%) of the public noted that physicians should maintain sole responsibility for decisions about medical treatment in the case of serious illness. Examination of the survey data indicates that the percentage of the general public indicating that patients should make final choices was moderated by age, income, health status, and education. Those in poorest health (17%), oldest (23%), least educated (27%), and with the lowest household income (32%) were far less likely to endorse patients' making medical decisions in the face of serious illness than were their healthier (51%), younger (56%), bet- ter educated (52%), or wealthier (47%) counter parts. In point of fact, among the oldest, poorest, least educated patients, as well as those in poorest health, there is a drastic change in the percent of respondents endorsing patients rather than doctors having responsibility for making treatment choices. Specifically, relatively large majorities of the oldest (60%), poorest (55%), least well educated (55%) patients and those in poorest health (61%) would prefer to have the physicians make the final choices about medical treatment. This is in stark contrast to relevant comparison groups who, almost without exception, manifest a clear plurality in favor of patient decisionmaking. In a followup question, the public was asked whom they would like to have make an important de- cision about their medical care if they were too sick to make such a decision. As might be expected, the majority of the public (57%) indicate that they would prefer that a family member make these de- cisions (Table 8-2). The second most preferred de- cisionmaker is "your doctor" (31%). Written Instructions Nrikten ‘nstructions Other sets of indicators of the extent to which the public would like to leave important treatment Surveys of Physicians and the Public 219 decisions entirely in the hands of the medical com- munity or, alternatively, to maintain control over such decisions, were derived from respondents' ans- wers to questions 39 to 41 -- "Have you given in- structions to anyone about how you would like to be treated if you ever became too sick to make deci- sions?" (Table 8-3), "Did you put these instructions in writing?" (Table 8-4), and "Do you think that the law should allow doctors to honor the written in- structions of their patients even if that means al- lowing them to die?" (Table 8-5). More than one-third (36%) of the general public sample report that they have left instructions regarding their treatment preferences. This large minority shows that there is strong sentiment that patients' preference should be at least part of treatment decisions and perhaps the sole basis for treatment decisions in instances where the patient is no longer competent. On the one hand it might be expected that the oldest patients, those in poorest health, the poor- est, and those least well educated, who have hereto- fore been characterized as least reactant to the medical decisionmaking milieu, would be least likely to leave written instructions, thus placing their trust in the hands of physicians. On the other hand, advanced age, poor health status, low house- hold income, and low levels of formal education have been observed to be highly interrelated as weli as individually related to the experience of life- threatening illnesses. Advanced age and poor gen- eral health are also actuarially related to death. These characteristics should therefore act as 'en- abling factors"; since the perceived probability of becoming "too sick to make decisions" among these groups should be highest, these respondents should be most likely to leave relevant treatment instruc- tions -- just as, all else being equal (i.e., in- come), they should be most likely to have last wills and testaments. Observed data support this general hypothesis. Each mortality-relevant enabling factor is related to having left treatment preference in- structions; the oldest respondents (42% and 39%), those in poorest health (54%), and those having experienced a life-threatening illness in the family (42%) are the most likely to report having given instructions (Table 8-3). Although education is inversely related to many of the mortality-related enabling factors, higher 220 Making Health Care Decisions: Appendix B educational achievement should function as an en- abling factor in its own right. This seems to have occurred. Achieved education is positively related to leaving instructions. College graduates are the most likely educational group to report having prov- ided another with their treatment preference in- structions (43%). These observations also help to explain the fact that those in excellent health reported (40%) having left treatment preference in- structions relatively often. Recall that survey data patterns indicate that achieved levels of ed- ucation and excellent health are related. Of those who left instructions regarding their treatment preferences, fewer than one-fourth (23%) have put them in writing (Table 8-4). As might be expected from the above analysis, education and poor health status indicators were related to having left treatment preference instructions in writing, i.e., the best educated (28%) and those in poor health (30%) were most likely to leave written instruc- tions. Variable interrelationships kept clear data patterns from emerging on age and life-threatening illness variables. Regardless of whether or not respondents left instructions, they were asked whether they thought the law should allow physicians to honor the written requests of patients, even if it meant allowing them to die (Table 8-5). The question was phrased to avoid issues of active termination or having the law require physicians to comply. The purpose was to focus respondent attention on the nature of physician-patient interchange and on the rights of both the physician and the patiant in such a situa- tion. The public, by a substantial majority (68%) felt that the law should indeed allow the physician to honor a patient's written instructions even if the patient's desire was to have treatment terminat- ed, thus accelerating the occurrence of natural death. Given the data patterns previously observed, it is not surprising to see that poor health status, advanced age, low educational achievement, and low household income are directly related to the ex- pressed view that the law should not allow physi- cians to comply with written patient instructions, and inversely related to respondents' saying that the law should allow such compliance. Although in no instance did a majority of people, regardless of Surveys of Physicians and the Public 221 group inclusion, say that the law should not allow physicians to comply with a patient's written re- quests, the orderly and large percentage differences in the responses of the various subgroups that com- prise the public sample are dramatic and fascinating. Differences of between twelve and seventeen per- centage points are observed among the subsamples rep- resenting the lowest and highest levels of age, health status, education, and income of those re- spondents saying that they think the law should allow physicians to comply with patients' written in- structions. The oldest respondents (54%), the least educated (58%), those in the poorest health (60%), and those with the lowest household income (58%) are each less likely to endorse the legal sanctioning of doctors complying with patients' written instruc- tions than their younger (71%), better educated (71%), healthier (75%), and wealthier (70%) counterparts. Decisions to Forego Treatment Physicians and the public responded to a series of survey iiems designed to tap their views related to decisions to forego treatment. Hypothetical cases were presented to physicians and they were asked to project realistically their behaviors and attitudes as a treating physician within these sit- uations. The public was sometimes asked to imagine they had a specific illness (e.g., cancer) and to project their behaviors and related attitudes, both as the patient with the condition specified and as a consumer of medical services. By analyzing these convergent views some insight can be gained into the actual preferences and practices of both the sup- plier and consumer of medical services with regard to issues of treatment termination. Disclosure of Condition The general public respondents were asked wheth- er they would want a physician to tell them if they had cancer or to administer treatment without tel- ling them of the diagnosis (Table 8-6). Virtually all respondents (96%) wanted to be told about the diagnosis. Only 3% of the public said that they would prefer to receive treatment without being in- formed of the nature of their condition. These data 222 Making Health Care Decisions: Appendix B are completely consistent with those previously re- ported indicating that the large majority of re- spondents (94%) report that they want their physi- cian to tell them everything about their medical condition even if it is unfavorable. Prognosis In a followup question, public respondents were asked to imagine that the cancer they had typically led to death in less than one year. This situation clarifies both the time frame and the illness's con- sequences (virtually inescapable), which were only two among many possibilities in the previous, more ambiguous situation. Respondents were then asked if they wanted to be given a realistic estimate of their remaining life span. Although this question is not directly comparable to the previous question, the attempt here was to gain by convergence some idea of the effects of anchoring concretely expected negative outcomes on the desire to have accurate diagnosis/prognosis information. It was expected that in general the response patterns observed be- fore would be greatly exaggerated. This hypothesis is completely consonant with expectations derived from the psychological literature on attributional biases. In general we see immediate support for a type of negativity bias. Whereas only 3% of the respon- dents wanted treatment without diagnostic informa- tion, in the previous situation, 14% of the respon- dents did not want to be informed about how long they had left to live. The aged are far more likely to desire no such estimate than are the youthful (22% vs. 14%); those in fair or poor health are far more likely than those in excellent or good health to want such information withheld (25% and 22% vs. 10% and 12%); the least well educated are more like- ly to want to remain uninformed than those better educated (20% vs. 15% or 5%); and those with the lowest household income are more likely to voice a preference for not being told than those with high levels of household income (23% vs. 11%). It is important to make it clear that the large majority of respondents (85%) would, under these circumstances, want a realistic estimate of the time they had left. Looking at the data again from this perspective in.terms of age and income, one might Surveys of Physicians and the Public 223 argue that those who have affairs to put in order (young to middle aged adults with at least moderate levels of household income) are most likely to say they want from physicians a realistic estimate of the amount of time they have left -- probably to make sure that they can put their affairs in order. Those without money or young families (e.g., elderly respondents) are least likely to want realistic es- timates of the time they have left. Physicians were asked a parallel question to determine the probable practices of disclosing a prognosis in a case where the patient will die of cancer (advanced lung cancer) in less than one year (Table 8-8). Direct comparison of physician and public responses does not yield clearcut conclusions. Looking at the data in terms of defining the re- sponse alternative of 'give a realistic estimate" presented to the public, we see that 41% of the phy- sician sample say that either they would most likely provide the patient with a straight statistically based prognosis for this disease (13%) or they would tell the patient that though the exact length of time left him is uncertain, it is no more than one year (28%). Recall that 85% of the public said they wanted a realistic estimate. However, only 41% of physicians surveyed noted that they would provide patients with a hard estimate. The most popular single response given by almost one-third of physicians (33%) regarding what to tell a patient dying from a rapidly progressing case of lung cancer was to say that 'you can't tell how long he (the patient) might live but stress that it could be for a substantial period of time.'" An additional 22% of physicians surveyed said that they would ref- use to speculate on how long the patient might live. Thus, under circumstances in which a pa- tient's death was certain, in a finite time period, more than one half of the physicians surveyed (55%) said that they would not provide a patient with sta- tistically accurate expectations concerning how long the patient would live. Whether this is because physicians feel that such disclosures will interfere with treatment or cause undue stress, because they are afraid of civil procedures if they turn out to ‘be wrong, or because they perceive that patients do not want to be told, one thing is clear if we are to believe the results of this survey: patients' pref- erences and desires for accurate information may be largely unmet in these most consequential of cir- cumstances. 224 Making Health Care Decisions: Appendix B Further scrutiny of these data provide a clear picture of the factors related to the likelihood that a physician would, under these circumstances, portray statistically the patient's prognosis. Clearly, younger (52%) and/or hospital-based physi- cians (50%), and those whose patients are seriously ill (44%), unable to understand treatment (42%), and/or poor (53%) are those more likely to provide patients with accurate information rather than to refuse to commit themselves to a statistically based prognosis, than their older (32%) and/or office- based (38%) counterparts and those whose patients are healthier (39%), better able to understand treatment (40%), and/or wealthier (39%). Treatment Decisions ~—realment Decisions Another set of matched items within a role-play- ing context was asked concerning the appropriate locus of decisionmaking between aggressive and pal- liative treatment for a case of advanced lung cancer that, as presented to the doctor, had previously been unresponsive to chemotherapy. When the ques- tion was asked of the public, previous chemotherapy was not mentioned, although it was emphasized that they would most probably die regardless of their choice of aggressive or "supportive" therapy. The public overwhelmingly feels that a decision between aggressive and supportive therapy should be made by the patient -- the respondent himself or herself (79%). Only 12% of the public feel that the physician should make such a decision unilaterally and 8% volunteered that they feel that the decision should be made jointly. The percentage of the public saying that doctors should make this decision was related to advanced age (21%), poor educational attainment (21%), poor health (33%) and low household income (25%). In the context of this measure of medical decisionmaking, where consequences are negative and inescapable, the medical-dependence-related variables work together to demonstrate a type of helplessness or hopeless- ness; this results in a large number of respondents’ complete withdrawal from the decisionmaking situa- tion and willingness to forego any control over their future. One-half of the physician sample was asked, "Who should make the decision?" The other half of the Surveys of Physicians and the Public 225 sample was asked, 'Who usually makes the decision?" (Table 8-10). Overall, the percentage of respon- dents endorsing patient input into this decision differs as a function of how the question was asked. When physicians described how the process should work, 67% volunteered that the choice should be made jointly. Another 24% noted that the patient should control the decision. Thus, 91% of the phy- sicians noted that the patient should have at least some say in making the decision to continue aggres- sive or to begin palliative therapies. When physicians were asked to describe the real- ity of "who usually makes the decision," 19% noted that the patient decides and 64% said that the de- cision is made jointly between the physician and the patient. Thus, 83% of the physicians noted that the patient usually has at least some say in making the decision to continue aggressive or begin palliative therapy. It is interesting to note that in this case the physician's view of the ideal situation places control of the decision more in the patient's hands than is described to actually occur. Observation: In general, there is a good correspondence be- tween the public's view that they should have input into the decision between aggressive and supportive therapy (87%) (Table 8-9) and physi- cians' reports of how much input they should (91%) and usually do have (83%). An interesting point to be made from these data is that almost twice as many physicians (15%) report that they usually make this decision as the number of phy- sicians who say they should make such decisions for their patients (8%). It is also interesting that while the public's view is strongly mod- erated by various demographic characteristics, physicians' views show only a slight and 1in- consistent tendency to vary, unlike much of the rest of the survey responses, where variation is clear and predictable by practice character- istics. These data are also consistent with previously reported data that indicated both the public's and the physicians' preferences for joint medical decisionmaking. Clearly, the po- tential consequences of the situation did not attenuate the general desire for mutual de- cisionmaking; in point of fact, they seemed to 226 Making Health Care Decisions: Appendix B sharpen stated preferences with regard to in- creasing the patient's role. Resuscitation A series of items related to resuscitation was asked of physicians in the context of a case study. Doctors were presented with a patient in the last stages of a degenerative disease. Initially, they were asked how likely they would be not to resusci- tate this patient in the absence of any guidance from the patient (Table 8-11). A majority (57%) said that they would be very likely not to initiate resuscitative procedures. An additional 15% said they would be somewhat likely not to resuscitate. In total, only about 1 of 4 physicians (23%) surveyed say that they would be at least somewhat unlikely not to resuscitate the patient in the absence of guidance from her, i.e., they would resuscitate her. Physicians were then asked if they would in- itiate a discussion with this dying patient re- garding her preferences concerning resuscitation (Table 8-12). A majority of physicians (52%) indi- cate that they would initiate this discussion, while 38% say that they would not initiate this dis- cussion. Almost 1 in 10 physicians are not sure (9%). Examination of the data reveals the clear and dramatic effects of physician and practice charac- teristics on willingness to initiate discussion of preferences concerning resuscitation. Hospital- based physicians (61%), those beginning practice most recently (72%), and those having a majority of patients who are seriously ill (59%), or poor (62%), or who have difficulty understanding treatment (59%) are far more likely to report that they would initi- ate this discussion, compared to their office-based (49%), older (40%) physician counterparts and those who have patients who are generally in better health (51%), have greater household income (50%), or are better able to understand treatment (48%). Again, one cannot say with certainty whether these vari- ables cluster together because of differential phy- sician training or values, because of the setting in which the physicians practice, or because of the na- ture of the physicians' clients (e.g., better edu- cated patients may themselves initiate this discus- sion, removing the need for the physician to broach the subject; office-based physicians may already Surveys of Physicians and the Public 227 know their clients' wishes because of familiarity with the client). In all likelihood, each of these factors serves as a contributory cause. The Effects of Verbal Instructions In an attempt to determine further situational delimiters of physicians' reported projected behav- ior -- that they would be at least somewhat likely not to resuscitate a patient in the absence of pa- tient guidance -- two sets of additional questions were constructed and administered to physician re- spondents. Initially, physicians were asked how likely they would be to order that a patient not be resuscitated if she had "asked you to do everything you could to maintain her life" (Table 8-13). In a complete reversal of stance from that observed in the situation where patients provided no guidance for doctors, in a case where a patient had requested that doctors attempt to do all they can to maintain life, more than four out of five doctors (82%) note that they would be at least somewhat unlikely to order that attempts at resuscitation not be made. Only 16% of the physicians sampled note that in the case where a patient had provided verbal instruc- tions or a plea that she be kept alive, they would remain at least somewhat likely to order that re- suscitation efforts not be effected. In general, physicians with patient populations that include majorities of the seriously ill (80%), the poor (80%), or those having difficulty under- standing treatment (80%) say less frequently that they would be at least somewhat unlikely to order that a patient not be resuscitated when she had asked to be kept alive than did physicians whose clients included a smaller percentage of the ser- iously ill (85%), the poor (81%), or those unable to understand treatment (84%). Physicians who received their degrees most recently also report more fre- quently than their older counterparts that they would be unlikely to order that this patient not be resuscitated (86% vs. 79%). Two observations can be made from these data. First, physicians' reports show that they are gen- erally sensitive and responsive to patient re- quests. Whereas almost three-quarters of the physi- cians surveyed note that they would be at least somewhat likely to order that a patient not be re- 228 Making Health Care Decisions: Appendix B suscitated in the instance where she had provided no guidance about her preferences, when she had asked that everything be done to keep her alive, more than 4 of 5 physicians say that they would be unlikely to order that the patient not be resuscitated. Secondly, especially in the latter case, physi- cian training and patient characteristic variables affected the data pattern observed. Older doctors, those primarily treating the seriously ill, the poor, or persons incapable of understanding treat- ment note more often that they would be likely, even in the face of the patient's verbal pleas to the contrary, to order that attempts to resuscitate not be effected more than would their younger counter- parts or physicians who treated primarily well in- dividuals, persons of higher socioeconomic status, or persons better able to comprehend their situation. The Effects of Written Instructions A final information set was provided to physi- cians in order to investigate the situational delim- iters of physician likelihood to perform resuscita- tive procedures. A split ballot technique was em- ployed. One-half of the physicians were presented with a situation in which the patient, now incompe- tent, had left written instructions that she be left to "die naturally” (Table 8-14) -- an explicit ex- tension of the first situation presented to physi- cians in the resuscitation series. The other half of the doctors were presented with a patient, also now incompetent, who had left written instructions that "she wished to have everything possible done to prolong her life." All physicians were asked if they would resuscitate this patient. Again, in both instances physicians are respon- sive to patient requests. Only about 1 of 20 physi- cians (5%) say they would resuscitate the patient if she had asked to be allowed to die, while more than half (52%) say that they would resuscitate her if the patient had asked that her life be prolonged, even though her hopeless condition had clearly worsened and she had become incompetent. Observation: Three conclusions can be drawn from this set of resuscitation data items. Surveys of Physicians and the Public 229 First, without specific instructions from the patient, the large majority of physicians will interpret the situation similarly to the patient saying she wants to die naturally -- a no code (nonresuscitation) situation. Secondly, physicians report they are responsive to patient instructions about their treatment, i.e., resuscitation, even when the situation is hopeless. Lastly, both verbal pleas and written instruc- tions are often heeded by doctors in the final treatment of patients. A clear comparison of the two cannot be made because the alternatives provided in this study were confounded with a worsening situation. Still it is worth noting that in the instance where no instructions were left by the alert but dying patient, 72% of the respondents said they would be at least somewhat likely not to resuscitate. When written in- structions were left to let the patient die and the patient became incompetent, 84% of the physicians report that they would comply and not resuscitate. Verbal instruction to keep the patient alive met with 82% acceptance of her wishes (at least somewhat likely to comply), but when she had become incompetent, even though she had left written instructions to prolong her life, physician compliance fell to 52%, which is still extraordinary given the patient's prognosis. Termination of Life Several survey items were designed to focus on issues involved in decisions to terminate life. In general, at least a plurality of the public feel that physicians should, and the law should allow physicians to: (1) administer drugs to a dying patient, thus relieving their painful symptoms, even though these drugs might abbreviate life; and (2) comply with dying patients' requests to have their life ended. Although the majority of the physicians surveyed agree with the first point, few physicians feel that it is desirable or permissible to comply with a dying patient's request to be terminated. dl 230 Making Health Care Decisions: Appendix B Palliative Treatment and Abbreviating Life The public and physicians were presented with the case of a dying patient, in severe pain, who had asked the physician to help ease the pain even though the patient fully realized that the use of narcotics, in doses capable of accomplishing pain reduction, might shorten his life. The general pub- lic and one-fourth of the physicians were asked if it were ethically permissible to administer such drugs. Another quarter of the physician sample was asked if they would be likely to administer the nar- cotics. First, large majorities of respondents, across samples and across question wording, noted that the administration of narcotics in this instance was completely acceptable (79% public, 82% doctors- permissible, 84% doctors-likely to administer) (Tables 8-15, 8-16, 8-17). Only 7% of all physi- cians asked did not endorse administration of drugs to ease patient pain (5% likely to administer, 2% admissable to administer); only 14% of the public answered this question in the negative. The usual clusters of variables related to pa- tients, physicians, and physician practice location moderate somewhat the level of agreement expressed in support of the administration of high doses of narcotics to relieve pain. Doctors trained most re- cently (93%) or generally treating patients with serious illness (87%) more often say that they would be likely to administer potentially life-abbreviating drugs than do their appropriate counterparts. As can be seen, the most striking differences were re- corded for recency of training. Somewhat contrary to expectations, those physicians who treat a large percentage of patients unable to understand treat- ment are less likely to say that they would adminis- ter narcotics than are physicians treating patients who generally understand treatment (76% vs. 87%). Thus, physicians treating patients who understand are more comfortable administering such treatment than physicians who view their patients as incapable of understanding even though the patient is said to know the consequences of their request. No differ- ences emerged as a function of the percent of poor patients or of practice location. When doctors were asked whether or not it was ethically permissible to administer these drugs, Surveys of Physicians and the Public 231 rather than being asked to describe their probable behavior pattern, the results observed previously were either attenuated or reversed in direction. Again, those trained most recently were more likely to endorse administering narcotics (85%) but the difference was slight in comparison to their coun- terparts practicing the longest (83%). Also, the large difference observed previously as a function of the percent of a doctor's patients thought cap- able of understanding treatment virtually disappears when the implicit consequences of the act are made conceptually remote and respondents are asked to react to an "ideal" situation. Thus, for physicians treating a relatively small number of those unable to understand treatment, 82% say it would be ethi- cally permissible to administer target drugs. This is little different from the 83% of physicians treating high concentrations of those having dif- ficulty understanding treatment who agree that it is ethically permissible to administer these drugs. In this "ideal" situation, the relationship pre- viously observed between administering these drugs and treating a high percentage of seriously ill pa- tients was reversed. Thus, those routinely treating few seriously ill patients are more likely to en- dorse the administration of narcotics (84%) than are physicians routinely treating high concentrations of seriously ill patients (80%). Observation: Regardless of doctors' focus on ethics or their probable behavior, the large majority of physi- cians note that symptom relief is acceptable even at the risk of abbreviating a dying pa- tient's life. Across question versions, more than 4 of 5 physicians (83%) agree to the ac- ceptability of such procedures. Also, it is clear that physicians do discriminate between their reports of hypothetical behavior and eth- ical or behavioral standards. Many of the re- lationships observed in the former are attenu- ated or reversed in the latter. For example, more recently trained physicians report more frequently than their older counterparts that they would be likely to administer drugs in the situation described. This difference diminishes substantially when the question asked focuses on medical ethics. 232 Making Health Care Decisions: Appendix B Also, whereas physicians treating primarily those capable of understanding treatment are far more likely to note that they would be likely to administer potentially life-abbreviating drugs, compared with physicians treating proportionally few patients who are capable of understanding, this difference disappears when physicians are asked if it is ethically permissible to adminis- ter these drugs. Because agreement is obtained on the issue of ethics, we must speculate that differences emerging when the focus of the ques- tion is probable behavior must be the result of some more pragmatic concern (e.g., fear of later legal liability). As in one version of the physician ques- tionnaire, members of the public were asked if they feel it is permissible for a physician to administer drugs designed to reduce pain, even if the dose re- quired might shorten the patient's life (Table 8-17). Overall, 79% of the public report that it should be permissible for physicians to engage in such activities. As expected, respondent age, health status, and education affect respondents’ agreement. The data indicate that respondents who are older (69%), least educated (73%), in poor health (74%), or poor (66%) are all less likely to say that doctors should be permitted to administer drugs to reduce pain, if the dose required might shorten the patient's life, than are their younger (77%, 86%, and 82%), more educated (80%, 81%), healthier (81%, 79%, and 76%), or wealthier (81%, 82%, 83%, and 84%) counterparts. Observation: Zwservarion Clearly, the large majority of the public (79%) agree with the physicians surveyed and say that doctors should be allowed some discretion in the final treatment of dying patients. However, the oldest respondents, those in poorest health, the least educated, and the poorest are most hes- itant to embrace this treatment possibility. Interestingly, the physicians who generally treat them are often the ones most likely to say that they would administer such drugs. Speci- fically, the youngest doctors and those treating the seriously ill are more likely to administer drugs than older physicians or those treating Surveys of Physicians and the Public 233 relatively few seriously ill patients. However, when physicians are asked about whether such activities are ethically permissible, these ob- served relationships are attenuated -- in the case of recency of starting practice --— or re- versed -- in the case of percent of patients with serious illnesses. Termination Both doctors and the general public were asked to express their views on the issue of actively ter- minating a patient's life. Respondents were pre- sented with a stimulus situation depicting a dying patient in severe, unabatable pain who asks to have his life ended. One-fourth of the physician sample were asked if they would be likely to comply with the patient's wishes. Another quarter of the phy- sician sample and the full public sample were asked if it would be ethically permissible or right for the physician to comply with the patient's requests (Tables 8-18, 8-19, 8-20). Only very few physicians note that they think it is ethically permissible to comply with this dving patient's wishes (4%). Another 6% say that "it would depend," and 6% say they are "not sure." Thus, at maximum, only about 1 of 6 doctors (16%) might feel such an act is ethically permissible. The large majority of physicians (81%) say that ter- mination is not ethically permissible. Inspection of data reveals that, in general, hospital-based physicians and physicians most recently beginning practice -- those most likely to deal firsthand with such cases -- are most apt to view termination as falling within the tolerance for ethical medical practice (10% and 8%, respectively). When physicians are asked if they would be like- ly to comply with this dying patient's wishes, the overall number of physicians responding affirmative- ly drops, even when hedges (i.e., "not sure' and "it depends") are included (Table 8-19). Only 2% of all physicians say that they would be likely to comply with a patient's request for termination. Addition- ally, 8% reply that their compliance would rest on some unspecified contingencies, while another 3% say they are not sure. Thus, a total of 13% of physi- cians say they might, at one time or another, ter- minate a patient in an instance such as the one pre- sented. 234 Making Health Care Decisions: Appendix B Similar to the findings discussed previously in this chapter under the topic of abbreviating life, trends in patient characteristic data observed when physicians are asked if the behavior is ethically permissible flip-flop when physicians are asked to describe their likely behavior. For example, al- though doctors treating a large number of those un- able to understand treatment are more likely to say that it is ethically permissible to terminate the stimulus patient (7%), relative to doctors treating few patients that cannot understand their treatment (2%), when doctors are asked about their probable behavior, those treating high concentrations of pa- tients unable to understand their treatment are less likely (0%) to report probable compliance than phy- sicians treating relatively few such clients (3% or 1%). Again, the discussion advanced under the ab- breviating life section of this chapter is germane. Unlike physicians, members of the public are far more tolerant and supportive in their stated views on compliance with a dying patient's request to have his life ended (Tabe 8-20). A plurality of the pub- lic (46%) say that it would be right for a doctor to comply with a dying patient's request to be ter- minated. However, this does not represent an over- whelming endorsement of euthanasia by the public. Forty-three percent of the public do not endorse termination; another 11% are not sure. Clearly, this minority could shift the majority of the pub- lic's level of approval either way. Further, it should be pointed out that the amount of public support for termination in this study falls far short of what might be anticipated from the results of previous survey efforts. The number (and in this case plurality) of per- sons endorsing doctor compliance with a dying pa- tient's request for termination varies as a function of respondent characteristics. Reported health sta- tus and household income are positively related to viewing physician compliance as ethical; respondent age is negatively related to holding this view- point. Specifically, those who are in poor health (37%), old (40%), or poor (41%) are less likely than their healthier (50%), younger (51%) or financially better off (47%) counterparts to view physician com- pliance in this situation as right. Observation: yoservarion Few physicians feel that complying with the re- quest of a dying patient in severe, unabatable Surveys of Physicians and the Public 235 distress to be put to rest is ethical. Still fewer physicians report that they would termi- nate such a patient's life upon the patient's request. In sharp contrast to the view of phy- sicians, the public generally views compliance with a dying patient's request as ethically per- missible. Life-Abbreviating Drugs, Termination, and the Law The previous survey items tapped respondents’ views of abbreviation of life and termination in the context of probable behavior and ethically permis-— sible behavior. Two additional questions were in- cluded in the survey to determine respondents’ views concerning possible legal sanctioning of termination. Initially, respondents were asked whether they thought the law should allow doctors to give pain- killing drugs that might shorten life to patients in severe pain with no hope of recovery, if the patient had requested the drug and understands the con- sequences (Tables 8-21, 8-22). One would expect that these data would parallel those obtained pre- viously on the issues of abbreviating life (Tables 8-15, 8-16, 8-17). In point of fact, one would ex- pect that the number of physicians endorsing legal sanctioning of administering pain-reducing (albeit potentially life-shortening) drugs might exceed the number saying they would be likely to administer or that it would be ethically permissible to administer such drugs without legal sanction. This is not the case. Recall that 84% of the physicians say that they would be likely to administer such drugs, while 82% say it would be ethically permissible to ad- minister such drugs. However, only 53% of the doc- tors surveyed feel that the law should allow them to administer pain-relieving drugs. This set of findings is provocative. The no- ticeable decline in physicians endorsing legal sanc- tioning of pain-reducing, potentially life-abbre- viating drugs may be the result of either their view that such a decision revolves around professional ethics and practices and thereby should fall outside the jurisdiction of the law, or the fact that the patient requested the administration of the drug, or both -- the patient's request, if legally sanction- ed, could coerce physician compliance. Regardless 236 Making Health Care Decisions: Appendix B of the reasons for the precipitous dropoff of sup- port for administration of palliative drugs de- scribed above, a majority of physicians still state that they think the law should sanction such activities. In contrast to the bare majority of physicians endorsing legal sanctions of palliation at the risk of death, a 3-to-1 majority of the general public (71% vs. 24%) feel the law should allow physicians to comply with a dying patient's request to have his pain eased. Clearly, the difference in perspective between the public and the physicians accounts for disparate results. Even though 79% of the public support the admin- istration of such drugs by a physician without his or her necessarily discussing it with the patient, and only 71% endorse the legal sanctioning of a phy- sician's compliance with a patient's request, the decrease in support is nowhere near as precipitous as that observed for physicians (82% to 53%). A review of the survey data reveals that poor health status, advanced age, and low educational attainment are each related to the lowest levels of public support observed for legal sanctioning of physicians administering potentially 1life-abbre- viating drugs. These data are wholly consonant with previous findings, which indicate that these factors serve as sensitizing agents disposing many of the respondents in each of these groups to support life-preserving and not life-abbreviating medical techniques. Still, it must be remembered that in each instance, a majority of the respondents support the proposed legal sanctions. A second question was presented to physicians and the public to determine their reactions to the idea of legal sanctioning of physicians complying with the wishes of a dying patient to have his or her life ended. Recall that when asked whether it is ethically permissible to comply with requests for termination, only a small fraction of doctors (4%) agree; an even smaller portion of the sampled physi- cians say they would be likely to comply (2%). When asked if the law should sanction such compliance, more than 1 of 4 physicians (26%) answer affirma- tively. Hospital-based physicians (34%), those who have begun practice recently (29%), and those with relatively high concentrations of seriously ill (35% Surveys of Physicians and the Public 237 and 26%) or poor (34%) patients are most likely to voice support for legal sanctioning of physician compliance with patient requests for termination. Observation: Whereas the proportion of physicians endorsing the administration of potentially life- abbreviating drugs decreases sharply when physi- cians are asked if the law should allow such activity, compared with when such activity is governed only by their own view of professional responsibility or ethics, the number of physi- cians endorsing legalizing outright termination increases sharply when legal sanctions are in- troduced in the stimulus situation. Clearly, the similarity of the situations -- a patient who will inevitably die in a short time who is also in severe distress -- should not support such dissimilar response patterns. Neither can statistical artifacts be invoked to explain away these dramatic differences. Perhaps the small uncertainty about the immediate cause of death that allows physicians to embrace administering almost lethal doses of narcotics to dying pa- tients is wholly absent in outright termina- tion. In a similar vein, perhaps the profes- sional ethic embodied in the Hippocratic Oath distinguishes between these two situations. However, this does not explain the observed in- teractive effects that the introduction of legal sanctioning has on physicians' responses. To explain this seeming dualism, we must utilize other concepts, such as the fear of losing con- trol or protection under the law. The use of such formulations as these are supported to some extent by the fact that physician and practice characteristics tend to predict adequately phy- sician views on these issues. Perhaps the most parsimonious explanation of this diverse set of findings utilizes or explains physicians' ethics as situationally specified. When doctors are asked whether they feel the law should allow them to comply with a dying patient's wishes to be terminated, the number of doctors voic- ing assent increases dramatically from that observed when they are asked if they feel it ethically per- missible to comply with their patient's wishes. In 238 Making Health Care Decisions: Appendix B contrast, the public's degree of endorsement in- creases only slightly -- from 46% to 53% across these two situations (Tables 8-20, 8-24). This mod- est increase may simply reflect the much higher ini- tial rate of affirmative responses among the public, compared with the physician sample. Alternately, by adding legal sanctions, some individuals' doubts, based on their view that such activity might be il- legal or immoral, or both, may have been assuaged., In this instance, a clear majority opinion (53%) is achieved. The public clearly feels the law should allow doctors to comply with the wishes of a dying patient in severe distress who asks to have his life ended. Again, poor health status (48%), advanced age (42%), low educational attainment (50%), and low household income (43%) are each related to relative- ly low rates of endorsement of legal sanctioning of targeted physician behavior. These results are com- pletely consonant with previous findings and anal- yses presented both in this chapter and elsewhere in this report to explain these patterns. G-37 Q.: If you were seriously ill, would you want the responsibility of making the final choices about your medical tr TABLE 8-1 RESPONSIBILITY FOR FINAL CHOICES IF SERIOUSLY ILL: want your doctor to make them for you? BASE PATIENT MAKES FINAL CHOICES DOCTOR MAKES FINAL CHOICES BOTH (VOL.) NOT SURE GENERAL PUBLIC GENERAL PUBLIC eatment, or would you 1251 z 43 38 14 4 482 z 51 32 14 3 REPORTED HEALTH STATUS TOTAL EXCELLENT GOOD FAIR POOR YEARS YEARS YEARS 550 z 43 38 14 5 EDUCATION LESS AGE THAN INCOME 65 HIGH HIGH $7,500 $7,501 $15,001 $25,001 MORE 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE OR TO TO TO THAN OVER GRADUATE GRADUATE GRADUATE LESS $15,000 $25,000 $35,000 $35,000 16 73 488 334 226 195 242 726 280 164 219 341 229 214 x x x 2 2 z 2 zx z x x x z z 33 17 56 45 29 23 27 45 52 32 40 49 41 47 46 61 28 36 47 60 55 37 27 55 41 35 32 33 17 20 14 15 17 12 13 14 16 11 14 12 18 16 3 - 3 3 6 2 5 3 5 2 4 3 4 3 orqng oy pue sueloIsAyd jo sheaing 6€C TABLE 8-2 TREATMENT DECISIONS WHEN PATIENT IS TOO SICK TO DECIDE: GENERAL PUBLIC Q.: If you were too sick to make an important decision about your medical care, who would you want to make the final decision for you -- a family member, a close friend, your doctor, or a lawyer appointed to protect your interests? ——— *LESS THAN 0.5%. TOTAL GENERAL PUBLIC —_— es. "UBL BASE 1251 x A FAMILY MEMBER 57 A CLOSE FRIEND 2 YOUR DOCTOR 31 A LAWYER 2 DOCTOR AND : FAMILY/FRIEND (VOL.) 6 OTHER * NOT SURE 1 ove fd xrpuaddy :suorstoa(] aren yieay Suryepy PROVIDED TREATMENT INSTRUCTIONS: TABLE 8-3 GENERAL PUBLIC Q.: Have you given instructions to anyone about how you would like to be treated if you ever became too sick to make decisions? GENERAL PUBLIC REPORTED HEALTH STATUS TOTAL EXCELLENT GOOD FAIR POOR EDUCATION LESS AGE THAN LIFE- 65 HIGH HIGH THREATENING 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE ILLNESS YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE YES NO BASE 1251 %82 550 144 73 488 334 226 195 242 726 280 385 861 % 2 rT x 2 % % 2 % z z z 2 YES, HAVE GIVEN 36 40 32 32 54 33 36 42 39 34 34 43 42 34 NO, HAVE NOT GIVEN 63 60 68 68 45 67 64 58 60 65 66 56 58 66 NOT SURE * * - = - - - - 2 * - - - * *LESS THAN 0.5Z. o1[qnd 8Y} pue suelIsAYd jo sfeaing ve G-40 TABLE 8-4 LEFT WRITTEN TREATMENT INSTRUCTIONS: GENERAL PUBLIC BASE: HAVE GIVEN TREATMENT INSTRUCTIONS Q.: Did you put those instructions in writing? GENERAL PUBLIC EDUCATION LESS THAN HIGH HIGH REPORTED HEALTH STATUS SCHOOL SCHOOL COLLEGE TOTAL EXCELLENT GOOD FAIR POOR GRADUATE GRADUATE GRADUATE BASE 451 196 166 47 41 82 254 115 % % z Z Z % z Z YES, PUT IN WRITING 23 23 22 17 30 24 19 28 NO, DID NOT PUT IN WRITING 76 76 77 83 70 74 81 69 NOT SURE * 1 - - - 2 - - *LESS THAN 0.5%. [444 fg x1puaddy :suorsia(] aren yieay Suryep G-41 TABLE 8-5 SHOULD LAW ALLOW PHYSICIANS TO HONOR PATIENTS' WRITTEN INSTRUCTIONS: GENERAL PUBLIC Q.: Do you think that the law should allow doctors to honor the written instructions of their patients, even if it means allowing them to die? GENERAL PUBLIC EDUCATION LESS AGE THAN INCOME 65 HIGH HIGH $7,500 $7,501 $15,001 $25,001 MORE REPORTED HEALTH STATUS 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE OR TO TO TO THAN TOTAL EXCELLENT GOOD FAIR POOR YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 482 550 144 73 488 334 226 195 242 726 280 164 219 341 229 214 z z 2 x x 2 x z x 2 z z z 2 2 x x YES, SHOULD ALLOW 68 75 65 62 60 71 3 67 54 58 70 71 58 64 72 76 70 NO, SHOULD NOT ALLOW 20 18 22 23 24 24 15 18 24 27 19 19 24 25 18 18 21 NOT SURE 11 7 12 16 16 5 12 14 22 15 11 9 18 11 10 6 8 o1[qng ay} pue SuedIsAyd Jo sheng e€ve G-42 TABLE 8-6 DESIRE TO BE INFORMED OF DIAGNOSIS OF CANCER: GENERAL PUBLIC Q.: Now I would like to get your opinions on some hypothetical situations. Suppose that a doctor found that you had cancer. Would you want the doctor to tell you the diagnosis, or would you prefer that he treat you without telling you what was wrong? TOTAL GENERAL PUBLIC BASE 1251 7% WANT THE DOCTOR TO TELL ME 96 PREFER TREATMENT WITHOUT TELLING 3 NOT SURE 1 vve fd xtpuaddy :suorstoa( aren) yjesy Sunyey G-43 TABLE 8-7 PROVIDE ESTIMATE OF LIFE EXPECTANCY TO DYING PATIENTS: Q.: Suppose that you had a type of cancer that usually leads to death in less than a year. estimate of how long you had to live, or would you prefer that he not tell you? GENERAL PUBLIC GENERAL PUBLIC Would you want your doctor to give you a realistic EDUCATION LESS AGE THAN INCOME 65 HIGH HIGH $7,500 $7,501 $15,001 $25,001 MORE REPORTED HEALTH STATUS 18-34 35-50 51-64 AND SCHOOL ~~ SCHOOL COLLEGE OR TO TO TO THAN TOTAL EXCELLENT GOOD FAIR POOR YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 482 550 144 73 488 334 226 195 242 726 164 213 341 yr z x 2 z z 2 2 z 2 z 2 z z 2 z z x GIVE REALISTIC ESTIMATE 85 89 8 13 15 84 92 86 75 78 84 93 76 81 91 87 87 NOT TELL ME 14 10 12 25 22 14 8 12 22 20 15 5 23 18 8 12 11 NOT SURE 2 1 2 2 4 1 1 2 3 2 2 1 1 1 2 1 1 o1[qnd 8Y} pue sueIISAY{ Jo sAoaing che P-26A TABLE 8-8 PROVIDE ESTIMATE OF LIFE EXPECTANCY TO DYING PATIENT: PHYSICIANS Q.: Now, I would like to get your opinion on some hypothetical situations. If you had a patient with a fully confirmed diagnosis of lung cancer in an advanced stage, which of the following would you be most likely to tell your patient? Would you (READ LIST)? PHYSICIANS PATIENT WITH OBTAINED SERIOUS PATIENTS MEDICAL DEGREE ILLNESS UNDERSTAND ING POOR PATIENTS PRACTICE 1966 1967 1973 MORE TREATMENT 252 51% LOCATION AND TO AND 0X TO 26% THAN 0X TO 70% 902 TO OR 26% OR TOTAL OFFICE HOSPITAL BEFORE 1972 AFTER 252 TO 50% 50% 69% TO 89% 1002 LESS TO 50% MORE BASE 805 533 232 441 97 266 367 229 193 150 270 384 500 143 107 z z 2 z 2 Z z 2 x 2 Z z z z 2 GIVE A STRAIGHT STATISTICAL PROGNOSIS FOR HIS CLASS OF DISEASE 13 14 12 14 14 11 12 14 14 11 14 14 12 14 17 SAY THAT YOU CAN'T TELL HOW LONG HE MIGHT LIVE, BUT STRESS THAT IT COULD BE FOR A SUBSTANTIAL PERIOD OF TIME 33 34 32 38 25 27 33 34 31 33 31 34 35 36 22 SAY THAT YOU CAN'T TELL HOW LONG HE MIGHT LIVE, BUT STRESS THAT IN MOST CASES PEOPLE LIVE NO LONGER THAN A YEAR 28 24 38 18 38 41 27 28 30 31 29 26 27 29 36 REFUSE TO SPECULATE ON HOW LONG THE PATIENT MIGHT LIVE 22 24 15 26 16 16 22 19 22 23 20 22 22 18 22 NOT SURE 5 5 3 4 6 5 5 4 4 2 7 4 5 3 4 9%C dg x1puaddy :suorsia( aren) yjesy Sune G-44 TABLE 8-9 DECISION BETWEEN AGGRESSIVE THERAPY AND SUPPORTIVE THERAPY: GENERAL PUBLIC Q.: Assume that the cancer has spread throughout the body. It is very likely that you will die regardless of what you do. There is a decision between ressive thera -- which will probably make you feel sick and will probably not help your condition -- and supportive therapy -= which will not help your condition, but will allow you to be comfortable. Who do you think should make the decision between aggressive therapy and supportive therapy? BASE DOCTOR SHOULD MAKE THE DECISION I SHOULD MAKE THE DECISION SHOULD BE JOINT DECISION (VOL.) NOT SURE *LESS THAN 0.5%. GENERAL PUBLIC REPORTED HEALTH STATUS EXCELLENT GOOD FAIR POOR 482 550 14 EDUCATION LESS AGE THAN INCOME 65 HIGH HIGH $7,500 $7,501 $15,001 $25,001 MORE 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE SEX OR TO TO TO THAN YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE MALE FEMALE LESS $15,000 $25,000 $35,000 $35,000 %88 334 226 195 262 726 280 610 641 164 219 341 229 214 z z ¥ 2 2 2 S12 1 z x 2 1 % z 7 9 18 21 21 11 5 12 12 25 14 9 6 6 87 84 68 65 66 81 86 79 80 65 76 80 86 86 6 7 12 9 8 8 8 8 8 8 9 10 7 8 - 1 1 5 4 * * 1 1 2 1 1 1 - o1[qnd 8Y}) PUB SUBIISAY JO SA2AING L¥C P-26C TABLE 8-10 WHO SHOULD MAKE TREATMENT DECISION: PHYSICIANS Q.: Ion this kind of situation who should make/usually makes the decision about treatment -- the doctor or the patient? BASE DOCTOR PATIENT JOINTLY NOT SURE TOTAL PHYSICIANS SHOULD MAKE DECISION USUALLY MAKES DECISION 401 % 24 67 404 Z 15 19 64 NOTE: SPLIT BALLOT ITEM, EACH QUESTION ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. 8%¢ fd xrpuaddy :suorsioa(y aren) yjesy Sune P-27A TABLE 8-11 RESUSCITATION OF A DYING PATIENT: PHYSICIANS Q.: Now, imagine you have a hospitalized patient in great pain in the end stages of a degenerative disease. You have been treating her for this disorder for a long time and know that she will never leave the hospital again. In the absence of any guidance from the patient, how likely would you be to order that she not be resuscitated in the case of cardiac arrest? TOTAL PHYSICIANS BASE 805 % VERY LIKELY 57 SOMEWHAT LIKELY 15 SOMEWHAT UNLIKELY 7 VERY UNLIKELY 16 NOT SURE 4 o1[qnd 8Y} pue SueIISAY JO sAdAING 6%C P-27B TABLE 8-12 INITIATION OF DISCUSSION WITH DYING PATIENT ABOUT RESUSCITATION: PHYSICIANS Q.: Would you initiate a discussion with this patient about her preferences concerning resuscitation? BASE YES, WOULD INITIATE NO, WOULD NOT INITIATE NOT SURE PHYSICIANS PATIENTS WITH OBTAINED SERIOUS PATIENT MEDICAL DEGREE ILLNESS UNDERSTANDING POOR PATIENTS PRACTICE 1966 1967 1973 MORE TREATMENT 25% 51 LOCATION AND TO AND OX TO 26% THAN 0% TO 70% 902 TO OR 262 OR TOTAL OFFICE HOSPITAL BEFORE 1972 AFTER 25% TO 50% 502 692 TO 89% 100% LESS TO 50% MORE 805 533 232 441 97 266 363 229 193 150 270 384 500 143 107 Z Z % % Z 2 % % % Zz % 2 z Z Z 52 49 61 40 56 72 51 50 59 59 54 48 50 57 62 38 40 32 51 32 18 40 41 33 33 37 42 40 34 3 9 11 7 10 7 10 9 9 8 8 9 10 10 9 8 0S? dg x1puaddy :suorsma( aren yiesy Surmyeiy P-27C TABLE 8-13 RESUSCITATION OF DYING PATIENT WHO REQUESTED LIFE MAINTENANCE: PHYSICIANS Q.: If this patient asked you to do everything you could to maintain her life, how likely would you be to order that she not be resuscitated in case of cardiac arrest -- very likely, somewhat likely, somewhat unlikely, or very unlikely? PHYSICIANS OBTAINED PATIENTS WITH PATIENTS MEDICAL DEGREE SERIOUS ILLNESS UNDERSTANDING POOR PATIENTS 1966 1967 1973 MORE TREATMENT 25% 51% AND TO AND 0% TO 26% THAN 0% TO 70% 90% TO OR 26% OR TOTAL BEFORE 1972 AFTER 25% TO 50% 50% 69% TO 89% 100% LESS TO 50% MORE BASE 805 441 97 266 367 229 193 150 270 384 500 143 107 % x z Z % % % Z % % 2 x 2 VERY LIKELY 9 12 5 6 8 10 11 9 10 9 10 5 10 SOMEWHAT LIKELY 7 2 7 7 5 9 8 10 7 6 7 5 10 SOMEWHAT UNLIKELY 12 9 9 18 11 11 15 17 11 11 11 13 16 VERY UNLIKELY 70 70 7» 68 74 67 65 63 n 73 70 76 62 NOT SURE 2 2 3 1 2 2 1 1 2 2 2 1 2 o1[qnd 8Y} pue SueIISAY{ Jo sAeaing 162 P-27D TABLE 8-14 RESUSCITATION OF DYING PATIENT WHO PROVIDED WRITTEN INSTRUCTIONS: PHYSICIANS Q.: Suppose the patient is now incompetent but had previously written instructions indicating that no extraordinary care be provided and that she be allowed to "die naturally." If this patient suffered a cardiac arrest, would you resuscitate her? Q.: Suppose the patient is now incompetent but had previously written instructions indicating that she wished to have everything possible dome to prolong her life. If this patient suffered a cardiac arrest, would you resuscitate her? TOTAL PHYSICIANS DIE NATURALLY PROLONG LIFE BASE 404 401 z x YES, WOULD RESUSCITATE 5 52 NO, WOULD NOT RESUSCITATE 84 27 IT DEPENDS (VOL.) 11 19 NOT SURE 1 3 NOTE: SPLIT BALLOT ITEMS, EACH QUESTION ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. SC dg xrpuaddy :suorsia(] aren) yjesy Sunyey Surveys of Physicians and the Public 253 P-28 TABLE 8-15 LIKELIHOOD OF ADMINISTERING NARCOTICS TO REDUCE PAIN: PHYSICIANS Q.: Assume that another patient who is in severe pain and with no hope of recovery has asked that you help ease the pain, knowing it might shorten his or her life. Under these circumstances would you be likely to administer narcotics to reduce pain, even if the dose required might shorten life? PHYSICIANS OBTAINED PATIENTS WITH PATIENTS MEDICAL DEGREE SERIOUS ILLNESS UNDERSTANDING 1966 1967 1973 MORE TREATMENT AND TO AND 0X TO 26% THAN 0% To 70% 90% TO TOTAL BEFORE 1972 AFTER _25% TO 50% 50% 69% TO 89% _100% 5 eae BASE 199 118 23 97 53 5 37 69 92 x x z x x Z 2 % x x YES 84 83 65 93 82 87 87 76 84 87 NO 5 7 4 2 4 6 4 5 6 4 IT DEPENDS (voL.) 9 9 22 5 11 6 9 19 7 6 NOT SURE 1 1 - - - 2 - - @ 1 NO ANSWER/ REFUSED 2 1 9 - 3 - - - 3 1 NOTE: SPLIT BALLOT, ASKED OF ONE-QUARTER OF PHYSICIAN SAMPLE. 254 Making Health Care Decisions: Appendix B P-28 TABLE 8-16 PERMISSIBILITY OF ADMINISTERING NARCOTICS TO REDUCE PAIN: PHYSICIANS Q.: Assume that ancther patient who is in severe pain and with no hope of recovery has asked that you help ease the pain, knowing it might shorten his or her life. Under these circumstances is it ethically permissable to edmicister narcotics to reduce pain, even if the dose required might shorten life? PHYSICIANS PATIENTS OBTAINED PATIENTS WITH UNDERSTANDING MEDICAL DEGREE SERIOUS ILLNESS TREATMENT 1966 1967 1973 MORE of 10 70% 90f TO AND TO AND 02 TO 262 TO THAN TOTAL _69% TO 802 _100X BEFORE 1972 AFTER 25% 50% 50% BASE 202 41 67 94 116 21 65 98 54 45 1 2 2 2 2 2 2 2 Z 2 YES 82 83 80 82 83 67 85 84 81 80 NO 2 - 4 2 2 10 2 4 - 2 IT DEPENDS (voL.) 9 10 8 11 10 10 9 7 9 11 NOT SURE “4 5 3 3 3 10 2 3 6 2 NO ANSWER/ REFUSED 3 2 4 2 3 5 3 2 4 4 —_— NOTE: SPLIT BALLOT, ASKED OF ONE-QUARTER OF PHYSICIAN SAMPLE. C-45 PERMISSIBILITY OF ADMINISTERING PALLIATIVE DRUGS: TABLE 8-17 GENERAL PUBLIC Q.: Assume that a patient who is in severe pain and with no hope of recovery has asked the doctors to help ease the pain, knowing that it might kill him. Under these circumstances should it be permissable for a doctor to administer drugs to reduce pain, even if the dose required might shorten the patient's life? YES, SHOULD BE NO, SHOULD NOT BE NOT SURE GENERAL PUBLIC EDUCATION LESS AGE THAN INCOME 65 HIGH HIGH $7,500 $7,501 315,001 $25,001 MORE REPORTED HEALTH STATUS 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE OR TO TO TO THAN TOTAL EXCELLENT ATR POOR YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE _LESS $15,000 $25,000 $35,000 $35,000 1251 482 550 144 73 488 33 226 195 242 726 280 164 219 341 229 214 1 1 Tx 2 2 1 1 2 2 2 2 2 1 1 2 79 81 79 16 14 1 86 82 69 73 80 81 66 81 82 83 82 14 12 15 18 13 18 4 11 18 16 13 14 21 14 11 13 11 7 7 6 8 13 4 10 7 1B 11 7 5 13 4 7 4 5 o1[qnd 9Y} pue SUeIISAYJ JO sAeAIng GG2Z TABLE 8-18 TERMINATION OF LIFE AS ETHICALLY PERMISSIBLE: PHYSICIANS Q.: [Imagine that a dying patient in severe distress, which cannot be relieved, asks to have his life ended. Under these circumstances is it ethically permissible to comply with the patient's wishes? PHYSICIANS OBTAINED PATIENTS MEDICAL DEGREE UNDERSTANDING POOR PATIENTS PRACTICE 1966 1967 1973 TREATMENT 25% 51% LOCATION AND TO AND 0% TO 70% 90% TO OR 26% OR TOTAL OFFICE HOSPITAL BEFORE 1972 AFTER 69% TO 89% 100% LESS TO 50% MORE BASE 202 137 59 116 21 65 98 54 45 134 31 29 % % x % % % z % z x % Zz YES 4 2 10 3 5 8 7 6 2 4 3 7 NO 81 84 74 84 86 75 78 78 85 84 81 76 IT DEPENDS (VOL.) 6 7 7 8 5 5 7 8 5 6 6 3 NOT SURE 6 6 5 4 - 11 2 6 7 5 10 7 NO ANSWER/ REFUSED 2 2 3 2 5 2 5 3 - 2 - 7 NOTE: SPLIT BALLOT ITEMS, ASKED OF ONE-QUARTER OF PHYSICIAN SAMPLE. 96¢ dg x1puaddy :suoisioa(q are) yjeay Surely P-29 Q.: Imagine that a dying patient in severe distress, which cannot be relieved, asks to have his life ended. Under these circumstances would you be likely to comply with the patient's wishes? PHYSICIANS OBTAINED PATIENTS MEDICAL DEGREE UNDERSTANDING POOR PATIENTS PRACTICE 1966 1967 1973 TREATMENT 25% 51% LOCATION AND TO AND 02 70% 90% OR 26% OR TOTAL OFFICE HOSPITAL BEFORE 1972 AFTER TO 69% TO 89% TO 100% LESS TO 502 MORE BASE 199 126 62 118 23 58 37 69 92 111 41 24 % % z % x % % % % % % % YES 2 2 2 1 “ 3 - 3 1 2 ~ 4 NO 85 83 87 86 70 88 89 84 84 84 88 83 IT DEPENDS (VOL.) 8 10 6 8 17 5 5 7 10 8 7 4 NOT SURE 3 3 3 2 9 3 3 3 3 3 5 4 NO ANSWER/REFUSED 2 2 2 3 4 - 3 2 2 4 - 4 NOTE: SPLIT BALLOT ITEM, ASKED OF ONE-QUARTER OF PHYSICIAN SAMPLE. TABLE 8-19 LIKELIHOOD OF TERMINATING LIFE: PHYSICIANS o1qngd 98Y} pue SUeIISAYJ jo sfaaIng LST G-46 TABLE 8-20 PERMISSIBILITY OF PHYSICIAN COMPLYING WITH A REQUEST FOR TERMINATION: GENERAL PUBLIC Q.: [Imagine that this patient asks to have his life ended. Under these circumstances do you think it would be right for a doctor to comply with the patient's wishes, or would it be wrong? GENERAL PUBLIC AGE INCOME 65 $7,500 $7,501 $15,001 $25,001 MORE REPORTED HEALTH STATUS 18-34 35-50 51-64 AND OR TO TO TO THAN TOTAL EXCELLENT GOOD FAIR POOR YEARS YEARS YEARS OVER LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 482 550 144 73 488 334 226 195 164 219 341 229 214 z % % 2 2 % % z z % % % z % WOULD BE RIGHT 46 50 43 48 37 50 46 40 40 41 45 46 47 47 WOULD BE WRONG 43 39 45 44 50 40 40 48 49 47 49 44 41 38 NOT SURE 11 11 12 8 13 9 12 12 12 11 6 11 12 15 862 fg x1puaddy :suorsioa(q are) yea Sunyepy Surveys of Physicians and the Public 259 TABLE 8-21 SHOULD LAW ALLOW ADMINISTRATION OF POTENTIALLY LIFE-ABBREVIATING DRUGS TO REDUCE PAIN: PHYSICIANS Q.: Do you think that the law should allow doctors to administer drugs that might shorten life to patients in severe pain with no hope of recovery, if the patient has requested the drug and understands the consequences? PHYSICIANS OBTAINED MEDICAL DEGREE POOR PATIENTS PRACTICE 1966 1967 1973 25% 51% LOCATION AND TO AND OR 26% OR TOTAL OFFICE HOSPITAL BEFORE 1972 AFTER LESS TO 50% MORE BASE 401 263 121 234 44 123 245 72 53 z 2 4 x Z x % x x SHOULD ALLOW 53 Si 58 49 50 63 54 54 62 SHOULD NOT ALLOW 35 36 32 38 39 28 34 35 28 NOT SURE 12 13 10 13 11 10 11 11 9 NOTE: SPLIT-BALLOT ITEM, ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. G-47 TABLE 8-22 SHOULD LAW ALLOW ADMINISTRATION OF POTENTIALLY LIFE-ABBREVIATING DRUGS: GENERAL PUBLIC Q.: Do you think that the law should allow doctors to administer drugs that might shorten the patient's life in severe cases with no hope of recovery, if the patient has requested the drug and understands the consequences? BASE SHOULD ALLOW SHOULD NOT ALLOW NOT SURE GENERAL PUBLIC EDUCATION LESS AGE THAN 65 HIGH HIGH REPORTED HEALTH STATUS 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE TOTAL EXCELLENT GOOD FAIR POOR YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE 1251 482 550 144 73 488 334 226 195 242 726 280 2 Zz z Z z % Z Z % Z x Z 71 76 71 63 52 78 74 63 56 60 75 69 24 20 23 27 44 19 22 29 32 33 20 25 5 4 6 9 4 2 4 8 12 7 S 6 092 dg xipuaddy :suorsioa(q are) yieay Suryep TABLE 8-23 SHOULD LAW ALLOW PHYSICIANS TO COMPLY WITH DYING PATIENT'S REQUEST TO END LIFE: Q.: Do you think that the law should allow doctors to comply with the wishes of a dying patient in severe distress who asks to have his life ended? BASE SHOULD ALLOW SHOULD NOT ALLOW NOT SURE PHYSICIANS PHYSICIANS PATIENTS WITH OBTAINED SERIOUS MEDICAL DEGREE ILLNESS POOR PATIENTS PRACTICE 1966 1967 1973 MORE 252 51% LOCAT ION AND TO AND 02 26% THAN OR 262% OR TOTAL OFFICE HOSPITAL BEFORE 1972 AFTER TO 25% TO 50% 90% LESS TO 50% MORE 401 263 121 234 44 123 195 107 90 245 72 53 Z x Z Z Z x Z Z % % z z 26 23 34 24 30 29 22 35 26 25 26 34 59 60 57 59 59 59 64 49 63 60 60 55 15 17 9 17 11 11 14 17 11 15 14 11 NOTE: SPLIT BALLOT ITEM, ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. o1[qnd 9Y} pue SUBIDISAYJ JO SAdAING 192 G-48 SHOULD LAW ALLOW PHYSICIANS TO COMPLY WITH DYING PATIENT'S REQUEST TO END LIFE: TABLE 8-24 GENERAL PUBLIC Q.: Do you think that the law should allow doctors to comply with the wishes of a dying patient in severe distress who asks to have his life ended? GENERAL PUBLIC EDUCATION LESS AGE THAN INCOME 65 HIGH HIGH $7,500 $7,501 $15,001 $25,001 MORE REPORTED HEALTH STATUS 18-34 35-50 51-64 AND SCHOOL SCHOOL COLLEGE OR TO TO THAN TOTAL EXCELLENT GOOD FAIR POOR YEARS YEARS YEARS OVER GRADUATE GRADUATE GRADUATE LESS $15,000 $25,000 $35,000 $35,000 BASE 1251 ~~ 482 550 144 713 488 334 226 195 242 726 280 164 219 341 229 214 z % x x 2 z 2 z z z 2 z 2 z z x z SHOULD ALLOW 53 58 50 50 48 61 54 42 42 50 54 53 43 51 57 55 55 SHOULD NOT ALLOW 38 34 40 39 40 33 36 46 43 40 37 38 45 42 35 37 34 NOT SURE 8 6 10 10 10 6 8 9 13 9 8 8 11 6 6 6 9 NO ANSWER/REFUSED 2 2 1 2 1 * 2 3 2 1 2 1 ) 1 1 1 2 2 *LESS THAN 0.5%. 292 dg xtpuaddy :suorstoa(] aren yjeay Sunyey Surveys of Physicians and the Public 263 CHAPTER 9: IMPACT OF PROFESSIONAL EDUCATION AND VALUES ON PHYSICIANS' ATTITUDES AND BEHAVIOR OF PHYSICIANS RELATED TO INFORMED CONSENT Introduction Throughout this report, we have seen that dif- ferences in physicians' affect, attitudes, and re- ported behaviors have been consistently related to physician characteristics, such as recency of begin- ning practice and type of medical school attended. These differences can be attributed in whole or in part to secular change over time or changes in the training of physicians. It is through this latter process that professional ethics, which might be expected to have a profound effect on physicians’ attitudes and behavior, are mainly transferred. To determine the impact of changes in the med- ical education system on physicians' attitudes to- ward and behavior regarding information disclosure and medical decisionmaking, components of physician training (e.g., completion of doctor-patient com- munication coursework, type of medical school from which graduated) were compared with responses on key indicators of propensity to disclose and preferred locus of decisionmaking. The results of this anal- ysis suggest medical education does have a rel- atively systematic impact on a physician's practice, but this impact is not dramatic. The remainder of this chapter presents illustrative data analyses. Background of Physicians Demographic variables are frequently the most powerful factors in explaining differences in the attitudes and behavior within a specified pop- ulation. Physicians are frequently treated as a homogeneous group because the range of variation among physicians on many important sociodemographic variables is quite restricted. Clearly, the char- acteristics of the physician population are dif- ferent from those of the general public. By def- inition, physicians are college educated, while only 22% of the adult population of the United States have a college degree. Virtually all physicians (94%) are men (Table 9-1). Although not investigat- ed in this survey, physicians' incomes are consider- ably higher than the national norms. 264 Making Health Care Decisions: Appendix B Despite the more restricted range among physi- cians on many important sociodemographic variables, it would be a mistake to assume that they necessari- ly represent a socially homogeneous group. First, physicians are distributed by age in a manner rough- ly equivalent to that of the general population dis- tribution, although the upper and lower ends of the adult population distribution are absent. Second, the religious background of physicians -- certainly an issue in a study of ethics -- is as diverse as that found in the general population. The survey finds that 42% of doctors were raised as Protes- tants, 26% were raised as Catholics, and 15% as Jews. Third, the socioeconomic background of physi- cians, as measured by father's educational attain- ment, is also varied. The survey finds that 24% of physicians have fathers with less than a high school degree, 27% have fathers with a high school degree, and 45% have fathers with a college degree. Al- though the survey finds that the average educational attainment of the fathers of current physicians is considerably higher than the educational attainment of a national cross section of adults, the survey also shows considerable variation in background among physicians. Indeed, we find that 6% of physi- cians were rated by interviewers as speaking English with difficulty, which is slightly more than the 4% of the public so rated (Table 9-2). Medical Educatjon of Physicians The norms of the medical profession are trans- mitted from one generation of doctors to the next primarily through the medical education process. The intensive training in medical schools represents the first stage of this process. Subsequent intern- ship and residency programs, where applicable, rep- resent the second stage. Differences in the medical education of doctors, consequently, could have an important impact upon their attitudes and behaviors. The age of the doc- tors reflects potential differences in their educa- tional experience. The type of medical school in which they were trained, as well as the type of courses that they took while in school, may also have an impact. The survey finds that 84% of the physicians were educated in American medical schools, while 16% are Surveys of Physicians and the Public 265 the graduates of foreign medical schools (Table 9-3). Most of the doctors who are not fluent in English are foreign medical school graduates (89%). However, only 31% of foreign medical school grad- uates were rated as speaking English with difficulty (Table 9-2). Nearly one-half of the physicians surveyed were educated at public medical schools in the United States (47%). An additional 38% were educated at private medical schools (Table 9-3). The current level of research funding for med- ical schools was calculated by the staff of the PCEMR and the medical schools were rated as among the top twenty, second twenty, or remainder in terms of their receipt of federal research dollars. The survey finds that 54% of physicians are graduates of medical schools with low research funding; 16% are graduates of medical schools with moderate research funding; and 15% are graduates of medical schools with high research funding (Table 9-3). (Foreign medical schools were not categorized by research funding.) Most of the physicians interviewed had been ex-— posed to formal training in the areas of medical ethics, medical law, or doctor-patient communica- tions while in medical school. One-half (50%) of the physicians report formal training in medical ethics while in medical school. More than one-half (54%) of these doctors have had formal training in doctor-patient communications. Over one-third (36%) report formal training in medical law. Formal training in medical ethics and medical law in medical school is not a new phenomenon. Nearly one-half of doctors who received their med- ical degree before 1967 (49%) report formal training in medical ethics while in school. Over one-third (35%) of doctors who graduated from medical school before 1967 report formal training in medical law. The proportions of post-1972 graduates who report formal training in medical ethics (58%) and medical law (41%) are only somewhat higher. The survey also finds a curious decline in the proportion of 1967-72 medical school graduates who report formal training in these areas. By contrast, there has been a clear increase in physician exposure to formal training in doctor-— 266 Making Health Care Decisions: Appendix B patient communication over the years. The propor- tion of doctors who report formal training during medical school in doctor-patient communication in- creases from 43% of those who graduated prior to 1967, to 53% of those who graduated from 1967-72, to 73% of those who graduated in 1973 or later (Table 9-4). There is little evidence that either the type of medical school, (i.e., public or private) or its research funding have any predictable impact on phy- sicians' exposure to courses in medical ethics or medical law. Doctors trained in public medical schools are somewhat more likely to have had formal training in doctor-patient communications than those trained in private medical schools, but the dif- ference is not dramatic. Similarly, there is little difference between foreign medical school graduates and American medical school graduates in the in- cidence of exposure to formal training in these areas during medical school (Table 9-5). Primacy of Values in Medical Practice Medical education has been given special atten- tion in this report for two reasons. It is assumed that one of the special marks of a profession, as distinct from an occupation, is that professions have a highly developed system of rules that govern the behavior of its members. This ethical system is instilled in new members of the profession during the professional education process. Thus, the type of formal professional training that is received by physicians should have an important impact on their attitudes toward issues of information disclosure and decisionmaking in treatment situations. Fur- ther, it would seem to be much more effective and less disruptive to focus on the medical education process, rather than the law, to achieve the goals of informed consent. The survey of physicians, however, suggests that the importance attributed to professional education in the development of norms governing medical prac- tice may be somewhat overrated. Physicians were asked: "Which of the following three items do you think has had the most influence on the way you practice medicine -- your personal values and be- liefs, your medical training, or your experience in practice?" The vast majority of primary care physi- Surveys of Physicians and the Public 267 cians -- 6 out of 10 (62%) -- say that their per- sonal values and beliefs have been most influential in the way they practice medicine. Fewer than 2 out of 10 physicians (18%) report that their medical training has been the most important influence on how they practice medicine, and the same proportion (18%) say that their experience in practice has been the most important factor. The age of the doctor has some effect on his or her beliefs in this area. Older doctors (21%) tend to put more emphasis on their experience in practice than do younger doctors (15%), while younger doctors put more emphasis on their personal values (65%) than older doctors (59%). The same proportion of doctors across age cohorts, however, feel that their medical training was most important (Table 9-6). The religious background of the doctors also has some effect on what they consider the most important influence on the way they practice medicine. Al- though majorities of doctors in every religious group (including '"none') report that their personal values and beliefs have been the most important in- fluence on how they practice medicine, the percent- age reporting this is higher among Catholics (70%) than among those in other religious categories (Protestants, 61%; Jewish, 57%; ‘'other," 58%; "none," 59%; "refused," 56%) (Table 9-7). Responsibility for Informing Patients Over three-quarters of physicians (77%) feel that it is primarily the doctor's responsibility to make sure that patients are fully informed about their condition and treatment. Formal training in ethics, law, and doctor-patient communications in medical school has no appreciable effect on this attitude (Table 9-8). Disclosure of Risk One important finding of the survey is that a significant difference exists between public ex- pectations and physician behavior in the disclosure of the more remote risks of treatment. No real dif- ference is found between physician behavior and pub- lic expectations when a 1 in 100 risk of death or serious injury is considered. But physicians become 268 Making Health Care Decisions: Appendix B less likely to initiate disclosure as the likelihood and severity of risk declines. The public, on the other hand, exhibits only a marginal decline in ex- pectations of physician disclosure of risk as the likelihood and seriousness of the risk decline. Have formal training in medical ethics, medical law, and doctor-patient communications increased physician sensitivity to public (and legal) expecta- tions of the disclosure of relatively remote risks? The survey suggests that medical education in this area may have had some effect, but not of the order of magnitude expected. Physicians who have had formal training in medical ethics are generally more likely to initiate discussion of risks, at every level of risk. Physicians with training in medical law appear to be more sensitive than other physi- cians to disclosures of remote risk, with 49% of doctors with medical law exposure saying they would initiate discussion of a 1-in-1,000 risk of tem- porary disability, compared with 40% among doctors without medical law exposure. However, physicians with formal training in doctor-patient communica- tions are less likely than those without such train- ing to disclose such low risk (40% vs. 47%) . Although overall, formal training in ethics, law, and communications in medical school does tend to increase physician willingness to disclose remote risk, the effect is slight, especially in comparison with the differences observed between public ex- pectation and physician practice in this area (Table 9-9), Withholding Information About Treatment Risk The formal training of a physician in ethics, law, or doctor-patient communications has no measur- able impact on the likelihood of withholding infor- mation from patients about treatment risks and al- ternatives. For 8 out of 10 physicians, instances where the doctor considers it advisable to withhold information about treatment risks and alternatives arise only a few times a year, if ever. There is no significant difference in the incidence of with- holding information about treatment from patients between those doctors with formal training in eth- ics, law, and doctor-patient communications and those without such training. Of course, if one con- siders the very low incidence of such occurrences, Surveys of Physicians and the Public 269 this finding cannot be viewed as conclusive evidence of medical education having no impact on medical practice (Table 9-10). Obtaining Patient Consent to Treatment Physician education in medical ethics and med- ical law has a consistent effect on the likelihood of obtaining formal consent from patients prior to treatment. For virtually every treatment investi- gated, doctors with training in medical ethics and medical law are more likely to obtain consent from the patient before initiating a medical procedure. For the medical procedures most closely associated with high risk to the patient -- e.g., inpatient surgery and the use of general anesthesia -- there are so few physicians who fail to obtain permission from the patient that variation among doctors cannot be measured. However, there is considerable varia- tion for low-risk procedures, such as blood tests and prescriptions. The incidence of physicians re- porting no consent for these procedures is substan- tially higher among those without medical ethics and medical law training than among those with such training. In other areas, the variation among doc- tors is too small to permit statistical analysis. However, the direction of the difference is con- sistent. Doctors without training in ethics and law are more likely not to obtain consent than those who have such training (Table 9-11). Patient's Role in Medical Decisionmaking Physicians' exposure to formal training in med- ical ethics, medical law, and doctor-patient commu- nications in medical school does appear to affect their attitudes toward the role of the patient in the medical decisionmaking. In general, physicians with training in medical ethics view the effects of increasing the patient's role in medical decision- making more positively than those without such training (59% vs. 55%). Similarly, doctors with medical law training (64%) are more positive about such change than those without law training (52%). However, the most dramatic difference in positivity of outlook is found between doctors with doctor- patient communications training (65%) and those without such training (48%) (Table 9-12). 270 Making Health Care Decisions: Appendix B Resuscitation 2esuscitacion Physicians differ somewhat in their response to the issue of patient resuscitation according to whether or not they have had formal training in med- ical ethics and doctor-patient communications. Take the case of the hospitalized patient in great pain in the end stages of a degenerative disease, who will never leave the hospital again. Among doctors with no formal training in medical ethics, 61% would be very likely to order that the patient not be re- suscitated in case of cardiac arrest, in the absence of any guidance from the patient, compared with 54% of physicians schooled in medical ethics. This same difference is found between doctors with no training in doctor-patient communication (62%) and those with such training (53%). Training in medical law, how- ever, has no appreciable effect (Table 9-13). Similarly, those doctors with training in med- ical ethics and doctor-patient communications would be more likely to initiate a discussion with the patient about resuscitation. Among doctors with medical ethics training, the survey finds that 57% would initiate discussion with the patient about preferences concerning resuscitation, compared with 48% of doctors without medical ethics exposure. Even more dramatic, 60% of doctors with doctor- patient communications training would initiate a discussion of the patient's preferences concerning resuscitation, compared with 44% of those doctors without such training. These findings suggest that medical school training in ethics and doctor-patient communication does increase the likelihood that a doctor will seek a patient's preferences concerning resuscitation, rather than make a decision on the merits of the case without such guidance from the patient. Taken together, these aspects of medical educa- tion do seem to have a systematic, but not neces- sarily dramatic, impact on the reported attitudes and behaviors of physicians regarding information disclosure and medical decisionmaking. Surveys of Physicians and the Public 271 P-F5,6,8,9 TABLE 9-1 PROFILE OF PHYSICIANS BY SEX, FATHER'S EDUCATION, RELIGIOUS BACKGROUND, AND ABILITY TO SPEAK ENGLISH Q.: Would you mind telling me what was the last grade or year of school that your father completed? Q.: Would you mind telling me in what religion you were raised, if any? Q.: Respondent's Sex. Q.: Respondent Speaks English. PHYSICIANS OBTAINED MEDICAL DEGREE 1967 1973 1966 AND TO AND TOTAL BEFORE 1972 AFTER BASE 805 441 97 266 2 2 x z SEX MALE 94 97 93 90 FEMALE 6 3 7 10 FATHER'S EDUCATION LESS THAN HIGH SCHOOL GRADUATE 24 33 16 10 HIGH SCHOOL GRADUATE 26 23 28 32 COLLEGE GRADUATE 45 37 52 57 NOT SURE 4 5 3 1 NO ANSWER/REFUSED 1 1 1 - RELIGIOUS BACKGROUND CATHOLIC 26 25 26 28 PROTESTANT 42 44 35 42 JEWISH 15 15 16 14 ALL OTHER 4 4 9 2 NONE 3 2 4 3 REFUSAL 10 9 9 1 SPEAKS ENGLISH WITH DIFFICULTY 6 6 6 4 TABLE 9-2 ABILITY TO SPEAK ENGLISH: PHYSICIANS Q.: Respondent Speaks English. PHYSICIANS _ TYPE OF MEDICAL SCHOOL PUBLIC PRIVATE OTHER OTHER TOTAL HIGH HIGH PUBLIC PRIVATE FOREIGN BASE 805 21 94 349 209 125 % % % % % % FLUENTLY 88 91 94 92 94 39% WITH DIFFICULTY 6 - 1 1 1 31 cle dg x1puaddy :suoisia(] aren yi[eay Suryep TABLE 9-3 MEDICAL SCHOOL ATTENDED: Where did you go to medical school? BASE PUBLIC, LOW RESEARCH §'s PRIVATE, LOW RESEARCH $'s PUBLIC, MEDIUM RESEARCH $'s PRIVATE, MEDIUM RESEARCH $'s PUBLIC, HIGH RESEARCH §'s PRIVATE, HIGH RESEARCH $'s ALL FOREIGN SCHOOLS NO ANSWER/REFUSED eC —C—— *LESS THAN 0.5%. PHYSICIANS PHYSICIANS OBTAINED MEDICAL DEGREE 1967 1973 1966 AND TO AND TOTAL BEFORE 1972 AFTER 805 441 97 266 % % % % 34 29 32 41 20 22 18 18 10 9 8 12 6 6 5 6 3 2 5 3 12 13 11 9 16 18 21 10 o1[qng oy} pue SueloIsAyd jo sfaAIng €L2 274 Making Health Care Decisions: Appendix B P-F3A,B,C TABLE 9-4 FORMAL TRAINING IN MEDICAL SCHOOL: PHYSICIANS Q.: While you were training in medical Q.: While you were training in medical Q.: While you were in medical school and residency, did you have any formal ethics? in medical school and residency, did you have any formal law? in medical school and residency, did you have any formal training in doctor-patient communications? PHYSICIANS OBTAINED MEDICAL DEGREE 196 973 1966 AND TO AND TOTAL _BEFORE 1972 AFTER BASE 805 441 97 266 % x 2 2 MEDICAL EHTICS YES 50 49 33 58 NO 46 46 62 41 NOT SURE 4 5 S 1 MEDICAL LAW YES 36 35 30 41 NO 60 60 67 58 NOT SURE 4 5 3 1 DOCTOR-PATIENT COMMUNICATIONS YES 54 43 53 73 NO 43 52 45 27 NOT SURE 3 5 2 - P-F3A,B,C TABLE 9-5 FORMAL TRAINING IN MEDICAL SCHOOL: PHYSICIANS Q.: While you were in medical school and residency, did you have any formal training in medical ethics? Q.: While you were in medical school and residency, did you have any formal training in medical law? Q.: While you were in medical school and residency, did you have any formal training in doctor-patient communications? PHYSICIANS RESEARCH FUNDING AND TYPE OF MEDICAL SCHOOL HICH RESEARCH $'s MEDIUM RESEARCH $'s LOW RESEARCH $'s TOTAL PRIVATE PUBLIC PRIVATE PUBLIC PRIVATE PUBLIC FOREIGN BASE 805 94 21 47 80 162 269 125 % % 2 % z x Z Z MEDICAL ETHICS YES 50 47 43 43 53 57 47 51 NO 46 50 52 51 46 38 51 44 MEDICAL LAW YES 36 29 19 53 43 30 38 41 NO 60 69 76 45 54 65 60 54 DOCTOR-PATIENT COMMUNICATION YES 54 54 57 49 64 48 59 47 NO 43 45 38 45 35 48 40 46 o1[qng dy} pue suedIsAyd jo sAaaIng GLC P-F4A TABLE 9-6 GREATEST INFLUENCE ON PRACTICE: PHYSICIANS Q.: Which of the following three items do you think has had the most influence on the way you practice medicine? PHYSICIANS OBTAINED MEDICAL DEGREE 1967 1973 1966 AND TO AND TOTAL BEFORE 1972 AFTER BASE 805 441 97 266 % % % 7% YOUR PERSONAL VALUES AND BELIEFS 62 59 66 65 YOUR MEDICAL TRAINING 18 18 19 18 YOUR EXPERIENCE IN PRACTICE 18 21 12 15 NOT SURE 2 2 3 2 9.2 gd x1puaddy :suorstoa(y aren) yjesy Sunyey P-F4A TABLE 9-7 GREATEST INFLUENCE ON PRACTICE: PHYSICIANS Q.: Which of the folowing three items do you think has had the most influence on the way you practice medicine? PHYSICIANS RELIGIOUS BACKGROUND TOTAL CATHOLIC PROTESTANT JEWISH OTHER NONE REFUSED BASE 805 211 339 121 33 22 79 7% 7% 7% 7 7% 7 7 PERSONAL BELIEF AND VALUES 62 70 61 57 58 59 56 MEDICAL TRAINING 18 13 18 17 15 18 29 EXPERIENCE IN PRACTICE 18 15 18 24 21 18 14 o1[qng dy} pue suedIsAyq jo sAdAINg LL P-24 TABLE 9-8 RESPONSIBILITY FOR FATIENTS' KNOWLEDGE OF CONDITION: PHYSICIANS Q.: Do you think that it is primarily the doctor's responsibility to make sure that the patient is fully informed about his condition and treatment, or the patient's responsibility to ask the doctor about these things? PHYSICIANS FORMAL TRAINING ETHICS LAW COMMUNICATION TOTAL YES NO YES NO YES NO BASE 805 402 374 293 484 434 345 % r % Zz % % PRIMARILY THE PATIENT'S RESPONSIBILITY 77 79 76 77 78 78 77 PRIMARILY THE PATIENT'S RESPONSIBILITY 2 2 2 4 2 2 3 EQUALLY RESPONSIBLE (VOL.) 19 17 20 17 20 19 19 NOT SURE 1 1 1 2 1 1 1 84¢ dg x1puaddy :suoisa(y aren) yyeey Sunyeiy TABLE 9-9 INITIATE DISCUSSIONS OF RISKS WITH PATIENTS: PHYSICIANS Q.: Another set of issues that may or may not be discussed with patients involves the risks of adverse consequences associated with treatment. In your practice, do you initiate discussion about each of the following things with the patient as a matter of course, or discuss it only if the patient asks about it? BASE RISKS OF DEATH OR SERIOUS DISABILITY (1 IN 100) RISKS OF DEATH OR SERIOUS DISABILITY (1 IN 1,000) RISKS OF TEMPORARY DISABILITY (1 IN 1,000) INITIATE DISCUSSION AS A MATTER OF COURSE PHYSICIANS FORMAL TRAINING ETHICS LAW COMMUNICAT ION TOTAL YES NO YES NO YES NO_ 401 190 194 151 233 209 178 z z z Z Z z z 81 83 79 82 81 84 76 52 56 50 54 53 55 49 43 46 41 49 40 40 47 NOTE: SPLIT-BALLOT ITEM, ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. o1[qng dy} pue sueIsAyd jo sAaaIng 64C TABLE 9-10 FREQUENCY DOCTORS WITHHOLD INFORMATION: PHYSICIANS Q.: Sometimes a physician feels it is advisable to withhold information about treatment risks and alternatives from a patient. eT i say this occurs once a day, once a week, once a month, a few times a year, or almost never? In your practice would you PHYSICIANS FORMAL TRAINING ETHICS LAW COMMUNICATION TOTAL YES NO_ YES NO YES NO BASE 805 402 374 293 484 434 345 x % 2 z % % ONCE A DAY 4 4 4 4 3 4 4 ONCE A WEEK 6 7 6 7 6 7 5 ONCE A MONTH 9 9 9 10 8 8 10 A FEW TIMES A YEAR 24 22 26 24 24 25 21 ALMOST NEVER 57 58 55 55 58 55 59 NOT SURE 1 1 * * 1 * *LESS THAN 0.5%. 082 fg x1puaddy :suoistoa(] aren) yj[eoy Sumyey Surveys of Physicians and the Public 281 P-33 TABLE 9-11 PROCEDURES FOR WHICH NO CONSENT IS OBTAINED: PHYSICIANS Q.: In vour practice, do you usually get written consent, oral consent, or no consent from patients before each of the following procedures? BASE INPATIENT SURGERY MINOR OFFICE SURGERY SETTING BONES GENERAL ANESTHESIA LOCAL ANESTHESIA DIAGNOSTIC X-RAYS INVOLVING INJECTIONS BLOOD TESTS PRESCRIPTIONS RADIATION THERAPY *LESS THAN 0.5%. 13 43 53 "NO CONSENT" PHYSICIANS FORMAL TRAINING —_— ETHICS LAW COMMUNICATION YES NO YES NO YES NO 402 374 293 484 434 345 x 3% % 3 % % * * - * - 1 4 9 5 1 6 8 3 4 3 4b 4 3 - 1 - 1 - 1 11 16 10 16 13 14 s 9 5 8 6 8 39 48 39 46 42 45 48 57 49 55 50 55 2 2 1 2 2 2 TABLE 9-12 INCREASING PATIENT'S ROLE IN DECISIONMAKING: PHYSICIANS Q.: Do you believe that increasing the patient's role in medical decisionmaking is likely to improve the quality of medical care, reduce the quality of medical care, or have no effect on it? BASE IMPROVE THE QUALITY REDUCE THE QUALITY HAVE NO EFFECT NOT SURE NOTE: SPLIT-BALLOT LTEM y PHYSICIANS FORMAL TRAINING ETHICS LAW COMMUNICATION TOTAL YES NO_YES NO ~ YES NO_ 401 190 194 151 233 209 178 % % ZZ 2 Z Z 56 59 55 64 52 65 48 16 15 16 13 18 12 18 22 20 25 18 25 18 28 6 6 4 5 6 5 6 ASKED OF ONE-HALF OF PHYSICIAN SAMPLE. 287 g xtpuaddy :suorsioa(] are) yjesy Sune P-27A TABLE 9-13 LIKELIHOOD OF NOT ORDERING RESUSCITATION: PHYSICIANS Q.: Now, imagine you have a hospitalized patient in great pain in the end stages of a degenerative disease. You have been treating her for this disorder for a long time and know that she will never leave the hospital again. In the absence of any guidance from the patient, how likely would you be to order that she not be resuscitated in the case of cardiac arrest -- very likely, somewhat likely, somewhat unlikely, or very unlikely? PHYSICIANS FORMAL TRAINING ETHICS LAW COMMUNICATION TOTAL YES NO YES NO YES NO BASE 805 402 374 293 484 434 345 z x z 2 2 x 2 VERY LIKELY 57 54 61 57 58 S54 62 SOMEWHAT LIKELY 15 16 14 15 15 15 16 SOMEWHAT UNLIKELY 7 7 71 4 8 8 6 VERY UNLIKELY 16 19 13 20 14 19 14 NOT SURE 4 4 4 3 4 5 4 o1[qnd dy) pue sueIdIsAyd Jo sAaaIng £87 P-27B TABLE 9-14 INITIATION OF DISCUSSION WITH PATIENT ABOUT RESUSCITATION: Q.: Would you initiate a discussion with this patient about her preferences concerning resuscitation? PHYSICIANS FORMAL TRAINING PHYSICIANS ETHICS LAW COMMUNICATION TOTAL YES NO YES NO YES BASE 805 402 374 293 484 434 7 7% ZZ % 7 YES, WOULD INITIATE 52 57 48 56 51 60 NO, WOULD NOT INITIATE 38 34 42 35 40 32 NOT SURE 9 9 9 9 9 8 NO 345 7% 44 45 11 ¥8¢ g xtpuaddy :suorstoa( are) yi[eay Suryeny Surveys of Physicians and the Public 285 APPENDIX A: PROFILES OF THE PUBLIC AND PHYSICIAN SAMPLES 286 Making Health Care Decisions: Appendix B TABLE A-1 PROFILE OF THE PUBLIC SAMPLE GENERAL PUBLIC NUMBER OF UNWEIGHTED WEIGHTED RESPONDENTS PERCENT PERCENT TOTAL PUBLIC 1251 100 100 2 2 2 REGION EAST 328 26 wk SOUTH 361 29 ** MIDWEST 331 26 hadad WEST 231 19 holed SIZE OF PLACE CITY 361 29 wk SUBURB 345 28 wk TOWN 191 15 x RURAL 354 28 *k SEX MALE 610 49 48 FEMALE 641 51 52 AGE 18-34 YEARS 488 39 41 35-50 YEARS 334 27 24 51-64 YEARS 226 18 18 65 AND OVER 195 16 16 EDUCATION LESS THAN HIGH SCHOOL GRADUATE 242 19 19 HIGH SCHOOL GRADUATE 726 58 58 COLLEGE GRADUATE 281 22 23 EMPLOYMENT FULL TIME 687 $5 55 PART TIME 103 8 8 UNEMPLOYED 65 5 5 RETIRED 176 14 14 STUDENT 43 3 4 HOUSEWIFE 160 13 12 DISABLED, TOO ILL TO WORK 22 2 2 ACTIVE DUTY IN ARMED FORCES 2 * * OTHER 1 * * *LESS THAN 0.5X. **NOT CALCULATED (CONTINUED) Surveys of Physicians and the Public 287 TABLE A-1 (CONTINUED) PROFILE OF THE PUBLIC SAMPLE TOTAL PUBLIC EXPERIENCE IN A HEALTH-RELATED OCCUPATION YES NO NOT SURE SIZE OF IMMEDIATE FAMILY 1 PERSON 2 PERSONS 3-4 PERSONS 5-6 PERSONS 7-9 PERSONS 10 OR MORE PERSONS HOUSEHOLD INCOME $7,500 OR LESS $7,501-$15,000 $15,001-$25,000 $25,001-$35,000 $35,001-$50,000 $50,001 OR MORE NO ANSWER HEALTH INSURANCE COVERAGE PREPAID HEALTH PLAN (HMO) BLUE CROSS/BLUE SHIELD INSURANCE COMPANY POLICY MEDICARE /MEDI-CAL/MEDICAID NO HEALTH INSURANCE ALL OTHER NOT SURE MAJOR MEDICAL COVERAGE ES NO NOT SURE RACE /ETHNIC ORIGIN ASTAN OR PACIFIC ISLANDER AMERICAN INDIAN OR ALASKAN NATIVE BLACK (NON-HISPANIC) WHITE (NON-HISPANIC) HISPANIC GENERAL PUBLIC NUMBER OF UNWEIGHTED WEIGHTED RESPONDENTS PERCENT PERCENT 1251 100 100 2 2 z 262 21 21 988 79 79 3 - - 101 8 8 205 16 16 505 40 39 286 23 n 126 9 10 27 2 5 164 13 13 219 18 18 341 27 21 229 18 18 111 9 12 103 8 10 84 7 8 212 17 17 485 39 38 493 39 39 222 18 18 73 6 5 188 15 15 38 3 4 932 75 74 278 22 22 38 3 3 19 2 2 24 2 2 122 10 10 1027 ] 82 79 36 3 5 288 Making Health Care Decisions: Appendix B TABLE A-1 (CONTINUED) PROFILE OF THE PUBLIC SAMPLE GENERAL PUBLIC NWBER OF UNWEIGHTED WEIGHTED RESPONDENTS PERCENT PERCENT TOTAL PUBLIC 1251 100 100 z x x RELIGION CATHOLIC 344 28 29 PROTESTANT 760 60 59 JEWISH 35 3 3 OTHER 25 2 2 NONE 36 3 3 RELIGIOSITY VERY RELIGIOUS 340 27 28 SOMEWHAT RELIGIOUS 614 49 49 NOT VERY RELIGIOUS 275 22 22 ENGLISH LANGUAGE SKILLS FLUENT 1210 97 96 NOT FLUENT 40 3 4 Surveys of Physicians and the Public 289 TABLE A-2 PROFILE OF THE PHYSICIAN SAMPLE PHYSICIANS NUMBER OF UNWEIGHTED RESPONDENTS PERCENT TOTAL PHYSICIANS 805 100 4 SPECIALTY INTERNAL MEDICINE 202 25 UROLOGY 31 4 OBSTETRICS /GYNECOLOGY 108 13 GENERAL PRACTICE 59 7 FAMILY PRACTICE 172 21 SURGERY 177 22 OTHERS 45 6 PRACTICE LOCATION OFFICE 533 66 HOSPITAL 232 29 PRACTICE CHARACTERISTICS MEAN NUMBER MEAN NUMBER OF HOSPITAL INPATIENTS/WEEK - 17.2 MEAN NUMBER OF OUTPATIENTS SEEN/WEEK - mn. LENGTH OF AVERAGE WEEK (IN HOURS) SPENT IN PATIENT CARE - 53.2 AVERAGE PERCENTAGE OF ADULT CLIENTS WHO ARE: MEAN PERCENTAGE AT LEAST 65 YEARS OLD - . COLLEGE EDUCATED - 25.1 NON-ENGLISH SPEAKING - 6.1 POOR - 24.6 SUFFERING FROM SERIOUS ILLNESS - 36.3 ABLE TO UNDERSTAND TREATMENT - 79.4 MEDICAL TRAINING INSTITUTION PRIVATELY HELD UNIVERSITY, AMONG TOP TWENTY RECIPIENTS OF FEDERAL RESEARCH DOLLARS 9% 12 (CONTINUED) 290 Making Health Care Decisions: Appendix B TABLE A-2 (CONTINUED) PROFILE OF THE PHYSICIAN SAMPLE PHYSICIANS NUMBER OF UNWEIGHTED RESPONDENTS PERCENT TOTAL PHYSICIANS 805 100 2 PUBLICLY HELD UNIVERSITY, AMONG TOP TWENTY RECIPIENTS OF FEDERAL RESEARCH DOLLARS 21 3 PRIVATELY HELD UNIVERSITY, AMONG SECOND TWENTY RECIPIENTS OF FEDERAL RESEARCH DOLLARS 47 6 PUBLICLY HELD UNIVERSITY AMONG SECOND TWENTY RECIPIENTS OF FEDERAL RESEARCH DOLLARS 80 10 PRIVATELY HELD UNIVERSITY, NOT AMONG TOP FORTY RECIPIENTS OF FEDERAL RESEARCH DOLLARS 162 20 PUBLICLY HELD UNIVERSITY, NOT AMONG TOP FORTY RECIPIENTS OF FEDERAL RESEARCH DOLLARS 269 34 FOREIGN SCHOOL 125 16 COURSEWORK MEDICAL ETHICS YES 402 50 NO 374 46 MEDICINE AND THE LAW YES 293 36 NO 484 60 DOCTOR-PATIENT COMMUNICATION YES 434 54 NO 345 43 YEAR COMPLETED MEDICAL SCHOOL 1945 AND BEFORE 118 15 1946-1965 311 39 1966-1970 70 9 1971-1975 113 14 1976-1980 192 24 GREATEST INFLUENCE ON PRACTICE OF MEDICINE PERSONAL VALUES AND BELIEFS 500 62 MEDICAL TRAINING 143 18 EXPERIENCE IN PRACTICE 145 18 (CONTINUED) Surveys of Physicians and the Public 291 TABLE A-2 (CONTINUED) PROFILE OF THE PHYSICIAN SAMPLE PHYSICIANS NUMBER OF UNWEIGHTED RESPONDENTS PERCENT TOTAL PHYSICIANS 805 100 2 NEXT MOST INFLUENTIAL ON PRACTICE OF MEDICINE PERSONAL VALUES AND BELIEFS 154 19 MEDICAL TRAINING 257 32 EXPERIENCE IN PRACTICE 364 46 LAST GRADE OF SCHOOL FATHER COMPLETED LESS THAN HIGH SCHOOL GRADUATE 190 24 HIGH SCHOOL GRADUATE 213 26 COLLEGE GRADUATE 365 45 RELIGION CATHOLIC 211 26 PROTESTANT 339 42 JEWISH 121 15 OTHER 33 4 NONE 22 3 RELIGIOSITY VERY RELIGIOUS 176 22 SOMEWHAT RELIGIOUS 369 46 NOT VERY RELIGIOUS 240 30 SEX OF RESPONDENT MALE 760 94 FEMALE 45 6 ENGLISH LANGUAGE SKILLS FLUENT 706 88 NOT FLUENT 44 6 Surveys of Physicians and the Public 293 APPENDIX B: ANNOTATED QUESTIONNAIRES 294 Making Health Care Decisions: Appendix B PHYSICIAN QUESTIONNAIRE LOUIS HARRIS AND ASSOCIATES, INC. FOR OFFICE USE ONLY: 630 Fifth Avenue 3312-0017 New York, New York 10111 EXPIRES 7/31/82 Questionnaire No. : Study No. 812041 (Physician Sample) —_— February 1982 - FORM Al Sample Point No. 10-11-12-13-14-15-16-17-18 Date of interview: Time of interview: Respondent's Name (PLEASE PRINT): Dr. Office or Hospital Address: City/Town: County: State: Zip: Telephone Area Code: Telephone Number: Mailing address for incentive reimbursement if different from above : r's Name (PRINT) Hello, Dr. . My name is from Louis Harris and Associates, the national public opinion research firm in New York City. Recently you were sent a letter from Mr. Harris describing a nationwide study of physicians we are currently undertaking for the President's Commission for the Study of Ethical Issues in Medicine. (IF PHYSICIAN DIDN'T RECEIVE LETTER OR DOES NOT REMEMBER, SAY: I'm sorry you didn't receive the letter. In that case, I'll briefly go over the details of the letter.) The purpose of this study is to learn more about the physician's views on the patterns of information disclosure and decisionmaking in medical settings. Your participation will be kept strictly confidential, and your responses will be analyzed only in aggregate statistical form. Your identity will not be disclosed to the commission or any other organization. You're certainly not required to respond, but your cooperation is needed to make the results meaningful and accurate. The interview should take about twenty minutes. For your time and effort, we would be pleased to provide you with an honorarium of $ - If it is not convenient now, I would like to make an appointment for a later time. Key: 19) *® 2S. 5% 0 = standard error; X = mean; all other entries are unweighted percentages rl- CARD 1 B812041-A1 1. Let me begin by asking you a few general questions about your practice. Including both inpatient and outpatient care, approximately how many hours do you spend on direct patient care activities in an average week? X = 53.2 0.6 (ASK Q.2) (NUMBER OF HOURS) (20-22) Less than 20 hours..(23( -1 (SCREEN OUT) Not BUre...csssvsurereese_* «2 (ASK Q.2) Refused. ..vves vinnie 3 (REFUSAL) 2. What is your primary specialty? x Internal Medicine 29 Ob/Gyn 13 (SPECIALTY) (24-25) Surgery 22 General /Family No specialty.........(26( -1 Practitioner 29 Other 6 Surveys of Physicians and the Public 295 -2- CARD 1 812041-A1 3. Is your main practice, that is, where you have most of your contact with patients, office-based or hospital-based? A270 66-1 Office-based....... Hospital-based..... Other (SPECIFY) we 4."3 No answer/refused.......cc.- -u 4. Approximately how many different hospital inpatients would you estimate that you see in an average week? Overall x =17.2 og=.6 (28-29) (NUMBER OF HOSPITAL INPATIENTS) None. ...eee Not sure... No answer/refused. 5. Approximately how many different patients would you estimate that you see on an outpatient basis during an average week? Overall x = 71.9 c= 1.9 (31-33) (NUMBER OF OUTPATIENTS) None... Not ‘sure.. 6. Now I would like to ask you a few questions about the kinds of patients that you see. Approximately what percentage of your adult patients would you say are (READ EACH ITEM)? Mean % Mean including | START AT "X"| Percent OJ None Not Sure "none" o ( ) a. Aged 65 or over.......... 33.4 $.7(38(1.2 -1 -2 35.3 28 (35-37) ( ) b. College educated......... 25.7 ¢ .6 (u2( 6.1 -1 -2 25.1 a1 (39-41) ( ) e¢. Non-English speaking..... 15.8 g .4 (4630.7 1 -2 _6.1 5 (43-45) ( ) d. Poor 25.9% .8(50(ge8 -1 =2 24.6 .8 (47-49) — TT Tr 7. What percentage of your total patients would you characterize as suffering from serious illness or health conditions? x=363¢g 0=1.0 51-53) Not sure. 8. What percentage of your patients would you say are able to understand most aspects of their treatment and condition, if reasonable time and effort are devoted to explanation? x=179.48 o0=.7 (55-57) Loo(58C * 1 -2 None..... Not sure. 296 Making Health Care Decisions: Appendix B -3- CARD 1 812041-A1 9. A lot has been said about the doctor-patient relationship, what physicians should tell patients and how medical decisions should be made. We are interested in your experience and views on these matters. First, how often do you find yourself in a situation where you must make a conscious and deliberate evaluation of how much to tell a patient about his or her condition or treatment -- would you say several times a day, daily, weekly, rarely, or never? Several times a day.(59( 27 -1 Daflye.essns 25 -2 20-3 25-4 35 -6 10. What are the primary factors that influence how much you tell a patient about his i condition or treatment? What else? 4 Patient's ability to understand his condition or treatment (60( “56 Patient's ability to cope with the news/emot ional stability (61( 31 Seriousness of patient's condition/prognosis (62( 30 Patient's desire to know (63( 25 11. How many patients who come to you for treatment want you to give them a candid assessment of their diagnosis and prognosis, even if it is unfavorable? Would you say all, most, some, few, or none? S(64C 17 -1 (SKIP TO Q.13) (ASK Q.12) 12. For those patients to whom you do not disclose bad news, how often is this because they tell you directly they don't want to know it? Would you say always, often, sometimes, rarely, or never? (N = 665 -- those not answering "1" to Q.11) (65( 12 -1 (These versions of Questions 13 and 14 were asked of one-half of the physician sample. N = 401) -l- CARD 1 81204)-A1 13. There are a variety of issues that doctors may or may not discuss with their patients regarding the patient's condition and possible treatment. In your practice, do you initiate discussion about each of the following things with the tient as a matter of course, or discuss it only if the patient asks about it? IREAD EACH ITEM = ALR ITEM] Discuss Never Initiate Only If Discuss Not Discussion Patient Asks _(Vol.) Sure a. Your diagnosis and prognosis of the patient's condition.............. (86(%8 3 ! -2 -3 *. b. Cost sociated with recommended procedure and whether they would be covered by patient's health insurance........(67(38 _) 56 «2 4 -3 2 _y c. The nature and purpose of the treatment option that you PECOBMEnds us snsssrsnrnssrrrnnnerrssnsnsases (BH 96 3 -2 -3 YL d. The pros and cons of the treat- ment you recommend compared to other alternatives................ eeeed(69( 84 -1 14 2 * 3 ll. e. The probable impact of the patient's condition and treatment on the patient's family 1ife, when there is a likely Impact. eit iiiie renee ee een (700 74 1 23 -2 1 3 2. f. The probable impact of the patient's condition and treatment on the patient's Job role, when there is a likely impact. een 79 21 20 -2 1 3 *. Surveys of Physicians and the Public 297 14. Another set of issues that may or may not be discussed with patients involves the risks of adverse consequences associated with treatment. In your practice do you initiate discussion about each of the following things with the patient a matter of course, or discuss it only if the patient asks about it? TREAD EACH ITEM] Discuss Never Initiate Only If Discuss Not [START AT *X"| Discussion Patient Asks _(Vol.) Sure ( ) a. Side effects, including inconvenience and pain, that are certain or fairly likely to attend the treatment.........(72( 95-1 4 2 * 3 ta ( ) b. Risks of death or serious di bility that are about 1 An 100....cceeeeeeesaa(73C 81 21 16 2 1-3 3a ( ) ec. Risks of death or serious disability that are about 1 in 1,000....c0c0cessce(T8( 57 1 40 I ( ) d. Risks of temporary disability that are about 1 in 21,0000 csvsvevreenenens (250 53-2 48 -2 4 __=3 _4L-M i682] (These versions of Q.13 and 14 were asked of one-half of the physician sample. N=404) =U CARD 1 8120M81-B1 13. There are a variety of issues that doctors may or may not discuss with their patients regarding the patient's condition and possible treatment. In your practice, how often do you discu ach of the following things with the patient -- always, usually, sometimes, rarely, or never? [READ FACH TTEM| Not READ EACH ITEM Always Usually Sometimes Rarely Never Sure a. Your diagnosis and prognosis of the patient's condition.........(66(56 -1 42 -2 2 -3 xb 5 __ -6 b. Costs associated with recommended procedure and whether they would be covered by patient's health INSUrANCe.sessrerensrnsensesesss(687(]15 -1 32 -2 34-3 15-4 4 -5 _1-6 c. The nature and purpose of the treatment option that you PeCOMMENd..ooassessenssnsensanee(688( g7 =1 34-2 4-3 -u -5 -6 d. The pros and cons of the treat- ment you recommend compared to other alternatives...... L..(69( 38 <1 46 _» 13 .% 3 ny - *_6 e. The probable impact of the patient's condition and treat- ment on the patient's family 1ife, when there is a likely AMPACt.cececrrrersnssnarocasaans(T0( 32 -1 46 -2 18 -3 4 -h *5 _*-6 f. The probable impact of the patient's condition and treat- ment on the patient's job role, when there is a likely impact...(71( 3g -1 50 -2 11-3 2-4 _*-5 _*-6 14. Another set of issues that may or say not be discussed with patients involves the risks of adverse consequences associated with treatment. In your practice, how often do you discus ch of the following things with the patient -- always, usually, sometimes, rarely, or never? IREAD EACH ITEM Not ISTART AT "X*| Always Usually Sometimes Barely Never Sure Side effects, including inconvenience and pain, that are certain or fairly likely to attend the treatment....(¥2(53 -1 40 -2 6 -3 1 4 _ 5 * 6 ( ) b. Risks of death or serious disability that are . about 1 in 100.. ..(73(52 -Y _30 -? 9 -3 6-4 1-5 _2-6 ( ) c. Risks of death or serious disability that are about 1 in 1,000... swonenrnee{JUCAG. =) 29 -? 22 -% 1a-h 4s 2-6 ( ) d. Risks of temporary dis- ability that are about 1 in 1,000.. L(15(25 1 32 -2 2) .3 15.4 2.6 2-6 298 Making Health Care Decisions: Appendix B (This version of Q. 15 was asked of one-half of the physician sample. N=401) -5- CARD 2 812041-A1 ‘15. Often there are significant uncertainties associated with medical decisions. When such uncertainties exist, do you initiate discussion about (READ EACH ITEM) with the patient as a matter of course, or discuss it only if the patient asks about it? Discuss Never Initiate Only If Discuss Not Discussion Patient Asks _(Vol.) Sure a. Uncertainties or reservations ADDU dIAGNOBYS. vv rersvrvesnives ..(10( 90 -1 8 -2 1 -3 1 4 b. Uncertainties in your own mind about the best course of treat- ment..ivencinnnnnnanas (11 66-1 27 -2 4-3 4 4 c. Different views within the pro- fession as to the best course of treatment for the condition..............(12( 75 -1 22 -2 3-3 _*.y d. Uncertainties about whether the treatment you recommend will be 3uCCesSful..uuuurnniiiiiiiiiiiiaaraanaaeeaaa(13( 80 -1 16-2 3 -3 2.4 16. In making a treatment decision, what proportion of your patients want you to choose the best alternative for them rather than simply presenting the alternatives for them to choose between? Would you say all of them, most of them, some, a few, or none of them? See Q. 16 All......(24¢( -1 (next page) None...... Not sure.. 115-2021 L7. When a patient disagrees with your recommendation for a particular procedure or rourse of treatment that you think is strongly medically indicated, do. you feel that it 8 your responsibility to try to persuade the patient to accept the medically indicated rourse of action, or do you feel you have no responsibility to do this? Responsibility to persuade. (21( -1 See Q. 18 No responsibility........ Depends (vol.)...... Not sure............. "(These versions of Q. 15 was asked of one-half of the physician sample. N=404) : =5- CARD 2 812041-B1 15. Often there are significant uncertainties associated with medical decisions. When such uncertainties exist, how often do you discuss (READ EACH ITEM) with the patient -- always, usually, sometimes, rarely, or never? IREAD EACH ITEMI Not Always Usually Sometimes Rarely Never Sure Sivays Usually Sometimes Rarely Never Sure a. Uncertainties or reservations about diagnosis.................(10(37 -1 47 -2 13-3 3-4 -5 -6 b. Uncertainties in your own mind about the best course of treat- meNt.uuuunniiiiiiinnnnes (QU 1 4g -2 26 -3 9 -4 3 -5 * -6 c. Different views within the pro- fession as to the best course of treatment for the condition..(12(23 -1 52 =? 19 -3- 5-4 1-5 1 -6 d. Uncertainties about whether the treatment you recommend will be successful....eeevaninnnnanana..(13(34 -1 43 2 16 -3 4 4 2 5 * _g 16. In making a treatment decision, what proportion of your patients want You to choose the best alternative for them rather than simply presenting the alternatives for them to choose between? Would you say all of them, most of them, some, a few, or none of them? -1 -2 1-3 -4 =5 -6 Surveys of Physicians and the Public (This question was asked of one-half of the physician sample. 17. decisions, running from 1 to 7. treatment decisions over which a doctor has no control. to discontinue prescribed medication would be a 1. represents a treatment decision over which a patient has no control. N-404) -6- CARD 2 299 812041-B1 1 would like you to think of a scale of doctor-patient influence over therapeutic At the low end of the scale, a 1 represents those For instance, patient decisions At the high end of the scale, a 7 For instance, emergency treatment of an unconscious patient is a decision over which a patient has no control and would be a 7. Thinking of this 1-to-7 scale of doctor-patient influence over treatment decisions, where 1 represents no doctor control and 7 represents complete doctor control, where do you think each of the following decisions should fall? |REPEAT AS NECESSARY] |READ EACH ITEMI| | SINGLE RECORD | Patient's Decision ISTART AT "X"| ( ) a. Whether to hospitalize a mild case of pneumonia..(15( 5 -1 _6 -2 ( ) b. Amniocentesis for pregnant women over thirty-five....(16(16 -1 15 -2 13-3 lay 0 Choice of antibiotics for treatment of strep throat......cccue teeee a Whether to initiate dialysis for end-stage renal disease.............(18( 7 -1 Whether to initiate estro- gen treatment for meno- pause problems.. Whether to amputate a gangrenous limb in a mentally alert patient....(20( 6-1 (This question was asked of the full physician sample. When a patient disagrees with your recommendation for a particular procedure or 18. ce QQI(r 1 x2 ...19( 6-1 12-2 18.3 27-4 18.5 5-6 Doctor's Decision 16.3 23.4 20.5 12. 12.9 9<8 Taf. Sy * -3 3-4 _9-5 30-6 54-7 8-2 12-3 254 13-5 10-6 10-7 6-7 7-2 s3 lay 17g 24g 18 N = 805) Not 7-8 7-8 course of treatment that you think is strongly medically indicated, do you feel that it is your responsibility to try to persuade the patient to accept the medically indicated course of action, or do you feel you have no responsibility to do this? Responsibility to persuade.(21( 75 -1 No responsibility. Depends (vol.). Not sure......... ‘18. or prognosis from a patient. 6 -2 18-3 1-4 wf CARD 2 once a week, once a month, a few times a year, or almost never? Once 8 dBYsesssssvneveveves{22( 2 A few times a year...... Almost never...... NOt Sure....eeeenceencennnns 19. risks and alternatives from a patient. Once a day.. Once a week. Once a month...... A few times a year. Almost never...... NOt SUr€....cecveranvessnnns 2.2 8120u41-A) Sometimes a physician feels it is advisable to withhold information about diagnosis In your practice would you say this occurs once a day, Sometimes a physician feels it is advisable to withhold information about treatment In your practice would you say this occurs once a day, once a week, once a month, a few times a year, or almost never? (23 4 3 FS J 3 7h _y 57-5 T 6 =~ = 300 Making Health Care Decisions: Appendix B 20. What are the most common reasons for you to withhold information about condition or treatment from a patient? What others? % Patient's ability to cope with the news/emot ional stability (2u( 34 Patient's ability to understand his condition or treatment (25( 28 The wishes of the patient's family (26( 21 Patient's physical or mental incapacity (27( 17 21. In those situations in which you withhold information about treatment from a patient, do you usually, sometimes, rarely, or never discuss the information with another family member? Usually....(28( gg -1 Sometimes. . (Quest fons 22 and 23 were asked of only one-half of the physician sample. N=401) =7- CARD 2 812041-A1 22. Do you believe that increasing the patient's role in medical decisionmaking is likely to improve the quality of medical care, reduce the quality of medical care, or have no effect on it? Improve the quality......(29( 56-1 Reduce the quality -2 Have no effect. . -3 Not sure...........c.... bu 23. Why do you say that? Improves patient's compliance (30037 Increases patient's knowledge and awareness of treatment (3103) Patients are not knowledgeahle enough to Make proper decisions (32(29 Patients cannot make objective decisions about their treatment (33 ° 24. Do you think that it is primarily the doctor's responsibility to make sure that the patient is fully informed about his condition and treatment, or the patient's responsibility to ask the doctor about these things? Primarily the doctor's responsibility...(34( 77 -1 Primarily the patient's responsibility...... 2-2 Equally responsible (vol.) ROL (UTE. sunvssnsevnnsnmesoitesss 25. To what extent do you rely on nurses or other nonphysicians to provide significant treatment information to patients -- do you rely on them heavily, somewhat, very little, or not at all? Heavily....(35( g -1 Somewhat. Surveys of Physicians and the Public 301 -8- CARD 2 812041-A1 26a. Now, I would like to get your opinion on some hypothetical situations. If you had a patient with a fully confirmed diagnosis of lung cancer in an advanced stage, which of the following would you be most likely to tell your patient? Would you (READ LIST)? ee | TSINGLE RECORD BELOW -- REREAD LIST, IF NECESSARY —_—— |READ IN THIS ORDERI JREAD AF I 1. Give a straight statistical prognosis for his class of disease... ena ra sens snrsamns raves 36C 13 1 2. Say that you can't tell how long he might live, but stress that it could be for a substantial period of time......_33 -2 3. Say that you can't tell how long he might live, but stress that in most cases people live no longer than a year......ceeeceececsece cepa vee SEY 4. Or refuse to speculate on how long the patient might live....cceceecneccnnnnnans tereeeresveas 22 <8 5-5 Not sure.. teeseesseesestessserretesvatsrs Erasers 26b. Assume that the cancer has metastasized and is not responsive to a first cycle of chemotherapy. Do you consider the decision about whether to continue aggressive therapy to be primarily a scientific, medical decision, or one which turns principally on personal values? Scientific.....(37( 21 -1 Personal.. Not sure.. (Data from the two versions of Q.26c are presented below) 26c. In this kind of situation who should make the decision about treatment -- the doctor or the patient? N=401 Who usually decides N=404 Doctor........(38( 8 -1 15 Patient...eeeesess _2' =2 19 Jointly (vol.) 87-3 64 Not sure......eeee TT. 2 27a. Now, imagine you have a hospitalized patient in great pain in the end stages of a degenerative disease. You have been treating her for this disorder for a long time and know that she will never leave the hospital again. In the absence of any guidance from the patient, how likely would you be to order that she not be resuscitated in the case of cardiac arrest -- very likely, somewhat likely, somewhat unlikely, or very unlikely? Very likely.. Somewhat likely. Somewhat unlikely. Very unlikely... Nob sures sssssivsvis Nob 27b. Would you initiate a discussion with this patient about her preferences concerning resuscitation? Yes, would initiate....(40( 52 -1 No, would not initiate..... -2 NOt SUr€..cceeececccnenne J -3 27c. If this patient asked you to do everything you could to maintain her life, how likely would you be to order that she not be resuscitated in case of cardiac arrest -- very likely, somewhat likely, somewhat unlikely, or very unlikely? Very likely...cceees 9 21 Somewhat likely. 7-2 Somewhat unlikely 12 -3 Very unlikely.. Not sure...... (Data from the two versions of Q. 27d. are presented below) 27d. Suppose the patient is now incompetent but had previously written instructions indicating that she wished to have everything possible done to prolong her life. If this patient suffered a cardiac arrest, would you resuscitate her? No extraordinary measures Yes, would resuscitate.....(l2( 52.1 5 No, would not resuscitate...... 5.2 84 It depends (vol.)..... -3 11 Not sure..... =u * (N=401) (N=404) 302 Making Health Care Decisions: Appendix B Quest fon 28 through 31 were asked of one-half of the physician sample. (N=401) =10- CARD 2 812041 ata rom wo versions © quest Tors an 28. Assume that another patient who is in severe pain and with no hope of recovery has asked that you help ease the pain, knowing it might shorten his or her life. Under these circumstances would you be likely to administer narcotics to reduce pain, even if the dose required might shorten life? N=199 To be ethically permissable N=202 Yes.ooooniniiiil (M3( 84-1 82 NOwi srvmarrvvvmrasionsa 5-2 2 It depends (vol.). 9-3 9 Not sure........ *-l 4 No answer/refused..... 2-5 3 29. Imagine that a dying patient in severe distress, which cannot be relieved, asks to have his life ended. Under these circumstances would you be likely to comply with the patient's wishes? To be ethically permissable N=202 Yes. srrrrrvnvrrnnn 4 Notirirtnsnnes . 81 It depends (vol.). 6 Not sure......... 6 No answer/refused..... 2-5 2 30. Do you think that the law should allow doctors to administer drugs that might shorten life to patients in severe pain with no hope of recovery, if the patient has requested the drug and understands the conspQuentes? Should allow......(4S( g3-1 Should not allow...... 35-2 NOt Sure. svoeevcannns 12-3 31. Do you think that the law should allow doctors to comply with the wishes of a dying patient in severe distress who asks to have his fife ended? Should allow... (u6( 26-1 Should not allow...... 59-2 Not sure.............. 15-3 32. Now I would like to turn your attention to a final set of questions about informed consent. What does the term "informed consent” mean to you? What else? Informing patient about his condition and recommended treatment (u7(s8 Informing/disclosing/explaining treatment risks to patient (48(47 Patient understands the nature of his condition and treatment (49(34 Patient gives permission/consent /agrees to treatment (50( 26 33. In your practice, do you usually get written consent, oral consent, or no consent from patients before each of the following procedures? |READ EACH ITEM| Written Oral Both No Mer atte Not Consent Consent (Vol.) Consent (Vol.) Sure a. In-patient surgery............(51( 62 -1 3-2 11-3 * 2s +g b. Minor office surgery..........(52( 22-1 _49 -2 8-3 b-4 14s * 6 c. Setting bones.. ceseeeln(53C 170 19 2 4g “ou 55 os a) d. General anesthesia............(54( 48 _) 2 2 _ 7-3 Pew 42 os a) e. local anesthesta..............(s5( 19-1 5! .» 6.3 13.4 _ 10.5 1 r CH A TI Wa Bo S53 _ Ta Ms g. Blood tests........... lel 2-1 4% 2 ty Bau bg 1 ho Prescriptions.................(58C '-1 42, 2.3 SB. _ 2.5 1. i. Radiation therapy.............(59( 28 _) 8 2 5.3 2 a 55 .g 2 -6 Surveys of Physicians and the Public 303 34, Three general standards have been proposed to define a physician's legal obligation to disclose treatment information to his patients. I will read you these standards and would like to know which one you feel would be the best disclosure standard. |READ LIST| |SINGLE RECORD BELOWI a. Physician should disclose the information that the average, reasonable practitioner would disclose under the same or similar Irene ances ssi vn dns san msewneennnennrns {EDL 26 -1 b. Physician should disclose the information that the average, reasonable patient would consider relevant to his decision whether or not to accept the proposed treatment.......... 21 2 c. Physician should disclose the information that the particular patient being treated would consider relevant to his decision whether or not to accept proposed treatment..... tesssnsae -3 No general standard of disclosure {s desirable (vOl.)uieeessenenncs 8 NOL SUP€.vevsessssrsssesssssssssnssastsscssssssasentesenssstoeesss -5 35. Do you know which standard is applicable in the state(s) in which you practice? Yes, Know...oeoes.(61( 23 =1 (ASK Q.36) No, does not know....._ 76 -2 (SKIP TO Q.37) 36. Which is that -- "average, reasonable practitioner," "average, reasonable patient,” or "particular patient"? (N = 188 -- those answering "yes" to Q. 35) Average, reasonable practitioner...(62( 66 -1 Average, reasonable patient............ -2 Particular patient.. 10 -3 Bl Not sSure........eeee 37. Written consent forms are now standard practice in many hospitals for surgery and certain other medical procedures. In your experience, what has been the effect of these forms? What else? Zz Improved patient awareness (63( 28 No effect (6u( 2? Legal protection for physicians and hospitals (65( 18 Improved doctor-patient communications 16 38. Now, I am going to read you a series of statements that have been made about informed consent, and I would like to know whether you tend to agree strongly, agree somewhat, disagree somewhat, or disagree strongly. IREAD EACH ITEMI| Agree Agree Disagree Disagree Not | START AT "X"| Strongly Somewhat Somewhat Strongly Sure ( ) a. Patients’ rights to informa- tion about treatment risks and alternatives should be protected by 1aW...o.eenvs-sn.o(66(38 -1 _38 -2 _10 -3 10 4 4 -5 ( ) b. The time spent discussing diagnosis, prognosis, and treatment with patients could be better spent in taking care of patients. .......(67C 7-1 14 =2 26 =3 S51 -4 2 -5 ( ) ec. The legal requirements for obtaining informed consent are clear and explicit.........(68(13 =] 19 -2 28 -3 30 -4 10 _-5 ( ) d. The primary purpose of consent forms is to protect physicians from lawsuits. L..(69(20 1 34-2 24 3 19 4 3 -5 ( ) e. Written consent forms are helpful to doctor-patient : COMIURI DALTONS: vusevessrvnwennsl0L26 1 38 -2 18 -3 15 -4 3 -5 ( ) f. The requirements of informed consent put too much emphasis on disclosure of remote risks..(71( 314 ( ) g. A patient's signature on a written consent form estab- l1ishes that the patient has given consent......cecces (7330 2 32 .2 20 3 14 uw 4s 173-80z 304 Making Health Care Decisions: Appendix B =14- CARD 3 812041-A1 Fl. Where did you go to medical school? (10-11) 2 (NAME OF SCHOOL) Public low research § 34 Private low research $s 20 No answer/refused..........(12( -1 Public med. research § 10 Private med. research $s 6 F2. In what year did you graduate from medical school? Public high research § 3 _% Private high research § 12 Pre 1945 - 15 (Year) All Foreign Schools 16 1946-1965 - 39 (13-10) 1966-1970 - 9 1971-1975 - 14 Not sure..........(15( -1 1976 + - 24 No answer/refused....._» -2 F3. while you were in medical school and residency, did you have any formal training in (READ EACH ITEM)? No, Didn't Yes, Had Have Not Sure a. Medical ethics........ eeea(16( 50-1 46-2 4-3 b. Medical law......eeiiiiennnne (17 36-1 60 -2 4-3 c. Doctor-patient communications. (18( 54 -) 43 2 3 -3 Fla. way you practice medicine? Fup. FS. Which of the followi —_——y 9 Fue TELORD BELOW] Influential Your personal values and beliefs.........(19( 62 -2 Your medical trafning.ceennnnniiiiininn.... 18 Your experience in practice..... Not ‘sure...svsnioss ng_three items do you think has had the most influence on the IREAD LIST, SINGLE RECORD BELOW| Q.Fup Next Most Influential (20( 19 -1 And, which has been the next most influential? [READ LIST, SINGLE RECORD ABOVE | —_— LS FRY AONE Would you mind telling me what was the last father completed? Fé. F1. Less than 8th 8th Grade . Some high school . High school graduate.. Some college. . Four-year college graduate.. Postgraduate...... . -2 grade.......(21( - Would you mind telling me in what religion you were raised (22-23) eeea(28( -1 seeeeeene 10 =2 Not sure. Refused... And, how religious are you today? Would somewhat religious, or not very religious? Very religious......(25( 22 -1 Somewhat religiou ee_46 2 Not very religious. -_30 -3 Not sure.... 1-4 No answer/refused.... 2 5 Thank you very much for your cooperation! grade or year of school that your |» » if any? x Catholic 26 Protestant 42 Jewish 15 Other 4 None 3 you describe yourself as very religious, Surveys of Physicians and the Public 305 INTERVIEWER: RECORD, DO NOT ASK: F8. Respondent's Sex: Male.......(26( 9% -1 Female....... oe -2 F9. Respondent Speaks English: Fluently.......(27( 88 With difficulty. [3 T2B-Bozl 306 Making Health Care Decisions: Appendix B GENERAL PUBLIC QUESTIONNAIRE LOUIS HARRIS AND ASSOCIATES, INC. FOR OFFICE USE ONLY: 630 Fifth Avenue . New York, New York 10111 EXPIRES 7/31/82 Questionnaire No. : trikes 13/8 Study No. 812041-C2 5-6-7-8 6 Sample Point No. 10-11-12-13-14-15-16-17-18 Date of Interview: Time of Interview: — —_— Interviewer's Name (PRINT) Date: February 1982 - Form C2 Hello, my name is from Louis Harris and Associates, the national public opinion research firm in New York City. We are conducting a national survey of public attitudes toward health care issues for the President's Commission for the Study of Ethical Issues in Medicine. The purpose of this study is to learn more about the public's view about information disclosure and dec ionmaking in medical settings. Your participation will be kept strictly confidential and your responses will be analyzed only in aggregate statistical form. Your identity will not be disclosed to the commission or any other organization. The interview should take about thirty minutes. You're certainly not required to respond, but your cooperation is needed to make the results meaningful and accurate. Are you 18 years of age or older? (IF NO, ASK IF ANYONE IN HOUSEHOLD IS 18 YEARS OLD OR OLDER AND ASK TO SPEAK WITH THAT PERSON. IF NO ONE IN HOUSEHOLD IS 18 OR OLDER, SCREEN ouT) (CONTINUE WITH QUESTIONNAIRE) . (NONELIGIBLE) 019) . (REFUSAL) FROM OBSERVATION: Sex Male.......(20( 48 -1 Female......... 83 =2 Region: 2 x East 76 Central Cities 29 South 29 Suburbs 28 Midwest 26 Town 15 West 19 Rural 28 Key: * = SZ sy, 0 = standard error; X = mean; all other entries are weighted percentages Note, region and size of place percentages are unweighted I= CARD 1 812041-C1 rr ———————————————————————————— LC I. 1 would like to begin with some background questions about your own health. First compared to others your own age, would you sav that your health is excellent, good, fair, or poor? Excellent....(21(38 -) Good... . 3 Fair. 12-3 Poor... 6 -4 Not sure * 5 2. Is there a particular clinic, health center, doctor's office, or other place that you usually go to when you are sick or need advice about your health? Yes, there is..... (22( 81 -1 (Ask Q.3) No, there is not.. 19 -2 (SKIP TO Q.4) Not us Surveys of Physicians and the Public 307 3. What kind of place is that -- a clinic, a health center, a hospital, a doctor's office, or some other place? Clinic.. Hospital Doctor's office. coves 8) _b Other (SPECIFY) pC CE Not sure........ seeee dk __=6 4. Have you been a patient overnight in a hospital during the past year? Yes, have been....(24( 15 -1 (ASK 0.5) No, have not been..... 85 ~-2 (SKIP TO Q.6) 5. How many nights have ou spent in the hospital suring the past year? (N=180 -- those saying "yes" to (25) (26) 27) a Nights ur activities limited in any way because cf a health condition or disability? 6. Are yo Yes, limited. (29( 17 -1 No, not limited..... -2 Not sure...... Pa 7. How many times have you seen a doctor about your health during the past year, not counting any hospitalizat ion? (30) (31) (32) < Times verall|x = 3.1 Jremm——————— Nome. ........(33( =] Not Sure......oee ¥o2 Ba. Including yourself, is there anybody in your immediate family who is chronically sick or who needs medical treatment or hospitalization fairly often? Yes, there is....(36( 16-1 (ASK Q.8b) No, there is not....._ 84-2 (SKIP TO Q.9a) (N = 208 responding "yes" to Q. 8a) 8b. Who is that? MULTIPLE RECORD, IF NECESSARY Self.. 9a. Have you or a member of your immediate family ever had a life-threatening illness or medical condition? Yes, have had...(36( 3) -1 (ASK Q.9b) No, have not had.... £9 ~ 5 (SKIP TO Q.10a) NOt SUT@....oveenese_ * (N = 385 answering "yes" to Q. 9a) 9b. Who was that? MULTIPLE RECORD, IF NECESSARY Self.... wk37( 3] = Spouse ev 22 2 Child. 12 3 Parent. ere 32 4 Other (SPECIFY) .“ 2 =5 308 Making Health Care Decisions: Appendix B 9c. What was the matter? JPROBE Malignant neoplasms (38( 23 Heart attacks and related CV diseases (39( 21 Other CV diseases (40( 16 Chronic degenerative diseases 7 10a. Have you ever undergone a surgical operation that required an overnight stay in the hospital? Yes, have undergone...(41( 61 -1 (Ask Q.10b) No, haven't undergone..... 39 -2 (sKIP To Q.11) (N= 774 responding "yes" to Q. 10a) 10b. When you agreed to the surgical operation, did you base your decision primarily on your trust in your doctor's judgment, or was it based primarily on your own understanding and judgment of the situation? Own judgment. . Equally (vol.). Not sure....... 11. Now, 1'd like to ask you some more general questions about health care. In light of your experience, how satisfied are you with (READ EACH ITEM)? Would you say you are very satisfied, somewhat satisfied, somewhat dissatisfied, or very dissatisfied? Very Somewhat Somewhat Very Not Satisfied Satisfied Dissatisfied Dissatisfied Sure 8. Doctors, in general....ovuiuseunis (43(35 -) 41-2 16 -3 6 -4 2-5 b. The doctors you and your family have used in the last few years... (44(49 -) 35 -2 11 -3 3 -4 2-5 c. Your doctor's concern for your well-being (45(56 -1 30 -2 9 =3 3 whi Deb d. The amount of time your doctor spends with you........... teareae (66(50 -) 29 -2 12 -3 6 -4 3-5 e. Your doctor's willingness to answer questions about your condition and illness............. (67(61 = 25 8-3 4 -4 2-5 f. Your doctor's honesty in his discussions with you.............. (48(66 23 6 -3 2 ~4 2-5 g. Your doctor's respect for your treatment preferences...... 12. How would you rate your doctors on keeping you fully informed about questions and decisions relating to your medical care that you consider important? Would you say that they have kept you fully informed on all important issues, merit important issues, only some important issues, few important issues, or none of the issues you consider important? (50( 37 -1 30 -2 18 -3 6 -4 3 -5 5 =6 13. Do you usually come awav from vour doctor feeling that you have understood the important issues relating to your treatment -- fullv, more than adequately, adequately, or less than adequately’ More than adequately. Adequately...... wieis Less than adequately NOt BUTE: vuvivin in trmme vnss Surveys of Physicians and the Public 309 14. When a patient doesn't understand his medical treatment, how often is it because the doctor did not explain things well? Would you say always, often, sometimes, rarely, or never? Not sure.. 15a. Are there times when you discuss questions about your medical condition and treatment with a nurse rather than a doctor? Yes, discuss. ..(53( 32 -1 (ASK Q.15b) No, do not discuss.... 68 -2 (SKIP TO Q.16) 15b. When is that? (N = 373 answering "yes" to Q. 15a) 2 To make appointments (54(30 Get background information when doctor is busy (ss(17 About colds and minor illnesses (56(14 16. For routine care, do you think that doctors generally tell their patients too much, too little, or about the right amount? Too much..... Too little About the right amount. 56 -3 NOt SUTE@...ecovevens own bo =l 17. 1f your doctor spent more time explaining things to you about your routine care, would you be willing to pay more? Yes, would be willing...(58( 38 -1 No, would not he willing... 54 -2 Depends (vol.) 7 Not sure.. 18. In cases of serious illness, do you think that doctors generally tell their patients too much, too little, or about the right amount? Too much....... L(59C 2-1 Too little..... er 33 -2 About the right amount.._ 53 -3 NOt SUT€....vusaensnn a § 19. Who do you think is the best judge of the amount of information that should be disclosed to the patient -- the patient or the doctor? Patient. Doctor. Depends (vol.). Not sure. 20. Do you think that some patients should be told less about their condition and treatment than other patients, or should doctors provide the same amount of information to all patients? Some told les Same for all Not sure..... L.(61( 56 -1 30 -2 wre 83 21. Do you want your doctor to tell you everything about your medical condition, even if it is unfavorable? Yes, want.......(62( 94 -1 No, do not want. ZL Not sure..... cee 310 ~~ Making Health Care Decisions: Appendix B 22. Sometimes physicians feel that it is advisable to withhold information about diagnosis, prognosis, or treatment risks and alternatives from patients. Under each of the following conditions, do you feel that a doctor would be Justified in withholding information about condition and treatment from a patient, or not? READ EACH ITEM Not Not START AT "X Justified Justified Sure ( ) a. If the patient's family asks the doctor not to tell the patient ...e. veses swvws sain ise (63(48 -1 42 -2 9 -3 ( ) b. If the patient tells the doctor he doesn't want to hear bad news L(64(68 -1 27 -2 4-3 ( ) c. If the information might significantly harm the patient's health and well-being (65( 68 -1 27-2 5-3 ( ) d. If the information might make the patient anxious or upset (66( 52 -1 41 -2 1-3 ( ) e. If the information might make the patient unwilling to undergo treatment the doctor thinks £8 MECOBEBTY.ois iio vuinvis was fin shams enmmnmmns savas sae (67( 38 -1 55 -2 1-3 23. Have you ever asked a doctor not to tell you bad news? Yes, have asked.....(68( 2 -1 No, have not asked...... =2 NOt Sure................ 24. Have you ever asked a doctor not to tell another family member something concerning their health condition or treatment? Yes, have asked..... (69( 8 -1 No, have not asked...... =-2 Not ‘Bure... ssvininnes soe 3 (This version of Q. 25 was asked of one-half of the public sample. N=625) “Re CARD 1 812041-C1 25. There are a variety of issues that doctors may or may not discuss with their patients regarding the patient's condition and possible treatment. Do vou feel that doctors should initiate discussion about each of the following things with the patient as a matter of course, or discusc it only if the patient aske< about it? READ EACH ITEM Discuss Never Initiate Only If Discuss Not Discussion Patient Asks (Vol.) Sure a. Their diagnosis and prognosis of the patient's condition......eevuuninnnnnnnnnnn.,... (70(75 -1 21 =-2 -3 4-4 b. Costs associated with recommended procedure and whether they would be covered by patient's health insurance.................... RS (711(69 -1 27 =2 x -3 4 -4 ¢. The nature and purpose of the recommended LYEALMBOL iis uit ili mimmaains wnarummime CHER SFR EE (72(85 -1 13 3 1% 3. 2-4 d. The pros and cons of the recommended treatment..(73(81 -] 16 -2 -3 3-4 e. Side effects, including inconvenience and pain, that are certain or fairly likely to attend the treatment ..iivvivssiivsuesas ERE. ..(74(88 -1 10 =3 -3 2-4 f. The probable impact of the patient's condition and treatment on the patient's family life, when there is a likely impact...........ooueuunnn.n.. (75078 -1 18 -2 * -3 hb -a 8. The probable impact of the patient's condition and treatment on the patient's job role, when there is a likely impact... JE (76(79 -1 18 -2 -3 3-4 77-802 Surveys of Physicians and the Public 311 (This version of Q. 25 was asked of one-half of the public sample. N=626) -8- CARD 1 812041-C2 25. There are a variety of issues that doctors may or may not discuss with their patients regarding the patient's condition and possible treatment. How often does your doctor explain each of the following things to you -- always, usually, sometimes, rarely, or never? READ EACH ITEM Not Always Usually Sometimes Rarely Never Sure a. Their diagnosis and prognosis of the patient's condition......... ven «(70052 =1 26 =2 13-3 5-4 2-5 3-6 b. Costs associated with recommended procedure and whether they would be covered by patient's health IMSUFANCE. + aeresrnnrnnnnrsrsssssnnnn (7125-1 _ 20 -2 Woy es 24-5 “6 ¢. The nature and purpose of the recommended; Creatment , .. ovine vows {22(52-1 26 -2 13 «3 boy 3.5 346 d. The pros and cons of the recommended treatment.......... ceeen(23(4341 24 -2 17 3 boy Hef Sep e. Side effects, including inconvenience and pain, that are certain or fairly likely to attend the treatment...... (7639-1 29 -2 15 3 7.4 6-5 4g 75-802 -G= CARD 2 812041-C1 26. Another set of issues that may or may not be discussed with patients involves the risks of adverse consequences associated with treatment. Do you think that doctors should tell their patients about each of the following things as a matter of course, or only if the patient asks about it? TREAD EACH ITEM Discuss Discuss Never As A Matter Only If Discuss Not START AT Of Course Patient Asks _(Vol.) Sure ( ) a. Risks of death or serious disability that occur in about 1 in 100 cases....(10( 75 -1 22 =-2 * -3 2 -4 ( ) b. Risks of death or serious disability that occur in about 1 in 1,000 cases. .(11( 64 -1 33 -2 * -3 3 -4 ( ) c. Risks of temporary disability that occur in about 1 in 1,000 cases....... (12( 65 -1 12 -2 * -3 3 -4 27. Often there are significant uncertainties associated with medical decisions. When such uncertainties exist, do you think that doctors should tell their patients about each of the following as a matter of course, or only if the patient asks about it? READ EACH ITEM Discuss Discuss Never As A Matter Only If Discuss Not 0f Course Patient Asks _(Vol.) Sure a. Doctor's uncertainties or reservations . about diagnosis ARE (3 75-1 23 -2 3 2 b. Doctor's uncertainties about the best * 2 course of treatment....... (14( 80 1 18 =2 -3 4 c. Different views within the profession as to the best course of treatment for n 3 the CONdition. ..eeesssssesssssssnnnnenes eens 76 1 21 -2 -3 -4 d. Doctor's uncertainties about whether the x 2 ommended treatment will be successful....(16¢ 8 -1 14 2 -3 -4 312 Making Health Care Decisions: Appendix B 28. Now, I'd like to ask you a few questions about treatment decisions. There are many ways in which decisions may be made about medical treatment. Which of the following approaches would you prefer: READ LIST SINGLE RECORD 1. The doctor presents all of the available alternatives and allows the patient to decide....(17( 7 -1 2. The doctor discusses alternatives with the patient and the two of ‘them decide together how to proceed... seertrtttttnanttrenenes £3 3, The doctor presents his recommendation to the patient for the patient to accept or reject.......... 12-3 4. The doctor decides what should be done and does it... © -4 Depends (vol.).......... Not sure.....covevuunns 29. Have you ever refused to undergo treatment recommended by a doctor? Yes, refused........(18( 21 -1 No, have not refused 79 -2 30. On your own, have you ever sought a second opinion from another doctor? Yes, have sought.....(19(51 ~-1 No, have not sought......49 =-2 31. Have you ever changed doctors because you disagreed with them? Yes, have changed....(20( 36 -1 No, have not changed..... 64 -2 32. Has a doctor ever suggested that you find another doctor if you don't agree with his diagnoses and treatment recommendations? Yes, has suggested...(21( 20 -1 No, has not suggested.... BU -2 33. Do you think that patients should have the right to all available information about their condition and treatment if they wish to have it? Yes, have right.......(22( 97 -1 No, do not have right..... 2 -2 Not sures. ovsvevrnens . T_-3 34. Do you think that it is primarily the doctor's responsibility to make sure that the patient is fully informed about his condition and treatment, or the patient's responsibility to ask the doctor about these things? 20 Primarily the patient's responsibility.. Equally responsible (vol.). Bot ‘sures. uesvvevssvvnvavsess Primarily the doctor's responsibility..(23( 66 ~-1 2 35. Do you believe that increasing the patient's role in medical decisionmaking is likely to improve the quality of medical care, reduce the quality of medical care, or have no effect on it? Improve the quality..(24( 64 -1 Reduce the quality....... =2 Have no effect... 77-3 Not sure....... Ta 36. Why do you say that? z Increasespatient knowledge and understanding (25¢( 28 Improves doctor-patient relationships/communication (26( 24 — Improves medical decisionmaking (27¢_13 37. If you were seriously ill, would you want the responsibility of making the final choices about your medical treatment, or would you want your doctor to make them for you? Patient makes final choices..(28( 43 -1 Doctor makes choices..... Both (vol.) Not surRicessnsssvavasns vosvnnvve Surveys of Physicians and the Public 313 38. 1f you were too sick to make an important decision about vour medical care, who would you want to make the decision for you -- a family member, a close friend, your doctor, or a lawyer appointed to protect your interests? A family member.. A close friend.. Your doctor A lawyer......ooeennrennnans cores Doctor and family/friend (vol.).._§ ~- Other (SPECIFY) pees mp Not Sure...essesnns JR 1 =? 39. Have you given instructions to anyone about how you would like to be treated if you ever became too sick to make decisions? Yes, have given.....(30( 36 -1 (ASK Q.40) Not sure.. No, have not given......_ 63 3) (SKIP TO Q.41) . . X_=3 40. Did you put those instructions in writing? (N = 451 answering "yes" to Q. 39) Yes, put in writing..(31(23 -1 No, did not.. [a 76-2 Not sure...... . * -3 41. Do you think that the law should allow doctors to honor the written instructions of their patients, even if it means allowing them to die? Yes, should allow..... (32( 68 -1 No, should not allow. 20 2 42. Now, I would like to get your opinions on some hypothetical situations. Suppose that a doctor found that you had cancer. Would you want the doctor to tell you the diagnosis, or would you prefer that he treat you without telling you what was wrong? Want the doctor to tell me.......(33( 96 -1 Prefer treatment without telling..... 3-2 Not sure.......«. 43. Suppose that you had a type of cancer that usually leads to death in less than a year for most patients. Would you want your doctor to give you a realistic estimate of how long you had to live, or would you prefer that he not tell you? Give realistic estimate..(34(85 -1 Not tell me.. 4-2 Not sure....... ener . sews 3 =3 44. Assume that the cancer has spread throughout the body. It is very likely that you will die regardless of what you do. There is a decision between aggressive therapy -- which will probably make you feel sick and will probably not help your condition -= and supportive therapy -- which will not help your condition, but will allow you to be Comfortable. Who do you think should make the decision between these two therapies -- the doctor or you? Doctor should make the decision..(35(12 -1 1 should make the decision....... J =2 Should be joint decision (vol.).. 3 -3 NOU SUT@...evnvnnnnnnnnansnns corer 1-4 45. Assume that a patient who is in severe pain and with no hope of recovery has asked the doctors to help ease the pain, knowing that it might kill him. Under these circumstances should it be permissible for a doctor to administer drugs to reduce pain, even if the dose required might shorten his life? Yes, should be....(36( 79 -1 No should not be...... =3 Not sure ceeean =3 46. Imagine that this patient asks to have his life ended. Under these circumstances do you think it would be right for a doctor to comply with the patient's wishes, or would it be wrong? Would be right..(37( 46 -1 Would be wrong...... -2 Not Sure.....eeeess « IT -3 314 Making Health Care Decisions: Appendix B 47. Do you think that the law should allow doctors to administer drugs that might shorten the patient's life in severe cases with no hope of recovery, if the patient has requested the drug and understands the consequences? Should allow.....(38( 71 -1 Should not allow. Not sure.......... 48. Do you think that the law should allow doctors to comply with the wishes of a dying patient in severe distress who asks to have his life ended? Should allow.....(39( 53 -1] Should not allow..... -2 Not sure..... . eo_B 3 No answer/refused.... 2 -4 49. Now, 1 would like to turn your attention to a final set of questions about informed consent. What does the term "informed consent" mean to you? What else? 2 Patient informed/told about nature of condition & recommended treatment (40( 43 Consent to treatment (41( 43 Patient decisionmaking (s2( 19 Patient understands nature of condition and method of treatment 9 50. Should a doctor be held legally liable for failing to fully and completely inform his patient of the risks associated with a medical procedure? Yes, should be liable....(43( 72 -1 No, should not be liable..... -2 Not Sure ssw vwv inv 51. Three general standards have been proposed to define a physician's legal obligation to disclose treatment information to his patients. I would like to know which of these standards you feel would be the best disclosure standard. READ LIST SINGLE RECORD a. Physician should disclose the information that the average, reasonable doctor would disclose under the same or similar Circumstances............... samme sevens ceenn(b4C 18-1 b. Physician should disclose the information that the average, reasonable patient would consider relevant to his decision whether or not to accept the proposed treatment..... 28 -2 c. Physician should disclose the information that the particu patient being treated would consider relevant to his decision whether or not to accept proposed treatment.......... trssesssasaes 46 -3 52. Written consent forms are now standard practice in many hospitals for surgery and certain other medical procedures. Within the last year, have you been asked to sign a written consent for treatment? Yes, have been asked..... L(45(24 -1 (ASK Q.53a) No, have not been asked...... . 16-2 (SKIP TO Q.54) 53a. Did the doctor explain the consent form to your satisfaction? (N = 288 answering "yes" in Q. 52) Yes, he did.......(46( 71 -] No, he did not........ -2 53b. After reading the consent form, did you feel that you fully understood the risks of the treatment you were going to undergo? (N = 288 answering "yes" in Q. 52) Yes, fully understood.(47( 72 -] No, did not.... ee Not sure...... 53c. Have you ever refused treatment because of what you learned about the treatment from the written consent form? (N = 288 answering "yes" in Q. 52) Yes, refused..(48( 5. =] No, did not.. . Not sure..... 54. Surveys of Physicians and the Public 315 Now, 1 am going to read you a series of statements that have been made about informed consent, and I would like to know whether you tend to agree strongly, agree somewhat, disagree somewhat, or disagree strongly. READ EACH ITEM Agree Agree Disagree Disagree Not START AT "X" Strongly Somewhat Somewhat Strongly Sure ( a. Patients' rights to informa- tion about treatment risks and alternatives should be protected by law.................(49(63 -1 26-2 5 -3 4 -4 3-5 ( b. The time spent discussing diagnosis, prognosis, and treatment with patients could be better spent in taking care of patients..........(50(22 -1 22-2 25 -3 28-4 4-5 ( c. The legal requirements for obtaining informed consent are clear and explicit........ eeel51(22 -1 31 -2 21 -3 11-4 16-5 ( ) d. The primary purpose of consent forms is to protect physicians from lawsuits. ....ceuvevnsnvrnnnns (52(48 -1 31 -2 12 -3 5 -4 4-5 ( e. Written consent forms are helpful to doctor-patient communications....... wevwrenrerenl33( ZY =1 36 -2 19 -3 11 4 5-5 ( f. The requirements of informed consent put too much emphasis ; on disclosure of remote risks....(54(13 -1 31 2 30 3 13, 124 ( g- A patient's signature on a written consent form establishes that the patient has given consent......... eereesa(35¢57 -1 29 -2? 9-3 3 4 2-5 Fl. These last few questions are for statistical purposes only. How old are you? 38.4 (median) (56-57) YEARS Not sure......... .(58( -1 Refused.......conunn ve Rk =2 F2. What was the highest grade of school that you actually completed? No formal schooling...... (59 * -1 First through 7th grade . -2 8th grade.......cconnnn Lh 3 Some high school... T2-u High school graduate 35-5 Some college.....eoovuuannnn 23-6 Four-year college graduate... 13-7 Postgraduate. . 8 Refused......... 9 F3. Are you presently employed full time, part time, unemployed, retired, a student, a housewife, or what? F&4. Employed full time....... Employed part time. Unemployed.......... Retired... Student. Housewife. . . Disabled/too ill to work (vol.)... On duty in Armed Forces (vol.).... Other (SPECIFY) Have you ever been employed in a health-related occupation’ Yes, have.. No, have not. Not sure...... (61(21 oy 316 Making Health Care Decisions: Appendix B FS5. How many people are there in your immediate family -- including children and people away from home at school, at work, or someplace else? 4.4 (median) (62-63) WRITE IN NUMBER) Not sure.........(64( * -1 No answer/refused.... -2 F6. Which of the following income categories best describes your total 1981 household income? Was it (READ LIST)? $15,001 to $25,000 $25,001 to $35,000 $35,001 to $50,000. . £50,001 or over..... tonae IT wh Not sure/no answer/ refused................. 8 7 F7. What kinds of health insurance, if an are you covered by? Do you have any other health insurance? TREAD LIST] TMULTIPLE RECORD, IF NECESSARY Prepaid health plan (HMO)...(66( 17 -1 ~Blue Cross/Blue Shield..... Mp ye Insurance company policy.. * -3 Medicare/Medi-Cal......... 184 No health insurance at all 5-5 Other (SPECIFY) 6-6 Not sure/no answer /refused.. 37 F8. Do you have major medical coverage? Yes, have..... (67( 74 -1 No, do not have... 22 -2 Not sure.......... 3-3 F9. What is your race or ethnic origin? Is it: READ LIST SINGLE RECORD Asian (Oriental) or Pacific Islander.. American Indian or Alaskan native.... Black, but not Hispanic White, but not Hispanic Spanish-American (Mexican, Cuban, Puerto Rican, Central or South American)....... 6 =5 Refused/not: sre ..s veses ruven saves smsevvn sno I w6 F10. Would you mind telling me in what religion you were raised, if any? % (69-70) Catholic 29 Protestant 59 LIC 3 A Jewish weaves -2 Other 2 _—_— None..... Refused. . F11. And, how religious are you today? Would you describe yourself as very religious, somewhat religious, or not very religious? Very religious....(72( 28 -1 =2 =-3 =4 =5 Not sure........... No answer/refused.. Thank you very much for your cooperation! RECORD THE FOLLOWING -- DO NOT ASK:] Te AW NG 7” DO NOT AsR:i F12. Respondent speaks English: Fluently...... ceeenn(73C 96 4 With difficulty......... 2 74-802 Informed Consent ( \ and the Structure \_J of Medical Care Charles W. Lidz, Ph.D.* Alan Meisel, J.D." with the assistance of Janice L. Holden, R.N.* * John H. Marx, Ph.D.} Mark R. Munetz, M.D." Introduction Over the last three decades, the process of making deci- sions about what medical treatments patients will get and whether or not persons will become subjects in biomedical and behavioral research has come under increasing legal regulation. The centerpiece of this regulation has been the requirement that the doctor or .esearcher obtain the “informed consent” of the patient or subject. From the earliest of times under the English, and later the American, common law, the consent of the patient to medical treatment has been required in all but exceptional circum- stances. The old requirement has metamorphosed into a more elaborate requirement of “informed consent,” under which a doctor is obliged not only to obtain the patient’s consent to treatment, but to make disclosure to the patient of certain rele- vant information about the treatment before obtaining consent and embarking upon treatment. In the absence of informed consent or a recognized exception to the requirement, a physi- * Associate Professor of Psychiatry and Sociology, School of Medicine (Western Psychiatric Institute and Clinic) and Department of Sociol- ogy, University of Pittsburgh, Pittsburgh, Pennsylvania. + Professor of Law, Psychiatry, and Sociology, School of Law, School of Medicine (Western Psychiatric Institute and Clinic) and Department of Sociology, University of Pittsburgh, Pittsburgh, Pennsylvania. ** Western Psychiatric Institute and Clinic, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. t+ Professor of Sociology, University of Pittsburgh, Pittsburgh, Pennsyl- vania. +t Assistant Professor of Psychiatry, Western Psychiatric Institute and Clinic, University of Pittsburgh School of Medicine, Pittsburgh, Penn- sylvania. August 1982 318 Making Health Care Decisions: Appendix C cian who provides treatment to a patient may incur civil lia- bility. The prototypical situation that the courts have focused on in formulating the various rules of operation of the informed consent doctrine envisions a single physician in private prac- tice administering a discrete hospital-based treatment—usually (but not always) a surgical procedure. The increasingly frequent legislative enactments and administrative regulations dealing with informed consent also implicitly operate on this pro- totype, despite the fact that informed consent is now required in a wide variety of situations that may not conform to this model. Thus, there is reason to be concerned that the rules governing informed consent may be unrealistic because they fail to take into account that medical research and practice is a widely diverse, not a monolithic, enterprise. At the same time that the law of informed consent may have become increasingly less reflective of reality—and possi- bly as a result of this process—informed consent in practice often amounts to little more than obtaining a signature on a consent form (Lidz & Roth, 1981). If informed consent is to be more than an empty formality—and possibly a legal trap for unwary medical practitioners and researchers—the rules gov- erning informed consent must take cognizance of the differing structures for delivering medical care, making medical deci- sions, and recruiting research subjects. The legal model of informed consent anticipates that, when we enter the clinic, we should find the following behaviors or conditions. (1) Information is “disclosed” to patients by their doctor(s) or other health care personnel delegated by the doctor to do so. (2) Patients “understand” the information they are given, or they ask questions to clarify what they do not understand, and health care personnel make efforts to determine whether patients understand and to clarify understanding. (3) Patients make a “decision” either to undergo (consent) or forego (refuse) the procedure in question. (4) Patients act “voluntarily” in making a decision in that no coercion, duress, or undue influence is imposed on them by health care personnel. (5) When patients are incapable of understanding the in- formation that is (or should be) disclosed to them— that is, when they are “incompetent” —health care personnel invoke some sort of proxy decisionmaking process. Ordinarily this will involve obtaining informed consent from next of kin. The current investigation grew out of the hypothesis that the manner in which informed consent ought to be obtained must vary in accordance with a number of variables that pertain Consent and the Structure of Medical Care 319 in the practice of medicine. Because the operation of in- formed consent does and should differ substantially, owing to the varied structure of medical practice, the traditional, mono- lithic legal model of informed consent may require rethinking. Whether this reevaluation should be minor or major and what form it should take we can not say a priori; rather, this investi- gation was undertaken to provide the empirical basis upon which such determinations can be made. Our hypothesis arose in part from our speculations about how informed consent might operate in clinical settings, but it had another basis as well. Although there is extensive literature reporting empirical findings about informed consent, there is very little data that sheds light on the entire decisionmaking process; therefore, it is hard either to validate or question the application of the linear legal model of informed consent in clinical settings (Meisel & Roth, 1981). However, previous stud- ies of the decisionmaking process in psychiatric settings con- ducted by a research group of which we are members (Lidz et al. forthcoming 1983) strongly hint that there is substantial divergence between the legal model and clinical practice. In particular, we were interested in finding out which char- acteristics of the structure of providing medical care are likely to affect the processes of making disclosure and obtaining con- sent. We started with four major hypotheses. (1) The setting in which care is provided—e.g., inpatient versus outpatient—will affect the way in which information is provided, decisions are made, or both. (2) The nature of the disorder for which a medical inter- vention—i.e., treatment or diagnosis—is sought will affect the decisionmaking process. (3) The professional identity and status of the care pro- viders will affect the extent of disclosure and the pressures that come to bear on patients in making decisions. (4) Whether the procedure recommended for a patient is accepted or experimental will also have an impact upon the form that the decisionmaking process takes. We thus undertook to study a number of different treatment settings staffed by a range of health care professionals who performed a variety of diagnostic and therapeutic procedures. We hoped to learn what patients are told about particular diag- nostic, therapeutic, and experimental procedures and by whom; what questions patients ask, and of whom; and the temporal dimensions of “disclosure,” that is, whether it is the event envisioned by law or a process that unfolds, perhaps slowly and in non-linear fashion, over time. We sought to dis- cover how much patients understand, and how their level of understanding is tied to the foregoing variables in the dis- closure process; whether patients are perceived by health care 320 Making Health Care Decisions: Appendix C providers as capable of understanding (i.e., in legal terms, “competent”), and how disclosure and decisional responsibil- ity may vary in accordance with this perception. We sought to learn how decisions are made; what roles are played by health care providers, by patients, by family members, and by others; and what the pressures and influences are that come to bear upon patients in making decisions. This was a rather grand —if not grandiose—research pro- gram. However, despite the fact that we have gathered literally reams of data on about 200 patients, 35 doctors, 20 nurses, and countless family members, in a handful of different clinical settings, we have been unable to answer most of the questions we posed for ourselves. This inability amidst a mountain of relevant data is, we think, not a problem with our research strategy. The explanation is simple: informed consent is largely absent from the clinic; it is almost exclusively a creature of law. This is, to some extent, an oversimplification of our find- ings. But for rhetorical purposes, it is close to the mark. In fact, information is sometimes provided to patients, and patients sometimes make decisions. But when this happens, and it does not very frequently, the explanation for it most likely does not lie in law, but in medical custom. And we are convinced that this medical custom is deeply engrained and to date has not experienced much change under the influence of law. Research Methods The research was primarily based on a version of “partici- pant observation,” which is a general research strategy employ- ing a series of different methods that are used when the research goal is to gain new knowledge about the everyday world (Denzin, 1970). Because we were trying to learn about a routine part of hospital life, we followed the basic outline of participant observation and chose the method that, in any given situation, least affected the interaction we were observing and that provided the necessary data. Thus, the observers sat and observed unobtrusively before participating in what they ob- served, listened to others talk in preference to asking questions, asked questions in conversational contexts before doing struc- tured interviews, and memorized incidents rather than taking notes (Whyte, 1955). However, when the passive procedure would not yield the necessary data, the observers did not hesi- tate to use a more active one. For example, since we were concerned to report dialogue as completely as possible, the observers almost always took notes when watching staff and patients interacting. In each of the settings the observers, after obtaining the cooperation of the staff, spent a couple of days observing the routine procedures, getting to know the staff, and becoming a Consent and the Structure of Medical Care 321 regular part of the setting. During this period they took notes only sporadically and concentrated on the overall ward func- tioning. After this period, the observers began systematically to focus on aspects of staff and patient behavior related to the treatment decisionmaking process and to interview patients after each decision, using a semistructured interview schedule (see Appendix I for a copy of the schedule). What we observed varied from setting to setting. In outpa- tient settings the process could be captured by observing a small number of encounters between patient and clinician. However, on inpatient wards decisions were sometimes made over a period of several days or even weeks and involved a large number of people, including physicians, nurses, families, and the patient. The participant observer had to spend many days on the ward, doing rounds with the staff, observing nurses getting forms signed, listening to patients expressing their con- cerns and worries, and talking with relatives. We used two observers in both wards and the clinic. In surgery, one observer began with morning rounds at 6:30 a.m. and ended after afternoon rounds. The other observer started in early afternoon and continued until consent forms were signed between 7:00 p.m. and 9:00 p.m. Both observers also attended the outpatient surgical clinic on the two half-days each week that it was operating. In cardiology, one observer was assigned to each resident's service and began with morning rounds and ended when the interns finished seeing their patients in late afternoon. Patient inteviews were conducted whenever there was a slack time. Obviously we did not see everything. Sometimes an intern would see one patient while we were interviewing another. Sometimes a decision was made in the middle of the night. We often did not see patients’ families, and in cardiology there were literally a dozen physicians who might make a decision about any particular patient. Especially in cardiology, what we ob- tained was a cross-sectional view of how decisions were made rather than an adequate longitudinal study of any case. The observers were trained in speed-writing and took al- most verbatim notes of all observed patient-staff interactions. For interviews with patients, a tape recorder was often used since such interviews required too much participation from the observer for him or her to record the discussion verbatim. Information Collected in Each Setting. In general the ob- servers collected the following information on each treatment decision: (1) General background information (from patient's chart) on (a) date of first treatment for the disorder, 322 Making Health Care Decisions: Appendix C (b) date of admission, (c) diagnosis and description of the disorder, (d) patients occupation, (e) patient's level of educational attainment, and (f) date(s) of decisionmaking; (2) staff and patient formulations of patients problems and treatment; (3) staff and patient beliefs about the patient’s proper role in treatment planning; (4) conversations between patient and staff about treat- ment plans; (5) which staff members talked to patients about treatment plans and at whose initiative; (6) efforts staff made to persuade patient of their ideas about appropriate treatment; (7) patient's expectations of consequences of refusal of treatment; (8) patient's expectations of benefits and risks of treatment; (9) persons to whom patients talked about treatment; (10) patient's efforts to persuade staff when patient had a differing view of the correct treatment; (11) length of time patient took to decide: and (12) when patient apparently decided. Data Processing, Organization, and Analysis. Each day’s field notes were dictated to be ready for transcription the next morning. They were transcribed and ready for review and dis- cussion by all members by 5:00 p.m. This allowed the non- fieldwork members of the team to maintain quality control and to assure that necessary data were gathered. At the conclusion of the fieldwork the data were organized and summarized by patient so as to provide a complete descrip- tion of each consent or refusal. Each summary contained the information necessary to decide whether or not a particular case was relevant to one of the topics for analysis. In all, we observed 194 cases (74 in cardiology, 120 in surgery). However, on only 124 (58 in cardiology, 66 in surgery) did we gather enough data to allow in-depth use of the data; data from the other cases assisted in providing background knowledge about the routine in each setting. The incomplete cases involved mostly outpatients who came to the clinic to have an incision checked after surgery or patients who were already well into the recovery period from a cardiac arrest when we began our study. We conducted in-depth tape recorded interviews with 101 patients (43 in cardiology, 58 in surgery), Consent and the Structure of Medical Care 323 and this group constituted the core of the cases we analyze in this report. The Setting and the Routine Surgery. The surgical ward—referred to as the “University Service” —we selected to observe was by no means a typical one, but it had a number of advantages from our point of view. The University Service is a remnant of the era of large public hospitals when most patients were “charity patients” and their care was provided by doctors who were in the final years of their training. The chief resident of the ward obtained his final training by having primary responsibility for patients who could not afford a private doctor. However, the advent of Medi- care and Medicaid has greatly reduced the numbers of these patients. (When the service is not filled with such patients, private patients are assigned to the empty beds.) Nonetheless * the position of chief resident and the University Service con- tinue to exist, somewhat transformed. University Service surgical patients are a hodgepodge of patients who, for one reason or another, have no other surgeon to whom they can turn. Dr. A, the chief resident in general surgery, described his patient population as one-fourth trauma patients (who come to the University Service from the emer- gency room with gunshot wounds, after auto accidents, etc.), one-fourth patients admitted from his outpatient clinic where he treats patients who have no insurance or nowhere else to go for surgical consultation, one-fourth patients referred from the renal service for the surgical work necessary to begin renal dialysis (this seems to be a matter of convenience, since the dialysis clinic shares the same floor with the University Serv- ice), and one-fourth referrals from other sources in the hospital. Because this was a “general surgery” service, it focused on abdominal surgery. The medical staff of the University Service consisted of the chief resident (Dr. A, who was in his seventh post-graduate year), the senior resident (Dr. B, who was in his fifth post- graduate year), and an intern (Dr. C). While we were observing, these three were supplemented by a junior resident from an- other surgical specialty who was doing a two-month rotation in order to learn about postoperative care and by two fourth-year medical students or “acting interns” who did simple routine medical work. The ward contained only 23 beds. The nursing service was divided into two “teams,” corresponding to the two corridors leading away from the nursing station. Team One was responsi- ble for 13 beds and on a normal day shift had two RNs or one RN and an LPN. Team Two had one RN and an LPN. The head nurse was not associated with either team and usually worked 324 Making Health Care Decisions: Appendix C the day shift and final part of the evening shift. Evenings the ward was staffed by two RNs and two LPNs. Nights were more lightly staffed with only one RN on the ward. The routine of the University Service's medical staff was quite regular. About 6:30 a.m., the intern, acting interns, and usually the junior resident would come in and do morning rounds. This consisted of checking each patient’s chart, waking the patients up, and checking on whatever their problems were. Bandages were changed, medications adjusted, and the patients were often reminded of any surgery, tests, or other events of the coming day. Sometimes patients used the opportunity to ask questions or to complain about their care. Sometimes the senior resident or the chief resident would join these rounds. About 7:30 a.m. surgery began and typically lasted until mid-afternoon. It was followed by afternoon rounds, which was the time for the primary review of the patient’s condition and for making plans for continued treatment. All the medical staff plus the head nurse or, in her absence, the charge nurse, would go from one patient to the next discussing the patient’s condi- tion. This usually began with either the intern or one of the acting interns reviewing the patient's history and the state of all current test results. The medical staff would discuss these, and their discussion would be interrupted by occasional asides to the patient. On the afternoon prior to surgery, Dr. A would usually take the opportunity to review with patients the infor- mation he felt necessary for them to have prior to their signing a consent form.* (On weekends, “afternoon rounds” were done in the morning directly following the interns’ working rounds.) Afternoon rounds were usually followed by consult rounds, which involved seeing patients on other wards of the hospital who were possible candidates for surgery. The large majority of these patients were renal dialysis patients who needed new peritoneal catheters for peritoneal dialysis, gortex implants, or arteriovenous fistulae for hemodialysis. Although these patients were often seen more than once, they typically viewed the University Service physicians as peripheral medical personnel, and their primary relationship was with their nephrologist. Patients’ signatures on surgical permits were obtained by the nursing staff in the evening, after 7:00 p.m. medications. This was the task of the charge nurse for the evening shift. Although just how it was done varied from one charge nurse to another, typically the consent form signing followed a “pre- * In an interview after our observations were over, Dr. A told us that the only way in which he was aware that his behavior changed because of our presence was that he did this “while you were there on rounds.” Otherwise, he often did that by himself because the junior staff thought it “real boring to stand there and listen to me tell somebody what their operation's gonna be.” Consent and the Structure of Medical Care 325 operative teaching” process in which the nurse described every- thing that would happen in the next 24 hours, except the operation itself, and what the patient should do to facilitate recovery from the operation. The Outpatient Clinic. Twice a week, on Tuesday after- noons and Friday mornings, the staff of the University Service conducted an outpatient surgical clinic in a building housing numerous outpatient clinics a block from the hospital. The clinic area consisted of a large modern waiting area, five small examining rooms, and an equally small staff room with three desks and equipment for dictating additions to patients’ rec- ords. Patients were escorted into one of the examining rooms by the nurses who staffed the clinic on a permanent basis. (Other groups of physicians used the same facilities at other times.) For each patient, the intern, the junior resident, or one of the acting interns reviewed the record, saw the patient, listened to his or her complaint, took a history, and did a brief physical. He or she briefly discussed the case with Dr. A or Dr. B, who then saw the patient. Most, but not all, of what patients were told took place in this final interview, when decisions would also be made. Usually between 10 and 15 patients were seen per day in this manner. About half had previously been operated on and were receiving follow-up care. Of the remainder, some were scheduled for inpatient admission, some had immediate minor surgical interventions, and another group were given medica- tions and told to return if they needed further care. Cardiology. The medical setting chosen for our study was University Hospital's 45-bed medical/surgical Cardiovascular Unit. There were 35 medical cardiology beds as well as a 5-bed coronary care unit. In addition there was a 4-bed special care unit for postoperative cardiovascular care as well as for over- flow coronary care patients. The remainder of the beds were for pre- and postoperative cardiovascular surgical patients. Our observations focused on the medical cardiology service, al- though, particularly when it came to the observation of nursing interaction, we did see some surgical patients. Typical patient problems included rheumatic heart dis- ease, coronary artery disease, congenital heart disease, and primary cardiac muscle disease. Patients were either admitted as emergencies (and therefore put in the CCU) or referred to the hospital by their attending physician for testing, observations, or a cardiac workup. The Cardiology Service was divided into two separate geo- graphical units: Unit 32 in the West Wing with 20 beds and Unit 33 in the North Wing with 25 beds. Each unit had a mixture of medical and surgical patients, its own head nurse, nursing staff, aides, and orderlies. Unit 33 included the 5-bed coronary care 326 Making Health Care Decisions: Appendix C unit (with its own charge nurse) and the 4-bed special care unit. All of the beds on the unit had cardiac monitors, and therefore the more critically ill patients were to be found there. Nursing staff for Unit 33 consisted of RNs only. A typical day shift on Unit 33 consisted of seven RN, including the head nurse, a monitor nurse (who maintained watch over a bank of monitors in the nursing station), a special care nurse, two cardiac care unit (CCU) nurses, and two floor nurses. A floor nurse and an aide were each assigned to all the patients on one side of the corridor of the Unit. As a patient approached discharge and was considered to be more stable, he or she might be switched to Unit 32, which had a smaller nursing staff consisting (for the day shift) of a head nurse and two other floor nurses and an aide or two as well. The nursing staff here, too, was divided into two teams such that one team was assigned the patients on one side of the corridor. Both units were more lightly staffed on the evening and night shifts with the exception of the CCU and Special Care Units. The house staff on the Medical Cardiology Service con- sisted of a senior resident, Dr. D (third-year post-graduate); a junior resident, Dr. E (second-year post-graduate); and four in- terns, Drs. F, G, H, and I. The residents rotated through the service every two months and the interns every five weeks. They were divided into two teams, each consisting of a resident and two interns. Patient assignment was arbitrary. Whoever was “on call” and “doing admissions” picked up the new patient. According to University Hospital policy, on the Medical Cardiology Service, as on all other services, the intern was considered the patient's primary physician. He or she was re- sponsible for the initial evaluation of the patient admitted to the service, initiating the diagnostic evaluation and management plan, and providing daily care and instruction for the patient.In these endeavors the intern was supervised by the resident and assisted by the attending physician and consultative physi- cians. The intern was expected to cooperate closely with private attending physicians. If a patient did not have an attending physician upon admission, he was assigned one in the Emer- gency Room from a list of physicians available. The residents had immediate responsibility for the intern’s formulation and evaluation of the patient's problems, and they were to be readily accessible to the interns when needed. There were nine full-time cardiology faculty members who supervised teaching and patient care. The ones who partici- pated in our study included Dr. J, Director of the Exercise Stress Lab; Dr. K, in charge of invasive clinical electrophysiology; Dr. L, Director of the EKG Lab; Dr. M, Director of Non-Invasive Lab (echocardiography, etc.); Dr. N, involved in the clinical teaching program; Drs. O and P, the Director and Co-Director of the Cardiac Catheterization Labs; and Dr. Q, Director of the entire Consent and the Structure of Medical Care 327 Cardiology Service. There were also eight cardiology fellows who were primarily responsible for the various specialty areas. Dr. R, for example, was in charge of the CCU, and Drs. Sand T did many of the cardiac catheterizations. The routine on the Cardiology Service was somewhat vague. Interns were expected to review their patients from 7:00 to 7:30 a.m. prior to morning work rounds. The latter were to be conducted from 7:30 to 8:30 a.m. In reality, however, on the team headed by the junior resident, the interns made more thorough rounds on their own patients and spent less time in more formal morning working rounds. Morning report was next on the agenda. The residents on the medical service met with the chief of service to review the previous day’s activities and to present new or problematic cases. Also in attendance were several faculty members and fellows. The daily routine progressed with coronary care unit rounds at 9:30 a.m., which were attended by all residents and interns and conducted by a faculty member. These rounds con- sisted of a presentation of all of the CCU patient's cases. A lengthy discussion of any new patients ensued in which the proposed treatment was discussed. This was followed by a short visit with the patient, with whom there was often some discussion. CCU rounds were followed by floor rounds led by a visiting attending physician or a faculty member and were meant to be “patient management” oriented. Both residents and interns at- tended, and they presented two or three interesting cases for the visiting physician to review. Generally the patient was seen in order to supplement the data with a clinical assessment, but these conferences were highly theoretical and there was little discussion with the patients. The rest of the day on the cardiology unit was unstructured for the house staff. They remained on the ward, saw their patients once or twice informally, reviewed charts, wrote orders for tests and medications, and if they were “on call” did “ad- missions” as they came in. Nurses gave medications throughout the day, changed dressings, got consent forms signed, and pre- pared patients for various tests as required. There were no official afternoon rounds that required attendance of either nursing or house staff. Attending physicians, however, were expected to make rounds. This tended to occur quite late in the day or early evening, varying from physician to physician. Some would come in and review the charts, talk with the house staff, and see the patient on a daily basis; others every other day or less. Frequently the physician would come in and see the patient but not the house staff and simply leave a note in the patient's chart. 328 Making Health Care Decisions: Appendix C Additional Cardiac Facilities. Adjacent to the CCU was the Exercise Stress Test Lab. For the most part, the patients under- going the exercise testing were outpatients, referred by their private physicians for more extensive diagnostic testing. The procedure followed by the Stress Lab personnel was to send the potential patient a cover letter and description of the stress test, its risks and benefits, accompanied by the necessary consent form. The patient then scheduled an appointment with the Lab, came in, and was asked by the nurse in charge of the Lab if there were any questions. She gave patients a more detailed description of the actual procedure and then asked them to sign the consent form. There were two cardiac catheterization labs on another floor of the hospital. Along with cardiac catheterization to diagnose coronary artery disease, a related procedure used car- diac catheters to convert such malignant rhythms as ventricular tachycardia. This latter procedure was done under the direc- tion of Dr. K. Patients undergoing these electrophysiological (EP) studies were admitted to the hospital as inpatients. Findings Informed Consent: For Which Treatments? Physicians, nurses, and other health professionals often ask, “Do I have to get consent for everything?” Viewing informed consent as syn- onymous with having the patient sign a consent form, they fear the need to have one signed before they can change a bandage. In spite of the mistaken concept of informed consent on which such questions are grounded, the problem is a real one—to what sort of treatment decisions does (and should) the doctrine apply? In this section we will look at how the question is currently answered at University Hospital. For, in spite of the vagueness in both ethics and common law as to what types of decisions the informed consent model of decisionmaking ap- plies, both hospital administrators and individual clinicians have made determinations about the circumstances under which informed consent is required. In this section we will describe the results of those determinations as well as patients’ perceptions of their own role in this process. University Hospital's 19-page policy on informed consent contains three partially contradictory statements about when informed consent must be obtained. First it states that informed consent is required for all invasive surgery, any procedure using an anesthetic, experimental procedures, and any “non-surgical procedures which the chief of the department has determined involve more than a slight risk of harm.” At another point it states that informed consent is required for “any procedure which the medical staff determines requires a specific explana- tion to the patient.” Finally, it states that except for emergen- Consent and the Structure of Medical Care 329 cies, incompetency, and the therapeutic privilege, no “health care service [will be] performed unless the patient or the pa- tient's legal representative has given prior consent to these services.” It thus seems that the hospital policy requires “con- sent” to all procedures and/or treatments of any kind and “in- formed consent” to all risky or invasive procedures with all but one of the recognized exceptions. In practice, however, the hospital staff makes many more subtle distinctions about from whom, for what, and how to obtain consent. In this section we will deal only with the question of how decisionmaking structures differ depending on the procedure or treatment in question. We found that the decisionmaking process differs in cases of surgical treatment, “major” diagnostic tests, “minor” diagnostic tests, and medica- tions, and will therefore organize the findings in these four categories. Surgery: the process of disclosure and decision. The de- cision that surgery should be performed is ordinarily viewed in law as a relatively clear-cut event in which the doctor reviews the risks and benefits with the patient and the patient then decides. In actuality the twin processes of disclosure and deci- sionmaking in elective surgery are often much more complex than such a model assumes, and patients may come to under- stand many of the critical issues much later in the process than at the point at which the formal decision is made. The basic decision to have surgery is often made long before any consent form is signed and before many of the critical issues are discussed. Before surgical patients are admit- ted to the hospital, they have often visited the outpatient surgi- cal clinic on one or more occasions, where a decision has been made to enter the hospital for surgery or for further tests to determine the advisability of surgery. Other patients were re- ferred by other physicians and the decision to have surgery was often already made before they were seen by the surgical team. Thus, it is difficult to locate a point when the decision to have surgery is made. In one case, the patient seems to have “de- cided” to have surgery almost immediately upon recommenda- tion in the Outpatient Clinic. Yet two of the surgeons were still arguing about the necessity for surgery two days before it took place and after the patient had already been admitted to the hospital. The signing of the consent form did not take place until long after both patient and staff had made up their minds to go ahead with the operation. Even after the consent form was signed, patients sometimes continued to receive information about the procedure and their understanding thus continued to evolve. Some of the changes in understanding resulted from the patient having been inadequately informed at the outset, but sometimes patients simply did not absorb information the first time it was told to them. 330 Making Health Care Decisions: Appendix C What makes decisions about surgery unique among our observations was that there was both a formal consent form signing and a relatively casual, trusting relationship between doctor and patient. This combination was not in evidence in other situations, which were characterized by either little or no disclosure or by a highly formalized ritual of informed consent between relative strangers. Major diagnostic tests. Although informed consent law has largely been built around surgery cases, the requirement to obtain informed consent did not play a particularly major role in surgical decisions, and the execution of consent forms was treated by patients and staff as a largely irrelevant formality. However, in decisionmaking about potentially dangerous diag- nostic procedures, informed consent played a major role in the relationship between the physician and patient. We observed consent being obtained for three different such procedures: cardiac catheterizations, exercise stress tests, and “elec- trophysiological studies.” All of these diagnostic procedures were typically, but not uniformly, preceded by extensive dis- closure and the signing of a consent form. However, the dis- closure and decisionmaking processes varied among all three. The ordinary procedure for a cardiac catheterization began with the patient’s admission to the hospital by a private physi- cian to have a catheterization performed by the staff of the catheterization laboratory. Dr. O, who headed the “cath lab,” claimed that most private physicians do not tell their patients much about the procedure beyond the simple fact that they should admit themselves to the hospital for a catheterization. Dr. O, however, had been (unsuccessfully) sued for failure to obtain informed consent and subsequently developed a pro- cedure whereby one of his assistants explained the procedure in detail to patients. Almost all the patients cared for by Dr. O’s lab received a detailed explanation. However, not all catheterizations at Uni- versity Hospital were performed by the same team. Another group we observed only told one patient that they planned to use an artery in the arm rather than the groin, that they might need to exercise the patient during the procedure, and that there was “some risk of the artery getting blocked up from the cath,” but that shouldn't be much risk because “you are a big enough man and should have a good artery.” Interestingly, this patient, Mr. S.R., told us that his major source of information was listening to one of the staff members of the Dr. O’s lab talk to his hospital roommate, who was also getting a catheterization. One patient, Mrs. A.C., said she could not remember the risks right after she was told them. By contrast, Mr. S.R. gave a good description of the procedure and the utility of a catheter- ization: “They've lost people doing catheterizations . . . . There is a certain amount of risk involved in any physical test. You Consent and the Structure of Medical Care 331 know, if it was a chance of 100 to 1, I don’t know what the odds are, but if it was as low as 100 to 1, I'd still want to have the exam if they thought it was necessary.” The comparison of Mr. SR. and Ms. A.C. is interesting because he understood the risks better than she did despite her having been better informed by her doctor. However, even Mr. S.R., who had a post-graduate degree, a brother with a similar problem, and a personal in- clination to learn about his treatment, was largely willing to leave the treatment decision to his doctor. However, two catheterization patients did object to having the procedure. One was a medic who was transferred to Univer- sity Hospital for a cardiac catheterization in the aftermath of acute pericarditis, and there was substantial staff disagreement about the need for the procedure. The other patient, Ms. S.W., was a 63-year-old woman with an active case of metastatic breast cancer. She was admitted with a myocardial infarction, and the attending physician decided that she should have a catheterization. When the resident asked her what she wanted to do, she said, “I don’t think I need it. If I can get out of here without it, Id rather. My veins are so small; my blood’s thick. I'd rather not. I figure I ought to leave well enough alone.” Nevertheless, she was persuaded to have it. Another major diagnostic procedure was the exercise stress test. Most of the patients in this group were outpatients referred by their private cardiologists who wanted to take advantage of the hospital's more sophisticated equipment. The lab’s standard procedure was to send a cover letter and consent form by mail asking the potential patient to read it at home. We cannot be sure how much effect these consent forms had on patient deci- sionmaking because all the patients we were able to interview were ones who agreed to have the treatment. However, the nurse who was in charge of administering the tests told us that she could not remember a patient ever refusing the procedure be- cause of what was on the consent form. Even assuming we trust her memory, however, this is questionable evidence since it is possible that some patients cancelled their appointments be- cause of what they read, but may have stated a different reason. The procedure for obtaining consent consisted of the nurse asking patients if they had read the form and then asking if they had any questions. Only one of six patients had a question, and that concerned which of three variations of the test she would be getting. As far as we can tell from interviews with patients, only one of the six really gave the problem any consideration beyond what their private doctors had told them. This patient, Mr. R.W., who said that the decision for him to get the test was up to his doctor, did call his two sons, both of whom were physicians, to ask their advice. The final major diagnostic test was the experimental elec- trophysiological procedure directed by Dr. K. The procedure 332 Making Health Care Decisions: Appendix C involved giving the patient different anti-arrythmia drugs and then electrically stimulating the patient's heart to try to recreate the dangerous ventricular arrythmia that brought the patient into treatment. If they succeeded in recreating the arrythmia they then stopped it electrically and tried another drug. The goal was to find a drug that would prevent them from creating the arrythmia. Dr. K clearly felt that it was important to tell patients everything about this procedure because it was so risky. He told us: “Given what I am doing to them, I damn well better get informed consent.” His standard procedure involved a disclosure reasonably similar in length and detail to that which Dr. O's assistants did for catheterization. There were, however, several differences. First, instead of delegating the process to a subordinate, Dr. K talked with the patient himself. Moreover Dr. K, at least, made a convincing pretense of warning patients of the risks and encouraging them to make their deci- sions. In spite of, or perhaps because of, his openness and frankness with patients, their decisions were probably based as much on their trust in him as on the information he gave them. We have focused considerable attention on the ways in which decisions about serious diagnostic procedures were made in part to show how they systematically differed from consent to more routine kinds of care (such as decisions about surgery or medication) in which the treating physician dis- cussed treatment with patients. These diagnostic procedures were usually but not always performed by someone who had no day-to-day responsibility for the patient. Decisions about these procedures were made with the formality of agreements be- tween strangers rather than the casual, informal discussions that characterized decisionmaking in surgery and in routine cardiological care. Consent to these two different types of procedures reflects two totally different kinds of doctor-patient relationships and is reflected in the two different uses of consent forms. The surgi- cal decisions were imbedded in a personal relationship. Infor- mation was both disclosed and understood in the context of a fiduciary relationship in which the physician largely made the decision for the patient. While some of the tests were conducted as part of the same fiduciary relationship that surgery was, the relationship between doctor and patient was more typically as close to the arm’s-length relationship of commercial law as medicine ever gets. However, it is important not to get carried away with this distinction. Even the stress tests, which were conducted on an almost purely arm’s-length basis, were done on the order of the patient's personal physician. Likewise, doc- tors occasionally spent a considerable amount of time with patients getting to know their medical history. Nonetheless, even the distinction between the role of informed consent in arm’s-length and fiduciary relationships is an important one. Consent and the Structure of Medical Care 333 In spite of some similarities, there were substantial dif- ferences in the way in which informed consent to the three procedures was approached, reflecting the views of the staff about the meaning of informed consent for their procedures. The head of the stress test laboratory expressed the view that informed consent is just a needless way of scaring people. Therefore, his staff's procedure was to treat making disclosure and obtaining consent as pure formalities in the manner of arm’s-length commercial transactions. Similarly, the head of the catherization laboratory saw the problem as a purely legal one, and his subordinates followed what he saw as the letter of the law. Only the doctor performing electrophysiological stud- ies, who viewed what he was doing as very risky, was moved by a sense of ethical obligation to obtain informed consent. We think that these differences reflect the shared moral judgment that what Dr. K was doing (electrophysiological studies) was much riskier than what Drs. J (stress tests) and O (catheteriza- tions) were doing. Dr. O's staff went through more elaborate motions than Dr. J's but only because of Dr. O's heightened anxiety over being sued. Finally, although our sample was too small to draw definite conclusions, we found no evidence that one or the other of these three groups of patients understood more about their diagnostic tests. Levels of understanding seemed to depend on factors other than the seriousness with which disclosure was made. Minor diagnostic procedures. Although there was some variation among physicians and from one procedure to another, almost without exception both patients and staff believed that decisions concerning “routine” tests and diagnostic procedures were the doctors’ prerogative. Indeed, often patients were told nothing at all about such procedures and rarely complained about their exclusion. It is almost impossible to define in a conceptually sound way which tests are “routine” and which not. Yet there was a shared opera- tional consensus among staff and patients about which tests were considered the doctors’ exclusive prerogative and which not. Thus, ultimately we are left with a circular definition: those tests are routine that are considered by staff and patients to be exclusively within the doctors’ decisionmaking domain. However, it is possible to give a picture of which tests were minor by enumeration. They included: x-rays, CAT scans, blood tests, urine tests, ultra-sound tests, EEGs, EKGs, etc. Medications. Few decisions about treatment are, in prac- tice, as exclusively the property of the physician as are medica- tion decisions. While the doctrine of informed consent has made some inroads into at least the format of the way decisions are made about surgery and major diagnostic tests, the doctor’s 334 Making Health Care Decisions: Appendix C authority to prescribe medications seems, from our data, to be unaffected. Medication decisions, particularly in cardiology, were often all but invisible. In fact, no one conceived of them as involving “decisionmaking.” They were occasionally men- tioned by one physician to another or discussed between super- visor and supervisee, but often they were simply written into the chart without a word being spoken about them. However, patients were sometimes told more when their cooperation was needed. The problem was only slightly different on the surgical service where the most prominent medications were pain kill- ers, usually narcotics. Patients getting pain medications knew that they were getting them and usually had some control over the quantity of pain medications they got depending on how much pain they reported. Only with patients about whose pos- sible addiction the staff was quite concerned was there any substantial resistance to increasing doses of narcotics when patients complained of pain. The only other circumstances in which patients had some control over the drugs they received (or for that matter, in which they attempted to exercise control) existed when a patient had experienced a failure with that drug in the past. For example, Ms. S.M. had been on the surgical service for almost a month when the staff told her that she would be able to go home in the next few days. She responded happily but said that she was still having pain and the pain pills were not helping. The doctor said that they could give her either Zomax or Motrin. Ms. S.M. took a few seconds to respond and then said that she had had Zomax and it had not helped. So the doctor told her that they would try Motrin. In spite of the lack of disclosure about medications, most patients knew what drugs they were getting. The source of this information was the nurses who dispensed medications. (The role of nurses in disclosure is discussed in more detail in Section “Medication Rounds” under “The Role of Nursing.”) Treatment without consent. Some medication decisions and minor diagnostic procedures were almost the only treat- ments administered without express consent. With a single exception, there were no surgical or major diagnostic decisions made without explicit consent from the patient or family (ex- cept in emergencies). There was not always adequate dis- closure, but patients did at least assent. However, as we have seen, minor procedures and medications were often ordered and sometimes executed without any apparent consent. The single surgical procedure done without the patients assent occurred with Mr. E.R., a 20-year-old outpatient who had previously had a rectal abscess drained. Although we did not Consent and the Structure of Medical Care 335 observe the original drainage procedure since we were 0C- cupied with other patients, his screams were audible through- out the clinic. Nonetheless, he returned for follow-up care, and Dr. A told him he was going to have to pack the wound again so that it would heal correctly and began the procedure. It went on for several minutes in spite of his screams and objections. The patient did not physically resist and afterwards told our inter- viewer that the doctor “probably had to do what he did.” Aside from this instance, the only treatments without con- sent were for medications and minor tests. Even in those cases, the nurse or technician typically told the patient beforehand what was going to happen, even if he or she did not obtain express consent. Thus, in a sense, patient's failures to object were taken for consent. Conclusions. The most elaborate disclosures that were made involved diagnostic procedures performed by staff who were not directly responsible for the patient's day-to-day care. These typically involved either elaborate ritualized disclosures or detailed consent forms. Disclosure for surgery tended to occur over a substantial period of time and to be much less formal. Minor procedures and medications were administered almost without any disclosure, and these decisions were viewed by both doctor and patient as being the exclusive province of the doctor. Two other things stand out. The most formal consent pro- cess occurred where there was the greatest “personal distance” between clinician and patient. Second, there was less patient participation in frequent events than in rare ones. For patients to participate in decisions about surgery or catheterizations did not require very much staff time because these procedures were infrequent occurrences. However, medication changes and new tests were ordered daily for each patient, and consequently it would have required much more staff time to involve patients in these decisions. Perhaps a more important, if less distinct, factor is that medications and tests seemed routine to both patients and staff. They did not stand out of the blur of day-to- day work in the way surgery and “major” procedures did. In light of the absence of patient participation in decisions about medications, one conclusion that should not be drawn is that there is greater patient participation in decisions about risky treatments than less risky ones. Although patients certainly do not view medications as especially dangerous nor do doctors act as if that is the case by involving patients in the decision- making process, nonetheless medication decisions often in- volve more serious risks than surgery or major diagnostic pro- cedures. Cardiology vs. Surgery: Is Decisionmaking the Same? One of our hypotheses was that the nature of the disorder affects the manner in which informed consent is obtained. An associated 336 Making Health Care Decisions: Appendix C question was whether surgical decisionmaking was in any sig- nificant way different from medical (in our case, cardiological) decisionmaking. We found that decisionmaking is somewhat less complicated for both patients and physicians on the surgi- cal service than on the cardiology service. In this section we will focus on the differences and try to account for them. Some appeared to be inherent in the practice of cardiology and sur- gery themselves while others were largely incidental. The nature of treatment and the focus of decisionmaking. One of the major differences that surfaced from our observa- tions of patients and staff was the focus of disclosure and decisionmaking. On the surgical ward, there was usually one particular event, the surgical procedure, on which both patient and medical staff focused their attention. The decision to un- dergo the procedure was frequently made prior to the patient's admission to the surgical ward. The remaining communica- tions focused on preparing the patient for this event: schedul- ing it, giving preoperative instructions, getting the necessary tests and blood work done, having the consent form signed, and answering patients’ questions. On the cardiology service, matters were not quite so sim- ple. Most treatment was a complex, often ambiguous process of managing an acute crisis and trying to prevent its recurrence. In the first place, it was not always possible to come up with a definitive diagnosis, given the symptomatology that the patient presented upon admission. As a result, a process began in which various diagnoses were considered and then either ruled out or ruled in. This involved the use of a large number of diagnostic measures including electrocardiograms (EKGs), chest x-rays, blood tests to determine the presence of cardiac enzymes, and fluid and electrolyte levels. This laboratory work- up was complemented with thorough clinical work-up, includ- ing a comprehensive physical examination and a complete background of the patients medical, family, and social history. Given the results of the tests and the clinical findings, the medical staff could sometimes begin more clearly to define the cause of the patients problem and finally to begin the phase of the process that dealt with long-term management and treat- ment. Although the diagnostic phase took at least several days, patients were, of course, receiving treatment during this period, based on the most likely cause of the cardiac impairment. The most common tentative diagnosis was “rule out myocardial infarction” (R/O MI). In such a case, even prior to a definitive diagnosis, the patients symptoms were treated with pain medi- cations, diuretics (to relieve congestion caused by excessive fluid accumulation), cardiac monitoring, complete bed rest, and constant observation. Consent and the Structure of Medical Care 337 Surgery. A more or less typical patient on the surgical ward was Mr. Y.H., a 23-year-old male with an eleventh grade educa- tion who came in for a colostomy repair. Mr. Y.H.’s surgery was scheduled for two days after his admission, the length of time it generally took to get back the results of the preoperative lab work. In morning rounds the day before the surgery was to take place, Mr. Y.H. was informed by the Senior Resident that it would have to be postponed for a day due to a blood shortage. They discussed the possibility that one of his family members would be able to donate some blood. Later that day the doctors discussed the results of his tests and concluded there should be no problem with the surgery. The next day in afternoon rounds, the chief resident told Mr. Y.H. of his plans for surgery. Dr. A: The basic plan for tomorrow is to go into the old incision and we will close the colostomy up. [He described the operation in a little more detail.] The main risk of the operation is infection. That is why we take all the pre- cautions to keep everything clean. I guess the worst that could happen is that we would have an infection and we would have to do another colostomy again. He also told him what time the surgery would be and how long the operation would last and how long it would take him to heal up again. Mr. Y.H. said nothing except, “I don’t want to watch you do this, okay?” Dr. A: If you can, there is something wrong and you'd better tell us, because you're supposed to be “out.” Later that evening Mr. Y.H.’s nurse came in to have him sign a consent form and to discuss with him preoperative and post- operative problems. Although the discussion took more than a half an hour, the topic never strayed from its focus on the surgical event. Mr. Y.H.’s surgery was successful and he spent the next week or so in the hospital while his incisional wound healed. There were no complicating issues; daily conversations dealt with resumption of diet, exercise, pain medications, and the like. The day he was to go home, he had this final conversation with the chief resident. Dr. A: Why don’t you come back and see us [in the Outpatient Clinic] on November 3. You get the same restrictions as before. I don’t want you to lift anything over 10 pounds. Basi- cally no jogging, no weight lifting, no heavy exercises. Mr. Y.H.: What about my arms? Can I lift weights? 338 Making Health Care Decisions: Appendix C Dr. A: Well, like I said, I don’t want you lifting anything over 10 pounds. You can stretch them and do some calisthenics, that sort of thing. Mr. Y.H.: Okay. Do I get the prescription? [He was re- ferring to his pain medication. Dr. A: Yeah, we’ll make sure you get that before you go. He walked out of the room as the patient thanked him for his services. In terms of the focus of his care and treatment, Mr. Y.H. was quite typical of most patients on the surgical ward. The pattern described above was repeated frequently. Staff actions, both pre- and postsurgery, were focused on ensuring its success. The patient was expected to comply with the same purpose in mind. Furthermore, it must be remembered that by the time the surgical patient arrived in the hospital, he or she had already agreed to surgery as a method of treating the problem. This contributed to the simplicity of decisionmaking and commu- nication patterns: one major problem and one major solution. Cardiology. The focus of decisionmaking and communica- tion patterns on the cardiology ward was considerably more complex than that on the surgical ward. To begin with, a sub- stantial number of the cardiology patients came to the hospital experiencing varying symptoms of cardiac distress: severe chest pain, shortness of breath, pulmonary edema or lung congestion (secondary to heart failure), and cardiac arrythmias (irregular heartbeats). Thus the entire process of assessment, diagnosis, and treatment occurred entirely after a patient's admission to the hospital. Measures to control pain discomfort and anxiety were taken (generally this involved administration of intravenous morphine), chest x-rays, EKGs, and blood sampling were done, and the patient was admitted to the CCU for continuous obser- vation and cardiac monitoring. Ms. H.R., 68 years of age, was a typical cardiology patient. She was admitted to the hospital's emergency room very early one morning with complaints of chest pain and shortness of breath. She had a history of previous heart attacks, chronic renal failure, diabetes, and hypertension. (Cardiac patients fre- quently have other major diseases, which adds to the complex- ity of their management.) Her interview, first with an intern and then with the resi- dent, lasted almost an hour. They, however, asked all the ques- tions. .The communication involved in that interview was not very different from surgical admission interviews except that the physicians told Ms. H.R. nothing except “We are going to put you in the CCU, okay?” Typically the surgeons would have said something about what they thought the problem was and Consent and the Structure of Medical Care 339 what they would do about it. While the physicians probably could have told Ms. H.R. more than they did, the disclosure still could not have been very definite. About all they knew was what she might have had, and the only treatment was to watch and try to keep her from having more chest pain by adjusting her medications. Ms. H.R. was given the ambiguous diagnosis “R/O ML” On the cardiology ward the focus of treatment was to stabil- ize the patient and to “manage” the case medically. Observa- tions on a day-to-day basis enabled us to see just how this worked. Patients were constantly observed and monitored, and daily blood tests were done. The results of these blood tests showed the levels of cardiac enzymes (indicative of the stages of the infarct), as well as whether other elements of the patient’s blood chemistry were in balance. It was the responsibility of the house staff to maintain these within “normal limits.” To do so required a certain amount of juggling of what the interns and the residents called the “numbers” —i.e., keeping the elec- trolyte levels and other elements of the blood chemistry in a balanced state. This was a rather tricky business, in part be- cause the addition of medications that affected cardiac func- tioning also influenced the balance within the body’s circula- tory system. Patients, as a rule, had little idea about any of these problems. Although the primary focus of decisionmaking and dis- closure on the surgical ward was on an event and the focus on the cardiology ward was on the process of diagnosing, stabiliz- ing, and treating, there was a certain degree of overlap. On the surgical ward, this was often due to medical complications in the case of the surgical patient (e.g., a chronic case of diabetes or a postoperative blood clot) but also to the need to monitor the postoperative course. In cardiology distinct events some- times arose when the doctor felt the necessity for some sort of special diagnostic procedure (a cardiac catheterization for ex- ample). Who is responsible—chain of command. The difference in focus was not the only thing that produced different dis- closure and decisionmaking patterns on the two wards. Dif- ferences in authority structure also affected informed consent. On the surgical ward, it was clear that Dr. A as chief resident and head of the University Service had the most au- thority and was highest in the chain of command. This was communicated to the patients by his interns in morning rounds when they told patients that Dr. A would talk to them later, and by the daily ritual of afternoon rounds in which Dr. A led his entourage through the unit, stopping to chat with each patient and having his staff give him a report on the case. Patients quickly learned that Dr. A was the man to deal with; the one who would determine their fate directly or indirectly. Then 340 Making Health Care Decisions: Appendix C came his senior resident, the junior resident, the intern, and lastly the medical students. The head nurse, too, was often on rounds, and patients would see that she was part of the team and presumably fell under Dr. A’s jurisdiction. The behind-the- scene process was not very different. Dr. A had the final say on an issue, though his staff usually had some input as well. The pattern of staffing and chain of command on the car- diology unit was more complicated, owing to both the size and the structure of the service. The house staff was divided into two teams, each consisting of one resident and two interns. Upon admission, patients were assigned to an intern who was, according to hospital policy, the patient's “primary physician.” The intern was directly responsible for maintaining the patient on a day-to-day basis, making decisions about when to order lab work, getting routine diagnostic tests done, and observing for clinical signs of change in status. As a result, he or she tended to see the patient several times a day and therefore knew the case best. The residents’ task was to supervise the interns and to guide their decisionmaking. They were available to them for- mally in rounds and informally throughout the day as needed. The degree of resident participation in decisionmaking varied. Both senior residents we observed (our observations overlapped a change in the ward residents) seemed much more active in their role than did either junior resident. They routinely made rounds with their interns, asked the patients questions, and examined them. On the other hand, one of the junior residents told our observer that he had resented having had too much supervision when he was an intern and felt he should now function as a consultant. Others also had substantial input into decisions. These included the various faculty and visiting cardiologists who conducted rounds, the faculty in their roles as directors of various testing laboratories, the cardiology fellows who were receiving post-residency training, and the entire nursing staff. Furthermore, every patient had an attending physician, and although some of the attendings had been assigned to patients upon admission to the hospital and had only a formal relation- ship with the patient, in other cases the attending was also involved in the decisionmaking process. The differences in authority structure had two effects on the informed consent process. First, who was responsible for telling the patient what was going on was quite clear on the surgical ward. While other staff talked with patients, Dr. A saw this as his responsibility and always told the patients some- thing about the proposed surgery. On cardiology that responsi- bility was so vaguely located as often to be completely over- looked. Second, and equally important, the patients on the surgical floor knew with whom to talk about what they wanted. Consent and the Structure of Medical Care 341 On the cardiology service, who decided was quite ambiguous, and many patients did not know where to turn. To what degree do different structures of daily routines facilitate patient participation? The organizational structure of a typical day in the hospital appeared to have some effect on patient participation in decisionmaking. Both the surgical and the cardiology services had daily routines that were followed by staff members. The day on either service began with working rounds. More often than not, the patient had to be awakened by the doctor. These rounds were of short duration; they involved checking incisions, listening to lungs and heart, etc. This clearly was not a good time for patient participation in deci- sionmaking; they were simply too groggy. For the doctors though, it was a different story. They made decisions about the day's medication orders and the need for various tests. Gener- ally, patients were not included in this process although occa- sionally they would be informed of some testing due later in the day or when a new medication might start. However, these morning rounds did provide patients (on both services) with the opportunity to interact with their interns and residents. The rest of the morning generally did not provide the patient with much exposure to physicians, unless of course there was either an emergency or a scheduled procedure. Floor rounds in cardiology were an occasional exception, but for the most part the focus of those rounds was teaching and the pa- tient’s role was an entirely passive one. On the surgical service, another planned opportunity for patient/doctor interaction was afternoon rounds, which were routinely done after either surgery or clinic was over for the day. The chief resident made it a point to have his entire staff plus the head nurse accompany him on patient rounds. It was under- stood by both doctors and patients that this was a time for questions, for disclosure, and for (some) decisionmaking; it did, to a degree, facilitate patient participation. On the cardiology unit, house staff did not do afternoon rounds. If needed, they were generally available to patients and usually interacted with them at least once or twice during the day for one reason or another. Of course, there are exceptions to every rule. One patient, Mr. Z.H., seemed to have no luck at all seeing the house staff during rounds or during the day. As his resident said, “He's my least favorite patient.” Attending physi- cians also would come in and generally make daily rounds on their patients. However, it was difficult to predict when, and in some cases even if, this would occur. Conclusions. Decisionmaking and disclosure patterns were quite different on the two services we observed. One reason for these differences is essentially the difference between surgical treatment, which focuses on one event and can thus more easily 342 Making Health Care Decisions: Appendix C be described by the staff and understood by the patients, and medical treatment, which has an ambiguous focus on the pro- cess of adjusting medications to signs and symptoms. Other critical differences had to do with the authority structures and daily routines of the two wards. Outpatients vs. Inpatients. One of our hypotheses at the outset of this study was that there would be substantial dif- ferences in informed consent in different organizational set- tings. Perhaps the most important example of systematic dif- ferences in organizational setting affecting informed consent was the difference we observed between inpatient and outpa- tient treatment decisions. Our observations support the hypoth- esis that inpatients, removed from their everyday environment and social support, seem to lack the autonomy necessary for determining their own treatment. The fact that inpatients are often in poor physical condition and dependent on the staff for such mundane matters as eating and urinating may contribute to the lower level of participation on their part than outpatients. In order to evaluate the hypothesis that there would be substantial differences in the degree to which inpatients and outpatients participated in treatment decisionmaking, two raters were asked to rank inpatients and outpatients along three different five-point scales. The patients so evaluated were the 12 surgical outpatients (whom we observed) who made, or had made for them, some substantial decision about a treatment procedure and were not admitted as inpatients.* A group of 15 surgical inpatients was selected at random as the comparison group. Two raters independently scored these patients on a one- to-five scale for autonomy, understanding, and participation in the treatment decision.** Overall interrater’s reliability was r = .82. The results of these ratings are presented in Table 1. The results of these scores on the small group of patients did not reach statistical significance. However, the direction of all differences for both raters supports the notion that outpa- * Some of the surgical outpatients we observed received no treatment or only came for follow-up care. Cardiology patients were ruled out because the only group we observed, those getting exercise stress tests, were not making a treatment decision per se. We have no doubt that including them would have strengthened the result. Furthermore, the surgical inpatients and outpatients were dealing with the same medi- cal staff whereas the cardiology patients were not. ** All three were global ratings based on the rater’s impressions of all the data. Autonomy was scored on an overall impression of the pa- tient's sense that he or she felt capable of participating in the direction of treatment. Understanding was scored with reference to the five parts of informed consent disclosures: the purpose, nature, risks, and bene- fits of and alternatives to treatment. Participation in treatment deci- sionmaking was evaluated totally on overt behavior. Consent and the Structure of Medical Care 343 Table 1: Surgical Inpatients versus Outpatients Mean scores on 1-5 scale for patient understanding, autonomy and participation in decisionmaking Inpatients | Outpatients All Patients Understanding 2.99 3.62 3.31 Autonomy 2.41 3.24 2.83 Participation in Decisionmaking 2.19 3.13 2.66 tients were more autonomous and understood and participated more than inpatients did. These differences are illustrated by a comparison of two patients whose scores on all three dimensions were close to the mean for their group. The outpatient, Ms. FR., had a total mean score of 3.33, exactly the mean for outpatients. The inpatient, Ms. E.L., had a mean score of 2.33, slightly below the mean for inpatients. Ms. E.L. was a 62-year-old woman who had esophageal and laryngeal strictures due to lye ingestion a number of years ago. She was admitted for a possible colon bypass of the esophagus, having previously had a tracheostomy and a gastrostomy feed- ing tube created. After the endoscopic exam they told her: Well, basically what we saw was that there's a lot of scarring down there. Dr. J. . . . and I both looked at the results carefully. What I think is that I'm just not sure that we would be able to get it up there without damaging the voice box. And I know you don’t want that. So I think that maybe the best thing to do would be to just leave it pretty much the way it is. . .. So maybe tomorrow we'll put a little vaseline and gauze and see if you have a lot of trouble talking or not. The patient said nothing except to nod. All of the other conversations between Ms. E.L. and the doctors were of the same character. She answered their ques- tions as well as she could, she did what she was told, and rarely asked much about what was happening to her. This pattern of passivity and acceptance is typical of the patients on both the surgery and cardiology inpatient services. The outpatients were clearly more active than the inpa- tients in participating in making treatment decisions, as ex- emplified by Ms. F.R., a 72-year-old woman with sparse educa- tion, who came to the outpatient clinic with her daughter. She complained about a hernia, for which she had had an attempted surgical correction some time ago. She was examined by Dr. C 344 Making Health Care Decisions: Appendix C who simultaneously questioned her about her symptoms. In the process he found that she had hemorrhoids, diverticulitis, and previous gall bladder and hernia operations. Dr. A then made a lengthy explanation to the intern about the problem she was having and the difficulties involved in surgical repair, all in her presence and in relatively nontechnical terms. He then turned to the patient and told her: Dr. A: And I think we can help that significantly if we can get you to take some antacids. Now I think if the, if the Mylanta II bothers you, there's a better antacid that we can give you and one in fact that’s better for people that have heart trouble like you have, that I think would probably help you significantly. And I think we oughta try you on some of that and see what happens, and see if we can’t make this pain better. Ms. FR.: Well what about the diverticulitis there, doc- tor? Dr. A: The only thing you can do about the diver- ticulitis is you need to eat bran and stuff like that . . . There then ensued a lengthy discussion between them, with Ms. FR.’s daughter interjecting comments occasionally, about various antacids and what they could do for her and why some were preferable to others. The doctor then told her that he wanted to see her in about three weeks, to which she re- sponded: Ms. FR. Yeah. Well while you're standing there, doc- tor, will you look at that toe . . . For the next several minutes, Dr. A and Dr. C examined her toe and explained that they could do little about it. As they were leaving, the patient again raised the issue of diverticulitis. They promised her a barium enema in three weeks if she was not better and left. While there are certainly some unusual features about Ms. FR., especially her seemingly avid desire to receive some sort of treatment, the differences between her and Ms. E.L. are quite clear. While Ms. E.L. seemed to passively accept any of the decisions that Drs. A and C made for her, Ms. FR. did not. She continually proposed problems for the doctors to treat and clearly had her own ideas about what should be done. The difference between Ms. FR. and Ms. E.L. are fairly typical of the differences between outpatients and inpatients. In general the outpatients seemed to believe that they were still in control of what was going to happen. They negotiated with the surgeons as though they were expert providers of a needed Consent and the Structure of Medical Care 345 service. While both information and prestige gave the physi- cians the dominant voice in determining what treatment would take place, outpatients seemed to expect to have a say in treat- ment. Thus, Ms. FR. seemed to want very much to have one of her many ailments attended to surgically. She actively pressed Drs. A and C to do so. While they managed to control the outcome and prevent any surgery, they did it only at the cost of having to agree to discuss the problems again in three weeks. Of course, it is possible that patients’ levels of autonomy, understanding, and participation are not the result of dif- ferences in the overall situation of being hospitalized or being an outpatient, but rather the result of the behavior of the staff in the two settings. We tried to control this by comparing patients who were treated by the same medical staff in both settings. We found that most of the specific differences seem to cut in the other direction, that is to encourage more participation and understanding by inpatients. For example, outpatients did not sign a specific treatment consent form, whereas inpatients did.* Furthermore, there is no evidence that outpatients were told more than inpatients prior to their asking questions. While this is hard to document, one way of looking at it is to consider what patients who were admitted to the hospital from the outpatient clinic were told in both places about their operation. Ms. A.S., a woman referred from a rural clinic for a possible partial thyroidectomy, received typical “outpatient” and “inpa- tient” disclosures. In the clinic she was told: Dr. A: Well, the thyroid tests have shown that you have a nodule here. Well, I can't tell you whether or not you have cancer until we've operated on it. My recommendation is that we bring you in and remove the right half of the thyroid gland. Probably we will not have to put you on thyroid medications because probably we will find that the other side is okay. Of course, if we find that there are any nodules on the other side, we will take them out also. Ms. A.S.: Can you tell me what could have caused this? Dr. A: Well, no one really knows. Now I want you to understand that you may end up having to take thyroid medications for the rest of your life if we have to take out the whole thing. But I don’t expect that we'll have to do that. The problems with the operation are that you may be a little hoarse for a week or a little * At registration both inpatients and outpatients signed a blanket consent form giving the hospital the right to treat them. 346 Making Health Care Decisions: Appendix C while, but that should go away. And you may have some problems with shortage of cal- cium after the operation if things don’t go right. You see, there’ a little gland up near the thyroid that regulates the calcium levels in your body, and sometimes that gets in- jured in the operation. [He went on to de- scribe the problems she'd have if calcium were absent. | The patient nodded affirmatively and then asked Dr. A whether there could ever be any more cancer if they took out both sides of the thyroid. He told her that would depend on the results of the test. She asked him how long she'd have to stay off work, and he told her probably a week after she got out of the hospital and how long she would probably be staying in the hospital. They then scheduled her admission. In the hospital Dr. A told her much the same thing: Dr. A: Basically, what the plan is is to go ahead tomorrow as we talked about in the office. They will take you down at 7:00 tomorrow. morning and we'll try to operate on you at 7:30. You ought to tell your family so that they'll be here, if they want to be, before 7:00. We'll make a smail collar incision and take out the right half of the thyroid and give it to a pathologist to have some tests on. (Here he described briefly the circumstances under which they might have to take out the other side of the thyroid, which the observer did not record clearly.) The risks of the operation are primarily the possibility of bleeding dur- ing the operation. We'll also have to keep a small drain in to make sure that there is no bleeding that gets stuck in there. Another thing is that the nerve that controls your voice box runs down behind the thyroid gland and sometimes it gets bruised a bit and you may be hoarse for a little while. The other thing is that the glands that regulate your calcium levels are on either side of the thyroid and sometimes they get injured; but since we are just taking out one side of the thyroid, I don’t think we’ll have any prob- lems that are serious with that. Ms. A.S.: I'm not going to be on pills or anything like that, am I? Dr. A: You may. It depends on why you have to have it out. Sometimes we have to give you pills to keep the other side from becoming like the Consent and the Structure of Medical Care 347 side we take out. Then you might have to take a pill, but it’s not a serious thing. It's only a little pill; and if you miss it once or twice, it’s not going to hurt you any. The inpatient disclosure is slightly more substantial, in- cluding the location of the incision, the risk of bleeding, and the need to insert a drain. Thus it does not seem likely that the greater autonomy, understanding, and participation in deci- sionmaking of the outpatients can be accounted for by greater disclosure in the outpatient clinic. Conclusions. Perhaps it is easiest to understand the dif- ferences in the activity levels of inpatients and outpatients in terms of the classic psychological and sociological observations about the “patient role,” which has usually meant the inpatient role. Psychologists have noted that taking on the role of an inpatient often involves a profound psychological regression. Part of that regression is the childlike dependency on authority and the expectation that one will be taken care of in important ways. Sociologists of medicine have often noted that medical and nursing care in hospitals are arranged so as to allow and even facilitate the development of regression. Passive depen- dency is the normatively expected behavior pattern. Outpatients do not participate in this role or the associated regression. Because they leave the doctor's office and go back to their ordinary everyday world, they cannot afford to enter such a regressed state, and the staff does not encourage it. The Nature of the Disorder: Acute vs. Chronic Problems. The chronicity or acuteness of a patient's illness, as well as the nature of the illness itself, plays a major role in determining the level of patient participation in treatment. Like nearly all vari- ables examined in our study, there are complex interrelation- ships between the nature of patients, their illnesses, and their physicians, which often make it hard to isolate the impact of one variable. Yet clear patterns did emerge with regard to the acute/chronic continuum. Before these patterns can be elucidated, our definitions of acute and chronic need to be clarified. For our purposes we are using the expectable duration of the illness to separate acute from chronic disease. We designate an illness acute if it can be expected to be resolved in a relatively brief time and chronic to the degree that it is expected to persist over a substantial period of time. Also, to be considered chronic, patients need to face repeatedly a fairly predictable set of problems over the course of their illnesses, so that long-standing i'lness is not by itself sufficient to be considered chronic. It is helpful to classify illness into four categories: (1) acute patient, acute disease—classic acute illness that is likely to be self-limited and of short duration, such as acute appendicitis; 348 Making Health Care Decisions: Appendix C (2) acute patient, chronic disease—new onset of what will become a chronic illness, for example, a patient whose kidneys have suddenly failed, necessitating the beginning of mainte- nance dialysis; (3) chronic patient, chronic disease—another episode of a chronic disease, for example, another bout of pain in & patient with a chronic pancreatitis; (4) chronic patient, acute disease—a patient with a chronic disease who develops a previously unencountered problem, for example, a patient with long-standing diabetes mellitus who suffers an acute myocar- dial infarction. While this classification scheme helps in looking at our data, it is not sufficient. The nature of the illness itself and its treatment requirements play an important role in the kind of patient participation we saw. A prime example of a unique patient population were those patients on renal dialysis who were more active participants than any other group of patients we observed. Why this is so will be discussed below, but it suffices to say that the nature of the treatment is a major deter- mining variable. Acute patient, acute disease. The acutely ill patient with a discrete, limited problem is the basis of the classic doctor- patient relationship and seems to be the model for informed consent theory. The assumption in acute/acute cases is that, most of the time, the patient will present a problem that he or she has never experienced before but that the doctor will quickly recognize. The doctor will then specify a treatment that the patient will follow and the problem will disappear. Acute infections often follow this course but, of course, some acute problems do not. The pattern of decisionmaking in acute/acute cases typ- ically involved a fairly passive patient who was primarily inter- ested in getting relief for acute symptoms. What the staff told the patient was definitely affected by the clarity with which staff were able to diagnose the problem and the type of treat- ment they were prescribing. At one end of this continuum was the surgical patient with a clear diagnosis who was told at least what was wrong and what would be done about it. A good illustration is the case of Mr. N.H., a 27-year-old foreign student at a local college, who came to the emergency room with severe right lower quadrant abdominal pain. Dr. B saw him and immediately considered acute appendicitis as the likely diagnosis but also found some reasons to suspect a possible kidney stone. However, after con- sulting the chief resident they decided it was appendicitis and told the patient the following: “I think you probably have an appendicitis, and I think all indications are that you have it. | think we can help you by removing it.” The surgeon then went on to describe the proposed surgery including discussion of the incision, exploration of the abdomen, and removal of the ap- Consent and the Structure of Medical Care 349 pendix. He predicted an operation of one-and-a-half to two hours and presented infection as the main postoperative com- plication. However, he did not go into alternative diagnoses with this patient. Nor did he explain what appendicitis actually is. This patient agreed to surgery and all went well. Later Mr. N.H. told us that he had been adequately in- formed. While he was rather confused about what appendicitis was all about, he remembered what he was told prior to surgery. He felt that medical decisionmaking is a joint venture between doctor and patient but that since his situation was urgent, he felt comfortable with doctors taking immediate action. These acute surgical emergencies are notable because there was relatively little debate as to what the best intervention was. In contrast, many cardiology patients evaluated for possible myocardial infarctions or angina were at the other end of the acute continuum. In many of these cases there was no clear-cut diagnosis. The doctor often had little positive and reassuring to say to the patient and thus little was said. The patient often had little or no understanding of the problem and left the decision up to the doctor. Mr. M.G. was typical of these patients. Al- though his intern was probably the most inclined to talk with patients of all those we saw, and although Mr. M.G. was himself a talkative and lively person, he was not told much about his treatment. He was told that he had had a severe heart attack and that he would have to stay in the hospital for a while. Aside from that his conversations with his doctors focused on symp- tom reports and discussions of when he could leave the CCU and the hospital. When he had been in the hospital for 10 days we interviewed him: Interviewer: What sort of treatment have they given you? What have they done for you? Mr. M.G.: What have they done for me? I don’t know all of the medications I'm taking, I would say I am on at least, hmm, six or seven different kinds of medication for the heart. Probably some blood thinners and probably also the, I just don’t have any idea. I haven't even both- ered. That is their business. It's none of mine. It is not my field of work. And however . . . put my faith and trust in them that they know what they are giving me. This sort of passivity and distance from the treatment deci- sions are typical of acute/acute patients. Chronic patients, chronic disease. Numerous chronic di- alysis patients were observed on the University Service, and they repeatedly proved to be extremely knowledgeable and very active in their treatment. Physicians spoke to them more casu- ally, like old friends, and information presented at bedside 350 Making Health Care Decisions: Appendix C rounds was often in the form of a conversation with the patient rather than the sotto voce “shop talk” between the doctors as occurred with so many other patient groups. In fact, patients often spoke the same jargon as physicians and not infrequently diagnosed their own problems. Ms. L.C. is a 28-year-old divorced woman on hemodialysis for six years. She was admitted to University Hospital with a clotted fistula. Dr. A knew her well and breezed into her room to examine her. Dr. A: Well, what happened? Patient: My blood pressure dropped. Dr. A: When'd it happen? Patient: Well, after dialysis and after I gave myself some lactulose. Dr. A: [Examining the fistula] Remind me, is this the first graft on this site? Patient: Yes. Dr. A: The question is if we should even bother to try to open this up. Patient: It was working just fine until my blood pres- sure dropped. Dr. A: Well, lets try to open it up again. We’ll make a small incision under a local and see if we can fix it. Dr. A seems to have decided to repair the graft on the basis of Ms. L.C.’s observation. Renal patients suffer a host of complications, and a rocky course is not unusual. But most of them seem knowledgeable enough to express their preferences in tough situations. Ms. Z.R., a veteran of chronic renal failure, is a good example of this. Only 27 years old, she had been on hemodialysis for four years and had already had two transplants, each of which failed. This time she was in the hospital with a probable infected graft. Dr. A explained that if they took the graft out it would take some time for the infection to clear before they could put it back in, and so they would need an alternative form of dialysis. He suggested peritoneal dialysis: Patient: I don’t want it. Dr. A: But you did have it before, didn’t you? Patient: Yes, but we took it out because I wanted to be able to swim. Dr. A: But it did work, so we can do it if we abso- lutely have to. Isn't that so? Consent and the Structure of Medical Care 351 This last statement was a genuine one. Dr. A spent several weeks waiting for the graft to heal and even then did not suggest peritoneal dialysis. Patients did not learn about their problems overnight, and sometimes it took considerable resourcefulness to obtain thor- ough information. In our interview Ms. Z.R. described her expe- rience: Interviewer: You obviously know a lot about this. Where did you learn it all? Patient: I just picked it up as I went along. Interviewer: Doctors and nurses; read anything? Patient: Oh well, I did read a lot, too. I was in the hospital in Florida, I was in there from De- cember ‘til March, and my room was like right across from my doctor's office, and I had read everything in that hospital, and so he told me, you know, here's the key to the outside, and there's the magazines and the books in there if you want to read them. And so instead of reading the magazines I just started reading his medical books. I listen to them, too. I don’t let them do anything with- out telling me why or why they want to do it, what it’s going to prove, stuff like that. I al- ways make them explain it to me. Interviewer: Do you think most patients can make that sort of decision for themselves? Patient: I don’t know. I think you should be able to. I think it’s better the more you know about it. I think too, like you should be able to read your own chart because it’s you that’s in there. Interviewer: Do you read your own chart? Patient: I do, but they don’t know I do . . . [laughter] A couple of the nurses know I do, but like if ... any of the doctors knew, they would have a heart attack. Some of the nurses take my chart off me once in a while, too. I feel should know about it. Among chronic patients with chronic disease the renal patients were unique in terms of the extent to which they were included in decisionmaking. Patients with chronic pancreatitis contrast with the renal patients. Several patients were seen on the University Service with recurrent bouts or severe abdominal pain from a chron- ically inflamed pancreas. Most of these patients were alco- 352 Making Health Care Decisions: Appendix C holics. While Dr. A treated them as best he could, privately he admitted they were undesirable characters who had brought on their own problems. Once the pancreatitis had developed there is not too much these patients can do to improve their condi- tion other than to give up drinking. Therefore, the staff's moti- vation to educate them is far less than with the renal patients. In fact, it seems that the chronic pancreatitis patients were dealt with as if they were somewhere between acute/acute and chronic dialysis patients. They were somewhat more knowl- edgeable than typical acute patients because of their chronic history. Having been hospitalized numerous times for the same problem, they accumulated a fair amount of knowledge and became quite familiar with their doctors. On the other hand, there was no concerted effort to educate these patients nor did they make much of an effort themselves. We believe that this is because of the nature of the treatment, which requires little active participation on the part of the patient. This raises the question of whether the differences are due to a patient's experi- ence with the patient role or to the demands that having a chronic disease places on people. A review of our data indicates that the latter, in combination with a serious effort by staff to teach patients, is the primary factor. This is apparent when we examine acute patients with chronic disease. Acute patient, chronic disease. Unlike the patient with appendicitis who routinely recovers from surgery and can stop being a patient, a person experiencing the new onset of a chronic disease is often confronted with a new way of life. For the hypertensive patient this may simply mean taking a pill or two every day and reducing the salt in his diet. For the patient with chronic renal failure, it means a lot more—a life of re- strictions and dependence on some sort of machine (or some- one else’s kidney). The patients with new onset of renal failure and who were beginning, or about to begin, dialysis treatments were dealt with very differently from the acute/acute patients. Information was presented to the acute/acute patients to assure enough understanding to obtain consent and prevent “problems.” The acute/chronic renal patients were inundated with information in what seemed like a genuine attempt to inform them about what they were getting themselves into. The reasons for this immense emphasis on information transfer is certainly multidetermined. We believe the main rea- son for this information flow to renal patients is the nature of the treatment. Patients on dialysis need to understand their disease and treatment because they need to cooperate much more actively than other patients. Not only are demands quan- titatively greater (e.g. fluid and dietary restrictions) but also qualitatively greater. Renal patients at University Hospital are encouraged to learn to run the dialysis machines themselves Consent and the Structure of Medical Care 353 with the hope that many of them will eventually dialyze at home. To dialyze oneself requires a great deal of knowledge. Family members are also frequently included because they serve as backups for home treatment. In addition, the information transfer may reflect some real opportunity for choice. New kidney patients at University Hos- pital were repeatedly told about, and encouraged to learn about, hemodialysis and peritoneal dialysis and then to choose the one they preferred. They were told that there was no belief among the doctors that one treatment form was better than the other. With renal patients, an interdisciplinary team descended on the new patient and family with information. Mr. S.V., a 54- year-old man with long-standing diabetes and about to start dialysis treatment, explained what happened to him as he was given a tour of the hemodialysis unit: “Between the social worker and the head nurse, who gave the tour, we had a very comprehensive understanding exactly of what kidney disease is and what is to be done, and how the machines function. We talked to people on the machines, and I think we were given a very good insight as to ...a very disturbing and upsetting situation . . .” In observing the manner in which these patients were dealt with compared to the more “cut and dried” acute cases, one gets the notion that these patients have been elected to a very select club. Suddenly they are treated more like equals in the treatment and decisionmaking processes. Like a group of first- year medical students they are deluged with information that they are not yet able to totally make sense of and are asked to make disturbing decisions. As Mr. S.V.’s wife said at the out- set,”We're going to have to learn a lot about this stuff.” Chronic patient, acute disease. Our thesis implies that it is the demands of the illness more than the patient's length of experience as a patient that determines to a large extent how active a participatory role an individual plays in his or her treatment decisions. If this is the case, we would expect a change in the way a chronic patient is dealt with in the face of an acute, unexpected, previously unencountered problem. This is precisely what we saw. The patients lost the active partici- pant role and were treated as if they were acute/acute patients. Often this was explained away by the doctors as a result of time pressures in an emergency. Yet these were not always genuine emergencies. This pattern was particularly clear among renal patients hospitalized for problems unrelated to their renal disorder. In all of these cases the patients were treated like, and behaved like, any other acute patient. It was as though their chronic disorder was treated in a special type of way that did not generalize to other types of medical care. 354 Making Health Care Decisions: Appendix C Conclusions. Considering that nearly all patients fall some- where on an acute/chronic continuum, it is extremely dan- gerous to oversimplify and overgeneralize from our data. No two patients, doctors, or illnesses are alike, and the many rele- vant variables are so complexly interconnected that to isolate only one dimension, that of acute vs. chronic, is difficult. Yet despite these caveats, we have gone ahead and tried to look at how patient participation and decisionmaking varied according to the nature of the illness. We conclude that the nature of the demands that the illness places on a patient is an extremely important independent variable affecting the nature of disclosure, patient participation, and decisionmaking. The dialysis patients are the prototype, although not the only active chronic patients. For numerous reasons, at Univer- sity Hospital, patients were rarely dialyzed passively. Training for home dialysis was strongly encouraged as was self-care in the hospital-run dialysis unit. These patients needed to under- stand the machines, the anatomy of their shunts or catheters, and the meaning of their phosphate binders, of their dry body weight, etc. This knowledge was needed not so they could.be active decisionmakers per se, but so that they could be active treaters. Therefore, when an ‘acute emergency occurred unre- lated to the dialysis, everyone behaved again like the patient had little need to learn, and a knowledgeable patient who earlier was literally teaching an intern about kidney disease was again treated like a regular patient. In other chronic diseases it is not always. as easy for either the doctor or the patient to discern the utility of the patient's participation in his treatment. Likewise most acute patients were treated as though they had little role in the decisions and even less in implementing their treatment. While other factors play important roles in determining the level of patient par- ticipation, the acute-chronic dimension is clearly an important one. Managing Emergency Care. In medicine, an emergency has traditionally been thought of as having three aspects: (1) the problem is unplanned and unscheduled; (2) it demands imme- diate action; and (3) the patient's life or health is at stake. The law has traditionally recognized that an “emergency” is a special case allowing a doctor to render treatment without the patient's informed consent. However, if informed consent is to be suspended in an emergency, it should be because the time it would take to make the disclosure and obtain the patient's decision would work to the disadvantage of some compelling interest of the patient (usually the patient's health). University Hospital's Policy and Procedure Manual seemed to follow the legal definition closely, noting that an Consent and the Structure of Medical Care 355 “emergency” . .. prevents consulting the patient for consent, and no one acting on the patient’s behalf (fam- ily, parent, legal guardian, the court) can be contacted to give consent without creating a delay which would jeop- ardize that patient’s life or health. However, practice is far more complicated than the defini- tion derived either from law or from medicine suggests. Two cases lying at opposite ends of the definitional spectrum of an emergency illustrate this fact. In the first, a patient who urgently needed treatment but who was conscious was treated pretty much as an object: little was told to him even though the process of treatment was lengthy and consent was never sought. In the second case, despite the fact that the patient was near death, zealous efforts were made to obtain her consent to treat- ment at a time when she was barely capable of communicating. Mr. E.H., a 23-year-old man, was sent to the University Hospital emergency room from his hometown hospital where he had been medicated for pain, x-rayed, and diagnosed as having an accidental, self-inflicted stab wound in the chest. Dr. B, the senior resident, was busy when he received a call to go to the emergency room (ER). He was told, however, that it was not “immediately serious” and so 20 minutes or more passed before he went to check on Mr. E.H. Dr. B examined and interviewed Mr. E.H. briefly. It was apparent that he was in substantial pain and that he had diffi- culty breathing and speaking. Within 10 minutes the ER resi- dent and Dr. B decided that the patient had to have a chest tube put in because the x-rays showed his chest cavity was filling with blood. They did not tell the patient that his condition was serious, but they did tell him that they wanted to put the chest tube in. As the procedure went on, they warned him of each step as it was about to occur, but they continually understated how much pain he would feel. Since an inexperienced intern was putting in the tube, the process took about five minutes. In the meantime, though, Dr. B had become concerned about the amount of blood loss and ordered replacement by transfusion. Another concern was the source of the bleeding. If it was from the intercostal artery it would stop soon, and Mr. E.H. could be admitted to the University Service for postoperative care. If, however, it was from the mammary artery they might have to open up the thoracic cavity and tie off the artery, a procedure that would ordinarily be done by another surgical service. The patient was also told nothing about this. Dr. B decided to seek advice, and within about 20 minutes Dr. A and a thoracic surgeon came to the ER. They reviewed the chest x-rays, assessed the bleeding, and decided that Mr. E.H. could be admitted to the University Service. Only after exten- sive further discussion did one of the physicians finally tell Mr. EH. 356 Making Health Care Decisions: Appendix C I think we can just keep an eye on you. What I think happened and often happens in cases like this is that you hit the intercostal artery, and you're bleeding a lot, but that should stop. I think we're probably best off just keeping an eye on you. If you start to bleed again, we'll go in and look; but in the meantime, I think we're better off just watching. A radically different example of how an emergency can be managed was the case of Ms. O.S., a 78-year-old woman who had been transferred to the CCU because of cardiac complica- tions after her orthopedic surgery. She was being given a breathing treatment when she suddenly suffered a cardiac ar- rest. Dr. D, the senior resident, was in the CCU at the time, along with Dr. N, a faculty member. Ms. O.S. was unconscious and her cardiac monitor showed a straight line. The hospital's cardiac arrest team was called and Dr. D immediately began car- diopulmonary resuscitation. Within minutes the CCU and Ms. 0O.S.’s room were swarming with people. Dr. D continued to do compressions on the patient's chest. Other treatments were tried as well: oxygen was given, various cardiac drugs admin- istered, EKGs run continuously, blood samples drawn to deter- mine whether the patient was getting sufficient oxygen, IVs started to assure her of adequate fluids and as a means of giving her intravenous medications. The atmosphere was tense and hectic. About 3 to 5 minutes passed before Dr. T began to use the defibrillator to electrically stimulate her heart and he even- tually succeeded. Ten minutes after the crisis had begun only 10 or 12 people remained. Periodically someone shouted in Ms. 0O.S.’s ear to try to get her to respond. Finally, after an additional 15 to 20 minutes of work, she uttered a grunt. Outside her room there was some discussion among faculty and house staff members in which it was decided that she would need a temporary cardiac pacemaker immediately so as to avoid a repetition of the incident, and within minutes Dr. S (a cardiac fellow from the catheterization laboratory) was by Ms. 0.S.’s bedside. Up to this point, the patient had been completely unable to participate. However, Dr. S behaved differently toward Ms. O.S. He explained to the others that he needed Ms. O.S.’s permission to insert the temporary pacer. This caused some expressions of surprise since she was still far from lucid. However, Dr. S had been trained to get informed consent from all his patients by his superior, Dr. O. He bent over her with a consent form in his hand trying to get her to understand what he was saying: Dr. S: [In a very loud voice] Ms. O.S., we need your permission to do this procedure, to put a pacer in to help your heart. Can you hear me? Can you understand me? [The patient nod- ded slightly as he waved the form in front of Consent and the Structure of Medical Care 357 her face.] Can you sign this? [She was still pretty groggy, and her hands were lying limply by her side.] [To others in the room] We've got to get her consent! Intern: Well, what about getting a “verbal” if we have a couple of witnesses? [They looked around helplessly at one another. Dr. S apparently decided that he’d have to go along with this and started to tell her about the procedure.] Dr. S: [In a loud voice] I want you to understand what we're going to do. I need your consent to do it. We're going to put in a pacer to help your heart. I'm going to have to numb your leg and cut into a vein and put a wire up into your heart and put in a temporary pacer. [He leaned further over her face] Mrs. O.S. what am I going to do? Do you understand? Ms. O.S.: [Very groggily] You're going to put in a pacer. [She mumbled something else which couldn’t be heard.] Dr. S: Yes, that’s right. [He seemed somewhat re- lieved.] Now Ms. O.S., there's a slight risk in doing it. It might irritate your heart, and what happened to you a little bit ago could hap- pen again. And we might have to shock you to bring your heart back to make it work again. But we feel we need to put this pacer in to help your heart work better so that this doesn’t happen again. Ms. O.S.: [inaudible monosyllable] As he began the procedure Ms. O.S. began to wake up. She complained throughout the entire procedure, continually urg- ing him to hurry as she was quite uncomfortable. The pro- cedure continued for about an hour due to various difficulties with the equipment. All the while she continued to complain and he continued to urge her to be patient. In the end the temporary pacemaker was installed without further incident. By any definition, the early part of the above episode con- stituted an emergency.* It was life-threatening and required “immediate” medical intervention. In obtaining consent to in- sert the pacemaker, Dr. S was perhaps legally correct that the matter was no longer an emergency. But, although the medical results were good, obtaining consent might have compromised the quality of care. * Even if no longer an emergency, the doctor would have been excused from obtaining Ms. O.S.’s consent if he determined that she was incom- petent, which under the circumstances was probably the case. 358 Making Health Care Decisions: Appendix C Which, if either, of the two cases was handled correctly from a legal, medical, or ethical standpoint is a matter that could be debated endlessly. What is interesting about these two cases is that Mr. E.H. was competent, and it is unlikely that his health would have been risked by taking the time to talk with him about what was at issue. On the other hand, Ms. O.S. was marginally competent, and a second heart stoppage was possi- ble at any moment. Yet Ms. O.S.’s doctor tried to get her consent and Mr. EH.’s did not. The pattern of dealing with these two cases reflects more the particular training and attitudes of the physicians than any fine distinctions about what constitutes an emergency. Perhaps since compliance with informed consent law is so spotty anyway it is unreasonable to expect that physi- cians carefully distinguish when the emergency exception ap- plies. Limited Capacity Patients. The doctrine of informed con- sent presumes that a patient has the capacity to understand information provided by a doctor, to the extent that a “reason- able person” would understand it. If this capacity is lacking, the patient is said to be incompetent and therefore unable to give valid informed consent. It is assumed that a competent person who is provided adequate information and under no duress will understand the information presented. We have found that generally physicians assume that pa- tients are competent and treat them accordingly. When there is a question of competency, the doctors make a “clinical” deter- mination. Incompetency was an issue in a substantial number of different types of situations. The comatose or stuporous patient. The comatose patient is easy for health care providers to deal with. There is no doubt that a person in a coma is incompetent. In the cases we ob- served, major decisionmaking was turned over to the closest relative. The case of Mr. S.S. is illustrative. He was an 89-year- old married man admitted to the hospital in a coma, with a diagnosis of ischemic bowel. Immediate surgery was recom- mended. His wife was consulted but at that point refused to grant consent for the surgery saying that her husband was terminally ill and should be left to die in peace. The surgeons honored her decision. Much to everyone's surprise, Mr. S.S. showed some gradual improvement. He awoke from his coma, and he and his wife were able to talk some about the situation. This led his wife to request that the surgeons proceed with the operation. However, Dr. A, who had been hesitant to perform the emergency opera- tion earlier, was even more hesitant to perform a risky operation now that Mr. S. S. was improving. However, Mr. S.S. never regained his full faculties. He was disoriented—believing he was at home—and barely able to speak. After examining the patient Dr. A told his wife, “ . . . I think that he really has only a Consent and the Structure of Medical Care 359 50% or less chance of surviving the operation, so I don’t want to operate unless I have to.” He described a number of complica- tions that might occur and that would lead him to operate immediately. She seemed to accept this without question and only requested notification if they did decide to operate. While Dr. A obviously considered Mr. S.S. incapable of participating in a major decision, he did try to talk to him and tell him that he didn’t “think we're going to operate. At least not right now.” With a comatose patient, the assumption was made that the next of kin was the proper party to provide proxy consent. Just as with a competent patient, the role of the next of kin was to approve or disapprove of a recommended treatment, not to choose among alternative treatments. Transient incapacity—the delirious patient. It is not un- common for severely ill patients transiently to lose full control over their faculties. Patients routinely receive central nervous system depressants, pain relievers, sleeping pills, or sedatives, which compromise mental functioning. Patients may also suf- fer various insults to their central nervous system as part of the disease process itself. Thus, a previously competent person may be found to be temporarily limited in his or her capacity to participate in decisionmaking. One such patient was Mr. K.K., a 61-year-old married man who had a cardiac arrest on the street near the hospital. For nearly a week he experienced an anoxia-induced delirium with periods of confused agitation in which he tried to get out of bed and remove his IV and his oxygen. At times he thought he was home, at the airport, or in a restaurant, and that the medical staff in their white uniforms were waiters and waitresses. The attending cardiologist repeatedly tried to reorient him and reas- sure him and his wife that the confusion was not uncommon under the circumstances. Meanwhile, he was physically re- strained and the staff discussed several ways of helping to diminish his confusion. His wife's cooperation was enlisted, although generally the doctors made the decisions as they would have for competent patients in the CCU. However, even as his delirium seemed to be resolving, Mr. K.K. continued to pull the oxygen mask off. It was difficult to assess whether he did it because of continued confusion or because of his increasing denial of his illness. Eleven days after admission, the medical staff began trying to explain his condi- tion and recommended a cardiac catheterization. While Mr. KK. insisted that “I understand, I understand, doctor,” it is not clear whether he, in fact, was capable of understanding. Four days later he refused the catheterization and would not con- sider surgery. Pointing out that he risked another cardiac arrest at any time, his attending physician and his wife both tried to talk him into having the test, but the staff admitted that they could not stop him from leaving. 360 Making Health Care Decisions: Appendix C We interviewed Mr. K.K. the day he signed out of the hospital. At that time there was no evidence of a residual confusional syndrome. He knew who and where he was and precisely what he was doing. However he did not acknowledge that he had a serious heart problem. His assessment of the problem was that: Patient: ... I was comin’ up the top of the hill and were goin pretty fast. and I just . . . blacked out a lot. And I didn’t remember what hap- pened . . . Interviewer: ... Do you know what caused that? What happened medically to you? Patient: Uh, seemed like I was just pushing my- self . .. Interviewer: ... Do you know what happened to your body? What was the problem that made you black out like that? Patient: .. . They really don’t know themselves, ex- actly. They took tests and all . . . but I feel strong now. So that’s good. No trouble at all. Interviewer: Did your problem have anything to do with your heart at all? Patient: A little bit, uh-huh. Just pushed myself too much. And I was smokin’. Give that up. Fight that right off. Mr. K.K. did not understand that he had severe aortic stenosis and was at repeated risk for life-threatening arrhythmia, but whether his lack of understanding might have been related to the anoxia is not clear. Mr. K.K. was mentally clear and his cognitive functions intact. Even if he had been incapable of understanding the events of the previous two weeks at the time they occurred, there was no evidence of this incapacity at the time he signed out against medical advice. The problem seems to have been primarily emotional, not cognitive. For whatever reasons, he insistently denied the nature of his illness. In general, medical- practice assessment of capacity does not usually take psycho- logical factors into account. In this case, the medical staff was aware of, and struggled with, his early incapacity and later intransigence, but often transient incapacity is totally missed or conveniently ignored. A brief case will illustrate this point. Mr. M.G. was a 66-year-old man who was treated in the CCU for a myocardial infarction. He was doing well and was transferred out of the unit to a regular room when he experienced severe chest pain and what turned out to be a second MI. During the acute episode Mr. M.G. was alert and quite helpful to the doctors in assessing his condition. Consent and the Structure of Medical Care 361 Because he was in severe pain, he was given a considerable amount of morphine. Three days later he was recovering in the CCU and during rounds indicated a complete loss of memory for the events prior to the second MI. He forgot even being transferred from the CCU back to the floor and back again to the CCU. This could have been a result of the stressful events and/or the morphine. There was no time when the patient could have clearly been said to be incompetent, but no one seems to have noticed the cognitive difficulty at the time. Physical and emotional disabilities. Some physical diffi- culties cause special problems for patients participating in medical decisionmaking. For example, deaf patients may have trouble listening to physicians and blind patients are not able to read consent forms. However, our observation of two blind patients showed other difficulties. Both of these patients were young men with severe visual complications of diabetes. In both cases the patients’ mothers played major roles in the treat- ment decisions and the physicians often ended up talking more with the mothers than with the patients. While such involve- ment may be atypical, the blindness striking these young men in late adolescence seems to have produced a renewed emo- tional dependence on their mothers to make decisions for them.* Only one patient in our study was said to be mentally retarded. Like so many cases that we observed of possible limited capacity, the case of Ms. R.F. was notable for the pres- ence of involved family members who served as proxies with the tacit approval of her physicians. In fact, when our observers asked Ms. R.F. to sign a consent form so they could interview her, she told them to discuss it with her daughter. Everyone involved, including Ms. R.F. seemed to feel that she was not capable of making decisions about her care. Her daughter was given that role with no questions asked. On rounds, little was said beyond words of encouragement and a quick “how are you feeling?” Two patients had long-standing significant psychiatric ill- ness, one chronic schizophrenia and the other bipolar affective disorder. Neither exhibited acute psychiatric symptoms at the time they were seen for their physical problems, and these patients were not treated differently because of their history of mental illness. On the other hand, neither patient seem to have understood the treatment as well as the average patient. But despite these difficulties, we have no reason to believe they lacked substantial capacity to understand. Age. As with psychiatric patients, there may at times be an assumption by providers that an elderly person is incompetent * These cases are discussed later in the section “The Role of the Family.” 362 Making Health Care Decisions: Appendix C on the basis of age alone. We saw several passive, older patients who seemed genuinely to prefer to leave all decisions to the doctor. But this seemed more to reflect a traditional doctor- patient relationship than a sense of particular personal incom- petence. Ms. Z.S. was a good example. Admitted for shortness of breath, she was 82 years old with long-standing heart disease. She was pleasant and cooperative with the doctors and had a good sense of humor. She showed a fair amount of knowledge about her problems, but she indicated that she relied on her private physician: Interviewer: Do you have any idea why you were retaining the water? Patient: No, I don’t. I'll tell ya, I have so much confi- dence in my doctor that I never ask him any questions. Whatever he tells me to do [laughter] I do. Interviewer: Who's your doctor? Patient: Dr. S. ...Ican’t do without him [chuckle]. Interviewer: You've been going to him for a while, huh? Patient: © Ohhh, must be about 20 years or so. My husband went to him before: my son-in-law also goes to him. So... we like him very much. Interviewer: So you get quite a bit of information then, about your condition, or what's going to hap- pen to you from Dr. S...? Patient: Well, if, whether he wants me to know, of course he will tell me. And what he thinks I shouldn't know, and I guess what I shouldn't know, he [laughs] doesn’t tell me. Conclusions. Physicians did not expend much effort trying to gauge the competence of patients to participate in treatment decisions. In certain cases patients were assumed to be incom- petent, but staff members never overtly applied any formal test of competency. Instead, ad hoc decisions on the basis of unar- ticulated criteria were the norm. This was an easy assumption to make when a patient was comatose or when a patient de- clared, for example, that her daughter made all her decisions for her. In practice when incompetence was assumed, a family member was automatically enlisted to make a proxy decision. This seemed appropriate to all concerned, and a legal ruling was never considered or pursued. We did not have the oppor- tunity to see what might happen when the family strongly objected to the doctor's suggestions or when no family member was present. Consent and the Structure of Medical Care 363 The most difficult cases we observed involved patients whose capacity was transiently affected. At times the inca- pacity was completely missed or ignored because the patient's temporary delirium was too subtle to be noticed. The presence of physical handicaps seemed to affect decisions primarily through the emotional dependence that they sometimes pro- duce. We did not have an opportunity, however, to see how a written consent form would have been handled with a blind patient. Neither patients with psychiatric histories nor the aged were automatically deemed incompetent, nor were there marked differences from other patients in their capacities for decisionmaking. The Role of the Family. The law makes little provision for the involvement of the family members in treatment decision- making. Only with limited-capacity patients, and only in a tentative fashion, does the law recognize that the next of kin may be involved in decisionmaking. The actual role of family members is broader, although not systematic, and the law’s vision of that involvement is quite different from that of the participants. While the role envisioned by law sees the family as only involved in special circumstances, the participants viewed family members as having the right to be told about the disease and treatment and the right to have their opinions considered about the proper treatment. Whether or not family members are involved with a pa- tient’s decisionmaking depends on a number of factors includ- ing the relationship between family member and patient. How- ever, whether or not they are involved with the care-giver is largely serendipitous. Family members became involved be- cause they happened to be with the patient at a time when information needed to be given to or obtained from the patient, when the patient wanted advice, or when a decision about treatment was needed. At a conceptual level, family involve- ment was instigated either by the patient, by the family, or by the staff. In practice, of course, these categories tended to over- lap. Patient-initiated family involvement. Patients often turn to others for information and advice, and family members were a major source of advice and counsel. Two types of family members can be distinguished, those in the health professions and “lay” family members. “Lay” family members. A patient on the cardiology service, Mr. S.R., a 43-year-old married public school teacher, agreed to have a cardiac catheterization on the recommendation of his doctors. Much of what he knew came from his brother, who had previously had a cardiac catheterization and a heart transplant. He explained that “the only reason I think I know more than maybe the average person is from my brother’ experience.” He talked at great length with his brother about his problem. He 364 Making Health Care Decisions: Appendix C decided to have the procedure, he told us, because “I have confidence in the people here, only because I've seen what they've done [for my brother).” Mr. YM., a 38-year-old unemployed laborer, was seen in the outpatient surgical clinic for bleeding hemmorhoids. While claiming to have postponed surgery for three years for financial reasons, it was also clear that he was somewhat frightened of the prospect of surgery. For information about his condition, he had turned to his mother, who “has medical books.” Patients also turned to family and friends not merely for “information,” but for “advice.” (Though the distinction be- tween information and advice is conceptually simple, it tends to blur quite a bit in practice.) An interesting example was Mr. H.Z., who some years earlier had had severe gastrointestinal problems for which some of his intestine was removed and an ileostomy performed. The surgeons were now suggesting that they reconnect his intestines and eliminate the ileostomy. He decided against this operation for fear that the condition that had required the ileostomy would recur. In making his deci- sion, he had talked to his wife and his parents, and reported that “everybody feels the same way as what I just said, that I could be getting into more problems” by having the surgery performed. Health care professionals in the family. When patients had family members who were health care providers, they almost inevitably sought out their advice. Mr. R.W., a retired auto salesman, was seen in the outpatient cardiology service. He had two sons, both of whom were physicians. Before deciding whether to undergo a stress test or a cardiac catheterization for diagnostic purposes, he talked to both of his sons. Their advice that the catheterization was unnecessarily risky played some role in his decision to have the stress test. When patients had family members in the health care pro- fessions who got involved in the decisionmaking process, it was ordinarily because their involvement was actively sought out by the patient, rather than just because they happened to be present. On one occasion, however, a health professional family member did happen to be present. When Mr. N.J. was brought to the emergency room by his son and daughter-in-law, one of the emergency room nurses happened to be the daughter-in-law’s sister. Explaining that she was related, she came over and asked the doctors what was happening, and acted as something of an intermediary in informing the family. She was able to obtain far more information from the doctors than the family was given, in part because she asked and probably in part because of her status. The family and dependency. There were also two rather extreme instances of family involvement initiated by patients. Consent and the Structure of Medical Care 365 These two cases are extreme examples because the two patients had suddenly gone, on account of blindness, from being rather healthy young people, fully able to direct their own lives, to being highly dependent upon others for almost all aspects of their well-being. Both Mr. G.T. and Mr. E.K. were single men in their twenties who had become blind, both as a result of di- abetes and subsequent kidney failure. With Mr. G.T., the doctors had become accustomed to dealing largely with his mother rather than with him, probably because of his passivity and her activity. However, on one occasion in which the doctor talked directly to Mr. G.T. because his mother was not there, the doctor wanted to know if Mr. G.T. had decided to stay in the hospital longer or to leave: Doctor: What have you decided? Patient: I think we decided that we are going home. That seems to be what my mother wants. | have to come back next Monday anyway so we can decide then when to do it. [Emphasis added.] Doctor: Well, you let us know what you want to do. Patient: Okay. For a college graduate, he seemed to have retained little infor- mation about his problem and the two kinds of treatment avail- able to it, though his mother was far better informed. Family-initiated involvement. Sometimes the initiative for family involvement arises from the family members them- selves. This sometimes occurs at the time of admission; it some- times occurs to protect patients from information the family considers harmful to the patient; and it often involves giving advice to patients.* Accompanying patient at admission. Sometimes the pres- ence of a family member was compelled by the nature of the patient's illness, particularly when decisions were made at the time of admission to the hospital. This was especially true with seriously ill patients who were not able to get to the hospital themselves. Mr. N.J. had metastatic cancer and angina, both of which caused him great pain. He was not able to completely care for himself, and his family was involved not only in his medical care but also in providing the other necessities of life. Thus he was accompanied by his son and daughter-in-law both during the admission process and during his transfer to the cardiology service. Although the family was not at all involved in deciding that he was to be admitted to the CCU, they made their views about treatment known to the staff. They told the * Family members may also be involved in decisionmaking when patients’ mental capacities are impaired. This problem is discussed in the section “The Role of Nursing,” and will not be reviewed here. 366 Making Health Care Decisions: Appendix C resident that they did not want any major anti-coronary pro- cedures done, something the staff accepted without question. In other words, they directed the staff to let Mr. N.A. die in a relatively painless way if the opportunity arose. Protecting patients from “harmful” information. In some cases, families became involved to shield patients from infor- mation about their illnesses that they felt would do harm. The adult daughters of Mr. Y.H., a cardiology patient who had had a myocardial infarction, told one of our observers that they did not want him included in our study because they did not want him to find out that he had had a heart attack and how bad his condition was. They claimed that one of the interns had pre- viously told the patient something about his condition that had upset him, and his family was now trying to see to it that he was not upset again. There was no reason, except for the relatives’ statement, to believe that he was upset by what the intern had told him; in our interview with him (which occurred prior to the daughter's request), he did not seem to have been upset. Giving advice to patients. The role family members played in the process did not always totally take over the patient’s role. Often family members merely urged patients to obtain further information. Ms. Z.Q., a woman in her late fifties who had had renal failure for six years, said that she had been sick for a long time not knowing what was wrong, but continually seeing a doctor. After the doctor hospitalized her for two weeks, one of which was in the ICU, she related, “my family told me that they were sick of it; they wanted to send me somewhere else. So they sent me to [University Hospital], and the next day they found my problem. Polycystic kidneys.” Likewise Ms. N.]J., a 23-year- old woman with abdominal pain, who turned out to have gall- stones, reported that she was urged by her mother to obtain opinions from different doctors before agreeing to surgery. Staff-initiated family involvement. On many occasions doctors involved family members in one way or another in providing care to patients. They seemed to feel that it was part of proper medical care to keep the family well-informed and they often did so, especially if members of the family were physically present, and occasionally they went out of their way to telephone the family. However, aside from simply keeping families up to date on the patient's condition, there were several other reasons why staff seemed to communicate with patients’ families: to get informed consent, to obtain permission to treat, and to assist in obtaining patient compliance. Information for decisionmaking. Occasionally, informing for decisionmaking seemed to be the motive for giving informa- tion to the family. One case that bore some resemblance to proxy informed consent involved Mr. A. K., a 65-year-old man who was brought to the emergency room by his wife because he was nauseated and vomiting, and they thought he was having a Consent and the Structure of Medical Care 367 heart attack. It turned out that he had had a stroke in the brain stem, and as a result had lost his ability to swallow. Thus, the doctors were faced with the problem of how to assure him adequate nourishment. They had three options open to them, but clearly preferred performing a gastrostomy and inserting a feeding tube to use for intravenous or nasogastric nourish- ment. Mr. A.K.’s son told us that the doctors talked to him and his mother and told them about the three options, and they chose the gastrostomy because it was the “best.” After the operation, the doctors also explained to them how they would test out the feeding tube to see if it was working properly. Another case in which it appeared that the staff might have been informing for decisionmaking involved the doctor's dis- cussion with Mrs. S.S. about surgery on her comatose husband for his ischemic bowel, discussed more fully previously. How- ever, aside from these two cases, we observed no cases directly where informed consent seems to have been the motive for talking with the family. Obtaining permission. More often than obtaining informed consent from families, staff sought to obtain permission from family members in much the same way that permission was ordinarily obtained from patients (that is, without providing substantial information unless family members asked for it). Seeking permission from family members always occurred where the patient was clearly incompetent or of dubious com- petency. The converse, however, was not true. That is, where patients were clearly incompetent or of dubious competency, permission was not always sought from family members. In- stead, treatment was sometimes administered without asking anyone for permission, much was the case with competent patients. Obtaining patient compliance. As we have previously dis- cussed, one of the major reasons why doctors provided infor- mation to patients was to obtain their compliance or coopera- tion with the doctors’ therapeutic recommendation. When the doctors were less than fully successful in obtaining patients’ cooperation, they sometimes enlisted the assistance of family members in persuading the patient. For example, Mr. N.K. was admitted to the CCU because of chest pain. He had probably had an MI and definitely had arrhythmia. Shortly after his admission, he announced that he was going to go home on Friday, which was far sooner than medically advisable. When the staff's efforts to convince him to stay did not appear to be working and he remained firm in his intention to leave, the attending physician instructed the intern to call the patients’ wife to enlist her assistance in getting the patient to stay in the hospital until discharge was warranted. Summary and conclusions. Law acknowledges a role for families in decisionmaking only if the patient is incompetent. 368 Making Health Care Decisions: Appendix C In such a case a family member might act as a proxy decision- maker for the patient. Clearly the staff, patients, and families of patients did not view the families’ roles in that way. Families were seen as having both more and less of a role to play than law recognizes. Neither staff nor families consistently acted as though the family should be making the decision for an incom- petent patient. Rather, as with most competent patients, the decisions were primarily left to the physician. On the other hand, families did play a substantial part in the relations be- tween physicians and patients in other ways. First, patients used family members, particularly those professionally in- volved with medical issues, as sources of information and ad- vice and even as intermediaries with the medical staff. More- over family members acted as though they had a right to provide such advice even when the patient did not request it. Second, when patients were unable fully to look after them- selves, family members often provided the necessary care. In this context they cooperated in the treatment and, to some degree, in the treatment decisions. If a patient was incompetent, a family member was typ- ically used to make decisions, but they usually did not play any greater role than the patient would have were he or she compe- tent. Interestingly, in two cases, we saw families try to reduce the flow of information to patients even though they were com- petent. Physicians typically acted as though the family had a right to know about the treatment and kept the family some- what informed if they were easily accessible to the doctor. They also, under some circumstances, used the family to intervene with uncooperative patients. Clearly patients, physicians, and family members saw the role of the family in a much more complex way than the law currently does. The Role of Nursing. We observed nurses in both surgery and cardiology as they went about their daily routines to deter- mine whether they were under less time pressure and whether this would allow them to disclose more to patients than physi- cians do. There were several distinct and predictable areas in which nurses and patients interacted. The first occurred shortly after a patient's admission to the unit, when the nurse assigned to the patient for that shift went to the patient’s room to perform a “Nursing Assessment.” This initial formal interaction be- tween patient and nurse was followed by the more casual day- to-day encounters, though certain schedules were followed too: medications were distributed at established times, dressings changed, and nursing care administered when it could be fit in. Finally, nurses and patients had an opportunity to talk in an- other somewhat structured setting known as “patient educa- tion.” Consent and the Structure of Medical Care 369 Nursing assessment. The assessment was intended to per- mit the nursing staff to collect relevant information about the patient such as the reason for coming into the hospital, past medical history, and current medications and allergies. Al- though intended to transfer information from staff to patient, it also provided patients with a chance to express their view of what the problem was, what the goal of the treatment was, and any particular personal matters that the staff needed to consider in deciding on nursing care. Medication rounds. Although intended as a time for the dispensing of medications to patients, because medication rounds require a set block of time be devoted to seeing each patient, they also provided an opportunity to talk on an infor- mal basis with patients. Discussion was often about medica- tion, but also about other aspects of the patients care. The following excerpts from medication rounds conversations with several patients illustrate how they served this purpose: Nurse: Hi. Here’s your pyridium. Ms. Z.F. [No comment.] The nurse left the room. Next came Mr. T.L.; the nurse had to give him some Digoxin, a medication for his heart. His family was sitting at his bedside. Nurse: I'm going to listen to your heart before I give you your medication. Mr. T.L.: Okay. Nurse: Okay. This will make your heart work a little better. It'll make your blood pump better. It’s called Digoxin. There were some family members sitting in the room. A woman, possibly his daughter, said “I'm sure he’ll recognize that.” Nurse: Was he taking Digoxin at home? Family Member: No, no he started that here. She sat him up and he took the pill. Nurse: He'll probably be going home tomorrow. Family Member: Oh, that’s really good. The patient had no further comment, and she left the room. In this sort of interaction the patient was usually told the name and the most general purpose of any new medication. It also provided an opportunity for giving information about other aspects of treatment, such as “he’ll probably be going home tomorrow.” 370 Making Health Care Decisions: Appendix C Occasionally, however, the administration of medications also provided an opportunity for much more substantial dis- closure on the part of the nurse and a much larger part in the decisionmaking for the patient. A good example was a nurse administering a strong diuretic to Ms. R.H. She told her: Nurse: Instead of giving you Lasix this time, which is what you've been getting, this is another diuretic. It’s called Edecrin. It’s much stronger than Lasix. It’s a one-time dosage. [She sat herself down beside the patient with a large syringe.] It's gonna take a little bit longer to give it to you in the IV, so I'm gonna be here a few minutes. There may be a little bit of burning that you might feel, and it'll probably work better than the Lasix. It'll work faster, it'll take less time to work on you. The doctors feel that you need a stronger diuretic right now. As the treatment proceeded Ms. R.H. began to experience severe burning. She complained with increasing vehemence until she finally asked the nurse to take the IV out. With only slight hesitation, the nurse agreed with the patient, and after reassur- ing the patient, called the physician to get his agreement. Here the nurse was able to play a role in implementing the patient's treatment decision and persuading the doctor to go along with it. Patient education. Another opportunity’ for nurses and patients to interact was “patient education.” Each patient on the surgical service, for example, was given a description of what to expect before surgery by the nurse on duty. The content of this teaching varied from one nurse to another as well as from one surgical procedure to another. One nurse was personally and professional strongly committed to the idea of patient educa- tion and an example of how she conducted preoperative teach- ing provides an illustration of what it is possible to do. The patient, Ms. X.P,, was going to have triple bypass heart surgery, which required more than the usual explanation of surgical procedures. In the next hour, the nurse covered the following points. (1) The anesthesiologist was going to come to explain the anesthetic. (2) She was going to be shaved and exactly where she was going to be shaved, and that the incision wouldn't cover all of the area she was going to be shaved, but the shaving was just to keep the bacteria level down. (3) She would have to take a shower twice with a special soap that they would give to her. Consent and the Structure of Medical Care 371 (4) Someone would come in that night to get her consent. (5) Her family should be there well ahead of surgery be- cause she would be taken down to the operating room a half hour early. (6) She wasn’t permitted to wear any jewelry, bobby pins, etc. (7) They would take her down on a cart, and she would go to the anesthesia area first. (8) Everyone would have on hospital gowns like she saw on television and that was normal, and she shouldn't find that upsetting. (9) They would take her into the operating room and put an oxygen mask on her and that would be the last thing she would probably remember. (10) There would be a big crowd of people all around her and that she shouldn't find that unusual or exceptional. (11) There would be different incisions made on her, where they would be, and what their different purposes were. (12) A description of the heart-lung machine. [The patient commented, “That is the scary part.”] (13) They would have to put a tube in her nose and what it was for. (14) She would have a chest tube that would drain the cavity in the chest and around the lung. [The nurse drew a diagram of the area to show her what it was.] (15) The nurses would have to come in to “strip” the chest tube to do away with any blood clots around it. She described how it was done and said that most patients thought this hurt and were uncomfortable with it but that if it were to make her uncomfortable she should ask the nurse for some more pain killer. (16) She would have a pacemaker and what its function was. (17) She would have an IV in both arms and maybe in her groin too. (18) Her legs would be wrapped in a special stocking. (19) She would have to be in the hospital between 10 days and two weeks. At this point, the nurse was called out of the room for something, and the patient said that she found this pretty scary: “It almost makes you think about changing your mind.” The nurse came back in and told her that now she was going to talk about the important things, which were the things that she was going to have to do for herself. She explained how she should 372 Making Health Care Decisions: Appendix C practice breathing and gave a demonstration of it. She showed her how to cough from her lungs and gave a demonstration. She showed her how to exercise her leg muscles to prevent blood clots. There were several other things she described along the same lines. After that she made the patient practice all the exercises, which seemed to make her feel somewhat better. They talked about the patient's fear of the operation, and the patient repeated the comment that it almost made her think about changing her mind. The nurse said that it was important. that she understand this now so that she wouldn't be scared when it happened and that she understand that this happens to all patients and that it doesn’t mean that there's anything wrong. At this point the nurse pulled out a loose-leaf book filled with pictures and cartoons. It was quite thick, and she and the patient went over it. She left the booklet with her to look over for awhile, and that also seemed to make the patient feel better. Preoperative teaching for open-heart surgery was much more complex than that for most other types of surgery, and this nurse took preoperative education more seriously than most other nurses. For the nursing staff in the cardiology ward, patient educa- tion played a much smaller role. As we discussed in “Cardiol- ogy vs. Surgery: Is Decisionmaking the Same?,” medical car- diological treatment did not involve an event that would provide as clear a focus for teaching as surgery did. However, we did observe a few examples of cardiac surgical patients who were given some preparation for an operation the night before surgery. Generally, this was in conjunction with the signing of the surgical consent form. The following is a typical example: Nurse: Well at 6:00 they'll come and get you ready for it. You'll be getting the shave and an en- ema, and I want you to be doing some show- ering tonight with this Betadine. And then, as I said at 6:00 they'll come and give you a preoperative anesthetic to make you more relaxed, not to put you out or anything but just to make you more relaxed when you are down there. Then you'll be taken down into the operating room, and they'll be putting in various pieces of equipment. You'll be get- ting an IV and that kind of stuff. And they'll let you know what they'll be doing as they go along. Any more questions? [He had already asked a few questions about the consent form itself.] Mr. X.S.: No. As is evident from this dialogue, patient education on the Consent and the Structure of Medical Care 373 cardiology surgical service generally was not as thorough as on the surgical service. However, certain medical procedures for chronic patients did provide a good opportunity for patient education on the cardiology ward. For example, one nurse discovered that Mr. N.T., who had recently been diagnosed as a diabetic, knew little about self-injection and she took her next free ten minutes to describe to him how to rotate the sites of the shots and why he should do so. While it is clear that patient education by nurses was the most substantial single source of disclosure, with the possible exception of some of the informed consent disclosures by doc- tors prior to major diagnostic tests, it is important to realize what it was not. It was not information for decisionmaking. Nor, by and large, did patients understand it that way. More impor- tant, it did not necessarily include the full information that the informed consent doctrine requires for patient decisionmaking. It emphasized heavily the nature of the procedure, but the purposes, risks (aside from some information about discom- forts), benefits, and alternatives were completely ignored. Conclusions. Few issues that we studied are as complex as the role and potential role of nurses in informed consent. Nar- rowly speaking, nurses’ roles in informed consent focus on their getting the forms signed. More broadly they are a major source of information about treatment. Moreover, talking to patients seems, in a vague way, to be more consistent with the nurses’ conception of their role than the physicians’ conception of theirs. Nonetheless, there are two major barriers to nurses taking a primary role in informed consent. First, while there is some evidence that this may be changing, most nurses are young and stay active in the profession only for a few years. Thus it may be difficult to develop fully their communicative talents or deeply inculcate a commitment to the values on which the informed consent doctrine rests. Moreover, their youth and inexperience does not help them stand up against a variety of other aspects of hospital life that undercut the patient's participation in treat- ment decisionmaking. Even more important is the fact that, in most situations, doctors, not nurses, make treatment decisions. Thus nurses cannot present information as part of a decisionmaking process without seeming to challenge doctors’ authority. The mere dis- cussion of risks that a physician might not have disclosed inevitably seems to nurse, patient, and physician as question- ing the doctor's recommendation. Nonetheless, while the nursing profession cannot be the panacea for the problems of the doctrine of informed consent, active nursing roles in patient education can be helpful in furthering some of the goals of informed consent. 374 Making Health Care Decisions: Appendix C Medical Uncertainty. It is hardly a unique observation that medicine is an enterprise filled with uncertainty. All the exist- ing medical knowledge cannot predict with complete certainty the outcome of even the most predictable diseases and treat- ments. Not only is the outcome of medical interventions uncer- tain, but many of the points along the way—diagnosis, etiology, treatment itself —are shrouded in uncertainty. Sometimes the provider of medical services has a fairly good idea of the nature and degree of uncertainty—either on the basis of his own expe- rience or on the basis of the experience of the profession gener- ally—but sometimes the only thing that is known about uncer- tainty is that it exists. The fact that uncertainty permeates medical practice—and especially the fact that the outcome of treatment is often uncer- tain—is information relevant to making a decision about treat- ment, and further it is information that patients who are in- clined to be extremely trusting of doctors may either not know or may selectively ignore. For these reasons, uncertainty argua- bly ought to be disclosed by doctors to patients. However, many physicians are uncomfortable about con- veying this uncertainty because it is widely thought that faith in the ability and knowledge of the physician plays a major part in the efficacy of a treatment. To convey to the patient the doubt and ambiguity with which the physician must live would thus undermine an important therapeutic feature of the doctor-pa- tient relationship. Even if this is not true, in the view of doctors it remains difficult to convey the uncertainty of success without making the patient feel that there is no chance of success. By and large, the medical staff we observed did not try to convey uncertainty but, if anything, tried to protect patients from knowing about it. The medical staff rarely conveyed the generalized uncertainty about the outcome of proposed treat- ment. If any uncertainty was mentioned, and this was rare, it was specific to the particular treatment. We paid particular attention to those cases where there was some disagreement among staff about what to do. We found a number of different types of uncertainty expressed. Uncertainty often manifested itself when there was some disagreement among staff about diagnosis, etiology, treatment, or outcome. Although patients were rarely aware of such disagreement, its existence was often apparent to our observers who had access to the “behind-the-scene” aspect of medical practice. Types of uncertainty. Health care providers were often uncertain about a patient’s diagnosis. Yet even when the diag- nosis was certain, the etiology of the disease sometimes was not. Furthermore, a given diagnosis does not necessarily carry with it clear implications for therapy; that is, staff may and often did disagree as to what should be done. Finally, even where there was agreement about what therapeutic intervention Consent and the Structure of Medical Care 375 should be undertaken, the outcome of the intervention was uncertain in varying degrees. Diagnosis. By far the most frequent expressions of doubt by the staff occurred when explaining to a patient that a procedure was necessary to eliminate uncertainty about the patient's diag- nosis. For example, when Ms. N.P’s x-ray showed a small growth in her intestine that might have been a tumor, her doctor told her that he did not know what it was and that they would have to examine it with a fiber optic scope. Another instance involved Ms. E.R., a woman with a lump in her breast. After examining her, the doctor said: It's hard to say what this thing is. Might be a little cyst, it might just be some fat . . . might be a tumor . . .. I ex- pect probably what we're gonna have to do is do a biopsy of this . . . as an outpatient. But what I'd like to do first is put a needle in there and see if there's any fluid in it. ‘Cause if we would put a needle in it and there’s fluid in it and it went away, then we wouldn't have to biopsy it. And we can do that today, right here in the office . . . . But if it’s a tumor, then we probably have to do some more operating. But we can’t tell until we, you know, let our pathologist take a took at it. Etiology. Etiology and diagnosis are sometimes overlap- ping and indistinguishable issues. Although doctors used their uncertainty about what was wrong with patients to encourage them to agree to a diagnostic procedure designed to remove that uncertainty, they were far less inclined to express uncertainty about the origins of a patient’s difficulty if that purpose was absent. For example, Mr. Z.S., a well-educated executive whose personal physician was one of the cardiology department fac- ulty, had what was officially diagnosed as recurrent stress- induced angina. Mr. Z.S. was extremely knowledgeable about his disease, all of the past treatments, and the options available. However, in spite of the diagnosis, at least several of the house staff and his attending physician were fairly sure that there was a very important psychological component to his disorder. While, because of the episodic nature of our observations we cannot prove that no one told Mr. Z.S. about it, we did not observe it, and he did not mention it in his interview despite being given a chance to, and the matter was continually dis- cussed among staff after they left his room. Similar to this is the case of Mr. E.H., the young man previously discussed, who came to the emergency room with an accidental self-inflicted knife wound in the chest. He was bleeding profusely into the chest cavity, and a chest tube was inserted. The character of the wound convinced all of the doc- tors who were caring for him that he was bleeding from an artery, but there was substantial disagreement as to whether it was the intercostal artery (as most suspected) or whether it 376 Making Health Care Decisions: Appendix C might be the mammary artery. In the former case a pressure dressing would stop it; in the latter case thoracic surgery would be necessary. Nothing was said to the patient to indicate that there was any uncertainty about the source of his problems. Treatment. Staff seemed to hesitate to express uncertainty to patients when they were themselves uncertain about what intervention to pursue. They told patients that they were unsure only after they had decided on a method for resolving the uncertainty and when there was no immediate danger from postponing the decision. We saw only three instances in which patients were told by their doctor that they did not know what to do. Two of them involved hemodialysis patients who had possible infected gortex implants. In one case the patient was implicitly referred to a senior doctor by the resident. In the other the doctor said that they would not be able to decide until they opened up the sutures and looked at the infection. In neither case was the patient given the impression that the staff did not know what to do, only that the decision was still in the making. Another way of approaching this problem is to look at those instances where there was disagreement among the staff about what should be done and compare this to how much was told to the patient about such disagreements. There were seven cases in which there were overt, prolonged, although not neces- sarily severe, disagreements between staff members about what to do for a patient. Only in the case of Mr. D.M. did anyone make any effort to tell the patient that the matter was not clear- cut, and in that case the physician who lost the argument about what to do told the patient only that the proposed treatment might not help him as much as he was expecting. One other patient, Mr. S.P, managed to get conflicting opinions about the meaning of the results of his catheterization from his doctors. However, when he asked the opinion of the intern, whose private opinion was that the staff cardiologist was wrong, the intern hedged by saying that he just didn’t know and that Mr. S.P. should talk to the staff cardiologist about it. Outcome. Sometimes a doctor simply said that he or she could not predict the outcome of an anticipated procedure or medication. In none of these situations was the uncertainty related to whether the patient or doctor should do something. These decisions seemed to have been made independently of the uncertain prediction. Typical is the case of Ms. AS. a middle-aged woman hospitalized to have part of her thyroid removed. She had been in the hospital for a day when, the afternoon before the operation, she asked Dr. A if she was “going to be on pills or anything like that.” His response was: You may, it depends on why you have to have it out. Sometimes we have to give you pills to keep the other Consent and the Structure of Medical Care 377 side from becoming like the side we take out. Then you might have to take a pill, but it’s not a serious thing. It’s only a little pill and if you miss it once or twice it’s not going to hurt you any. In this and the other case in which the doctor expressed some uncertainty about what the future held, the decision to proceed in a certain manner had already been made, and the uncertainty was presented in such a way as to minimize its impact on the decision. Informed consent and the disclosure of uncertainty. Med- ical and nursing staff were not particularly enthusiastic about discussing the uncertainty of their enterprise with patients except where doing so would be likely to encourage patient compliance. However, in a way, this understates the problem: for although uncertainty is built into the medical model of decisionmaking, it is not easily accommodated in the linear informed consent model. At the point at which consent is obtained, the patient, in cooperation with the doctor, is sup- posed to decide on what the treatment should be. The medical decisionmaking process, unlike the legal model, is a nonlinear continuously branching tree. At any given point there may be a large number of possible outcomes. It the patient is to be told the options and given adequate information to weigh the possi- ble choices, disclosure can become unavoidably complicated. Nonetheless, because uncertainty is a pervasive aspect of medical practice, and one that is either not generally known to patients or ignored by them, it is highly material to making an informed decision about treatment. On the other hand, there may be a “negative placebo effect” associated with the disclosure of uncertainty. Telling a patient, for example, that a medication may not correct the problem for which it has been prescribed may create a self-fulfilling proph- ecy that “causes” the medication not to work. We were not in position to study this postulate, but in light of the strong evi- dence of a positive placebo effect, the negative placebo effect merits study. The Functions of Medically Relevant Information for Pa- tients. Although many patients expressed the desire for infor- mation about treatment, they did not necessarily want it to make decisions about treatment. Disclosure of medically rele- vant information served four different functions for patients. These functions are closely associated with the extent to which patients saw themselves as having a meaningful role in their treatment decisions. Information for decisionmaking. Some patients wanted, and sometimes actively solicited, information that would facili- tate rational decisionmaking with respect to treatment. These patients viewed themselves as active participants in their medi- 378 Making Health Care Decisions: Appendix C cal care and they tended to view their relationship with treat- ment professionals as collaborative rather than as passively compliant. This group of patients will be discussed in detail in the section titled “The Perspective of the Most Active Patients.” Information for veto. Sometimes patients used informa- tion to exercise a veto over some decision that the physician had already made. These patients viewed it as perfectly appro- priate and reasonable that their physician would decide what would best facilitate recovery, and they simply wanted to be briefly informed of what procedure the physician had decided on in case this decision involved something that they strongly wanted to avoid. As one patient put it: “Just tell me what you wanna do so I know somethin’ about it, and I'll almost always go along with it because I know that you know best for me. But I guess if it were somethin’ I really, for some reason, didn’t wanna do, I guess then I'd just tell ‘em not to go ahead.” Another patient, Ms. D.D., with rheumatoid arthritis compli- cated by a severe circulatory disorder explained: “I tell ‘em when I don’t want ‘em to do it. But like I say, I can’t say anything really, because the majority of dealings I've had with the hospital have been pretty good. But when I tell ‘em no, I don’t like this blood test, I don’t go for that, they have to take it. I know that.” She saw the physician as responsible for taking the initiative and determining the appropriate treatment pro- cedures on her behalf. She expected a brief synopsis of the plans, but both she and her doctors understood that, in almost all cases, she would ratify the proposal. These patients were not altogether clear about what were legitimate grounds for rejecting a proposed course of treatment. For Ms. D.D., the main criterion for vetoing procedures pro- posed by her physicians seemed to be that she either did not understand what they proposed to do or did not see how it would benefit her. Thus, she claimed; “I ask . .. [questions] . when they get to talkin’ over my head and you know, using those big words that I don’t know, and I just say ‘break it down to my level and tell me that my leg is going to get better or it's not going to get better, and whatcha going to do about it.’ ” However, other patients seemed unable to specify when they might object or for what reasons. Like the first function of information that we described, this function of information was held largely by chronic pa- tients with long-term involvements with their illnesses. How- ever, these patients were, by and large, less well-educated and older than patients who used information for decisionmaking. Both of these functions of medically relevant information be- come clearer when we compare them with two other functions that information-serves for patients. Information for compliance. For the majority of patients, information given to them by physicians facilitated their com- Consent and the Structure of Medical Care 379 pliance with treatment decisions and thereby promoted their health. These patients had typically left responsibility for deci- sionmaking with their physicians—usually on the grounds that only physicians have the necessary specialized, technical ex- pertise and experience treating illness. They desired informa- tion primarily to obtain the most benefits from the doctor's decisions. Ms. K.H. was a 70-year-old diabetic who was seen on the surgery ward. The decision to put her on IV insulin and take her off oral medication for her diabetes was made without consulting her, and she was given no information relevant to that decision. On the other hand, she was given considerable information, which she was very interested in, about the new nutritional requirements that were necessary because of the switch to IV medication. Ms. K.H. said she was very satisfied with the amount and kind of information she had received about her treatment—despite the fact that she was informed about the treatment decisions only after they were imple- mented—but noted that she would have liked more extensive and clearer information about the new dietary requirements that the change in her treatment required so that she could comply with them exactly. She had little to say about the change in her treatment except, “Whatever reasons they had . .. I'm sure that they're doing what's best for me.” Many patients who conceived of the role of information primarily as facilitating compliance, sought and expressed the desire for more extensive information. Thus, Mr. N.F.,, a 60-year- old married man, repeatedly requested more information per- taining to how best to comply with the treatment course deter- mined by his physicians. He was being treated for a pulmonary embolism caused by having his leg in a cast, which he believed might have been prevented if he had known earlier that it was a possibility, what its warning signs were, and that he should seek a doctor's attention when they occurred. His wife also actively sought to obtain more detailed information to assist him in complying with the treatment. However, both she and Mr. N.F. expressed complete satisfaction with the way in which the treatment decisions had been made by the physician. Information as a courtesy. The final function of informa- tion given to patients by physicians—and the one least antici- pated by us—is what we refer to as “information as a courtesy.” Like the patients in the previous two groups for whom the function of information was to permit them to veto unwanted treatment or to assist them in complying with wanted treat- ment, there was another group of patients who were perfectly content to permit their physicians to make medical decisions for them. However, at least as far as we were able to learn, their reason for wanting information about the treatments that the doctors decided on was less instrumental than the other pa- 380 Making Health Care Decisions: Appendix C tients. Rather they thought it was owed them as patients to be told what was going to be done to them. A patient who nicely exemplifies those patients who viewed information as a courtesy is Mr. R.K., who was brought into the hospital in atrial flutter and with chronic lung disease. He expressed little curiosity about what was being done to him, including electrocardioconversion, which he explained as an attempt “to get my heart back into rhythm.” He assumed that whatever decisions had to be made would be made exclusively by his physicians. He was not at all bothered by this and, as the following dialogue makes clear, was not interested in making decisions: Interviewer: Can you tell me how you came about to get- ting that done, whatever it was? Mr. R.K.: You mean, gettin’ my heart put back in rhythm? Interviewer: Yeah. Mr. RK.: The doctors decided it. They had me on medication for, say from Monday noon when I was admitted until Tuesday morning. Tues- day afternoon he decided to take me down there. Interviewer: What kind of medication? Mr. R.K.: I don’t know. Whatever they shoot into you to try to get the heart back in rhythm. They said something like the medication didn’t work. Interviewer: So, well how did you find out, then, about what you had done during this other pro- cedure? After the medication didn’t work? Mr. RK.: Well, they just came in and told me that they were going to take me down somewhere and this doctor, he came in and explained it to me . .. Interviewer: Well as far as say, you know, like your treat- ment’s concerned, who do you think should be making the decisions about it? Mr. R.K.: Well, I'm a firm believer that there should always be two opinions with anything, and that’s why I like what they do here where they got 3 or 4 doctors that make the deci- sion. Interviewer: And what about yourself? Do you have any involvement in the decision? Mr. RK.: No, not particularly. Consent and the Structure of Medical Care 381 What is most interesting about Mr. R.K. and many other patients we interviewed is their conception of the function of the disclosure of information. Mr. R.K. was particularly indig- nant about the information he was not getting: Interviewer: Well, do you feel, then, that you're getting enough information about . . . ? Mr. RK: Well, I'm getting information about what'’s wrong and everything. The only thing I don’t like is you never know from one minute to the next what's coming next. Just like this morning in here, they were talkin’ about this test and that, but they don't tell you they're gonna do it. And I didn’t even know it. Then the nurse come up and she says ‘well, you're gonna’ go down there.” Of course, maybe with them it’s impossible to do that. The information that is of central interest to him and of which he feels sorely deprived concerns not the rationale for or risks associated with various tests and procedures but their schedul- ing. The information that he saw as significant was “what they're gonna do and when they're gonna do it.” Patients who conceive of information as primarily relevant to treating them with courtesy typically trust their physician’s judgments and feel that it is the professional’s responsibility to make medical decisions. Mr. R.K. was a 60-year-old retired steel worker and wid- ower whose highest level of education was the 11th grade; Mr. Y.M. was 29 years old and had a graduate degree, yet the two patients share a strikingly similar conception of the role of information and kind of information seen as relevant and desir- able. But despite many personal, demographic, and medical differences between them, they have the same complaint about the kind of information that is lacking and the same underlying conception of the function of information dislosure in medical settings. Consider the following dialogue: Interviewer: And so far as treatment decisions are con- cerned it’s . ..? Mr. YM. Anything like the blood test or x-rays or any- thing; they want them, they can get them. The only thing I'd like, well, it really ticks me off a little that they don’t tell me that they want blood work and x-rays, like when they left here . . . . I knew that they would have to do some x-rays and I pretty much figured that he would increase, you know, the medication. So it’s like, they're assuming. And it’s like, 382 Making Health Care Decisions: Appendix C the relationship is there that they can as- sume, but I don’t think that it should be done, you know, in every case. I think that they should tell, you know. It is important to note that what we are describing is one conception of the nature and role of information. This is en- tirely different from the normal array of complaints from pa- tients concerning the routine indignities, discomforts, and dis- turbances of hospitalization. What this means is that some patients believed that the kind of information most relevant to them concerned neither medical decisionmaking nor com- pliance with some aspect of a treatment regimen but informa- tion that would render their institutional environment less unpredictable—information that would enable them to intro- duce some order into their daily schedule. Most patients “griped” about the food, being awakened at inconvenient times, and the boredom of hospitalization. But for patients like Mr. RK., Mr. YM,, and three or four others on each of the services we studied, information that increased the courteous- ness with which they were treated and the orderliness and predictability of their environment surpassed all other kinds of information in significance. The Perspective of the Most Active Patients. Most of the patients we interviewed and observed talking with the staff of the hospital seemed to want information primarily to know what to expect next and otherwise seemed content to let doc- tors make decisions, and the physicians seemed eager to com- ply. Thus, if the legal requirement of informed consent is intended as “therapy for the doctor-patient relationship” * that will make patients active co-participants in treatment deci- sions, law is not achieving its goal. Alternatively, however, if the goals of informed consent are viewed less ambitiously—e.g., as providing patients with the opportunity to participate in treat- ment decisionmaking—then the goal of the law may be closer to realization. In our observations we found about 10 patients, depending on how one draws the line, who were very interested in actively involving themselves in the decisionmaking. Since these patients provide an empirical justification for the doc- trine, they are of special interest to us. As a group, the most active patients were those on renal dialysis. Their knowledge of the options for treating kidney failure and of the details and problems of both the disease and its treatment were sometimes stunning. For instance, it took several hours of interviewing Ms. L.C.—a 28-year-old woman with an almost endless list of medical problems who was com.- pleting college between hospitalizations—to capture a part of * See Note, Restructuring Informed Consent: Legal Therapy for the Doctor-Patient Relationship, 79 YALE L.J. 1533 (1970). Consent and the Structure of Medical Care 383 her knowledge about these issues. The following is a part of her discussion of one of the possibilities that the doctor had men- tioned to her earlier in the day: Interviewer: Ms. L.C.: Interviewer: Ms. L.C.: Interviewer: Ms. L.C.: Interviewer: Ms. L.C.: So what is going to happen next? He said he might put an Uldall on my neck. I don’t want an Uldall. . . . If this graft doesn’t open he might put an Uldall on my neck and into the jugular vein, and it threads into your heart. A chick just had a massive throm- bosis . . . Who's this, someone you know? Yeah, she just died. She had one of those done, and her heart burst and they all know why. She had one of the Uldalls. There wasn’t anything wrong with her ... It threads into your heart ... and it might puncture. [After discussing some of the other options] . . . I can’t do [peritoneal dialysis] because I'm in school. I'm missing too much time out anyway. And I ain't walking around with anything hanging on my neck either. What do you mean? The Uldall. . . Idon’t know if it’s a one time thing; do they take it out after dialysis? If they take it out I might . . . consider it. You didn’t say anything to Dr. A this morning about that . . . Yeah. I'll tell him! I'll tell him. If there is nothing else he can do, he can believe he'll hear from me. [laughs] As we have noted before, renal dialysis patients predomi- nate among the most knowledgeable and most active patients, but they are certainly not the only ones in this group. Although most cardiology patients had relatively poorer understanding of their treatment than even the nonrenal surgical patients, there were exceptions. For example, Mr. Z.S., who was a hospital administrator, treated the doctors simply as better informed colleagues, and they treated him as a nonexpert insider. His grasp of the issues was excellent, as illustrated by his descrip- tion of the rationale for his not undergoing a cardiac catheter- ization: Mr. Z.S.; “I had 3 catheterizations. . . . The first one was to show that the bypasses took all right. The second one [was] when I was having a lot of chest pain, and they thought the by- passes were occluding and there was some evidence that it was happening. And the 384 Making Health Care Decisions: Appendix C third . .. confirmed the fact. To do a cath now would be no advantage. I just talked to the surgeon who said there is no more they can do for me. There is a certain amount of risk in doing a cath and there is no sense taking it if . . . Interviewer: What is the risk from a cath? Mr 2.S.: You can puncture an artery, the probe can go through your heart, you can pick up a piece of plaque and have a stroke. So there’s a cer- tain amount of risk. Mr. Z.S.’s detailed knowledge was, of course, strongly af- fected by the fact that he was something of an insider. Moreover, he was a well-educated man. However, not all patients who were very active in treatment decisionmaking were well-edu- cated, insiders, or even middle-class (and vice versa). Ms. N.F., a 45-year-old high school graduate on welfare, is an interesting example of just how completely a chronic patient can come to manage her own treatment. Fifteen years earlier she had had bowel bypass surgery in order to lose weight. The surgery went badly as did subsequent efforts to correct the problem. The result was that she had periodic bowel obstructions. Her de- scription of the development of the problems and how she treated them and how she knew when to hospitalize herself took 45 minutes for her to recount. However, her description of her current admission is particularly revealing: Interviewer: What did the doctors in the Emergency Room tell you so far is going on in there? Ms. N.F.: They didn’t tell me. They . . . asked me what the problem was and I told them. And like they said to me, how did I know it was a bowel obstruction and, like I told you, if you throw up stuff that is awful smelly green and smells like shit, it probably is. ... The [doctor] I had when I first went in, . . . told me what we'll do. He’s gonna send me over for x-rays and whatever, and I said to him, put the stomach tube in first. Because that is my biggest problem. I can go over and get my x-rays done, I can stand on my head, but get the swelling out, because the pain was un- bearable. Interviewer: Did he do that? Ms. N.F.: Yeah . . .. Except when dealing with physicians who were expert on this type of condition, Ms. N.F. directed much of the treatment herself. She personified the patient who played a major role in Consent and the Structure of Medical Care 385 treatment decisions because she understood her problem in great detail and could manage much of the treatment herself. Although not all patients who participated heavily in treat- ment decisionmaking had a good understanding of treatment, most did. For example, Mr. M.R., a 54-year-old man with a serious disease of unknown diagnosis and origin, participated substantially in decisionmaking. When the surgeon told him that they wanted to remove a large mass in his intestine in spite of the fact that a “superficial biopsy” did not show any evi- dence that it was a tumor, Mr. M.R. responded: “Well, I want to talk to my regular doctors about this. I also think I've got to talk to my family and make a decision with them . . .. and, quite frankly, unless it's very important, I have to get home for a while. I have a lot of personal and business things that I've got to arrange.” Conclusions. While informed consent has not transformed the doctor-patient relationship, there are patients who actively pursue control over their treatment and want to know about it in great detail. Some of these patients seek to become coordina- tor and overseer of their treatment. While there are certain dominant patterns in who takes on this role, and under what circumstances, there are exceptions too. Most of the patients who became deeply involved in their treatment decisionmak- ing tended to be younger, well-educated, chronic patients, whose treatment was clear-cut and demanded specific coopera- tion from them, but this was not uniformly true. Consent Forms. For many clinicians, obtaining informed consent involves nothing more than getting a consent form signed. In theory consent forms are merely a means to the goal of making patients more informed and better able to participate in decisionmaking, not an end in themselves. Do consent forms contribute to that end? To answer this question was one of the tasks we set for ourselves. The consent form at University Hospital. We observed the use of two consent forms in the hospital (see Appendix II for copies). The most commonly used was a multipurpose form entitled “Consent to Operation, Anesthetics and Other Medical Services.” This form contained three blanks that were to be filled in with the patient’s name, the name and description of the procedure, and the name of the physician who would be conducting or directing the procedure. It went on to assert the patient's consent to: (a) whatever might be necessary in an emergency; (b) anesthesia if needed; (c) the observation and photographing of the procedure; and (d) the disposal of tissues or parts removed during the procedure. Finally, the form said: “The nature and purpose of the operation, possible alternative methods of treatment, the risks involved, and the possibility of complications have been fully explained to me. No guarantee or assurance has been given by anyone as to the result that may be 386 Making Health Care Decisions: Appendix C obtained.” Below these statements was an instruction to “cross out any paragraphs that do not apply.” The form was to be signed by the patient and a witness. The other form was one specifically used by the Exercise Physiology Laboratory for patients who were to undergo one of several types of exercise stress tests. It was sent to the homes of prospective patients accompanied by an information sheet de- scribing the various tests in detail. The patients were told to read the forms and bring the “permission form” when they came for the test. The actual signing of the consent form was done in the lab in the presence of the nurse, after which she gave another short description of the procedure and answered further questions. The multipurpose form was by far the most frequently used one in the hospital. Whenever a “procedure” * was proposed, the physician wrote an order on the patient’s chart to have a consent form signed. This was the duty of the nursing staff. Unwritten hospital policy was that the physician was supposed to obtain informed consent from the patient, and the nurse was supposed to witness the patient's signature on the form. This role differentiation was said to have been designed to reduce the likelihood of coercion of the patient by the physician. The “consent-form signing” was generally the responsibility of the evening charge nurse. The consent form as part of the nursing routine. The nurses tended to regard the consent-form signing as just an- other job in a busy schedule. On the surgical ward the form was usually taken into the patient’s room and either read to or by the patient. If the patient looked confused or did not seem to understand the terms, the nurse would paraphrase it. Mr. N.J., an 80-year-old man with a third-grade education, who seemed to have substantial trouble with his sight, was an example of such a patient. In many ways, Mr. N.J.’s response to the consent form was fairly typical: Nurse: I'm here to have you sign a consent for sur- gery. Well, it isn’t really an operation but a cystoscopy. [She handed him the form.] Can you read this? Mr. N.J.: Well, I'll try. Nurse: Would you like me to read it for you? Mr. N.J.: Yes. Nurse: [Read the first part of the form.] That just means that they do not know what they may have to do. * For a discussion of the meaning of this term, see the section, “In- formed Consent: For Which Treatments?” Consent and the Structure of Medical Care 387 Mr. N.J.: They ain’t going to cut my heart or my liver out? [Joking] Nurse: No, no, of course not. It also says, “I consent to the administration of such anesthetics as may be considered necessary or advisable by the physician responsible for this service.” He has already come up and told you about that, right? Mr. N.J.: That is what he is doing here? [pointing to our observer] Nurse: No, no, he’s doing something else. [She then read the rest of the form, providing one sen- tence summaries of each part as she went along.] Mr. N.J.: Okay. [He signed the form.] Nurse: [Simultaneously with his signing] That they can’t guarantee anything but that they’ll try real hard. Do you have any questions? Mr. N.J.: No questions. In the hall the nurse told our observer, “I don’t think he under- stood very much of that!” Often the whole procedure was even less substantial than that. Mr. W.G. was a 40-year-old man who was to undergo a vasectomy reconnection. Nurse: Time to sign your consent. I'd like you to read it. [She gave him the consent form and left the room. While she was gone the pa- tient, with his wife looking over his shoul- der, read the consent form.] Nurse: [Upon returning] Any questions? Mr. W.G.: Well, I need a pen. [He took it and signed.] On the cardiology ward, consent-form signings were less frequent. Of those observed, most involved getting the form signed by patients scheduled for cardiac surgery, which was not included in our study. However, the process was much briefer than that on the surgical service. The nurse often simply went into the patient’s room, and informed the patient that she had to get the form signed and that she would leave it to be read, and returned in a short while to witness the signing of the form. Often the consent-form signing was embedded in other events. Mr. N.B. was a 64-year-old, retired man who was suffer- ing from exertional angina. As a result of post-catheterization complications, he had to have a right brachial artery explora- tion done. The evening prior to the surgery, the evening charge nurse gave the form to Mr. N.B.’s nurse and told her to get the 388 Making Health Care Decisions: Appendix C consent form signed and to get him ready to go downstairs. They had brought a stretcher up, and it was outside his room. Nurse: Hi, sweetie. Let's pull this out before you go [referring to the IV pump]. Mr. N.B.: OK. [doing what he could to help.] Nurse: Mr. N.B., here’s the consent form so that you can get this procedure done. [She showed him the form and read.] Right Brachial Ar- tery Exploration. Did they explain it already to you? Mr. N.B.: Yeah, a good bit. Nurse: OK, well, I'm just here to witness your signa- ture. If you feel you understand it and don’t have any questions, go ahead and sign. But don’t sign it if you feel you have some ques- tions or you don’t understand it. Mr. N.B.: No, that’s OK. [He signed the form.) The consent form as an opportunity for discussion. On two occasions we saw instances in which the consent-form signing became an opportunity for the patient to raise ques- tions. On one of these occasions, although the patient asked relevant questions prompted by his reading of the form, the answers he received from the nurse were evasive. The other instance involved Ms. TF, a middle-aged patient who had an ankle affected by cancer that had to be immobilized in a cast to promote healing as a “last ditch” effort to save her leg. Ms. TF. had many questions and doubts that she chose to share with her nurse but not with her physician. The conversa- tion went as follows: Nurse: Mrs. TF, can I get you to sign this consent? Ms. TF. Depending on what it is. Nurse: [Laughing] Oh, I'm asking you to sign your life away! [She pulled up a chair and sat down beside the patient’s bed.] What you're consenting to is this. [She started reading what the nature of the procedure was. She said that they were going to put on a short leg cast. | Ms. T.F.: A short leg cast? What is going to happen to the drainage? Nurse: Well, there might be a window in it, then we can change dressings through it. Ms. TF. What's the idea of the cast? Nurse: Um, did they explain it to you? Consent and the Structure of Medical Care 389 Ms. TF.: Nurse: Ms. TF: Head Nurse: Ms. T.F.: Head Nurse: Nurse: Ms. TF. Nurse: Well, I was pretty sleepy when they came by this morning. I don’t remember much. My leg isn’t broken. Why would they want to put on a cast? To be perfectly honest, I don’t know what it’s for. I didn’t check your chart. I didn’t have time before I came in. Do you want me to call them, and you can have them explain to you more about it? No, it’s all right, if that’s what they want to do. I guess they know. [At that point, the Head Nurse came in, and they asked her “why the cast.”] They want to immobilize your leg to give your leg a rest, so that it will be more com- fortable for you, and so that it doesn’t get more irritated when you move. Will I be able to walk? It depends. They'll tell you. [Then she left the room.) Here, you are consenting that you will allow your doctor to use his better judgment of how to do it, that is, the procedure. For example, if he has to put a window in it [the cast], that’s the kind of decisions they’ll be making here. And here, you're agreeing to the anes- thesia, though in your case it doesn’t apply. [The patient was quiet and read along with her.] The next part was that there might be some students, and possibly, if they're going to use photography, that is included here, though they're likely to ask you about it first. [The patient was looking at her and nodding but not saying anything.] Here you're agree- ing that the doctor talked to you about this and that you understand. [At this point she handed the patient the form and a pen.] Any questions? I'll do whatever the doctor says. He knows what he’s doing. But how will it drain if there's a cast on? The doctor never seen it open. [She was referring to the wound.] Well, I don’t know exactly. If there’s a win- dow, which would be around the wound site, then we’ll be able to change the dressings through that, and the drainage will come onto those dressings. Any more questions? 390 Making Health Care Decisions: Appendix C Ms. TF. What about pain medication? Nurse: Well, we have something ordered . . . . [Hesitantly and with doubt in her voice] They talked to you about it didn’t they? Ms. T.F.: Yes, he said something about a window. Nurse: [Interrupting in a more comfortable voice] Oh, there is going to be a window. Then we will be changing dressings. That's why the window's there. [The patient then signed the form.] Of course the consent form could also act as a substitute for discussions. In the section, “Informed Consent: For Which Treatments?,” we discussed how consent forms were used in the exercise stress laboratory to explain the procedure. Since there was no discussion of the risks, benefits, purpose, or alter- natives of the procedure, and since patients asked almost no questions, it seems fair to say that both patients and staff used the form as a substitute for a discussion.. Conclusions. Consent forms do not seem to play much of a role in determining the extent to which patients have informa- tion about their treatment nor the extent to which they actively participate in decisionmaking. While the legal and ethical obli- gation to obtain informed consent encouraged disclosure, the form itself seems to have been useful only in relieving the anxieties of physicians and hospital administrators about the prospect of a lawsuit. In only one case (Ms. T.F.) did we see any positive contribution toward encouraging discussion made by the consent form, and in that case it did not bring the physi- cians into discussion with the patient but rather indicated the extent to which such prior discussion had either not occurred or had left the patient confused. Almost uniformly both pa- tients and staff treated the consent form as a largely meaning- less piece of legal paperwork. Summary of Findings Our primary finding is that the manner in which decisions are made about which treatments patients will get bears little resemblance to the decisionmaking process contemplated by the informed consent doctrine. This is generally true in all the settings we observed although in a few situations the decisions were reasonably close to what the doctrine envisions. Before discussing some of our more specific findings a few cautions are in order. Although our overall sample of patients was large, it was not a systematic one. Moreover, our “subjects” were not only patients but medical personnel, and here our sample size was much smaller. Furthermore, not only was the number of medical personnel that we were able to observe fairly Consent and the Structure of Medical Care 391 small, but the number of different settings in which we ob- served them was far smaller. We observed only a handful of different settings, and all of these were in a single hospital complex. Thus there is no way of knowing whether or not our data are representative. Furthermore, because of the large number of potentially significant variables in the decisionmaking process and the relatively small number of patients, we were not able to perform a multivariate analysis. Thus, it is often difficult to be sure just what it is that accounts for particular behaviors exhibited by doctors and patients in the decisionmaking process. The only way to achieve a higher level of certainty is to undertake a far more extensive study than we were able to conduct within the limits of our fiscal and temporal constraints. As a result, most of our findings—with the exception of the primary general one that medical decisionmaking rarely bears much resemblance to the legal model—must be viewed with some tentativeness. Indeed, many of our findings should be treated as hypotheses for further investigation. The Decisionmaking Process in General (1) There was a great variety in what doctors told patients, what patients learned from other sources, what patients under- stood, and how decisions were made. (2) The ways in which decisions were made varied from rather close conformity with the legal model in a few situations, to a type of decisionmaking that bore almost no resemblance to the legal model. There were a variety of intermediate patterns, some involving a lesser degree of “disclosure” than legally required and some involving a different temporal order of events than contemplated by the legal model. (3) To the extent that actual decisionmaking processes ap- proximated the legally contemplated model, they included one important variable—time—of which the legal model does not take account. Informing and deciding often took place over time, and the greater the degree of patient participation in the process, the more this was so. (4) In general the physician was clearly the dominant actor in terms of making decisions about what treatments, if any, a patient was to have. Both doctors and patients saw the process this way. (a) The doctor's ordinary role, in practice, was to de- cide what was to be done and to inform the patient of that decision. Ordinarily this information came in the form of a recommendation; though depending on the treatment in- volved or the personalities of the doctor of the patient, it might be better characterized along a spectrum running from an “order” at one end to a neutral disclosure of alter- natives at the other end. 392 Making Health Care Decisions: Appendix C (b) The patient's ordinary role, in practice, was to ac- quiesce in the doctor’s recommendation. Patients played a more active role when the doctor presented alternatives without placing any preference on them. Sometimes pa- tients objected to a recommendation: occasionally they vetoed it. But on balance the typical patient role was one of passive acquiescence. The Role of the Medical Setting in the Decisionmaking Process (1) The manner in which decisions were made was af- fected by the setting in which they were made. (a) Surgical outpatients were more active participants in the decisionmaking process than surgical inpatients. They asked more questions: they tended to negotiate more for treatment, often having a better idea of what they wanted than inpatients; and doctors seemed to take se- - riously the outpatient’s ability, literally, to walk away if they were not satisfied with the “negotiations.” (b) Doctors spontaneously gave less information to out- patients than to inpatients but this probably resulted from the fact that outpatients had fewer contacts with the doctor than inpatients did. (c) Although surgical outpatients were more active par- ticipants than inpatients, doctors were still the dominant participants in the medical decisionmaking process, largely as a result of their expertise and authority. (2) In settings in which there was one person clearly iden- tified as responsible for patients’ care (e.g., the inpatient surgi- cal service), patients were more likely to ask questions and receive information about their problem and its treatment than where the lines of responsibility and authority were less well- defined (as was the case on the cardiology service). Where responsibility and authority for medical care were less well- defined, responsibility for providing information was also less well-defined, and consequently patients tended to receive less information. This is the problem of “floating responsibility,” which we have previously identified on an inpatient psychi- atric service (Lidz et al. forthcoming, 1983). (3) Opportunities for contact among patients on the inpa- tient services were low. Thus, patients did not seem to learn much from other patients. This is in sharp contrast to an inpa- tient psychiatric service where patients not only have a multi- tude of opportunities for contact with other patients, but also may be encouraged to do so for therapeutic purposes (Lidz et al. forthcoming 1983). (4) Patients on inpatient services did not learn a great deal from family and friends either. Outpatients seemed to consult Consent and the Structure of Medical Care 393 other sources of information (people and medical literature) more frequently then inpatients did, largely reflecting their greater mobility and ability to do so. The Role of the Nature of Medical Care in the Decision- making Process (1) The nature of the medical procedure is a major deter- minant of the degree of information that patients receive about that procedure. (2) For purposes of analyzing informed consent, medical care falls into two general categories: routine and non-routine. Non-routine care is generally referred to as a “procedure.” (3) The consequence of classifying something as “rou- tine” was that informed consent was not required; the conse- quence of classifying care as a “procedure” was that informed consent was required. (4) The determinants of what was classified as a “pro- cedure” (for which informed consent had to be obtained) are difficult to pin down precisely. (a) A procedure is something that is performed infre- quently and is discrete. It is usually, but not always, invas- ive (e.g., stress tests); however, not all invasive treatments are “procedures” (e.g., medications, X-rays). (b) There is a strong association between a “procedure” and surgery (witness the common usage “surgical pro- cedure”), but not all “procedures” are surgical. (c) There is also no relationship between what is classi- fied as a “procedure” and the degree of risk entailed. Many medications are quite risky, but are never thought of as procedures. Perhaps if risk has anything to do with classifi- cation, it is risk as perceived by laymen. (d) Routine care is easier to exemplify than to define or even describe. Routine care includes having blood drawn for laboratory tests, x-rays and scans, and medications. (5) Whatever definitional problems there may be in classi- fying a treatment or diagnostic act as “routine,” it is clear that there was a shared agreement by doctors and patients that patients were to play no role in these decisionmaking pro- cesses, other than the residual one of objecting. (6) Decisions about routine care are “low-visibility” deci- sions. Such decisions are so “second nature” that those making the decisions do not even conceptualize them as decisions. (7) Decisions about which medications patients will get— and in what quantities—best exemplify the decisionmaking process for routine care. (8) Medication decisions—and decisions about other kinds of routine care—were viewed both by doctors and by 394 Making Health Care Decisions: Appendix C patients as “medical” decisions, that is as being almost ex- clusively within the province of doctors. There seemed to be only two common exceptions to this. (a) Patients sometimes objected to taking a particular drug, usually as the result of previous experience with that drug. (b) Patients often requested analgesic medications. (9) Surgical inpatients received more information than did cardiology inpatients in large part because of the more discrete nature of surgical practice focusing on a single event—sur- gery—as compared with the more complex, continually chang- ing, and process nature of cardiology practice, which made cardiological issues more difficult to describe. (a) On the inpatient cardiology service, a large propor- tion of effort went into diagnosis. Thus decisions had to be made about which diagnostic procedures to do before fur- ther decisions could be made as to which therapies to administer. Thus, practice on the cardiology service was far more conceptually complex than in surgery. We believe that this complexity was a disincentive to informing pa- tients about the nature of their illness, and the nature of the procedures being contemplated or undertaken to diagnose and treat it. (b) In those instances where cardiology practice was event-focused—i.e., where a particular special procedure was being considered—cardiology patients got at least as much information as surgery patients did about surgical procedures. The Role of the Nature of the Patient's Disorder in the Decisionmaking Process (1) The nature of the patient’s disorder, in a general way, affected the way in which the doctors dealt with the patients, including the amount of information given and the extent of patient participation in decisionmaking. (2) On the surgical service, where the nature of the pa- tient’s problem was usually fairly certain and where there was a consensus in the medical profession about the best approach for dealing with that problem, patients received more informa- tion about their illness and its treatment than on the cardiology service, where the nature and etiology of the problem, and thus the best method for dealing with it, were more uncertain. (3) Even where there was a high degree of uncertainty about the nature and etiology of the patient’s problem, an effort was made by medical personnel to convey a sense of certainty. (4) Whether a patient was suffering from an acute or chronic illness appeared to be a major determinant of the pa- tients role in the decisionmaking process, including what the Consent and the Structure of Medical Care 395 patient was told by medical personnel, what the patient under- stood, the degree of self-perceived autonomy, and participation in making decisions. (5) Patients with chronic illness who had been receiving treatment for that illness for a substantial period of time (“chronic-chronic” patients) came closest to playing the role envisioned by the informed consent doctrine. They were given more information by medical personnel, they understood it better, and both they and medical personnel viewed them as having a greater say in what treatment they received. These patients were sometimes so well-educated that they spoke about their illness in the same jargon as the medical personnel. Their understanding of information relevant to their illness and treatment grew over time. The temporal dimension both of informing and understanding was most apparent in the case of chronic illness. (6) Much of the information given to patients with chronic disease was not necessarily intended to assist them in par- ticipating in decisionmaking, though it sometimes had that consequence. Information was given to persons who were newly diagnosed as having a chronic disease (“acute-chronic” patients) to assist them in coping with a new way of life and in complying with the treatment recommendations made by medi- cal personnel. (7) Patients suffering from an acute illness, who were not also suffering from some chronic problem simultaneously (“acute-acute” patients), in contrast with chronically ill pa- tients, were given enough information so that they could agree to the therapeutic recommendations made by medical person- nel and so that they would not be too surprised and unable to deal with untoward results. This, for example, was the main purpose of preoperative teaching by surgical nurses. (8) The best test of the acute/chronic distinction is the way in which chronic patients who suffered an acute illness (“chronic-acute” patients) behaved. When chronic patients suf- fered an illness that was unexpected and about which they were previously ignorant, they reverted to a role similar to that which acute-acute patients played. Thus, it is not so much the charac- ter of the patient, but the character of the illness, that deter- mined the degree of participation. (9) We have no evidence that the prospect of death does or does not play a role in a patient's ability to participate in the decisionmaking process. The Role of Incompetency in the Decisionmaking Process (1) Staff's assessments of incompetency were not made in any systematic way. Determinations of incompetency were made in an informal and extralegal manner, with little or no manifest awareness that they were even being made. Resort to 396 Making Health Care Decisions: Appendix C the judicial system (or even to legal counsel) was extremely rare. (2) Incompetency was viewed by medical personnel largely from a behavioral perspective. That is, patients who were capable of functioning in the normal patient role were viewed as competent, others as incompetent. (3) The patient's status per se commanded little attention in the determination of incompetency. Status was relevant, if at all, only insofar as it affected functioning. Thus, patients who were terminally ill were not treated as incompetent unless they were incapable of communication; patients who were heavily sedated were not treated as incompetent (even if their cognitive processes were clouded) if they were able to engage in conversa- tion; patients who were comatose or stuporous were always treated as incompetent; and patients of transient competency were treated as competent when they were capable of commu- nication, and vice versa. (4) The “test” of competency employed seems to have been a test of general competency (Roth, Meisel, & Lidz, 1977; Meisel, 1979). That is, whether or not one was incompetent depended on one’s generalized functioning in the patient role. (5) The doctors’ perceptions of a patient's ability to under- stand was probably the dominant determinant of general com- petency. (6) Doctors adjusted what they told patients based on their perception of the patient's ability to understand. This was most evident with patients of certain incompetency, but also oc- curred under other circumstances where there was a suspicion of an inability to understand. (7) When a patient was viewed by the staff as incompetent, permission was often sought from family members if they were available and if permission would have been sought from a competent patient. (8) When the competent patient's role involved merely ac- quiescence in or compliance with treatment recommendations made by medical personnel, the fact of incompetency had no practical significance. Family Participation in the Decisionmaking Process (1) Medical personnel did not often make systematic efforts to involve family members in the decisionmaking pro- cess. When family members were available, they were some- times called upon to act in the patient's stead. When family members did participate, it was often because they manifested an interest in doing so either to the patient or to medical personnel or both. (2) Family members were inclined to ask more questions of staff members than patients, especially than inpatients. Consent and the Structure of Medical Care 397 (3) Family members occasionally sought to prevent pa- tients from being given information by medical personnel that they thought would upset the patient. (4) Patients’ relatives who were health professionals were frequently sources of advice and information for patients and sometimes intervened to aid in the decisionmaking. (5) Patients often involved family members in decision- making on their own initiative. The Role of Consent Forms in the Decisionmaking Process (1) Consent forms were not viewed either by doctors and nurses or by patients as a part of the decisionmaking process. They were seen primarily as a bureaucratic formality imposed by hospital administrators, hospital lawyers, or the law. (2) In most cases, a preprinted consent form was used that had minimal information inserted in the blank spaces to give the name of the procedure the patient was to undergo and the name of the doctor who was to perform the procedure. (3) Rather than including specific information required to be disclosed by the informed consent doctrine, the most often used consent form contained a conclusory statement that the general categories of legally material information had pre- viously been “fully explained.” Thus this type of consent form did not perform any informing function. (4) Patients were asked to sign consent forms after the decision to have the treatment in question was actually made. (5) Patients were asked to sign consent forms the evening before surgery was performed. Thus, even if the forms had been more informative, they came at a time when there were substan- tial inhibitions against changing the decision. (6) Consent forms were used only for “procedures.” (7) Nurses were responsible for having consent forms signed. They did not usually view their role (and indeed there may be legal inhibitions on their viewing their role) as includ- ing the responsibility or authority for providing patients with information material to decisionmaking. (8) Consent forms may be used as a formal substitute for the kind of conversation between doctor and patient envisioned by the informed consent doctrine (Meisel, 1982). In one in- stance, a consent form other than the standardized form was used—for exercise stress tests. In this situation, the form was the sole source of information for the patient (other than any information that the patient may have received from his private referring physician) unless the patient chose to ask questions. The consent form was used as a substitute for informed con- sent, rather than merely a memorialization. 398 Making Health Care Decisions: Appendix C The Role of Nurses in the Decisionmaking Process (1) Most of the information directly relevent to decision- making that patients got from medical personnel came from physicians rather than from nurses. (2) The nurses viewed disclosure as the doctor’s responsi- bility, not their own. (3) Nurses sometimes provided important background in- formation through “preoperative teaching.” The purpose of pre- operative teaching was not informing for decisionmaking, but alleviating patient anxiety and promoting patient cooperation. The content of this information was often relevant to decision- making, and if it were to come earlier than the evening before surgery, it might prove helpful to patients in deciding. (4) Most information about medications came from nurses, a form of treatment that they were responsible for administer- ing. This was often the only information that patients received about their medications. The information usually included the name of the medication, and sometimes some explanation of its side effects and purpose. (5) The information that patients obtained from nurses about medications came after the decision was made by the doctor that the patient was to have the medication. The infor- mation was not intended to inform the patient for decisionmak- ing. Despite this, the information may have been useful for decisionmaking since medications are usually taken on more than one occasion, and thus patients were in a position to object to a medication. The Operation of the Decisionmaking Process in an Emergency (1) In emergencies, medical personnel typically rendered necessary medical care without the patient’s informed consent. If a family member was at hand, and if sufficient time was available, permission was sought from the family member. (2) What constitutes an emergency in practice is difficult to specify. There is a great deal of diversity in which situations are treated as emergencies. (3) The operational definition of an emergency appears to have included the existence of a perceived immediate threat to the patients life, and the belief that, from a medical perspec- tive, the only alternative to immediate treatment of the kind being adminstered was to let the patient die or suffer unduly. (4) When a situation was deemed to be an emergency, the consequences were variable. In some situations, no effort was made to discuss treatment with the patient, even when there was time to do so; but in other situations, efforts were made to obtain a patient's signature on a consent form despite the fact that the patient might have died at any moment. Consent and the Structure of Medical Care 399 Conclusions Law's vision of how medical decisions should be made is embodied in the informed consent doctrine. The doctor-patient relationship is to be a collaborative one. Each participant brings to bear a unique contribution: the doctor, medical expertise; the patient, a personal value system. Though the two participants are not equal in terms of their expertise, they are moral equals, which imposes upon the doctor the moral obligation, now embodied in law, to provide the patient with the one thing that creates the inequality in expertise: information. The patient who is given medical information thereby becomes not only the doctor's moral equal, but his or her “medical equal” too. Once the information is transmitted, patients are in a position to bring their own personal values to bear on this information and to instruct the doctor as to how to proceed. A reasonable inference from law’s vision of the medical decisionmaking process is that there will be a single occasion on which the doctor will talk to the patient about his or her condition and its treatment or diagnosis. The doctor will ex- plain the alternative ways of dealing with the medical problem, and the risks and benefits of each. The patient will then make a decision, and if the decision is in favor of one or the other of the procedures, the doctor will proceed with therapy. Or, the pa- tient may conclude that the one alternative that is always avail- able—to do nothing—is preferable, and will refuse treatment. In its pristine form, law's vision does not exist in reality. This should come as no surprise; the ideal is rarely to be found in operation in any context. What is more surprising, though, is that the actual deci- sionmaking process generally diverges from the ideal much more than might be expected. Certainly there are exceptions. There is no single pattern by which all medical decisions are made, but the great run of medical decisions diverge substan- tially from the model. In practice, the medical decisionmaking process is gov- erned far more by the logic, the logistics, and the ethos of medical practice than by the legal ideal of individual auton- omy. Medical diagnosis involves the rapid formulation of a small number of potential causes for the patient's particular malady. Early diagnostic work leaves behind a primary hypoth- esis as to the cause. Diagnostic and treatment decisions are based on that hypothesis until further results force a reassess- ment. Much of the decisionmaking that doctors engage in takes place at a preconscious level. Only relatively rarely does sub- stantial uncertainty about causation persist. Quite early in the process the physician reaches a diagnosis and a decision about a preferable treatment. Seldom does the doctor see a series of alternative possible treatments. Rather, for each problem there 400 Making Health Care Decisions: Appendix C typically exists a medically preferable treatment, not a series of alternatives from which the patient may choose. It does not seem to the doctor to be a decisionmaking process but simply a question of persuading the patient to accept proper treatment. The decision has been made—by the doctor. It is now, in the medical view, time to make a recommendation to the patient. In addition to making a recommendation, the doctor's self- perceived role is to get the patient to go along with this recom- mendation if there is any hesitancy on the patient's part. This is done by some explanation about the need for the recommended treatment and the consequences of not heeding the recommen- dation. But in the doctors view there is no decision for the patient to make, except whether or not to get proper medical care. Thus, the patient's role is a passive rather than an equal one. The patients role is first to acquiesce and then to comply. An occasional patient refuses to conform to this role, but be- cause of the manner in which the process has operated up to this point, he or she is left with little to do but object to or veto the doctor’s recommendation. Only the exceptional patient at- tempts to push the doctor back into the role envisioned by the informed consent doctrine by vigorously questioning the doc- tor. Such patients are few and far between. Most patients accept or even insist on this role relationship with the physician. In outpatient settings where patients have more real options, they may, instead of attempting to get the doctor to act in accordance with law's vision, merely seek another doctor. (We have insuffi- cient data to know how frequently this happens in practice.) When patients admit themselves to the hospital or visit the outpatient clinic, they sign a blanket permission form. (See Appendix II.) Such forms have long been viewed by courts as having little legal value. But in fact they accurately reflect the way in which almost all doctors and patients view the decision- making process. In essence, the patient entrusts him or herself to the doctor's care. Admission to the hospital carries with it a de facto presumption of cooperation by the patient with the doctor. The patient may back out of this agreement, but it is mutually presumed that he will not. Rather than the treatment requiring consent, treatment takes place unless there is an ob- jection. Thus, there is patient acquiescence in medical control until such time as the patient may change his or her mind. This stands the traditional doctrine of informed consent on its head. Instead of a presumption of bodily integrity that pre- vails until permission is granted by the patient for a “touching,” there is a presumption that doctors may ply their trade unless patients affirmatively object. Even the vocabulary of the in- formed consent doctrine is inappropriate to a description of the medical decisionmaking process as it actually operates; an en- Consent and the Structure of Medical Care 401 tirely different vocabulary must be employed if reality is to be more accurately portrayed: (1) “Disclosure” does not typically occur. Rather patients learn various bits of information, some relevant to decisionmak- ing, some not, from doctors’ and nurses’ efforts to obtain com- pliance and from “situational etiquette.” (2) “Decisions” are not made by patients. “Recommenda- tions” are made by doctors to patients. (3) “Consent” does not exist. Instead what we find is “ac- quiescence,” the absence of “objection,” or occasionally a “veto.” Law’s vision of medical decisionmaking involves an im- plicit assumption about the practice of medicine that is either generally invalid or invalid in a substantial number of situa- tions. This is an assumption that medical practice is discrete— that is, broken into distinct parts, or decision units—and that there can be consent by the patient to each of these individual parts. In fact, much of medical practice is so complex, uncer- tain, and unknowable in advance (except in the most general terms) that there really can be nothing like informed consent. Because of the complexity of practice and the structure of medical logic, there rarely exists a set of alternatives from which the patient could choose. Moreover, the explanation of the consequences of an “alternative” is complicated by the fact that the results of any procedure may reveal that yet another procedure might need to be performed. The role of the concept of medical alternatives in the doc- trine of informed consent has developed largely as a result of the fact that the doctrine has been formulated in the context of a procedure that has gone wrong. In considering cases, the courts have compressed several sequential decisions into a set of con- temporaneous choices. For example, from the physicians’ point of view, the situation looked like a series of decisions having the structure: “If X test shows A, we have to do M; if it shows B, we have to do N.” However, the courts compressed the same set of facts into the statement: “The physician could have done M, N, O, or P” Thus there has been a failure to understand the organic nature and the complexity of medical practice, a failure that might not have occurred if lawsuits focused (as they do not and probably cannot) on a longitudinal process rather than on a cross-section of that process. Lest we be accused of overstating the divergence between theory and practice, it is necessary to emphasize the point made earlier that medical decisionmaking occupies a range of points along a spectrum defined on one end by something like the informed consent doctrine and on the other end by the foregoing description. Why is it that, in some situations, doc- tors behave more as law envisions? Is it that some doctors are 402 Making Health Care Decisions: Appendix C more cognizant of their legal obligations and more willing to abide by them? Although some doctors do take legal obligations qua legal obligations more seriously than others, we do not think this is the primary explanation for the differences in behavior. What we believe is more important in determining the correspondence between clinical practice and the informed consent doctrine is not any of the variables we postulated, but the “nature” of the doctor-patient relationship, by which we mean whether or not the relationship is a long-standing one based on trust, or a rather new and often ad hoc one. In those cases in which the doctor and patient have had some opportunity to get to know each other over a period of time, they tend to act more in the manner previously described. By contrast, when a patient is referred to a doctor for a particular procedure and where it reasonably appears at the outset that their relationship will not be an enduring one but will exist only for purposes of the performance of this particular procedure, the behavior of the participants moves closer to the “informed consent” pole of the decisionmaking spectrum. In other words, we believe that the greater the degree of trust in the doctor by the patient, the less that occurs that looks like the “disclosure” envisioned by the informed consent doctrine. This explanation, so far, only identifies the contexts in which doctors do behave more in conformity with legal obliga- tions, but does not explain why they do. Where trust is a major component of the doctor-patient relationship, doctors giving patients information and patients asking questions of doctors are fundamentally upsetting to the maintenance of trust. The latent message from the questioning patient is “I don’t trust you”; the latent message from the informing doctor is “I think you don’t trust me, so I'm giving you information to show you how open and honest I am.” Thus, in the relationship in which trust plays a large part, there is less substantive communication and more faithful obedience. Conversely, where there is more communication, it may be a sign that trust is breaking down or has never existed. Paradoxically, those doctors who do explain more to their patients often thereby create a more trusting atmo- sphere. Thus, disclosure functions not to produce autonomy, as the informed consent doctrine contemplates, but to produce trust that, once it exists, will then permit the doctor to dispense with further explanation. There is yet one other reason why doctors talk to patients besides the need to elicit acquiescence and build trust. They also talk to patients for the same reason that two strangers thrown into a.common situation talk to each other—because they're there. This is what we refer to as “situational etiquette.” Especially on an inpatient service, and especially with more alert patients, doctors and patients constantly come in contact Consent and the Structure of Medical Care 403 with each other. When they do so, they feel compelled by the etiquette of the situation to say something, and the one thing they have in common to talk about is the patient's illness and its treatment. Thus, much of what is said by doctors to patients is essentially idle conversation, but it sometimes serves to pro- vide patients with information they would otherwise not ob- tain. Thus, even when patients do learn from doctors the kinds of information that is required to be disclosed by the informed consent doctrine, they do so in a haphazard way, rather than in the neat, linear fashion envisioned by law. The primary emphasis in this discussion has been upon the fact that the behavior of physicians does not often reflect that which is contemplated by the legal doctrine of informed consent. We have largely taken this focus because the obligation that law imposes is upon the doctor, not the patient. Indeed, the doctrine recognizes (through the “waiver” exception) that the patient is as free not to participate as to participate, that any requirement to participate by listening to information that one does not want to hear or by making a decision one does not want to make is equally as paternalistic as being denied the opportunity to be informed and to decide. Our findings also strongly suggest, however, that even if doctors were acting in the way anticipated by law, decisionmak- ing would bear little resemblance to the legal model. Both from our data in this study and from that obtained in a similar study in three different settings in a psychiatric hospital (Lidz et al., forthcoming 1983), we have been struck by the fact that over- whelmingly, even when patients are given information about their treatment and treated as if they had decisional authority, they act in a passive manner. When asked, most patients seemed happy with the amount of information they were get- ting and even when they wanted more it was rarely in order to make decisions about treatment. Even when they said they wanted information to make treatment decisions, they often acted as if they would rather have the doctor decide. For the most part, patients were not very interested in much of what was told to them. Even when they were interested in the infor- mation, they still often acted as if the final decision ought to be left to the doctor. Fundamentally, patients feel that they are unequal to the task of making medical decisions, even when provided infor- mation to do so. They feel no more equipped to decide which treatment among several alternatives to choose than the average person does to choose between several ways to build a wall. If we want a wall built, we tell the contractor to use brick and where to put it. The contractor chooses the brick and the mor- tar, hires the assistants, and decides whether to use two or three layers of brick. If a sick person wants his illness attended to, he feels that he is well-advised to hire a physician, osteopath, or 404 Making Health Care Decisions: Appendix C chiropractor to deal with it. The patient chooses the doctor and presents the problem. The rest is the doctor’s concern. Although it is sheer speculation, it is possible that what deters doctors from obtaining informed consent in conformity with the spirit of the doctrine is that they know, and have long known, what we have found in this study: that patients are not interested in, nor do they believe that they are capable of, playing the role assigned by law. Knowing this—and believing, too, that patients are not capable of playing that role—doctors do not take the time and effort themselves to comply with the legal dictates addressed to them. While these implications of our research are hardly good news for the doctrine, before we can conclude what should be done we must first return to the ethical justifications of in- formed consent. There are two primary traditions in current ethical thought: deontological and utilitarian. The deontologi- cal tradition derives from the work of Immanual Kant, whose categorical imperative was intended to be a nonreligious, abso- lute grounding for moral judgment. The importance of Kant’s position for present purposes is that the moral quality of an act is said to be independent of the outcome or consequences of the act. Right and wrong are categorical judgments of types of acts, not situationally specific judgments or results. In contrast, the utilitarians, whose views were best expounded by Jeremy Bentham and John Stuart Mill, held that the key to the morality of an act is to be found in its consequences. Whether or not it is right to lie, for instance, cannot be settled categorically but must be determined by reference to whether or not the conse- quences of the lie are mostly positive or negative. If the benefits exceed the costs, then a particular lie—indeed, any given ac- tion— is morally justified. The doctrine of informed consent has been justified on the basis of both these ethical theories. Most deontological analyses of informed consent start from the value of autonomy. Since the autonomy of the individual is a highly regarded value and one to be fully implemented, at least in all spheres where it does not enter into substantial conflict with other significant values, the requirement of informed consent is justifiable to the extent that it is reasonably calculated to promote the autonomy of indi- viduals. A further deontological justification for informed con- sent arises from the fact that it also promotes another important and widely held value—rationality. Informed consent provides the patient with information that facilitates a rational decision. This study has, of course, focused on an evaluation of the consequences of informed consent. While we can not find much to justify informed consent in the consequences of its implementation, this does not touch the deontological ground- ing of the doctrine. Short of findings that show that informed consent produces severe harm to medical care, we can not Consent and the Structure of Medical Care 405 imagine that American law would jettison the doctrine, because autonomy is such an important American value. Moreover, one other argument can be mustered in favor of the doctrine. Although informed consent has done little to reform the doctor-patient relationship, we did see some pa- tients, particularly chronic patients, who felt a strong desire to play a major role in directing their treatment. Even though they are a small minority of patients, it seems important that both law and medical ethics recognize their right to be active partici- pants in their treatment. We were also impressed by the success of preoperative teaching performed by nurses. Perhaps this can be broadened into an effort directed at the somewhat different end of involv- ing patients in the decisionmaking process. Law also ought to take more specific account of the exis- tence of patients’ family members and the fact that they do play some role in the decisionmaking process. Families, especially those of inpatients, may be better equipped than the patient to play a more active role, not being compromised by illness in all its medical, psychological, and sociological ramifications. The common law and statutes as they currently exist do not mandate the use of consent forms (with the exceptions of federal regulations governing the conduct of research with human subjects, and state mental health laws). This is probably as it should be. Consent forms do not often enhance patient participation in decisionmaking and may sometimes impede it. The fact that a patient has signed a consent form, whether detailed in its content or wholly conclusory, is no real indica- tion that he has been informed or that he has understood. Those provisions of state statutes conferring a “presumption of valid- ity” on such writings are therefore unwise (Meisel & Kabnick, 1980). Finally, serious thought must be given to alternative means of encouraging patient participation in decisionmaking, and perhaps to the nature of that participation, in those situations in which informed consent has proved most unsatisfactory. One potential alternative is that suggested by the Oregon informed consent statute, which is unique in its requirement that doctors make only limited and general disclosure, but then ask the patient if he or she wishes additional information. If the patient responds affirmatively, the doctor's duty to disclose is thereby expanded (Meisel & Kabnick, 1980). Informed consent has not been legal therapy for the doctor- patient relationship. It has not substantially affected the man- ner in which most doctors practice medicine under most cir- cumstances. From time to time, patients do go through a pro- cess that bears some resemblance to that contemplated by law, but this is the exception rather than the rule. Although these 406 Making Health Care Decisions: Appendix C exceptions should be encouraged, they seem unlikely to be- come the dominant pattern merely as a result of changes in the law. References Denzin N (1970), THE RESEARCH ACT: A THEORETICAL INTRODUCTION TO SocioLoGicAL METHODS. Chicago: Aldine Publishing Co. Lidz CW and Roth LH (1983), The Signed Form—Informed Consent? in SOLUTIONS TO LEGAL AND ETHICAL PROBLEMS IN APPLIED SOCIAL RE- SEARCH, RF Boruch et al. (eds.) New York: Academic Press. Lidz CW et al. (forthcoming, 1983), INFORMED CONSENT: AN EMPIRICAL STUDY OF DECISIONMAKING IN PSYCHIATRY. New York: Guilford Press. Meisel A (1982), More on Making Consent Forms More Readable, IRB 4(1):9. (1979), The “Exceptions” to the Informed Consent Doctrine: Striking a Balance Between Competing Values in Medical Decision- making, WiSCONSIN Law REVIEW 1979:413-488. Meisel A and Kabnick LD (1980), Informed Consent to Medical Treat- ment: An Analysis of Recent Legislation, UNIVERSITY OF PITTSBURGH Law REVIEW 41:407-564. Meisel A and Rqth LH (1981), What We Do and Do Not Know About Informed Consent, JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION 246:2473-2477. Roth LH, Meisel A, and Lidz CW (1977), Tests of Competency to Consent to Treatment, AMERICAN JOURNAL OF PSYCHIATRY 134:279-284. Whyte W (1955), STREET CORNER SOCIETY—THE SOCIAL STRUCTURE OF AN ITALIAN SLUM. Chicago: University of Chicago Press. APPENDIX I: Interview Questionnaires (1) Interview Questionnaire: Patient Already in Treatment 1. To begin with, could you tell me a bit about your medical problems? (If necessary probe for more detail.) 2. Would you explain to me as best you can what your treat- ment consists of? 3. How is it going so far? What do you think it is (or might be) doing for you? 4. Do you know of any risks, side effects, or other discomforts that might occur as a result of your treatment? 5. What else do you think might be done for your problem? Have you tried anything else for it in the past? 6. Can you tell me a bit about how, when, and why you decided to come here for treatment? (If necessary: Who did you talk with about coming here?) 7. What have you been told about your condition and its treatment by the staff and other medical personnel? Consent and the Structure of Medical Care 407 10. 11, 12. . Do you feel you get enough information about your treat- ment? Whom do you ask about this? . Whose job is it to make decisions about your treatment? How do you feel about the way you have been treated so far? What are the high and low points? If relevant: What happens if you wanted to stop treatment? There are a few more questions about your personal back- ground. (a) How old are you? (b) What is your highest level of education? (c) What is your religious affiliation? (d) Are you married (single, divorced, etc.)? (e) What is your occupation? (2) Interview Questionnaire: Patient Not Yet in Treatment 1 2. Could you tell me a bit about your medical problem? Would you explain as best you can what your treatment will be? _ What are the benefits of this treatment? What about disad- vantages? Are there any? . Do you know what else might be done for your problem? Has anything else been tried in the past? . What have you been told so far by (the) staff and other medical personnel about your treatment? Who is it that will make the decision about your treat- ment? What is your role? . Do you have as much information about treatment as you want? If relevant: What happens if you want to stop treatment? I have a few more questions about your personal back- ground which you may choose not to answer if you prefer: (a) How old are you? (b) What is your highest level of education? (c) What is your religious affiliation? (d) Are you married (single, divorced, etc.)? (e) What is your occupation? 408 Making Health Care Decisions: Appendix C APPENDIX II: Consent Forms UNIVERSITY HOSPITAL CONSENT TO TREATMENT FORM Consent to Treatment 'I, being aware that I have a condition requiring diagnosis, hospital care, and medical ireatment and being desirous of such diagnosis, hospital care, and treatment at miversity Hospital, do hereby voluntarily consent to such diagnostic procedures, hospital ‘are, and treatments including blood transfusions by my attending physician, his assistants nd designees, and hospital employees, as is necessary in their Judgement. This form has 'een fully explained to me, and I am fully aware of its contents and understand the same. I 1so acknowledge that no guarantees have been made to me as to the result of any treatments T examinations received by me in the Hospital.” As a condition of my admission to the hospital, I agree to leave the hospital upon discharge y my attending physician, or if my attending physician so recommends, I agree to transfer to nother medical facility, and I consent to and authorize the transfer of the medical records ‘ept by the hospital to the said medical facility." Patient's Signature Or By Legal Representative If Patient Is A Minor, (Relationsh: Incompetent, Or Otherwise Unable To Consent. f patient is 21 years of age or older and is mentally competent, he must sign for himself. f patient is under 21 years of age, check the appropriate sections: Oa Patient represents that he is 18 years of age or over and/or has made the representation which under Pennsylvania Law entitle him to effectively consent to wedical, dental, or health services without the consent of his parents or legal guardian. EPRESENTATION OF MINOR: ave you graduated from High School? Year of graduation f "Yes", state the name of the school Address ive you ever been married? Female answer: Eave you ever been pregnant? lese representations are made by me knowing that others; in good faith, may rely on them. Witness Signed 0 Patient is unable to consent for himself under existing law and the signature above is tha of the closest relative or legal guardian. DISCEARCE AGAINST MEDICAL ADVICE am leaving the University Hospital, against the counsel, advice d consent of the authorities, attending surgeon or physician of the-said hospital. I.there= re do hereby release ard absolve from all responsibility whatever the authorities, surgeons, d physicians of said hospital of any consequences or ill result that cay follow zy voluntary parture from said hospital. TNZSS MY EAND AND SEAL THIS DAY QF » 19, TNESS SIGNED 13-866 Rev. 10/78 Consent and the Structure of Medical Care 409 .UNIVERSITY HOSPITAL CONSENT TO OPERATION, ANESTHETICS, AND OTHER MEDICAL SERVICES Date Time Bt 1. I authorize the perfor upon of e following operation to be srformed under the direction of Dr. ; 2. 1 consent to the performance of operations and procedures in addition to or different from those now sntemplated, whether or not arising from presently unforeseen conditions, which the above-named. doctor or is associates or assistants may consider necessary or advisable in the course of the operation. 3. I consent to the administration of such anesthetics as may be considered necessary or advisable by 1e physician respoasible for this service. 4. I consent to the photographing or televising of the operations or procedures to be performed, including ppropriate portions of my body, for medical, scientific or educational purposes, provided my identity is not evealed by the pictures or by descriptive texts accompanying them. 5. For the purpose of advancing medical education, 1 consent to the admittance of observers to the op- rating room. 6. I consent to the disposal by hospital authorities of any tissues or parts which may be removed. 7. The nature and purpose of the operation, possible alternative methods of treatment, the risks involved, md the possibility of complications have been fully explained to me. No guarantee or assurance has been given ay anyone as to the results that may be obtained. (CROSS OUT ANY PARAGRAPHS ABOVE WHICH DO NOT APPLY) Signed Relationship! Patient, Parent, Guardian Witness 410 Making Health Care Decisions: Appendix C UNIVERSITY HOSPITAL EXERCISE PHYSIOLOGY LABORATORY CONSENT FORM FOR EXERCISE STRESS TEST 1, authorize Dr. of the University Hospital and such other physicians as may be assigned by him, the nurses, technicians, and other persons employed in or associated with the Exercise Physiology Laboratory to administer and conduct the exercise stress test. This test will measure my exercise capabilities, possibly determine the presence or absence of heart disease, and/or help evaluate the effectiveness of my current therapy. I understand that I will walk on a motor driven treadmill or be asked to lie on a table and pedal a bicycle. During the performance of physical activity, my electrocardiogram will be monitored, and my blood pressure will be measured at periodic intervals. The degree of exercise will be progressively in- creased until I have gone as far as I can 80, attain a pre-determined end-point of the test, become distressed in any significant way, or develop any abnormal response the physician considers significant, whichever occurs first. In addition, I understand that I may ask that the test be discontinued at any time. Every effort will be made to conduct the test in such a way as to minimize discomfort and risk. However, I understand that just as with other types of diagnostic tests there are potential risks (approximately 2 to 3 per 10,000) associated with an exercise test. These include episodes of dizziness, fainting, chest discomfort, leg cramps, and very rarely strokes, heart attacks or sudden death. I further understand that the laboratory is properly equipped for such situations and that its professional personnel are trained to administer emer— gency care. I voluntarily accept the risks associated with the above procedure, and I have asked all of the questions which I thought were important in deciding whether or not to undergo the exercise stress test. I also authorize, request and consent to any diagnostic procedure, medical and/or surgical treatment that may be advisable or beneficial for my health should such become necessary during or after the exercise test, I CERTIFY THAT I HAVE RECEIVED THE EXERCISE TEST INFORMATION FORM AND READ AND FULLY UNDERSTAND THE ABOVE CONSENT STATEMENT, THAT ALL OF MY QUESTIONS HAVE BEEN ANSWERED TO MY SATISFACTION, THAT ALL BLANKS REQUIRING INSERTION OR COMPLE- TION WERE FILLED IN PRIOR TO THE TIME OF MY SIGNATURE, AND THAT THIS CONSENT IS GIVEN FREELY, VOLUNTARILY, AND WITHOUT RESERVATION. Signature of Patient Witness ———————————————————————— Date 10/81 Treatment Refusal D in Medical Hospitals Paul S. Appelbaum, M.D.* Loren H. Roth, M.D., M.PH." Refusal of treatment in medical hospitals is a phenomenon that raises important questions about the practice of medicine today. Do such refusals reflect the alienation of patients from their physicians and from an increasingly technological and impersonal health care system? Or is refusal a sign of healthy eagerness on the part of patients to participate more actively with their physicians in making decisions about the care they receive? Is refusal the result of patients’ irrational impulses, or the end product of a careful weighing of risks and benefits? And, in particular, what are the implications of refusal for the relationship between doctors and patients? This paper reports the results of a study of treatment refusal in medical hospitals that was designed to address these questions. Any consideration of the meaning and implications of treatment refusal must begin with an examination of the inter- personal context in which it occurs. The doctor-patient rela- tionship has been the subject of considerable critical comment in recent years. The traditional model of a benevolent, but authoritarian, doctor interacting with a passive patient has been condemned on a variety of grounds: as being disrespectful of the autonomy of the patient and his uniqueness as a human being; as contributing to the alienation of many people from the medical profession, thus leading them to seek alternative, and * Assistant Professor of Psychiatry and Law, University of Pittsburgh Schools of Medicine and Law, Pittsburgh, Pennsylvania t+ Professor of Psychiatry, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. Both authors are affiliated with the Law and Psychiatry Program, Western Psychiatric Institute and Clinic; Dr. Roth is Director of the program. July 1982 412 Making Health Care Decisions: Appendix D potentially less useful, forms of health care; and as tending to lessen the quality of medical care, by derogating the potential contribution of the patient, thereby encouraging misunder- standing and non-compliance. (1-3) A number of remedies have been suggested for the ills that have plagued relations between doctors and patients. On one hand, doctors have been urged to pay increased attention to the psychological and psychosocial needs of their patients and to their own psychological responses to those they treat, in the hope that this will result in more sensitive and individualized care. (4-7) On the other hand, patients themselves have been encouraged to play a more assertive role in their own treatment, demanding a part in decisionmaking, refusing to go along with undesirable forms of treatment, and caring for their own medi- cal needs whenever possible. (8-12) This conviction that patients ought to be exercising more control over their medical care has become quite widespread and is reflected in a number of initiatives on the judicial, legislative and regulatory levels. The most notable of these innovations has been the judicial promulgation of the doctrine of informed consent. (13) Advocated as a means of equalizing power between doctors and patients, and thus contributing to a more collaborative relationship (14), informed consent requires physicians to provide patients with information on the treat- ment or procedures they are proposing, the possible risks and benefits of the treatment, any reasonable alternatives, and the risks and benefits of those alternatives. Informed consent has become, at least theoretically, a prerequisite to all medical care in the United States. (13) Legislatures and rule-making agencies have been even more active than the courts, although not always for the same reasons, in seeking to alter the balance of power in the physi- cian-patient relationship. Informed consent is required by stat- ute and regulation in many states where legislatures did not wait for the issue to be raised in court. On the other hand, legislative initiatives have often been directed towards limiting the extent of judicially-ordered changes. (15) Other steps have been taken as well. Patients now have access to their medical records in many jurisdictions and often have been provided with an impressive list of rights guaranteed to them by the state or the treatment facility. (16) Yet, the belief persists that all is not well as far as doctors and patients are concerned. As an example of the continuing ferment, the courts have been asked to rule with increasing regularity on instances in which patients and doctors have disagreed about the need for or propriety of treatment. A num- ber of these cases have involved attempts by physicians or hospitals to obtain authorization from the courts to administer Treatment Refusal 413 treatment over patients’ objections. Among the more notable cases of this type, which have resulted in conflicting opinions, have been several instances of refusal of blood transfusions by Jehovah’s Witnesses and instances of refusal of recommended surgery by elderly patients. (17-20) Analogous situations have arisen when others have sought to refuse treatment on behalf of incompetent patients, including the patients’ families (21-22), friends (23), and treatment teams. (24) In addition, a number of courts have recently defined a right to refuse treatment, of varying proportions, for psychiatric inpatients (25-27), resting their holdings on principles that could just as easily be applied to the medical hospitals. Outpa- tient refusal, too, has attracted the attention of the courts, both when patients have refused medical and psychiatric treatment on their own behalf and on behalf of their children. (28-29) The focus of these recent legal cases on situations in which recommended treatment is refused suggests that a study of this phenomenon might provide a unique window on problems in the medical care system and, in particular, on continuing diffi- culties in the doctor-patient relationship. This is particularly true insofar as refusal represents an extreme situation of strain in the relationship and thus is likely to cast existing problems in bold relief. Yet, despite the potential lessons that such situa- tions have to offer, there have been no previous empirical stud- ies of the phenomenon undertaken in the general medical set- ting. Prior discussions have been based entirely on anecdotal data. (30-34) Only one published study exists of the related problem of treatment refusal in the psychiatric inpatient setting. (35) It found the problem to be rather common, provoked by a variety of stresses, and usually short-lived. A group of long-term re- fusers was identified, however, whose refusal appeared to be the result of their psychotic illnesses, and who refused for prolonged periods of time, to the detriment of their long-term care. It was also found that refusals tended to be handled as a matter between patients and nurses, with intervention by phy- sicians relatively rare, except in prolonged refusals. It is not known, however, how typical the setting studied was of psychi- atric hospitals, much less how close the parallels are to refusal in medical settings. The present study was undertaken, therefore, to provide preliminary data on the following issues: the incidence of treat- ment refusal in medical hospitals; the factors that appear to be involved in motivating refusal; and the natural history of re- fusal, particularly as it relates to responses of medical person- nel. The implications of the data for the doctor-patient relation- ship, and from a legal and ethical point of view, for the role of the state in structuring that relationship, will also be examined. 414 Making Health Care Decisions: Appendix D Methods Refusal was defined for the purposes of this study as the overt rejection by the patient, or his or her representative, of medication, surgery, investigative procedures, or other compo- nents of hospital care (e.g., daily weighings, bed restrictions, etc.) that were recommended or ordered by the patient's physi- cian. A single episode of refusal was defined as beginning with the initial refusal of a treatment or diagnostic procedure (or a class of procedures, such as nursing care) and ending when the treatment or procedure was accepted or the recommendation for treatment withdrawn. Simultaneous refusals of multiple treatments or procedures, or refusals occurring as part of dis- crete behavioral sequences, such as admission physical exam- inations, were counted as single episodes. Refusal so defined encompasses a broader variety of inter- actions than an alternative definition might. For example, were this study to be limited only to cases in which patients never received the treatment or procedure they refused (i.e., exclud- ing all patients who later changed their minds and accepted treatment), nearly half of the refusals recorded in this study would have been eliminated from the sample. Although such a limited sample (whose dimensions are defined in the section on “Outcomes” below) might be of particular interest to those who are concerned about the “bottom line” results of re- fusal—those instances in which recommended care was never received—much of importance would be lost from the analysis. The factors that bring many doctor-patient relationships to the point of stress at which refusal occurs, many of the reactions to refusal, and the means by which refusals are resolved would pass through a sieve so coarse that it caught only unresolvable cases. The instances of refusal identified here, however, represent cases that are distinctly different from doctor-patient interac- tions in which refusal does not occur. In the cases discussed below, the patient (or his or her representative) has interrupted the dialogue with the physician and/or the course of treatment by overtly refusing a therapeutic or diagnostic intervention. This requires that the patient take an important step beyond merely expressing hesitations or asking for additional informa- tion (although, as noted below, refusal was sometimes the only way that additional information could be obtained). Thus, al- though the definition of refusal employed is in some sense a broad one, it demarcates a unique group of patients, who have the potential for telling us a great deal about the current state of medical care. The study was conducted in three stages. Stage 1. Seven wards in four different medical hospitals were studied in this stage: a medicine, a surgery, and a neurol- Treatment Refusal 415 ogy ward in a university-affiliated teaching hospital; a gynecol- ogy ward in a university-affiliated women’s hospital; an ophthalmology ward in a university-affiliated, specialty eye and ear hospital; and a medical and a surgical ward in a large community hospital. Each ward was visited by one of the inves- tigators four times, at roughly one-week intervals, with an effort made to visit the ward on a different day of the week on each occasion. The investigator met with the head nurse of the ward to review with her all known instances of treatment refusal (as defined above) that had occurred during the preceding 24 hours (from the beginning of the 7 AM-3 PM shift on the previous day to the end of the 11 PM-7 AM shift on the day of the visit). Previous experience had suggested that the nursing staff would be more likely than any other members of the hospital staff to know about instances of treatment refusal; this was confirmed in later stages of this study. The head nurses had been informed in advance of the dates of the visits and were asked to inquire about instances of treat- ment refusal in their daily “report” sessions with the nursing staff. When an instance of refusal was identified, the head nurse and the nurse most familiar with the refusal were asked about the treatment refused, the staff member who elicited the refusal, and the reason for the refusal from the perspective of the nurs- ing and medical staff. In addition, the following data were collected from the patient’s chart: demographic data (age, race. sex, marital status, religion, occupation, education) and medi- cal data (number of previous hospitalizations, diagnosis, sever- ity of illness, hospital course). An attempt was then made to interview the patient to ascertain the reason for the refusal from his or her perspective. Patients were told of the purpose of the interview and were given an opportunity to refuse to partici- pate. Only one patient refused to be interviewed. Interviews varied in length from 5-45 minutes, averaging approximately 15 minutes. Stage 2. The second stage of the study was designed to allow the gathering of data about the longitudinal course of treatment refusal in a smaller number of settings. One of the investigators made daily rounds with the house staff on two of the wards studied in stage 1, the medicine and the surgery ward of the university-affiliated teaching hospital. Rounds were made on the surgery ward for a three-week period, and on the medicine ward for an eight-week period. Information was ob- tained about instances of treatment refusal from the house staff’s daily discussions of each patient on rounds. In addition, the same investigator continued to meet with the head nurse of the ward on a daily basis to review with her, as in stage 1, those instances of refusal that had come to her attention. The same data as in stage 1 were gathered from the nursing staff and from the chart about each instance of refusal. Additionally, more explicit data were collected about the house staff's view of the 416 Making Health Care Decisions: Appendix D basis for refusal, and observations were made of the interactions between refusing patients and the house staff on daily rounds. The investigator making rounds with the ward team did not approach the refusing patients directly. After a refusing patient was identified, the second investi- gator approached the patient later in the day, explained the purpose of the interview, and asked the patient for consent to explore the basis of the refusal. All patients in this stage con- sented to the interview. Patients were then asked: why they refused the recommended treatment; how they understood the nature of the treatment or procedure refused; what they viewed as the potential consequences of the treatment or procedure; what they viewed as the potential consequences of the refusal: and what role, if any, was played in their decision to refuse by their doctors, the nursing staff, their own past medical experi- ences, the past medical experiences of those they knew, and the opinions of their family and friends. The interviews averaged about one-half hour in length. Stage 3. The third stage of the study was intended to offer an opportunity for in-depth exploration of individual cases of treatment refusal, utilizing a “case study” approach. Refusals were identified by monitoring, for a three-month period, the computerized report sheets used by the three nursing services (medicine, surgery, and intensive care) at the same university- affiliated teaching hospital studied in stage 2. These sheets contained hand-written notes of the nursing supervisors, com- menting for each patient on important developments, including refusals, in the preceding 24 hours. Although not every refusal was noted on these sheets, and not every refusal noted was followed up, the nursing sheets served as a convenient access to the more significant cases of refusal throughout the hospital. A small number of additional cases was identified from referrals to the psychiatric consultation service. When a refusal was identified, one of the investigators would obtain the approval of the attending physician to ap- proach the patient for informed consent to participate in the study. If the attending physician refused to give consent, for whatever reason, the patient was not entered into the study. This occurred in three cases. No patient whose physician had already agreed refused to participate. When both physician and patient agreed, the procedures outlined above for stage 2 were followed. In addition, an attempt was made to conduct as much of a clinical psychiatric interview with the patient as possible, including the elicitation of a personal history, family history, social history, and a brief mental status examination. Whenever possible, at least one member of the family was contacted (with the patient's consent) to verify the patient’s account and to shed additional light on the patient's functioning prior to admission and on factors involved in the patient's refusal of treatment. The Treatment Refusal 417 patient's chart was monitored on a daily basis and an effort was made to interview the patient at least every other day for the remainder of the hospital stay. These subsequent interviews frequently offered an opportunity to obtain information that was not available in the initial encounter. Results The results of the study have been organized into the fol- lowing sections: epidemiology of treatment refusal, treatments refused, reasons for refusal, responses to refusal, outcomes of refusal, the effects of refusal on the doctor-patient relationship, and family participation in treatment refusal. Summary. Treatment refusal, as defined in this study, was found to be a relatively frequent occurrence in the hospitals studied, with its incidence ranging between 0.7 and 11.4 re- fusals per 100 patient-days. The types of treatment refused were almost evenly divided between medications, diagnostic pro- cedures, and adjunctive care; refusal of surgery was much less common. Although most refusals were not of life-threatening dimensions, refusals that threatened to seriously impair the patient’s medical care were observed. Multiple factors appeared to interact to cause most pa- tients’ refusals. Among the common elements involved were failures in doctor-patient communications, problems in trust, psychological and psychopathological factors, patient weari- ness of treatments and procedures (“hospital fatigue syn- drome”), patients’ idiosyncratic beliefs, legitimate objectives to the proposed treatments, and a desire for “death with dignity.” Physicians’ (and other care-takers’) responses to refusals tended to vary along a spectrum of activity and coercion with the seriousness of the refusals for patients’ care; responses were generally not individualized as a function of the reasons for patients’ rejection of treatment. Thus, although refusal most often ended in ultimate acceptance of treatment, outcome fre- quently appeared to depend on a serendipitous association between the patient's psychological motivation and the physi- cian’s response; the persistence of the physician in seeking to reverse the patient's decision also played an important role. Refusal often created a sense of guilt in patients, but occasion- ally positive effects were seen, as well. Physicians, too, had difficulty in dealing with refusals. Their sense of rejection sometimes aroused anger and led them in turn to reject the patient. The role of families in the process was especially prob- lematic. They were often torn between empathy for the patient and a desire to see their own needs fulfilled, which often required that the patient accept the proffered care. Families tended to be even less informed decisionmakers than patients. 418 Making Health Care Decisions: Appendix D Epidemiology of Treatment Refusal Stage 1. Thirty-one refusers accounting for 33 episodes of refusal were identified in stage 1 (see Table 1). The refusing sample was predominantly white (68%), female (68%), and married (65%). Forty-five percent were Protestant and 32% Catholic (of 29 patients for whom data was available). Mean age was 50.7 years (range 22-79). Mean level of education was 11.8 years, with 67% having at least a high school education (of 21 patients for whom data were available). Table 1: Refusals in Stage 1, by Site Refusals Number of | Number of per 100 Site Refusals [Patient Days | Patient Days University Medical 15* 132 11.4 University Surgical 9 135 6.7 Gynecology Ward (Women’s Hospital) 3 73 4.1 University Neurological 2 54 3.7 Community Medical 3 132 2.3 Community Surgical 1 151 0.7 Ophthalmology Ward (Eye & Ear Hospital) 0 39 0.0 Total 33 716 4.6 * Two patients refused twice. Refusals in this stage occurred at an incidence of 4.6 per 100 patient days (see Table 1). This finding suggests that treat- ment refusal is a relatively frequent occurrence; based on this data, for example, a medium-sized hospital with 400 inpatient beds might expect 18 refusals to occur on an average day. As noted in Table 1, however, the rate of refusal varied widely from hospital to hospital and from ward to ward, rang- ing from a high of 11.4 per 100 patient-days on the university- affiliated teaching hospital’s medicine ward to a low of 0.7 per 100 patient days on the surgery ward of the community hospi- tal. (No refusals were reported on the ophthalmology ward, but the small sample size there makes interpretation difficult; the nurses on this ward reported that refusals do, in fact, occur.) Although the wards studied in this stage represent too small a sample to allow any firm conclusions to be drawn about the relative distribution of treatment refusals, it is possible to point to some interesting trends. The highest rates of refusal were, found on the wards of the university-affiliated teaching hospitals (excluding the ophthalmology ward); the wards of the community hospital had the lowest rates of refusal. Looking only at teaching institutions, the general hospital wards (medi- Treatment Refusal 419 cine and surgery) exhibited higher rates than the specialty wards within the same hospital. The medicine wards displayed markedly higher rates of refusal than the surgery wards in both university and community settings. Stage 2. During three weeks of observation on the surgery ward, ten refusers were identified, accounting for 13 episodes of refusal. Given the small number of patients involved, and their demographic similarity to the 45 patients on the medicine ward, who accounted for 60 episodes of refusal over a period of eight weeks, the data from the two groups will be combined in this analysis. The sample in stage 2 was predominantly white (78%), male (53%), and married (55%). Forty-seven percent were Prot- estant and 41% Catholic (of 49 patients whose religion was known). Mean age was 60.8 years (range 23-89). Mean level of education was 12.2 years, with 72% having at least a high school education (of 39 patients for whom data were available). The incidence of refusal in this stage was 3.0 per 100 patient-days on the surgery ward and 4.9 per 100 patient-days on the medicine ward. These represent substantial decreases from the rates on these wards reported in stage 1 and are, in fact, closer to the average rate of refusal for all wards in stage 1. The differences between stage 1 and stage 2 incidence figures point out the need for extreme caution in basing comparative analy- ses on the stage 1 data, but data from both stages confirm that refusal is a common event in medical hospitals. Another way of demonstrating the frequency with which refusal occurs, and the magnitude of the problem, is to estimate the percentage of all patients who refuse at one time or another during the course of their hospitalization. The most compre- hensive data available for this purpose are from the medicine ward in stage 2. Two hundred forty-two patients passed through the ward during the eight weeks of intensive study. Of these, 45 (19%) were noted to have refused at least one treatment or diagnostic procedure. In all likelihood, this represents an un- derestimate of the percentage of patients who refused at some point in their hospital stay: those patients who were in the midst of their hospitalizations when the study began and ended, as well as those patients who were transferred to and from other floors of the hospital, may have had refusals that were not recorded as part of this study. Additional refusals occurring on the study ward may also have been missed. Treatments Refused. The types of treatment refused in stages 1 and 2 are shown in Table 2. Nearly one-third of refusals involved medications. Approximately one-half of the medica- tions refused were prescribed to increase patient comfort or to treat minor discomforts (e.g., laxatives, anti-emetics, analgesics, and some antacids) rather than to treat primary illnesses. A 420 Making Health Care Decisions: Appendix D Table 2: Refusals in Stages 1 and 2, by Treatment Refused Treatment Refused Stage 1 Stage 2 Total Medications 15 20 35 (31%) Laxatives 4 1 5 Antacids 3 1 4 Anti-Emetics 3 3 6 Intravenous Fluids 0 6 6 Cancer Chemotherapy 1 3 4 Other 4 6 10 Diagnostic Tests 7 32 39 (34%) Blood Drawing 2 7 9 Weighing 2 5 7 X-rays 1 8 9 Other 2 12 14 Adjunctive Care 12 25 37 (32%) Enemas 2 1 3 Nasogastric Tubes 2 5 7 Nursing Care 2 8 10 Other 6 11 17 Operative Procedures 1 2 3 ( 3%) Total 35* 79* 114* * Totals exceed total number of refusals in each stage because some refusals involved more than one category. number of cases, however, involved refusals of medication re- quired for care of acute or otherwise life-threatening conditions. These included four refusals of cancer chemotherapy, a refusal of potassium in a patient who was severely hypokalemic, and several refusals of intravenous fluids in severely dehydrated patients. Diagnostic tests, which accounted for just over one-third of refusals, were, for the most part, not related to urgent issues of patient care (e.g., weighings and many blood drawings). Here, too, however, there were a number of instances in which re- fusals of diagnostic procedures potentially placed the patients’ lives in jeopardy. These included refusals of a CAT scan for acute head trauma, workup for a fever of unknown origin, a chest x-ray for cardiac failure due to possible pericardial effu- sion, a lumbar puncture for a confusional state, nasogastric tubes for acute upper gastrointestinal bleeding. All three opera- tive procedures refused were diagnostic biopsies: an axillary node, mediastinal nodes, and the synovium of the knee. The first two of these cases involved suspected carcinoma; the last, unbeknownst to the patient, was for research purposes. Treatment Refusal 421 Adjunctive care refused, accounting for another one-third of cases, was frequently routine, but nonetheless significant for patient care. Included here were dressing changes, care of pres- sure ulcers, bed restrictions, placement of nasogastric tubes for gastric decompression, and enemas. Reasons for Refusal. Data in this section are drawn from all three stages of the study. It became clear as the study progressed that it was usually impossible to identify a single factor (or even a group of factors) that could definitively be said to have “caused” a given refusal. Patients, doctors, nurses, and the investigators frequently differed as to their views on why an individual patient refused. Patients themselves often had diffi- culty offering reasons for their refusals based on introspection, and, even when they were able to point to causative influences, other factors frequently appeared to be involved, as well. There are, in addition, considerable epistemological problems in infer- ring causation merely from the presence of a particular factor in any given case. For all of these reasons, a statistical enumera- tion of the number of cases in which one or another causative factor was present is problematic. An effort has been made, however, in Table 3 to separate apparent primary and secondary reasons for refusal in each stage of the study. Only the first refusal of each subject was included, to avoid the distorting effects of multiple refusals by the same subject occurring for the same reason. The arbitrariness of identifying one factor as the primary one, and the likelihood that many of the interrelated factors were never identified, argue for extreme caution in inter- preting the numbers offered in Table 3. Rather than emphasiz- - ing a strictly quantitative approach, therefore, this section will present a description of the most common factors that appeared to be associated with treatment refusal. Case examples will be used to illustrate these factors, and to emphasize that it was uncommon for any one factor to stand in isolation. The im- plications of this multi-causal model will be discussed below. Problems in doctor-patient communication Inadequate provision of information to patients Patients not informed about treatments or procedures or- dered. The investigators frequently observed treatments or diag- nostic procedures being ordered without patients being aware of this fact. This occurred, for example, on daily ward rounds, when the team of house officers agreed, after seeing the patient, that a particular test or change in medication was required. Orders would then be written without the patient being told. Only sometime later, when the person implementing the order, often a nurse or orderly, appeared at the patient's bedside, would the patient discover what was planned. This common practice was a factor in 8 of the 105 refusals in Table 3. 422 Making Health Care Decisions: Appendix D Table 3: Reasons for Refusal Stage 1 Stage 2|Stage 3 Reasons PrS*=*i1P 3 P S [Total Problems in Communication Patients Not Informed About: Treatments or Procedures Ordered og 3 1 3 0 1 8 Purpose of Treatments or Procedures 4 2 |6 11 2 2 27 Risks of Treatments or Procedures 01 (1 0 |0 2 4 Conflicting Information Provided to Patients 3 0 (3 0 |1 0 7 Difficulties in Conveying Medical Uncertainty 0 0 |0 1 [1 1 3 Problems in Trust Lack of Trust Due to Previous Bad Experiences 1 1 (2 2 (2 0 8 Lack of Trust of a Particular Physician 2 1 (1 0 |0 2 6 Psychological and Psycho- pathological Factors Delusions 0 0 |0 0 [1 1 2 Depression 0 09 j2 0 12 0 4 Denial 1 0 [3 1 1 0 6 Characterologic Factors 8 312 5 1 3 22 Other Psychoses (including OBS) 2 2 n5 2 2 0 23 Situational Reaction 0 3 1 1 0 5 Hospital Fatigue Syndrome 1 0 (3 0 [1 0 5 Other “Legitimate” Refusals 7 0 [5 0 (0 O 12 Idiosyncratic Refusals 01 (3 0 (1 oO 5 “Death with Dignity” Refusals| 1 0 [4 0 |1 1 7 Other 0 0/0 0 [1 0 1 Unknown 1 0 (1 0 [1 oO 3 t Only the first episode of refusal for each patient was included. * Primary reasons for refusal. ** Secondary reasons for refusal. Case example #1—A 32-year-old housewife with a 12- year history of ulcerative colitis refused a chest x-ray several days after a partial colectomy had been per- formed. Questioned afterwards about the reason for her refusal, she reported that no one had told her that the x- ray had been ordered or why it was being done. After the Treatment Refusal 423 head nurse on the ward came to speak with her, explain- ing the purpose of the radiologic examination, she agreed to have it performed. Although inadequate information about the procedure was a recurring theme in our discussions with patients, it was apparent that this factor was by itself an inadequate explanation for refusal. The failure to inform patients of treatment or diag- nostic procedures that had been ordered was an extremely common practice. Yet, the vast majority of patients willingly cooperated with procedures, even when they lacked advance knowledge of them. Failure to provide information appeared to trigger refusals in patients who had already been “primed” to refuse for some other reason or group of reasons. For example, the patient described above had previously refused pre-opera- tive enemas. Exploration at that time revealed a highly am- bivalent attitude to the proposed surgery. At surgery, a sigmoid abscess was found, thus making completion of the originally planned colectomy impossible. Instead, a partial procedure was performed, with plans made to complete the operation in six months’ time. The patient was upset about the complications and resulting delay, and felt that she was not getting adequate support from her surgeons. It was in this context that the refusal of the x-ray occurred. This observation will recur repeatedly in the data pre- sented below. A precipitating factor in the patient's refusal could often be identified, but it almost always was impossible to attribute refusal to that factor alone. Rather, the precipitating factor often seemed to be the “last straw” for the patient, whose relationship with his or her care-givers previously had been subjected to a variety of other stresses. Patients not informed about purpose of the treatment or procedure. Patients were often told that a procedure was going to be performed (e.g., “We'll be drawing some blood from you for tests”), but in such a manner that they had little idea of the purpose of the procedure. Asked to trade off their own comfort for a procedure whose value was obscure to them, some pa- tients (27 of 105 first refusals) refused to allow the procedure to be undertaken. Case example #2—While in the intensive care unit, an 89-year-old woman with congestive heart failure as a result of arteriosclerotic disease of her coronary vessels refused to permit arterial blood gases to be drawn. The technician who had approached her to draw the blood appparently responded to her query about the necessity for the procedure by saying, “You can refuse if you want to.” The patient thereupon refused, explaining her action by noting, “No one would want to have those stabs, those sticks. Would you?” Further exploration of the issue re- vealed that she was uncertain, because of the technician's 424 Making Health Care Decisions: Appendix D answer, whether her doctors actually had ordered the test. She knew neither the purpose of the test (hypoth- esizing that perhaps drawing the blood would make her “more relaxed”), nor the possible consequences of refus- ing it. When these were explained to her by the house officer and her attending physician, she consented to the procedure. A complicating factor in this patient's refusal was the infor- mation she received from the technician, which she interpreted to mean that the test really was not necessary. The problem presented by multiple sources of information will be consid- ered again below. Patients not informed about risks of the treatment or pro- cedure. Even when patients were told clearly about the plans for their care and the reasons behind them, failure to discuss the potential risks of the procedure sometimes created stum- bling blocks to patient consent. This was the case in 4 of 105 first refusals. Case example #3—This 60-year-old woman, a Jehovah's Witness with chronic obstructive pulmonary disease and congestive heart failure, was told by the medical student who was responsible for her care that she would be scheduled to have a test to “see how well your heart moves.” The patient agreed to undergo the procedure. Later in the day, however, after discovering in a conversa- tion with a nurse that the test (a MUGA scan) involved the injection of a radioactive isotope into her vein, she refused to go with an orderly to the nuclear medicine suite. The patient described herself subsequently as “a woman who likes to know what is going to happen to me, before it happens.” She explained, that her refusal stemmed from the discovery that radioactive materials, which she knew were potentially harmful, were involved in the test. When the medical student returned to reas- sure her that the amount of radiation involved was no more than would be generated by two chest x-rays, she concluded that it would be safe to have the scan and the procedure was performed. In this case rather simple information was sufficient to allay the patient’s anxieties. Rather than being calmed by omis- sion of the potential risks of the procedure—the argument of- fered by many physicians for scaling down the scope of their disclosures—this patient was subject to considerable anxiety when she discovered from other sources that radioactive mate- rials were involved. Conflicting information provided to patients. In a modern medical hospital responsibility for patient care is divided among a large number of care-givers from a variety of disci- plines. Patients may be seen by an attending physician, who has Treatment Refusal 425 overall responsibility for their care; a referring physician, with whom they may have a long-standing relationship and who, despite having surrendered control of their inpatient care to a hospital-based physician, may continue to visit them; a house officer (intern or resident) or medical student, who takes re- sponsibility for coordinating their care; several other house officers on the ward team, who may jointly participate in deci- sions about treatment and diagnostic procedures; a number of consulting specialists, each of whom may be preceded by the resident or fellow attached to their service; at least three nurses per 24-hour period assigned to their room; and a variety of personnel from other health professions, including physical therapists, respiratory therapists, IV technicians (who may also draw blood), and social workers. The result of contact with such an imposing array of professionals, almost all attired in indis- tinguishable white coats, is often profound confusion over who, in fact, is responsible for their care. Or, as one patient put it in the middle of a prolonged admission process, “Everybody here has to examine me, but no one has the authority to give me a pill for my pain.” All of the care-givers enumerated above convey informa- tion to the patient. The large number of care providers, many of whom are not in direct communication with each other, makes conflicts in information inevitable. Since patients are often unsure exactly who is making the final decisions about their care, they are frequently uncertain whose information to trust when facts conflict. Such confusion contributed to 7 of the 105 first refusals. One instance of contradictory communication leading to refusal was described above in case example #2; a further example follows. Case example #4—This patient was himself a physician with chronic low back pain. He had undergone many previous operative procedures and was now referred in for bilateral decompression of the affected nerve roots. The procedure had been explained to him in detail by an orthopedic specialist in his home town. When he arrived at the hospital, he found that the neurosurgeon who was to perform the operation was ill and would be unable to visit him for several days. In the meantime, the house officers on the service told him that the best procedure for dealing with his type of pain was electrocoagulation of the nerve roots, and that this was what he would be scheduled to have. He had never heard of this procedure and found the sudden switch extremely anxiety-provok- ing. At one point he sought out a resident—from which service he was uncertain—who was passing by on the floor. The resident told him (apparently incorrectly) that electrocoagulation was an imprecise procedure that might result in a total loss of sensation in his legs. This made him even more anxious. 426 Making Health Care Decisions: Appendix D Finally, after three days of hospitalization, the neu- rosurgeon appeared. As the patient described it, “He swept in with a retinue of residents and said that he would look at my x-rays and get back to me.” There was no opportunity (even for a fairly assertive physician- patient) to ask questions. The next day, when the neu- rosurgeon returned, he informed the patient that he would, in fact, undertake an operative procedure, but that it would be substantially different from the one the patient’s orthopedist had originally described. The pa- tient was again upset by this switch. Shortly thereafter, a resident came to talk to him about the risks of surgery. He began by saying that he assumed the patient, being a physician, was already aware of the risks. The patient urged him to inform him about them anyway. The resi- dent said blithely (according to the patient), “Well, we could cut your nerve with the air hammer we use, which would leave you paralyzed, you might get osteomyelitis, or you might bleed to death.” The patient was further shaken by this news. As he explained it, “I had carefully weighed out the risks and benefits of having the procedure, and now, the day before the surgery, the whole balance was changed by an en- tirely new procedure with risks I had never heard about before.” That night the patient became convinced that the planned operation was improper and that an entirely different procedure, involving repair of a previous fu- sion, was really what he needed. He asked the nurses to contact the attending neurosurgeon and was told that he was out of the hospital and unavailable. The same was true for the chief resident. The only resident who was in the hospital was handling two emergencies downstairs and could not come up to talk with him. Shortly thereaf- ter, a nurse approached him with the consent form for surgery the next morning. He refused to sign. The patient in this example presents the paradigmatic case of conflicting information leading to refusal of treatment. His referring physician, attending physician, and house officers had all given him different information about his condition and about which treatment would be best. Even though he himself was a physician, and had read extensively about his condition in the medical literature, he was left confused and frightened by this process. His attempts to seek out still other sources of data, such as the resident who was passing in the hall, left him even more anxious and bewildered. Of importance, as well, was the lack of opportunity to have his questions answered and his fears relieved; this was true for several days preceding the planned surgery and, most crucially, on the night before the procedure, as well. The nurse who had been assigned (as was the practice in this hospital) to obtain the formal consent for Treatment Refusal 427 surgery, was utterly unable to allay his fears, knowing less about the surgery than the patient himself. Conflicting communications were clearly the precipitating factor in this refusal, but once again there were underlying factors present. The patient had an extremely rigid, obsessive character structure. He carried with him a file of reprints from the medical literature on his condition, which were dog-eared and extensively underlined. It was important to him to under- stand every aspect of his condition and its treatment. This predisposition made the subsequent communication difficul- ties even more intolerable to him. Nonetheless, before writing off this case as reflecting largely the patient’s characterologic difficulties, it must be noted that, in the end, the patient was right. After his refusal, the neurosurgeon came to see him and spent a good deal of time answering his questions. He agreed to call in an orthopedic consultant to review the patient's x-rays and examine him. The orthopedist confirmed that the patient's explanation for his pain, namely the instability of the previous fusion and move- ment in supporting rods, was most likely true. Instead of the previously planned procedure, therefore, the patient under- went an operation for removal of the loose rods and regrafting of the fusion site. Although at all times avowing profound respect for his neurosurgeon (if not for the house officers), the patient himself noted that there was something about the hospital structure that appeared to facilitate communicative miscues of the sort he endured. Difficulties in conveying medical uncertainty. To attribute difficulties in communication entirely to physicians’ failure adequately to inform patients is entirely too facile an explana- tion for a very complex set of phenomena. The investigators observed a number of occasions on which patients claimed to be unaware of important informtion, while the responsible phy- sicians maintained, and observation confirmed, that the infor- mation had been conveyed. Some of these instances can be accounted for by certain psychological phenomena, such as patients’ denial, as will be discussed below. In other cases, however, patients’ difficulties in grasping the uncertainties in- herent in medical diagnosis and the largely probabilistic nature of decisionmaking in medical care appeared responsible for the problems that arose. Three of the 105 first refusals fell into this category. Case example #5—This 56-year-old, widowed black woman had originally been hospitalized for workup of a suspected brain tumor. During angiography, she suffered laceration of her iliac artery, requiring an emergency surgical repair. Postoperatively, she developed a pro- longed fever that may have been due to abscess formation somewhere in the surgical field. She was first seen in 428 Making Health Care Decisions: Appendix D this study when she refused to have blood drawn for several diagnostic tests. The patient was clearly angry that “mistakes” by the surgeons had prolonged her hospitalization. She was convinced that too much blood was being taken from her, although she did not know what was to be done with the blood or what the consequences might be of not having the tests performed. Furthermore, she and her family, of whom at least one member was always present, believed that the doc- tors were deliberately withholding treatment in order to justify doing more tests. The patient knew that antibiot- ics were used to treat “fevers” and believed that if she were given antibiotics she would get better. She refused to accept the house officers’ explanations, which were observed by the investigators and were offered repeatedly and clearly, that they could not institute treatment until they had definitively identified the underlying condition causing the fevers. After the initial refusal of blood work, the patient intermittently accepted and refused blood drawings. Ul- timately, three weeks after the operative procedure, with fevers continuing and no treatment begun, the patient refused any further diagnostic evaluation and was dis- charged from the hospital. As always, multiple factors were involved in this patient's potentially quite damaging refusals. She and her family, who were black, appeared distrustful of the entirely white team of house officers. In addition, she was angry at the mishap, for which she blamed the surgeons, that resulted in emergency operation and led to the subsequent fevers. But an important component of the evolving difficulties between the patient and the treatment team appeared to stem from the problems she had in recognizing that the physicians truly did not understand the cause of her fevers and were quite uncertain how to proceed. Case example #6—The patient was a 24-year-old man who had suffered from severe chronic pancreatitis for three years. He had had three previous surgical interven- tions, none of which relieved the pain that had made him dependent on narcotics. This hospitalization had been arranged for the purpose of anastomosing two ducts in his pancreas to drain a presumed pseudocyst. Although this was the procedure that the patient had expected when he entered surgery, while he was on the table a 95% resection of his pancreas was performed. One week postoperatively, when he was started back on oral fluids, his pain recurred. He was placed on IV fluids and, when entered into the study, had not had anything to eat for nearly 2-1/2 weeks. Treatment Refusal 429 The index refusal was for the placement of a central venous line for hyperalimentation (intravenous provi- sion of basic nutrients). At first the patient explained his refusal on the grounds of previous difficulty with insert- ion of central lines. It soon became clear, however, that he was tremendously angry at his care-givers. He said that he was upset, in part, because no one ever told him anything “for sure” about his condition. This was not a complaint that his physicians failed to talk to him, but rather that they could never tell him what caused his pancreatitis, whether any of the procedures they pro- posed would definitively resolve it, or what the future course of his illness would be. The patient’s attending surgeon confirmed this view by noting that the initial recurrence of the patients pain postoperatively was provoked by his obtaining illicit milkshakes (which stimulated pancreatic secretion) from other patients. He noted that the patient could not under- stand the need for a prolonged recuperative period. The patient wanted to be well at the end of one week after the operation, and thus refused to adhere to dietary re- strictions or to consent to the placement of a central line that would allow his gastrointestinal tract a prolonged period of rest. (Other aspects of this patient's anger will be considered below.) In our experience, failure of patients to grasp the uncertain- ties inherent in medical care could not be laid exclusively at the feet of physicians. Characterologic factors (as in case #6) were often involved. At other times, patients appeared to have intel- lectual difficulty with the concept of uncertainty. In particular, it was hard for them to envision a diagnostic paradigm that called for taking progressive steps to reduce uncertainty. Given the problems that medical students and many young physi- cians have in learning to tolerate uncertainty, this response is not surprising. In fact, physicians learn to deal with uncertainty more by means of experience than by any intellectual mastery of the concept. This may account in part for the difficulties they have in explaining such situations to patients; that is, physi- cians may know what to do when faced with uncertain situa- tions, but they may not be able to articulate the reasons for the choices they are making. (36) Thus, the uncertainties in the situation handicap both doctors and patients. Problems in trust Lack of trust due to previous bad experiences. The law characterizes the doctor-patient relationship as “fiduciary” in nature. That is, by virtue of superior knowledge and given the trust with which the patient endows the relationship, the physi- cian is obligated to act at all times in the best interests of the patient. This conceptualization is an acknowledgment that ef- 430 Making Health Care Decisions: Appendix D fective provision of medical care depends on the patient’s trust of the physician. In the absence of such trust, despite the physicians presumed expertise, the patient will not ordinarily go along with the physician's recommendations, and medical care will be stymied. One of the most frequent reasons for a failure of trust is the patient’s previous bad experiences with the medical care system. This occurred 8 times in the 105 first refusals studied. Case example #7—The patient, a 51-year-old married man, was admitted after suffering multiple injuries in an auto accident. He had a history of diabetes mellitus and alcoholism. At the time of admission he was intoxicated. Because of trauma to his ribs and lungs, immediate oper- ative repair of his fractured leg could not be performed under general anesthesia. It was planned to use spinal anesthesia instead. The patient's refusal of spinal anes- thesia occurred despite his knowledge that this refusal would mean a delay of up to 10 days, until he would be a suitable candidate for general anesthesia. He also under- stood that the delay meant a greater chance of lasting complications at the fracture site. The patient's refusal was based on the experience with spinal anesthesia 14 years previously. At that time, he recalled, he had nearly passed out on the table, while the doctors raced about madly trying to revive him. From his description it appeared that he had suffered a hypo- tensive episode. Regardless of the amount of encourage- ment he received to consent, he was unable, based on his previous experience, to accept his physicians’ assurance about the safety of spinal anesthesia. He continued to refuse and underwent a successful operative procedure under general anesthesia one week later. Previous bad experiences with recommended procedures appeared to be one of the most powerful determinants of pa- tients’ decision making in this study. A related factor was the previous bad experiences of other people known to the patient. For example, one woman refused chemotherapy for cancer be- cause it had not helped her daughter-in-law. On the other hand, she was more than eager to have radiation therapy. Patients who refused for these reasons appeared to hold by the maxmim, “Experience is the best teacher”; they were unshakable in their determination to resist their doctors’ recommendations, Lack of trust of a particular physician. A variant on this phenomenon was the refusal of patients because of a previous bad experience with the individual physician in question, even if the procedure this time was different. The patient in case #6, for example, who was refusing placement of a central venous line, appeared to harbor an underlying suspicion of his surgeon Treatment Refusal 431 because of the unexpected switch in the operative pro- cedure—from a ductal anastomosis to a subtotal pancreatec- tomy—while he was on the operating table. Teaching hospitals are uniquely susceptible to failures of trust induced by patients’ uncertainty about the capabilities of the house officers who are responsible for their day-to-day care. Case example #8—This 33-year-old woman with long- standing, insulin-dependent diabetes mellitus refused to have a blood sugar drawn at 3 AM. When she was inter- viewed, she demonstrated a great deal of anger at the house officers who had been taking care of her. She said that they sent off a large number of unnecessary tests, “for their benefit, but not for mine.” She maintained great trust in her private physician, however, and was angry that the intern now on her case would not listen to her physician’s suggestions. Subsequently, a new intern rotated on to the service, and he and the patient's private physician agreed that blood would be drawn only twice daily for blood sugar determinations. The patient pro- nounced herself satisfied with this arrangement. Mistrust of house officers extended to more significant diagnostic and operative procedures, as well. It was not uncom- mon for patients to refuse to sign consents for surgery after the house officer explained the procedure to them. This was true even when the house officer indicated that their attending phy- sician had endorsed these plans. The next morning, however, when the patients were able to confirm that the attending had, in fact, ordered the procedure in question, they generally signed without any hesitation. Failure of trust was not limited to the house staff. Patients often responded similarly to unfamiliar hospital-based attend- ing physicians, demanding the endorsement of their primary physician before they would agree to undergo a procedure. Overall, this phenomenon was seen in 6 of the 105 cases of first refusals. Psychological and psychopathological factors. Psycho- logical factors are ubiquitous in treatment refusal. As noted above, refusal is almost always the result of extrinsic factors (e.g., impaired communication) acting on patients predisposed by their cognitive and affective states to reject the recom- mended procedures. To attempt to denote particular refusals as “caused” by psychological factors, therefore, is difficult. None- theless, the investigators observed 44 of 105 first refusals in which psychological, and in the extreme psychopathological, factors appeared to play a predominant role, with situational elements making a relatively lesser contribution. In 18 other cases, psychological elements played a significant secondary role. 432 Making Health Care Decisions: Appendix D In some cases, there was little question about the presence of psychopathology. Seven cases (from all three stages of the study) involved patients with senile dementia; many of these patients were grossly disoriented, combative, and resistive of all care. In two other cases, patients were suffering from delirium tremens (DT’s) during acute withdrawal from alcohol addiction. Also observed were two cases of acute brain syndrome and a patient who experienced a grand hysterical episode. Yet, it was not always easy to identify cases in which less dramatic psychopathological factors directly affected patients’ decisions. The cases presented below represent a gradation of severity of psychological disorder, ending with examples of characterologic difficulties that many might consider to be within the spectrum of “normal” functioning. Delusions Case example #9—This 63-year-old man, who was being treated for a pulmonary abscess, had refused bron- choscopic drainage of his lesion. Despite recurrent epi- sodes of shortness of breath and a fever, which had per- sisted for six or seven months, the patient maintained that he had no problem in his lungs at all. Rather, he insisted, the problem was in his nose. Exploration with the patient revealed that this man had become convinced several years ago that something was stuck in his nose. A normal ENT examination did not persuade him that this was untrue. Instead, he began trying to dislodge the imagined foreign body with pieces of straw from a broom. So persistent was he that he punctured the nasal septum. He now believed that all of his problems re- sulted from pieces of straw that were still lodged in his nose; if his doctors could remove them, his fever and breathing difficulties would disappear. All efforts to per- suade him otherwise failed. He died several days later. This was a relatively clear-cut case of refusal based on delusional premises. Yet, as illustrated by the next case exam- ple, it is not always so easy to link causally a patient's psycho- logical problems with the refusal. Depression Case example #10—Eight months after amputation of her left leg due to vascular complications of diabetes, this 63-year-old married woman was refusing amputa- tion of her gangrenous right leg. When interviewed, this woman said quite directly that she wanted to die. The last year had been very difficult for her. Once a socially active person, she had become noticeably withdrawn after the first amputation. Her medical problems had multiplied and she had spent a good part of the year in the hospital. Now she was faced with the prospect of Treatment Refusal 433 losing her other leg, a situation that she was sure would result in her being confined to a wheelchair for the rest of her life. She discounted her physicians’ reassurances that other people with similar disabilities had learned to use double prostheses to walk independently. The patient was considered by those care-givers who knew her best to be depressed. This was evidenced by her withdrawal from those around her, slowed speech, sad affect, and disturbed sleep. Nonetheless, her physi- cians were uncertain whether her choice of death was an “irrational” one. Several of them noted that they might make the same choice themselves, were they in her situa- tion. An experienced legal-psychiatric consultant specif- ically noted that he did not believe that this woman'’s decision was so clearly rooted in her depression that a court would find her refusal incompetent. This case points out the difficulty in establishing causal links between a patient's mental state and the decision to refuse treatment. Although this paient was clearly depressed, many of her physicians felt that they could empahize with her choice on “rational” grounds. Only after treament did it become apparent how much her decision was a manifestation of her depression. Case example #10 (continued)—Following several visits from a psychiatric consultant and three days after treat- ment with an antidepressant was initiated (which or- dinarily takes 7-14 days before an effect is evident), the patient consented to amputation of her leg. Her mood was clearly brighter. She explained that her decision had been precipitated by an attempt to walk on her leg, which caused a great deal of pain and led her to recognize how badly the gangrenous leg smelled. She no longer wanted to die, noting that even if she spent her life in a wheel- chair, it would be less painful than the state she now endured. She also spoke hopefully of learning to walk with artificial limbs. Psychiatrists are taught in their training about the dangers of establishing a “false empathy” with depressed patients; that is, they are warned how easy it is to be convinced by the depressed person that things really are as bad as the depression makes the patient feel. Such a false empathy was apparently established here. Despite several physicians’ conclusions that had they been in the patient's situation they would have de- cided similarly, when the patient's depression improved, the refusal resolved. At least one other case in this study displayed the same phenomenon, as physicians backed off from challeng- ing a depressed physician-patient’s refusal of resuscitation on the basis that it represented a realistic assessment of his posi- tion. Only over a period of several weeks did it become appar- ent how influential the patient’s depression was in his decision. 434 Making Health Care Decisions: Appendix D In this case, identification with another physician appeared to hinder the assessment of the patient's mental state. (The prob- lem here will be discussed further in the section on responses to refusal.) Of all the psychopathologic processes associated with re- fusal, depression is often the most difficult to recognize, be- cause it masquerades as, “Just the way I would feel if that happened to me.” It is also the most difficult to link causally to the refusal, because, unlike the grossly delusional patient in case example #9, the depressed patient is frequently able to offer “rational” explanations for the choices that are made. Denial. Denial is an essential mechanism of defense. If we were not able to put out of our conscious awareness most of the potential difficulties we might face, our ability to act would be paralyzed. When denial itself forestalls action, however, partic- ularly action that is required to preserve life, it has assumed clearly dysfunctional dimensions. Case example #11—This 52-year-old executive had re- cently been diagnosed as having cancer of the colon that had spread to nearby lymph nodes, which were removed. The usual treatment, although its effectiveness is uncer- tain, called for chemotherapy to follow surgery, on the grounds that other small metastatic foci probably re- mained. This patient, however, refused to go along with the recommended chemotherapy, at least immediately. He said his reason for delaying treatment for up to two months, which increased the chances of a recurrence, related to the 25% chance of significant nausea and vomiting that might accompany medication protocol. The patient was looking forward to spending some time with his family and did not want to feel unwell as a result of the side effects. Although this decision appeared “rational” at first glance, further discussion revealed that the patient did not acknowledge the previous presence of the malignant lymph nodes or the possibility of spread elsewhere in his body. He chose to view the. recommendation for chem- otherapy as “preventive” medicine, not as a response to existing tumors. His view was that the surgery had cured him. Thus, he could recognize no danger in delaying the initiation of chemotherapy for up to two months. This patient's denial of the seriousness of his illness might be seen by some as a positive affirmation of his will to live in the face of life-threatening illness. Because of the potential consequences of delaying treatment, however, it also repre- sented a maladaptive response to his situation. Denial often appears in patients with strong needs to re- main in control of their situations. These patients also fre- Treatment Refusal 435 quently manifest other elements of an obsessive personality structure, being intense, hard-working, perfectionistic, and moralistic. The following case demonstrates the primacy of such characterologic factors in a patient in whom denial also played a role. Characterologic factors Case example #12—We first became aware of this pa- tient when he refused a blood transfusion. He was a 75- year-old engineer, who was admitted in a debilitated state of sepsis secondary to metastatic carcinoma of the bladder of 13 years duration. The patient had quite an involved explanation for his refusal. He was convinced that the intern who had recommended the transfusion “had not studied my case in depth.” He reiterated over and over again how important it was for doctors to know their cases thoroughly. At that, he pulled out a log book containing the results of every blood test he had under- gone since his condition had first been diagnosed 13 years previously. He pointed out the maximum and mini- mum levels of blood hemoglobin he had attained and explained why the intern’s goal, which had been ex- pressed as “returning your blood to normal,” was impos- sible to achieve. When the patient’s wife noted that his hemoglobin was now below the range in which he previously had been transfused, the patient countered that the test had been performed on a Sunday and it was well known that one could not trust the results of weekend laboratory tests. The patient also noted that he was angry at the intern because he had not permitted him to see his cur- rent chart so that he might copy down the latest labora- tory values for his log. In addition, he pointed out that the intern had failed to warn him of the risks of transfu- sion, particularly hepatitis, which he previously had ex- perienced. Through all of this discussion, the patient maintained that there was no more cancer in his body. He attributed his weight loss, kidney problems (resulting from the obstruction of a ureter by tumor), and other symptoms to the after-effects of radiation therapy he had received many years before. This patient presented an extreme case of an obsessive personality whose need to maintain control of his situation led him to distrust his intern, refuse the recommended transfusion, distort the meaning of the laboratory results he received, and deny the reality of his widely metastatic carcinoma. He was dreaded by the nurses and laboratory technicians on the floor, because they knew that he would give them an argument about any procedure they had been ordered to do. As an exasperated technician said, after the patient refused to have an intravenous 436 Making Health Care Decisions: Appendix D line inserted, “They called me to put an IV in, not to rationalize with the man.” Patients’ characterologic difficulties were not always so blatant. Obsessive desires to stay in control, or dependent de- sires to be cared for often seemed to play a role in refusals, even when the ostensible reasons for refusal related to discomfort, lack of need for the procedure, or mistrust. One patient, for example, who refused to cooperate with a bowel regimen that involved moving from his bed to a bedside commode, objected on the grounds that the regimen was ineffective and the pro- cedure uncomfortable. The nursing staff, however, perceived a great reluctance to take responsibility for his own care, man- ifested by his stated desire to have the nurses manually disim- pact him when he became constipated. It should be noted that to point to patients’ characterologic structures as causative elements in refusal (as it was in 22 of 105 first refusals) is not to label these patients as mentally abnor- mal. Each individual's personality reflects a mixture of a small number of characterologic factors, with one of them often pre- dominant. Only when such factors prevent flexible responses to new situations do they become maladaptive. The value of recog- nizing that behavior is rooted in these factors lies in the pos- sibility of responding in ways that allow maladaptive behaviors to be relinquished. (See section on Responses to Refusal below.) Hospital fatigue syndrome. As this study progressed, it became increasingly apparent that many of the refusals that lacked rational justification were being seen in patients who had undergone long hospitalizations with multiple treatments and diagnostic procedures. Their refusals of important, as well as trivial, procedures tended to be accompanied by the angry disclaimer, “I don’t care what happens to me. I'm just fed up.” These patients appeared to be suffering from what we came to call “hospital fatigue syndrome,” a response to prolonged hos- pitalization that was manifested by a seemingly irrational rejec- tion of their doctors’ suggestions and an overriding desire to return home, regardless of the cost. Case example #13—This was a 61-year-old woman whose complicated medical history extended back 23 years. Over that time she had suffered disseminated tu- berculosis, chronic renal failure, hypertension, breast cancer resulting in mastectomy, and low back pain for which she had medicated herself with intramuscular narcotics. The, presenting problem was two years of pro- gressive pain and swelling of her foot and leg. She was first seen when she refused a lymphangiogram as part of her workup. She expressed a fear of having to lie still on the table in the x-ray suite for several hours, noting that this would provoke severe back pain. Immediately prior Treatment Refusal 437 to her refusal, she had had a painful set of x-rays for another purpose. This patient ultimately agreed to have the lymphangiogram when the radiologists told her that they would modify the procedure to provide support for her back, and that she would receive analgesics. The impres- sion at the time of the interview was that this was a passive-dependent woman, with great concern about avoiding pain. Although the patient went through with the lym- phangiogram, she subsequently refused to have a workup (involving barium contrast studies) for gastroin- testinal bleeding, and an arthroscopy of her knee to rule out the possibility of active tuberculosis. At this point, she was reinterviewed. The patient described herself as “tired,” noting that she had suffered more than 20 years of illness, as well as multiple interventions during this hospitalization. She expressed a willingness to have the recommended workups at a later point in time. In a phone interview, the patient's husband described heras a very strong woman who had coped well with her many illnesses, and who had never previously refused recom- mendations for treatment. The patient was discharged shortly thereafter, planning to return for further workup at a later date. This case demonstrates many of the factors associated with hospital fatigue syndrome. The patient had been admitted for what she assumed would be a simple workup of arthritis in her leg. She was at first surprised and then exasperated by the multiple tests that were performed. Patients who expect to be hospitalized briefly, only to face prolonged stays, may be at special risk for hospital fatigue. In addition, this was a woman who had been through many hospitalizations over a 23-year period. The cumulative effect of illness and hospitalization over the years seems to have a bearing on patients’ tolerance for additional procedures. Finally, it should be stressed that this woman's presenta- tion when first interviewed was quite different from her base- line, as described by her husband and outlined in her chart. The danger is evident of drawing conclusions about charac- terologic styles on the basis of a cross-sectional evaluation dur- ing a period of considerable stress. Fatigued patients behave quite differently from the way they would under less stressful circumstances. A variant of hospital fatigue syndrome involved refusals by patients of tests that had previously been performed at other hospitals. The reluctance of tertiary referral centers to accept the validity of tests performed elsewhere often infuriated pa- 438 Making Health Care Decisions: Appendix D tients, who felt that their previous discomfort had gone for naught, and led to refusal of the repetitive procedures. One such patient, a “take charge”-style business executive with a probable lung tumor, was so upset by the need for repeat x-rays that he investigated the possibility of transferring his care to another facility. Ultimately, as in both cases of this sort that we observed, the attending physician persuaded him to go along with the radiographic studies. Hospital fatigue appeared to be a factor in a number of refusals in which characterologic elements and simple un- willingness to undergo discomfort were also present. If, in fact, it is true that patients are more willing to have needed pro- cedures early in their hospital stays than after many weeks of confinement, there may be lessons to be drawn for the utility of shorter hospitalizations with planned readmissions to com- plete lengthy diagnostic evaluations. Other reasons for refusal. The preceding enumeration of reasons for refusal is in no sense exhaustive. Given that a variety of causative influences are at work in most cases of refusal, the number of factors involved can be multiplied al- most without end. Nonetheless, the major factors, with the exception only of a few reasons that do not fit into any neat conceptual scheme, have been presented above. The remaining major factors are described in this section. “Legitimate” refusals. There were a number of refusals that everyone involved recognized as “legitimate.” These included cases in which patients with diarrhea refused laxatives, and patients who had been taking one formulation of a medication for years refused to accept a substitute. In such cases, the patients’ doctors were more than willing to accommodate their orders to the patients’ desires. One must ask, however, about the situations that precipi- tated these refusals. How does a patient without need for a laxative come to have one prescribed? Why must patients refuse a medication in order to obtain the preparation they prefer? In large part, these legitimate refusals appeared to stem from less than optimal communication between patients and doctors. It is unclear, however, which party was at fault here; that is, whether patients failed to express their desires and to inform their doctors about their conditions, or whether doctors failed to ask. That both factors often interacted was suggested by the case of a surgical patient who refused a laxative because of severe diarrhea. She entered the hospital with normal stools, but despite a worsening problem, she continued to take the routinely prescribed laxative every day without complaint. Only when the problem could no longer be ignored, did she inform the nurse about her condition. Patient reticence, com- bined with the unfortunate practice of writing routine medica- Treatment Refusal 439 tion orders for such items as laxatives, produced, in this case, an easily avoided iatrogenic complication. Another class of “legitimate” refusals involved patients for whom an aggressive attending physician had recommended further treatment, despite a general perception on the part of others involved in the patient’s care that the potential benefits were outweighed by the risks. Case example #14—This patient, a 61-year-old man with end-stage, metastatic thyroid carcinoma, refused a sec- ond course of chemotherapy for his illness. He explained that the proposed chemotherapy was unlikely to help him a great deal. “It’s something new. The doctors can’t tell you about the benefit.” In addition, it might cause further breakdown of some very painful bed sores. Given the likelihood of distressing side effects and the remote possibility of benefit, the patient decided to forego the treatment. Interestingly, the intern on the case agreed with his decision. He noted, “He’s the kind of guy I could talk into having it; I could probably talk him into anything. But I just can’t see the point of giving him another course of therapy.” The patient's refusal stood until his discharge to a hospice the following week. Decisions of this sort to refuse high-risk, low-benefit treat- ment in this study were limited to cases of cancer chem- otherapy (although there is no reason to believe that they do not occur in other serious illnesses, as well). Patients opted for a higher quality of life in the time left to them than would be possible if they accepted the treatment. Such decisions repre- sent the prototypical balancing of risks and benefits envisioned by the doctrine of informed consent. It is somewhat ironic that this process was limited almost entirely to terminal patients. Idiosyncratic refusals. Some patients had reasons for refus- ing treatment that can only be classified as unusual. Overt psychopathology did not appear to be involved in these cases. Case example #15—This was an instance in which the family’s idiosyncratic views were determinative. The pa- tient was a 74-year-old man who was quite hard of hear- ing. As a result, his wife mediated all contact between him and the outside world. Originally admitted for a cornea transplant, the patient was found on a closed needle biopsy to have a squamous cell carcinoma of the lung. His doctors recommended open lung biopsy or mediastinoscopy to determine if the lesion was resecta- ble, i.e., if there had been spread to the mediastinal lymph nodes. If metastasis had taken place, treatment would be limited to radiation, and death from the tumor was inevitable. If no metastases were found, surgical 440 Making Health Care Decisions: Appendix D resection of the primary lesion, with a chance of cure, might be undertaken. The patient's wife and daughter, who spoke for him, strongly rejected the recommended procedures. When interviewed, the wife noted that she had decided in- stantly on hearing of the tumor that “there would be no surgery.” Her reasons for this was the belief that “if any air got into the lung it would make the tumor worse.” She maintained that she had discussed this belief with the intern (although it was not clear from discussions on rounds that he recognized this as the basis of the refusal) and that he had told her that although what she feared might happen with some tumors, it would not happen with this one. She remained unconvinced. Her husband was discharged a few days afterwards, scheduled for outpatient radiation therapy. Somewhat less dramatic examples of idiosyncratic beliefs affecting refusal occurred when patients became convinced that a certain medication was causing side effects. One patient, for example, maintained that Mylanta, an antacid, was the cause of a burning sensation in her stomach. No amount of persuasion could convince her that Mylanta was prescribed to treat her gastric distress and could not be the cause of it. The patient shrugged off these attempts to reason with her by claiming, “My body is different from other people’s bodies. Things that work one way for them, always work just the opposite for me.” “Death with dignity” refusals. Several cases were observed that fit the model of many of the court cases referred to in the Introduction above; that is, cases that concerned the rights of patients or their families to refuse treatment intended to sustain life in clearly hopeless situations. Case example #16—Not long after admission for re- moval of a cast and physical therapy, this 69-year-old woman suffered a brainstem stroke that resulted in a “locked-in syndrome.” Although she was conscious and apparently able to understand what was going on around her, the patient could move no muscles in her body except those controlling her eyes. Efforts to communicate using a system of eye blinks, however, were unsuccessful (i.e., the results were ambiguous). The day following the stroke, the patient’s daughter arrived and insisted that no further supportive care be given her mother. The daughter explained that her mother’s brother had suffered a similar stroke and, at that time, the patient had expressed a clear desire not to be allowed to linger if she ever faced a similar fate. Thus, the daughter was quite sure that all the mother desired now was to be allowed to die. The attending on the case, however, who had not Treatment Refusal 441 known the patient previously, had reservations about following the daughter's wishes. He agreed that no ag- gressive measures should be taken, but he was reluctant to stop all intravenous fluids and even wanted to insert a nasogastric feeding tube. Ultimately, his plan was to transfer the patient to a nursing home after a permanent feeding tube had been inserted via a surgical procedure known as gastrostomy. His view seemed to be that he would not allow the patient to die of starvation or de- hydration, even if that required quite intrusive measures, but that he would not treat other complications of her state. The attending physician and the daughter (sup- ported by other family members) struggled over the issue of supportive measures for several days. At one point the family threatened to dismiss the physician from the case if he refused to withdraw IV fluids. Ultimately, he agreed to abide by their wishes. The IV was removed and the patient died two days later. This case was similar to many other cases of this type reported in the medical, ethical, and legal literature. It empha- sizes, however, how difficult it is for physicians to resist the imperative to maintain the life of someone who is in their care. This was observed in other cases, as well. For example, a clearly terminal cancer patient was treated with anti-arrhythmic drugs and carotid massage for a supra-ventricular tachycardia, even after it had been agreed that no measures beyond simple sup- portive care would be taken to sustain his life. For a variety of reasons, physicians have great difficulty abstaining from action when patients develop treatable complications, even if their death from other causes is imminent. Responses to Refusal. Data on responses to refusal have been drawn from all three stages of the study. The multiple responses that frequently followed a single refusal make analy- sis of these data problematic. Nonetheless, an effort has been made in Table 4 to categorize the kind of primary responses that occurred and to attempt to compare their relative frequency. As previously, only initial refusals have been recorded in Table 4; all numbers in this section are taken from this table. Responses have been grouped according to the stage of the study in which they occurred; responses from stages 1 and 2 obviously repre- sent the most valid sample for statistical purposes. Even here, however, the limitations of our methodology, which resulted in uneven accumulation of data about individual cases, make it likely that we have not been able to identify every response to each refusal. The data, therefore, are more useful for a qualita- tive understanding of how care-givers respond to refusal than for a rigorously quantitative description of the phenomenon. 442 Making Health Care Decisions: Appendix D Table 4: Primary Responses to Initial Refusals Response Stage 1|Stage 2|Stage 3| Total Forced Treatment Competent Patients 0 3 0 3 Incompetent Patients 1 11 3 15 Forceful Persuasion 0 2 2 4 Coax and Wheedle 1 3 2 6 Reinformed Patients 7 15 7 29 Refusals Permitted Proper 8 4 1 13 Undesirable Patients 2 2 0 4 Tool in Management 3 1 1 5 Cost/Benefit 0 4 0 4 Denial 1 0 1 2 Other (including unknown) 7 11 2 20 In the discussion that follows, responses have been orga- nized along a continuum of behavioral activity, ranging from coercive treatment over the objections of the (usually incompe- tent) patient to denial on the part of the physician that refusal had occurred. Regardless of the ultimate reaction, however, we were impressed that responses tended not to depend on the specific reasons for the patient's refusal. Rather, a number of factors related to the medical desirability of treatment and the possibility of treatment over the patient’s objections were usu- ally the determining variables. These factors included: the urgency of the need for the treatment or diagnostic procedure; its potential risks and benefits for the patient; whether the physicians felt themselves ethically compelled to treat despite the patient's refusal; whether the procedure was minor and routine or a major intervention; and whether formal consent, including a signed consent form, was ordinarily required. This does not mean that care-takers’ responses to patient refusal were determined solely by medically oriented consid- erations. As will be illustrated, patients’ psychological states or reasons for refusing did at times influence the staff’s response to refusal. In addition, it was clear that the staffs attitudes about the patient also sometimes influenced their responses. The interactions between the factors will be demonstrated in the case examples that follows. Forced treatment of the incompetent patient. Treatment over a patient’s objections was restricted in 15 out of 18 cases to patients who, in the layman's sense, appeared to be “generally incompetent.” These patients were usually disoriented or in- coherent, due to dementia, delirium, head trauma, or alcohol withdrawal. They were often restrained or treated with oral or intramuscular medications, such as Haldol, for purposes of sedation. The purpose of this non-consensual treatment was usually (14 of 18 cases) to prevent harm ensuing to the patient Treatment Refusal 443 (e.g., as a result of falling out of bed, inducing other types of self-injury, or interfering with treatment, as by removing intra- venous lines). Some patients (4 of 18 cases) were also restrained or sedated because they wandered through the hospital cor- ridors or shouted loudly from their rooms. Once restraints were applied, treatment with medication of the patient's primary medical problem also proceeded non-consensually. Except in the case of surgical procedures, efforts were usually not made to obtain substituted consent. Case example #17—A demented 79-year-old man who had refused oral fluids at home was hospitalized in a severely dehydrated state. He had a history of multiple brain infarcts. The patient consistently pulled off a con- dom catheter that had been placed on him in order to prevent him from being incontinent in bed, and also to measure his urinary output. The staff viewed this behav- jor as a result of the patient’s state of disorientation. He was placed in four-point restraints and a Posey vest to keep both his body and his limbs tied down to the bed. The condom catheter was replaced. When an attempt was made to interview the patient, he could not give coherent responses to questions. Forceful persuasion. Four patients were rather forcefully persuaded to receive treatment, usually in semi-emergencies, even though tiiey were able to talk and reason and were, from the ordinary perspective, competent to make treatment deci- sions. Such persuasion was an immediate and forceful, albeit verbal, exercise of physicians’ authority. Case example #18—We first learned about this 53-year- old obese male who had respiratory difficulties and pleural effusion, due to lung cancer and tuberculosis, when he refused daily weighing, saying he was tired of being weighed. The patient’s refusal to be weighed was permitted by the staff. Later in the patient's hospitaliza- tion he became increasingly short of breath. It was thought that his pleural effusion had worsened, but the patient refused to go down for an x-ray, claiming that he was too sick. He wished instead to have the x-ray techni- cian come to his bed to take the film. The patient's first refusal of x-ray was permitted, but when he refused the next day, the intern told him that he absolutely must have the film and that he could not refuse it. He therefore agreed to be taken down to the x-ray unit. Forceful persuasion of the patient was not limited to imme- diate attempts to change the patient’s mind about treatment refusal, but also involved efforts over time. Case example #19—This 72-year-old woman, with a pre- vious history of a colonic tumor, was admitted to the hospital for a workup of explosive diarrhea. She refused 444 Making Health Care Decisions: Appendix D sigmoidoscopy, an important part of her workup, noting simply, “I'm afraid of it.” She characterized her fears as “irrational.” Several physicians spoke with her firmly about the need for the procedure, but the patient con- tinued to refuse, despite understanding the conse- quences of refusal. She meanwhile agreed to have an upper GI series. Shortly before her discharge, after hav- ing been repeatedly informed by her physicians about the necessity of having the procedure, the patient finally consented. The physician wrote on her chart, “I spoke to the patient at length about procto and she finally re- lented.” The patient reported to us that the pressures she felt from the doctors and her daughter persuaded her to have the treatment; it was “their decision” and she sim- ply went along with it. Forceful and repeated efforts at persuasion by her doctors, who also orchestrated similar pressures from her family, even- tually allowed this patient to see herself as a helpless victim of other peoples’ decisionmaking. She thus agreed to go along with the sigmoidoscopy that she dreaded and to which she could not agree of her own volition. Interestingly, she reported afterwards that she did not find the procedure to be as dreadful as she had anticipated. If she required sigmoidoscopy again in the future, she would “just have it and get it over with.” The coax and wheedle technique. When patients’ refusals involved less urgent or serious matters, patients were “coaxed and wheedled” to accept treatment. Six cases in which this was the primary response were observed. Staff responses to these patients—who were not always clearly or consistently refusing, but who were often periodically uncooperative, required reas- surance or encouragement, or otherwise appeared to want their emotional needs attended to—involved more clear-cut “psycho- logical” management of refusals. Such psychological manage- ment was usually done by the nursing staff, as opposed to other hospital staff who were not so continuously in contact with the patient. But in the case below, the patient's physician, too, was sensitive to her psychological problems. Case example #20—This 53-year-old woman intermit- tently refused medication, including Kayexalate, a po- tassium-lowering substance. The patient had chronic pyelonephritis, deteriorating kidney function, and was on the verge of requiring renal dialysis. The nurse noted that the patient generally refused to take medication, claiming nausea, but then allowed herself to be coaxed into it. Her refusal was viewed by the nurses as more of a delaying tactic, wherein the patient garnered attention. The patient's physician noted that the patient did not like to lose control of her situation (as she had during Treatment Refusal 445 this hospitalization), and saw her behavior as an attempt to establish some measure of control. The patient, how- ever, had also been noncompliant with medication at home in the past, manifesting behavior similar to that shown in the hospital. The medical student who evalu- ated the patient viewed the patient’s refusal in a still more specific psychological framework, believing that the patient was trying to get back at her daughter by not taking medication, thus making the daughter responsible for her deterioration. This patient represents an important group of patients, namely, those who find some gain in the manipulation that attends their refusal of medication. Both the nurse taking charge of this case, as well as the patient’s physician, were sophisticated about these problems. They continued to monitor the patient's treatment and to encourage her to take medication, despite her characterological problems. Reinforming the patient. Another technique used in non- emergency situations involved giving additional information to the patient. This was one of the most characteristic responses we observed (29 of 105 initial refusals). Case example #21—this 25-year-old woman with a his- tory of renal stones had a surgical procedure in which a stone was removed from the pelvis of her kidney. Two weeks postoperatively, the patient refused a renal sono- gram, a diagnostic test ordered because she was having continued abdominal pain. Her physicians wished to see whether there was a blockage of her ureter that might be leading to her pain. The patient told us that she had decided to refuse all tests that day because she believed she wasn’t being told enough about her condition and she was, on the day of her refusal, in a disgruntled, “fed- up” mood. The doctor later came by to tell her that the purpose of the sonogram was to “look at her kidney.” On the basis of this partial information she agreed to have the test, agreeing, in part, also because she felt very guilty at having refused. On the other hand, even though she consented to have the test, she still did not understand what its purpose was, nor what the doctors were looking for, nor how the sonogram might help in the diagnosis or evaluation of her pain. She complained to us that all during the hospitalization she had been given a lot of medications and tests whose purpose and effects she did not understand. It must be noted that the information provided to refusing patients, as in this case, was frequently presented in a stereo- typed fashion, with no effort made to tailor the explanation to the patient's particular concerns. Thus, although most often this type of response resulted to patient agreement to undergo the 446 Making Health Care Decisions: Appendix D procedure (19 of 29 cases), full or specific information was not always given, resulting in a compliant but still not fully under- standing, and often dissatisfied patient. Patients often appeared to be responding to the interpersonal intervention per se, rather than to the specific information they received. Refusals permitted—no attempt made to persuade the patient to have the procedure. The frequency with which physicians made no attempt to persuade patients to abandon their refusals (26 of 105 first refusals) was one of the most surprising findings of this study. The reasons for this response varied and were often complex. They ranged from the staff’s harboring a dislike for the patient and desiring to be rid of him to the staff permitting a given refusal in an effort to obtain patient compliance with other, more important aspects of the treatment plan. Of course, refusals were permitted to stand untouched when they were viewed by the treatment staff as the “proper” response; e.g., when a patient with diarrhea refused a laxative, and especially when they involved non-essential ele- ments of medical care. The undesirable patient Case example #22—This 45-year-old man, with a his- tory of alcoholism, was admitted the day after Christmas for treatment and evaluation of a gastrointestinal bleed. Although the patient initially allowed the staff to put down a nasogastric tube to lavage his stomach (which yielded bright red blood), he pulled out the tube several times and ultimately refused to permit its reinsertion. Later he refused to have a nasogastric tube hooked up for drainage of the stomach, and also refused an upper GI series, telling the staff that he had them in the past and they had been normal. Discussion with the patient on hospital rounds did not address his refusals, but concerned the possibility of his being transferred to a Veterans Administration Hospi- tal. Although the patient had not been treated regularly in VA hospitals, he did have veterans’ benefits and the staff believed this might be the “more appropriate” place for such a troublesome patient to be hospitalized. The patient, however resisted the transfer, telling the hospital staff that he didn’t think he would be staying in the hospital very long anyway. Later that day, the patient signed out AMA. We observed a somewhat similar interaction involving a chronic schizophrenic patient who early in his hospitalization had to be treated involuntarily with intravenous antibiotics to combat septicemia. Later in the hospitalization the patient re- fused an upper GI series meant to workup the cause of blood in his stool, as well as Amphogel, an antacid. The patient was Treatment Refusal 447 confused about what he had refused, and about the implica- tions of his refusal. This patient had been a chronic problem in terms of compliance with staff recommendations, and he had little insight about the nature of his medical problems. The patient’s refusal of the upper GI series was permitted to stand, with the house staff's primary concern being discharge plan- ning. Permitting refusal as a tool in patient management Case example #23—A 26-year-old chronically “difficult to manage” woman with severe rheumatoid arthritis re- fused daily weights. The head nurse viewed the patient's refusal as evidence that the patient desperately wanted to control her own care. She was striving to deal with the severity, chronicity, and irreversibility of her illness, all of which had contributed to making her a virtual invalid. In the opinion of the head nurse, weighing was the only thing that the patient knew she could refuse without the staff pressing her to accept. The controversy about weighing represented the patient's only potential mecha- nism for exerting control over her fate while in the hospi- tal. In response to the patient's refusal of weights and unhappiness at early morning procedures, a contract was negotiated with her to permit her to be able to sleep until 9 AM in the morning, and also to allow her to refuse daily weights. Following this compromise, the patient was much more affable and cooperative. When inter- viewed by the investigators, this patient initiated a lengthy discussion about the impersonality of modern hospital life, indicating she didn’t want to be just a number, “bed X, floor Y.” The hospital's initial request for early morning weighings of all patients, she ex- plained, did not take into account the problems that patients like her have with rheumatoid arthritis, with its characteristic early morning stiffness. The patient was extremely pleased with the staff's response and their understanding of her individual problem, and in- creasingly cooperative with other aspects of her care. This case demonstrated optimal management of a patient's refusal. The staff recognized the psychodynamic roots of the patient's behavior, namely her desire for some measure of con- trol over her illness and its treatment, and responded in a manner that allowed her that control and at the same time enhanced her cooperation with other aspects of her care. Al- though this particular response was facilitated by the non- essential nature of the procedure in question (weighings), there is often some way to provide patients with the sense of control and participation they desire. One patient, for example, who refused all his medications because he believed that the nurses were not bringing them precisely every three hours as his doc- 448 Making Health Care Decisions: Appendix D tor had ordered, was given his daily dosage of medication each morning and allowed to take the pills on his own at the proper time each day. Permitting refusal as a cost/benefit decision. Refusals were often permitted to stand when the physician decided that the effort required to reverse the refusal or the discomfort to which the patient would be subjected were not worth the potential benefits. In essence, the decision was made that the knowledge that could be gained from a particular diagnostic test or the benefits of a given treatment were not critical for the patient's care. Case example #24—A 29-year-old female refused to have a repeat gastric washing performed to workup the possibility of tuberculosis. The test required having a tube inserted in her stomach. The first gastric washing had involved great pain and discomfort; the patient had felt as if she were choking. When the tube was put down, the patient was temporarily left alone in the room, at which point she became extremely anxious and pan- icked, because she “did not know what was going to happen next.” She thus found this diagnostic procedure very aversive. She was, furthermore, surprised when the doctor appeared to do the second gastric washing be- cause she had thought she was only going to have one. In addition, contradictory information had been given to her about the chances that she had tuberculosis. The patient refused the repeat washing, not understanding, however, how collecting stomach mucous related to mak- ing a diagnosis of tuberculosis, or why repeated gastric washings were necessary. She was permitted to refuse by a house officer who thought TB was an unlikely diag- nosis. A PPD skin test was later negative, ruling out the possibility of TB. Consequences of permitting refusal. When physicians indi- cated that they accepted a patient's refusal, patients often took that as confirmation of the correctness of their position, rather than as a sign of the physician's accommodation to the failure of his or her efforts to persuade them otherwise. In part, this phenomenon is another illustration of problems in communica- tion between doctors and patients. The following case, already briefly discussed (see above) illustrates this phenomenon. Case example #15 (continued)—The wife of a 74-year- old man who was extremely hard of hearing and who did not participate in the decision, stated that the family was opposed to exploratory surgery or open biopsy to deter- mine the extent and treatability of her husband's lung tumor. The basis of her refusal was at least in part idio- syncratic; she refused surgery because she did not want “air to get into the lung” to make the tumor worse. Some Treatment Refusal 449 discussion between the patient’s wife and the physicians then ensued. The patient’s wife “slept” on her decision. When she awoke, the answer was clear. There should be no more biopsies. She came to the hospital, therefore, contemplating taking the patient home, or even transfer- ring him to another hospital, noting that she was also angry at his physicians because it had been so difficult to arrange a meeting time with them to discuss this matter further. When she arrived at the hospital, believing she would refuse the open surgical procedure, she learned that the surgeon had written a note in the chart in which, according to her, he had agreed that an open biopsy would be “pointless,” that nothing more could be achieved. The patient's wife therefore concluded that everyone was in accord that open biopsy would not be helpful and that the patient should receive x-ray treat- ment. In fact, what the surgeon had written in the chart was that it would be “pointless” to do an open biopsy if further surgery was not contemplated (because the fam- ily would not permit it) and that therefore x-ray treat- ment would be given instead. We saw several other examples in which, from a desire to accommodate the patient, alternative treatment or procedures were instituted, especially when it was the physicians judg- ment that it was not absolutely necessary for the patient to have the originally recommended treatment or procedure. But be- cause the physician was willing to alter his recommendation, or to institute an alternative treatment, the family then concluded that the “best treatment” had been decided upon after all. Certainly, from the perspective of informed consent doctrine, there is nothing problematic (even something quite desirable) about a process of negotiation, assuming that each party to the negotiation has a good understanding of why the other party has reached its decision. In the absence of adequate commu- nication between doctors and patients, or as a result of mistaken communication, alternative treatments were sometimes de- cided upon which may not, however, have been in the patient’s best interests, despite the patient's strong desire to believe that this was the case. Denial of refusal. Physicians were sometimes unwilling to acknowledge that any refusal actually had occurred, thus pre- cluding any further response to the situation. Case example #25—A 48-year-old woman who suffered from lymphocytic lymphoma intermittently refused a number of treatments, including radiation therapy, part of her chemotherapy regimen, and intravenous fluids. The patient described herself as a person who had “bent” towards the needs of others all her life. Now, in 450 Making Health Care Decisions: Appendix D contrast, with the end approaching, she wanted to be in control herself. She said she did not have the strength to withstand the nausea and vomiting caused by the chem- otherapy and radiotherapy and insisted that she not be forced to tolerate it anymore. She also complained to us that the distressing side effects of radiotherapy (nausea and vomiting) had not been explained to her in advance. She was tired, she said, of being treated like a child. The hospital staff considered the patient to be “trou- blesome,” because of her fluctuating willingness to go along with treatment. The physician responsible for tak- ing care of the patient, however, explained to us that he did not regard her as a refuser. Instead he described her as emotionally labile and dependent, someone who needs a great deal of support, who is very childlike and therefore has to be dealt with as if she were a child. Rather than acknowledging (as had most of the other persons caring for this patient) that the patient was a refuser and exploring the reasons for the patient’s refusal, this physician saw his main responsibility as dealing with the patient in a manner that would produce com- pliance. In general, denial of refusal indicated an unwillingness on the part of the physician to recognize the basis for the patient's act. It is of particular interest that this patient’s underlying complaint, that she had never been able to exercise autonomous control over her life, was validated by her physician's own description of how he dealt with her. The physician's failure to acknowledge the reality of her refusal was also implicitly a failure to recognize the patient as an autonomous person who was capable of refusing, thus reinforcing the frustration and sense of powerlessness that motivated her rejection of treatment in the first place. With death approaching, the patient’s ability to come to terms with her life and place it into perspective was also thereby impaired. Psychiatry and the refusing patient. Psychiatric consul- tants were called upon to examine patients who had refused in two circumstances: first, when the patient had a known psychi- atric history, and second, when the patient appeared to be suffering from an obvious psychiatric disorder and the refusal was potentially serious. Psychiatric consultants were used in- frequently, however, in the absence of these two conditions. We observed only seven refusals in which psychiatrists were con- sulted. House officers and attendings rarely believed that psy- chiatrists could add anything to the management of refusal if the patient weré not a “psychiatric case.” The following case, the only one we witnessed in which a consultant was called in the absence of clear-cut “psychiatric indications,” demonstrates the unfortunate consequences when the psychiatric consultant, too, adopts this narrow view of the role of psychiatry in refusal. Treatment Refusal 451 Case example #26—A 71-year-old woman had a dissect- ing aortic aneurysm repaired. Postoperatively she expe- rienced numerous problems and it was difficult to with- draw her from the respirator. It was not clear whether her failure to breathe was due to metabolic imbalance, or whether the patient had simply given up, as was sug- gested by the patient's periodically saying that she wanted to die. She was extremely uncooperative and gave stereotyped responses to all who came into her room, telling them to leave her alone, to “get out of here.” Despite the fact that the patient refused to eat, and intermittently refused respiratory therapy, these things were pressed upon her forcefully. A psychiatric consulta- tion was obtained in which the psychiatric consultant, a resident, recommended additional metabolic workup, for example checking on the patient's arterial blood gases, electrolytes and digitalis levels, and medicating the patient with anti-psychotic drugs. He gave no consid- eration to psychological factors. The patient's primary medical physician, however, noted that this woman had had a quite severe personality change following surgery. She previously had been quite pleasant. He believed that all that had happened to her in the hospital had ex- hausted her and angered her. Nevertheless, he saw her problem as not being “primarily psychiatric” but simply that she was “fed-up with procedures.” Having been sick so long had caused her to become angry and uncoopera- tive. The patient's son agreed. The cause of the patient's abnormal mental state never became fully clear during the hospitalization. Shortly before she was scheduled to be discharged to a nursing home, however, the patient became more respon- sive. She indicated that she was worried, that having to go to a rest home is the kind of thing “that makes one very sad.” She appeared clearly depressed. In retrospect, this patient's depression certainly seemed to have contributed to her refusal of treatment, but even though psychiatric consultation was obtained, this patient's psycholog- ical concerns were not addressed in the hospital. Outcomes of Refusal. The outcomes of initial refusals in all three stages of the study are summarized in Table 5. It must be noted that the completeness of this characterization of out- comes is limited by the lack of post-discharge follow-up of our patient sample. Thus, this analysis reflects the short-term out- comes, prior to discharge, of patients who refuse. Given the finding that the staff’s responses to refusal were primarily determined by their views of the medical issues in- volved, one might expect that the outcomes would also be a function of the medical necessity for the treatment refused. If 452 Making Health Care Decisions: Appendix D Table 5: Outcomes of Initial Refusals Outcome Stage 1|Stage 2|Stage 3| Total Delay and Acceptance 8 17 8 33 Substitute Equivalent Treatment 4 4 A 9 Substitute Inferior Treatment 1 3 2 6 Substitute Inadequate Treatment 0 0 1 1 No Treatment Given 16 16 4 36 Forced Treatment 1 14 3 18 Unknown 1 1 0 2 this were the case, one would expect to find that relatively minor refusals were likely to persist, whereas refusals of more important procedures were generally resolved in favor of treat- ment. Although it is our impression that there was a tendency for this to occur, the picture is rather more complex. It will be recalled that responses were generally not “fine- tuned” to patients’ underlying motivations for refusal. The out- comes, therefore, were often dependent on the fortuitious reac- tion of the patient to a relatively undifferentiated response. Another key variable appeared to be the persistence of the physician in seeking to reverse the patient's refusal. Regardless of the reasons for refusal, repeated attempts to persuade the patient to accept treatment were usually successful. Some phy- sicians, however, lacking an understanding of the reasons for a patient's refusal, and often hurt by the implicit rejection of their ministrations, stopped pursuing the issue after an initial at- tempt at persuasion. This occurred in serious as well as minor cases and accounts in large part for the failure of outcome to correlate with the seriousness of the refusal. The interaction of these three factors—the reason for the patient's refusal, the nature of the response, and the persistence of the responding staff member—usually determined the out- come of the episode. Delay and acceptance. One of the most common out- comes of patient refusal (33 of 105 first refusals) was delay culminating in later patient consent. Refusal often meant a lapse of a day or two until tests could be rescheduled, medica- tions administered, or procedures performed. Delays thus re- sulted in lost time, increased expense, and in some cases a period of time during which patients suffered unnecessary pain or other discomfort. The following case was typical. Case example #21 (continued)—This patient, whose case is described in detail above, refused a renal sono- gram for the diagnosis of renal colic. She objected to tests being scheduled without her being informed in advance of their purpose. After additional information was pro- vided (although not enough, as noted, to afford her a Treatment Refusal 453 good understanding of the purpose of the test), she agreed to go ahead with the test, which was rescheduled for the following day. She remained angry at her physi- cians for leaving her in the dark about the course of her medical treatment. We witnessed no cases in which treatment was later ac- cepted where it was possible to identify serious sequelae as a result of the delay. The possibility exists, however, that more prolonged follow-up might have identified such cases. Substitution of equivalent treatments. The outcomes of nine initial treatment refusals involved substitution of alterna- tive treatments or procedures equivalent to the treatment the patient had refused. For example, one patient received Tinactin, an antifungal medicine he preferred as a substitute for another antifungal preparation, Desenex. Some patients’ refusals of oral fluids or medication culminated in their receiving the medica- tion intravenously. The converse was also true; oral administra- tion was provided as a substitute for intravenous therapy or instead of utilizing a nasogastric tube. In some instances, the patient’s preference of an alternate treatment was based on the patient's past experience with the same medication. The most common examples in this area involved patients who were receiving cancer chemotherapy and who preferred one type of antinausea drug, (e.g., Droperidol) to another (e.g., Compazine). Substitution of an equivalent treatment sometimes permit- ted the patient to avoid having to undergo a more serious or invasive procedure. Case example #27—After transfer from an outlying hos- pital, this 32-year-old man with ulcerative colitis and rectal bleeding developed a recto-vesicle fistula (a com- munication between the bowel and the bladder). Treat- ment required that the bowel be given a chance to rest. The patient, however, refused to have an intravenous line placed in his superior vena cava for hyperalimentation, a way of providing concentrated nutrients without the need for oral feedings. The patient previously had had a negative experience with hyperalimentation and wanted to avoid it if at all possible. Because of his insistence, the house staff was willing to try “low residue feedings,” at least for a few days, as an alternative to the hyperalimen- tation. The patient did very well with the alternative treatment and the fistula healed. Substitution of inferior treatments. Six cases of initial refusals resulted in the application of what the physicians con- sidered “second-best” treatments (i.e., treatments that were less likely than the preferred one to accomplish the stated aim). This usually occurred when the physicians were unable to shake the patient’s resolve concerning the refusal, yet the patient's condi- tion demanded that some treatment be instituted. 454 Making Health Care Decisions: Appendix D Case example #28—A 62-year-old woman who was re- ceiving radiotherapy for treatment of cancer of the lung was admitted to the hospital with shortness of breath. About fifteen months previously she had been found to have an infiltration of her lung which did not improve with antibiotic treatment. The patient resisted having a biopsy done for three months. Then when the biopsy showed cancer, she refused chemotherapy even though, as she told us, it was strongly recommended by her physicians, who continued to come to her room on a daily basis to indicate that she should have chem- otherapy (pressure the patient resented). It was her belief, however, that “chemotherapy kills,” “it does not heal.” The patient had known several people who had had chemotherapy, one for lung cancer, but who had, nev- ertheless, died in a short period of time. She furthermore noted that chemotherapy causes one’s hair to drop out and it causes nausea. She was quite convinced that it was better to have the “good treatment,” radiotherapy, rather than the “bad treatment,” chemotherapy, despite her doc- tors’ suggestions to the contrary. The patient therefore was treated with radiotherapy. Six months later radi- otherapy was given to a painful leg metastasis permitting her to walk again. When we talked with the patient she knew that she had a “growth on her lung,” but she avoided discussing the issue of cancer. Even though she was doing poorly, she indicated that under no circumstances would she consider chemotherapy; however, she was willing to have more radiotherapy, “50 or 100 treatments,” “how- ever many it may take to be well.” She showed some evidence of denial of her condition, and anger at her physicians, even blaming her continuing “growth on the lung” on the original needle biopsy of the tumor (which she had not wanted), indicating that the doctors must “have messed up something in there.” The family had considered pressing the patient to have chemotherapy, but decided not to when they learned from the doctor that at this point, chemotherapy might only prolong life for a few additional months: the patient’s prognosis was presently very poor no matter what treatment was re- ceived. This patient received a treatment—radio- therapy—her physicians clearly considered “second- best,” although not without some potential efficacy. Whether her survival would actually have been pro- longed by chemotherapy is, of course, unknown. Substitution of inadequate treatment. One patient's objec- tions culminated in what was a clearly inadequate treatment (i.e., a treatment that could not accomplish the stated aim) being prescribed. Again,the relationship between the strength Treatment Refusal 455 of the patient's refusal and the vigor of the efforts to overturn it was important here. Interestingly, in the case that follows, the physicians felt a need to do something, even if they knew it was inadequate, rather than stand by impotently in the face of re- fusal. Case example #29—Following an operative procedure on his vertebral discs (laminectomy), a 62-year-old man was suffering from a probable infection of his disc space. The patient was immobilized in a full body cast, while he was being treated with antibiotics. The patient, how- ever, refused to continue to take heparin, an anti- coagulant used to prevent thrombophlebitis that might occur because of the patient's immobilization. The pa- tient stated he had noticed a connection between receiv- ing heparin and feeling nauseous a half-hour later. Al- though considerable efforts were made to convince the patient that it was unlikely the drug was the source of his nausea and of the necessity of the heparin, he would not consent. The patient was then placed instead on Persan- tin, an oral aspirin-like drug that the house officer ac- knowledged is a very poor substitute for heparin, and which was unlikely to prevent venous thromboembol- ism. No treatment given. In a large number of cases, (36 of 105 initial refusals) no treatment was ever provided as a substitute for the original treatment refused. This occurred in a variety of circumstances. Of course, when physicians failed to respond to the patient's refusal, (26 cases) for all the reasons noted in the section on responses to refusal, the refusal stood. These in- cluded those cases that can be called “death with dignity” cases (see example #14), when patients made a decision with which the physician could empathize to avoid the discomforts of treatment in their final days. There were, in addition, ten cases in which efforts to overturn the refusal were unsuccessful and no suitable alternative was available. Apart from the clearly terminal cases, in which death was the patient's desired end point, we saw only two cases in which a successful refusal was temporally associated with the pa- tient’s death: the patient with a pulmonary abscess described in case example #9; and the patient who refused a blood transfu- sion, reported in case example #12, who ultimately succumbed to sepsis. The refusal may have been a causal factor in each case. In other instances, patients were discharged from the hospital, and thus lost to follow-up, after refusal of treatment for potentially life-threatening conditions (e.g., case example #5). Serious consequences may have ensued from these refusals, but they were beyond the capacity of our methodology to detect. Effects of Treatment Refusal on the Doctor-Patient Rela- tionship. Refusal of treatment arises out of the context of the 456 Making Health Care Decisions: Appendix D doctor-patient relationship and in turn can be expected to influ- ence the nature of that relationship. Effects on patients’ attitudes. Although some patients felt justified in refusing treatment, often because of anger at their physicians, a more common reaction was for patients to feel guilty about what they had done. Thus, when patients were interviewed, they were often apologetic, explaining that all they had desired was to obtain more information before a treatment or test was performed. Case example #30—A 23-year-old male, who had a frac- ture of the tibia that had failed to heal, possibly because of osteomyelitis, refused application of a cast-splint. The patient balked because he assumed (incorrectly) that ap- plication of the cast-splint would interfere with the plans for curretage and culture of the bone at the site of the presumed infection. When the patient refused, the resi- dent eliciting the refusal called in a second resident, who had examined the patient on admission, and with whom the patient had a better rapport. The second resident explained to the patient why the two procedures would not be incompatible and the patient consented. When we questioned the patient initially about his refusal, he was very apologetic, explaining that the first doctor had talked too fast for him to understand and that he had been confused when he had voiced his refusal. He added, apparently assuming that we were investigat- ing the first resident’s performance, that he had not meant to cause trouble for the doctor, and that his own lack of comprehension was to blame for the episode. Another patient who refused weighing apologized to us for having caused problems for the nursing staff, indicating that she was never the type to complain. If she had known how important if was for her to have the weighing, she said, she would have agreed to have it in the first place. Still other patients who refused minor procedures (but who were consid- ered by the staff to be “difficult” patients) were upset by the idea that they might be considered to be uncooperative, even becoming tearful when the issue of their previous refusal was discussed. Even patients or families who were convinced that refusal had been the proper step were, nevertheless, apologetic, indicating that they hoped they had not hurt themselves or their medical care by refusing. In most cases, they then reaffirmed their faith in the medical profession and their doctor. Some patients became panicky about refusing treatment or tests, feeling that if the treatment were insisted upon they might have to choose between accepting the proposed treatment, ob- taining a second opinion, or signing out of the hospital. As many previous case examples have suggested (see e.g., case Treatment Refusal 457 example #21, the patient who refused a sonogram because she wanted to know the purpose of tests before they were per- formed), refusing patients frequently felt concerned enough about the refusal to consent to the procedure when asked to again, even though the problem that provoked the refusal re- mained unresolved. The reasons for this display of guilt and renewed willing- ness to cooperate are of interest. Many patients may still be socialized to view passivity as the norm, with any deviation from that accustomed stance leading to considerable anxiety. They may, in addition, sense the anger that their refusal has provoked in their caretakers and, given their dependent posi- tion, realistically fear the consequences of that anger being expressed. Lastly, some patients may recognize their refusal as the expression of their own angry impulses, the projection of which on to their physicians may lead to fears of retaliation. Whatever the etiology, patients’ guilt often resulted in renewed compliance, though often not in more rational modes of deci- sionmaking. Of the 105 patients in this study who refused treatment or procedures, we witnessed only 5 instances when patients, at least briefly, considered obtaining a different physician or sign- ing out of the hospital against medical advice. Case example #31—This 42-year-old man with an en- larged spleen, a low blood count, possible leukemia, and a history of ulcerative proctitis of several years duration, refused a complete electrocardiogram (EKG) the day after admission. The patient was quite aggravated when the technician came by to do the EKG, noting that an EKG had been done at the time of admission the previous day. The patient did not understand the reason for having both the EKG on admission and the full EKG, which is routinely done the following day and “officially” inter- preted. In the background of this refusal was the patient’s anger at what he perceived as the inefficiency of his attending physician, who had failed to bring the patient’s most recent medical records to the hospital. The patient had hand-delivered all his records, including his most recent blood counts, from a medical specialist to his attending physician and had expected that these would be at the hospital when he was admitted. When they were not, despite the patient's efforts, the patient feared that his admission would be prolonged, causing him to worry about finances. When the full electrocardiogram was proposed (which the patient viewed as evidence of the hospital's inefficiency and performance of unneces- sary tests), the patient told the technician to “get out” and explored the possibility of being hospitalized else- 458 Making Health Care Decisions: Appendix D where. He went so far as to call another physician, who said he would be willing to accept the patient's case, noting, however, it was up to the patient to tell his physi- cian at the hospital that he wished to change doctors. The patient never took that step. Of great significance for this study of treatment refusal is that only a single refusal resulted in a patient’s signing out against medical advice (AMA). An alcoholic patient (see case example #22, above) signed out shortly after the staff raised the possibility of his being transferred to a Veterans Administration hospital. More than any other, this finding indicates that treat- ment refusal is something quite different than discharge against medical advice, as both a social and psychological phe- nomenon. The numerous case examples presented suggest that what most refusing patients wanted was not to refuse care altogether, but instead to refuse some aspect of it. Treatment refusal (of the type we have described here) did not represent an absolute rejection of either the hospital or its physicians, or the patient’s other care-takers such as nurses. In fact, even if some patients communicated anger at their care-takers, they were usually at the same time also anxious to tell us both how much they admired their physicians and how helpful their medical treatments .had been. They did not want to be cut off from further medical care. The brief, in-hospital follow-up of our sample makes it difficult to say what effect their refusals may have had on subsequent relations with their physicians and, in particular, on the likelihood of subsequent refusal. In discussing their reasons for refusal, however, a few members of our sample indicated that past episodes of refusal or non-compliance had influenced their present behavior. The following case was an example of this phenomenon. Case example #11 (continued)—This 52-year-old man rejected—at least for the present—chemotherapy for colon cancer, regarding the recommended treatment as “preventive,” and thus safely delayed. In discussing the potential outcome of his refusal, the patient talked about an episode of non-compliance two years previously, when he had not followed his doctor’s recommendations after a heart attack. Following three weeks of hospitaliza- tion, his physicians had told him he must be extremely cautious in driving the several hundred miles to his home. He was advised to make the trip in a leisurely fashion with frequent stops. He did not follow the doc- tor’s instructions, driving almost non-stop. The patient observed that, contrary to the doctors’ predictions, every- thing had come out all right. By implication, he seemed to conclude that this would again be the case, despite his present refusal of recommended treatment. Treatment Refusal 459 The likelihood of patients drawing conclusions about the invalidity of medical advice is enhanced by the tendency of patients, noted above, to assume that their physician's ultimate acquiescence in their refusal represents a concurrence in their decision. It was not possible, however, to characterize the multi- ple refusers in this sample as differing either according to the responses to their initial refusals, or to the factors underlying their refusals. Many of the effects of refusal on patient attitudes thus far enumerated might be considered negative; there were also posi- tive effects on patients. The most outstanding examples here involved instances when patients, through refusal, were able to gain some increased actual or perceived mastery over their situations. Case example #23 (the 26-year-old woman with rheumatoid arthritis who refused weighing, but who later felt much better about herself and her hospitalization when her refusal was permitted) was such a case. Another example fol- lows. Case example #4 (continued)—This patient, himself a physician, had refused a decompressive surgical explo- ration of his nerve roots. Subsequently the patient, essen- tially through joint management of his case with his physicians, underwent a procedure to remove unstable rods from his back, which were the probable cause of his pain. Further, while discussions and negotiations be- tween this physician-patient and his physicians were taking place, the patient’s neurosurgeon asked the pa- tient to consult on the care of another one of the neu- rosurgeon’s patients who was having medical problems in the physician-patient’s area of expertise. These ma- neuvers gave the patient a sense of mastery and control of his situation, the absence of which had previously in- creased his anxiety. In this case, effective staff manage- ment of the patient's refusal culminated in an apparent strengthening of rapport with the physician. As a result of the extended decisionmaking process occa- sioned by patients’ refusal, instances of treatment refusal culmi- nated in the patient's greater sense of involvement and commit- ment to a difficult decision. Case example #10 (continued) —This was the most dra- matic case observed of this phenomenon. This 63-year- old woman had initially refused to undergo amputation of a second gangrenous limb. In our discussions with this patient, we learned that she had some regrets about her previous leg amputation, and that at least intermit- tently she believed that this prior decision had been made too quickly. Once she finally decided to have the second limb amputated (a decision made in part because of the agony and pain she suffered when she tried to 460 Making Health Care Decisions: Appendix D walk on the limb), she became “committed” to the deci- sion and her spirits rose somewhat. When we discussed her decision with her following her operation, she was positive about her choice. The decision had been made carefully. Furthermore, it was one that she had essen- tially made herself; even her husband had not insisted upon the surgery. No one had forced her to have the amputation. Effects on physicians’ attitudes. It is somewhat more diffi- cult to specify the effect of patients’ refusal of treatment on the patients’ care-takers—especially the patients’ physicians. Cer- tainly, as we have already discussed, one finding was that patients who refused procedures (particularly when they did so repetitively) were seen by physicians as troublesome, “diffi- cult” patients. On the other hand. at least episodically, the patient's refusal of treatment offered an opportunity for treat- ment staff to recognize, to identify with, and/or to address some of the psychological problems or conflicts that patients were undergoing, thus contributing to more satisfactory management of the patient. (Note, however, the earlier discussion of the problem of false empathy.) As a consequence of patient refusal, most hospital staff with whom we talked, including physicians, had developed some sensitivity to what we have previously labeled the “hospital fatigue syndrome”; most treatment staff did recognize that hospitalization was often not easy for the patient and that many patients were exhausted by their ill- nesses and by the complicated treatment and diagnostic pro- cedures they underwent. It must also be stressed that while physicians and other treatment staff complained to us or among themselves about “difficult patients,” and were occasionally “sharp” or directive with patients who refused, we witnessed no instances in which treatment staff, either nurses, house officers, or attending physi- cians, became overtly angry or acted in a directly “punitive” manner toward patients. This does not mean that occasional antagonism of the treatment staff toward patients (see case example #22, above) did not result in some patient refusals being acceded to when, arguably, more efforts might or should have been taken to persuade the patient to have treatment. Physicians’ sympathies were most clearly extended, al- though somewhat ambivalently, to dying patients or to their refusing relatives. Here physicians were often heavily involved in discussions about the futility of continuing treatments and routine procedures—though often indicating through their be- havior that they believed it professionally necessary that rou- tine treatments continue. The following patient presents such an example. Case example #32 This 77-year-old long-suffering widowed lawyer was dying from metastatic cancer of the Treatment Refusal 461 prostate. The patient knew that he was dying and why, and he reviewed his life in a positive manner. Although he did not welcome death (in fact, he accepted blood transfusions, which he hoped would improve his condi- tion and make him more comfortable), he was also quite realistic about his condition, recognizing the inev- itability of imminent death. We saw the patient because he refused intravenous fluids, all medications, and re- fused to eat. He said that he did not wish to cause his death per se, but pointed to the limits of pain and dis- comfort he could tolerate and his disgust with having so many procedures. The patient's attending physician was quite sym- pathetic to the patient's plight and psychological state, specifically stating that he did not regard the patient as “depressed.” He even went so far as to cancel a psychi- atric consult that had been requested by one of the house officers, because of his conviction that the patient was not a “psychiatric case.” He also communicated to us his respect for this patient, indicating that he did not intend to push on the patient in any way in light of the medical facts. Nevertheless, despite the inevitability of death and the wishes of family members who were constantly at the patient’s bedside, the patient was given intravenous treat- ment to prevent him from becoming dehydrated. When he had cardiac arrhythmias (already having become som- nolent), he was treated for these with both carotid mas- sage and medication that was altered on a daily basis to strive for maximum therapeutic effect. This case graphically demonstrates the dilemma of treat- ment refusal from the physician's point of view. Trained to treat illness and socialized to believe that passivity in the face of treatable disease is undesirable (perhaps even unethical), phy- sicians find it difficult to accept patients’ refusals, even when they appear to be justified. Thus, even when physicians came to believe that impending death was unlikely to be stayed by further efforts, it was obvious that the decision not to intervene was hard for them to reach and required continued self-justifi- cation. Family Participation in Treatment Refusal. The role that families play in treatment refusal is not simple or unconflicted. Families (and close friends) are often torn between two emo- tionally charged positions: strong identification with and sym- pathy for the refusing patient, and concern that their loved ones might not receive needed treatment. This latter concern is moti- vated both by family members’ altruistic desires to see their loved ones spared unnecessary illness and suffering, and by more personalized fears that the patient's refusal of treatment might jeopardize their own needs. 462 Making Health Care Decisions: Appendix D Family members, including spouses, siblings, children, and in-laws, participated in treatment refusal in a variety of ways. On occasion, a family member was the primary person who refused for the patient. (e.g., case example #15 above). In one such instance a wife was instrumental in refusing a lumbar puncture necessary to evaluate the possibility of treatable de- mentia in a 74-year-old man; the patient seemed to believe that the procedure had something to do with chiropractic, while the reasons for the wife’s refusal were unclear. In most instances, however, the family role in encouraging treatment refusal was not nearly so direct. Instead, families were supportive of the patients point of view, verbalizing their concerns about the quality of care that the patient was receiving, even if like the patient, they did not always understand the full rationale for the proposed treatment. Some family members, however, especially spouses, sided with those who were recommending treatment. Patients were often acutely sensitive to the nature of this “alliance.” Thus, a 45-year-old woman who had refused workup of a coin lesion (possible cancer) of the lung told us that when she discussed this matter with her husband, he suggested that she have the x-ray. She then went on to note that her husband's approach was the same as the doctor's, namely, suggesting “one little thing at a time.” If she agreed to the x-ray, she believed, then “the needle biopsy would come next.” Because her husband was upset by her decision, she believed that he and her other relatives would probably “work on her” to have the additional tests. Similarly, a 59-year-old patient with metastatic gastric car- cinoma refused chemotherapy in part because he realistically believed there was little hope for him regardless of the treat- ment received. He told us that after his wife had spoken with the doctor, she believed he should do whatever the doctor recommended. The patient added, however, that “she doesn't understand these things.” The patient's wife believed the pa- tient should, if necessary, even have another operation (which clearly was not feasible or medically indicated). Family members were sometimes more successful in per- suading the patient to accept treatment; see, for example, case example #19, in which the patient's daughter helped persuade her to have a sigmoidoscopy. In one instance family “persua- sion” was still more direct, amounting to clear, if understand- able, coercion. When a 30-year-old man had his brain injured in a fight and then in a confused manner refused x-ray examina- tion, his brother personally took him to the x-ray suite to have a CAT scan. The examples so far have demonstrated families taking clear positions on one side or the other of the question of refusal. In many cases, however, familial ambivalence about the Treatment Refusal 463 issues involved led them to assume less consistent or firm positions. Case example #37—A 63-year-old executive who had a mass in his mediastinum and who had recently lost thirty pounds in weight, refused CAT scan examination. He had previously had this test performed at another hospital. When he arrived at this hospital, however, he was told that the CAT scanner was better at this hospital and that his surgeons would need their own CAT scan prior to performing any surgery. The patient claimed only to be upset about the need for repetition of the test, but underlying issues included his uncertainty about his diagnosis, the possibility of malignancy, and his mistrust of physicians because a previous biopsy (performed at another hospital in a manner with which he did not agree) did not yield a diagnosis. The patient's wife was concerned by the patient's refusal and initially favored completion of the test. But she, too, became angry at his care-givers when, after his doctors spoke with him and he agreed to have the test, it then had to be delayed because of machinery breakdown. The patient's wife's emotions moved among anger, distress, anxiety, and fear. Perhaps because of her own feelings, she was quite sym- pathetic to the patient's changing emotional responses to his plight. As his wife put it, “he needs to take control of the situation.” This was the patient’s usual style and she was trying to cope in the same way. Despite their concerns, family members often had reserva- tions about taking too direct a role either in attempting to force treatment on patients or in bargaining with the patients’ doc- tors. Part of the reluctance was rooted in family members’ fears of taking responsibility for a major treatment decision. Case example #10 (continued)—Despite the patient’s husband's hope that his wife would eventually consent to have her second leg amputated, he would not, for several days, declare his felt desire that the patient con- sent to surgery. He explained that his wife had always been an independent person. It was not his practice to tell her what to do. As he later noted following her successful surgery, he was reluctant to “assume the ma- jor responsibility” for the patient's treatment decision. Although eventually the husband did recommend to his wife that she have the procedure, neither of them ever regarded his request as determinative of her decision. Other families also demonstrated this reluctance to take responsibility for a previously independent person. Even in life-threatening emergencies, it appeared to be difficult to alter patterns of familial interactions, including allocation of deci- sionmaking responsibility. 464 Making Health Care Decisions: Appendix D This matter was, however, quite complicated. For while family members recognized that it was the patient who would suffer the pain or discomfort of treatments (or who must live with the consequences of treatment refusal) family members had their own needs to consider as well. The decisionmaking in the following case involved such an instance. Case example #13 (continued)—This 61-year-old woman, exhausted from years of illness and intermit- tently ready to die, refused potassium, a gastrointestinal workup for bleeding, arthroscopy, and at one point a lymphangiogram. She, nevertheless, agreed to have an A-V fistula put in her arm to permit renal dialysis. She informed us she had agreed to the fistula because of pressure from her family, primarily her husband and sister. They were very concerned and wanted her to con- tinue to receive care; she described her decision as an accommodation to their pressure. The husband in partic- ular pushed the patient along, telling her that she was a strong person who had survived all these years and who could do it again. The patient described her sister as somewhat more understanding and, lately at least, as more willing to let her decide what she can tolerate. The husband's needs for the patients survival were obviously stronger than the sister's. In another case, the obviously concerned daughter of a 69- year-old woman became extremely upset when the hospital permitted her mother to terminate a pulmonary angiogram be- fore it could be completed. It was the daughter’s view that the patient's consent was irrelevant. The procedure should have been finished no matter what the mother’s desires. This case was interesting because even though the patient was not incom- petent, it had been the patient’s husband, not the patient, who had signed the consent form for the pulmonary angiogram. A striking finding in our study, however, was that families did not necessarily have a better understanding, and were per- haps even more poorly understanding, than patients about the medical rationale for the refused treatment. Thus, to the extent that family members concurred with the patient or attempted to override the patient's refusal, their actions were largely based on their own psychosocial concerns, their idiosyncratic fears, or their partial understanding of the risks or potential conse- quences of nontreatment, rather than any good appreciation of the medical issues involved. Case example #3, previously dis- cussed, was such a case. The family as well as the patient failed to understand what was involved in a heart test utilizing radi- oisotopes. See also case example #15 above. In another case also previously discussed (#11), a patient refused chem- otherapy treatment for colon cancer. The patient’s wife was dissatisfied with the patient's care, and may have played a role Treatment Refusal 465 in the patient's refusal because of her belief in faith healing. Family members also had problems in understanding issues of medical uncertainty, a fact that seemed to play a role in both the patient’s refusal and the family’s support of that refusal. The previously discussed case example #5, in which a patient re- fused workup of a fever of unknown origin, is an example of this. In these and related cases the patient's family members, often not fully understanding the rationale for treatment recom- mendations, came to share the patient’s anger about the care that had been proposed. Among the more subtle family reactions we observed was the recognition by family members that while, because of their usual relationship to the patient, they could not play too active a role in treatment decisionmaking, they could sometimes by virtue of this relationship “set limits.” Thus, when one patient (see case example #27) refused hyperalimentation, his wife gently indicated to him that if the low residue feedings had not worked after several days, then, of course, the patient would have hyperalimentation. Families sometimes acted directly as proxies for patients in ways seemingly not recognized by informed consent doctrine. Thus we witnessed at least five cases, in which the family signed the patient into the hospital, or even signed for a diag- nostic procedure, even though the patient was not incompetent to do so. The effect of this phenomenon on later consent and refusal is unknown, though as previously discussed, some of these patients did subsequently refuse treatment. The subtle blend of interests in which families advocated for patients, while taking their own needs into account, is perhaps best exemplified by the “death with dignity” cases. Here we witnessed several instances in which families at- tempted, with varying degrees of success, to act as proxies for patients to ensure that they were treated as they would have wished, assuming they had been competent. In the following case, the family was successful in that goal. Case example #16 (continued)—Following a brainstem infarct, which rendered the patient paralyzed except for eye movements, the patient's family refused feeding tubes, intravenous fluids and all other interventions. The patient’s daughter was convinced that her mother did not want to live. The daughter interpreted the patients visual inspection of the intravenous set-up to mean “No, [don’t want it.” In addition, the patient had previously told her daughter that in the event of severe disability, she should not be lied to nor should she be “allowed to linger.” Such discussion had taken place on more than one occasion, especially following the death of the patient’s husband two years previously of a brain tumor. The patient’s brother-in-law also noted that it would be a great mistake 466 Making Health Care Decisions: Appendix D for this patient to continue to receive any active treat- ment: “She would dislike us for it, and we would dislike ourselves.” Some of the “death with dignity” cases, however, illus- trated the limits of proxy decisionmaking in such situations. First, despite families’ attempts to refuse all treatment for their dying family members, doctors were not always willing to with- hold what they regarded as conventional or routine treatments. The most striking example here was when, despite the objec- tions of the family of a patient with terminal breast cancer to her having a catheter inserted in her bladder, the patient was deliberately, covertly catheterized. It is probably of some sym- bolic significance for the doctor-patient relationship that medi- cal treatment staff were usually unwilling to overturn the objec- tion of family members who refused treatment for patients who were not dying, while being willing to do so when patients were dying. Not only treatment staff, however, but even well-inten- tioned families were not always able to fulfill patients’ requests, even when such requests were clearly manifestations of prior competent choice. Thus, not only the staff, but the patient’s family overcame the prior wishes of the patient with terminal breast cancer mentioned above. The family took her to the hospital so that her pain could be controlled, because their attempts to care for her at home had been unsuccessful. The patient had wanted to die at home, feeling very strongly about this. But, as the patient began vomiting blood, the family had to consider their own needs, as well as the patient’s; thus they took the patient to the hospital to die. Another example of this phenomenon involved the family members in attendance when a 77-year-old former lawyer, an elegant, intelligent, dignified man (see case example #32), re- fused intravenous fluids, oral medications, and other pro- cedures, because he could no longer tolerate his pain and dis- comfort, and because he knew he was dying. Although the patient’s daughters desired that this patient be allowed to die in a dignified manner (and, in effect, as soon as possible), they too found it impossible to accede completely to the patient’s wishes. The daughters visited him, at least initially, over his objection and despite a notation on the patient’s chart that he did not want ‘to see them or his sister. When the daughters visited, however, the patient did not ask them to leave, illustrat- ing the problems that dying patients have in adhering to their own previous, competent wishes—as well as the difficulties in determining the extent to which third parties should real- istically be expected to adhere to such requests. Although family members sometimes played important roles in the decisionmaking of refusing patients, these roles were subtle and often conflicted. Given the problems in com- Treatment Refusal 467 munication between patients and their doctors, let alone be- tween families and doctors, to suggest that families be encour- aged to participate in more meaningful ways as decision makers when their loved ones refuse treatment would require a signifi- cant reorientation of the structure of medical care. For while family members often played significant roles in treatment re- fusal, like patients, family members often did not play informed and understanding roles in these matters. Discussion Refusal of treatment and diagnostic procedures in medical hospitals is a common phenomenon that has received little previous attention in the medical literature. The reasons for this neglect are of interest. In obtaining consent to undertake this study, the investigators were told by a number of senior physi- cians that, although we were welcome to visit their wards, they were convinced that no instances of treatment refusal would be found there. “I do not recognize the phenomenon you are de- scribing,” one such physician said. He then noted, apparently quite sincerely, that since patients were always presented with a variety of options concerning their care, there was no reason for them to refuse; they need only select the option they preferred. In contrast, when the investigators approached the head nurse on a ward under this physician's supervision, she re- sponded to a description of the purposes of the study with, “Oh, that happens all the time.” Other nurses told us how pleased they were to see someone studying a phenomenon with which they had to cope daily, but about which they had re- ceived little instruction or support. The divergence between the views of senior physicians and of nurses on the wards is striking. Part of the explanation for this discrepancy undoubtedly lies in the multiple layers of insulation that separate the two positions. The more removed a physician is from the minute-to- minute experience on the wards, the more likely that someone else will have heard about and dealt with a patients treatment refusal. Thus, a certain percentage of refusals are managed by the nurses, without ever coming to the attention of a physician. Even when refusals fail to be resolved, they will frequently, due to the press of information to be communicated, remain un- discussed with the physicians on the case. Of those that are passed on, many will be resolved by the house staff. Only a fraction of cases, usually those involving more significant re- fusals that are resistant to initial attempts at resolution, will come to the attention of the attending staff, and even fewer will be passed on to the administrators above them. Yet refusals that threatened to significantly affect patient care were not rare in this study. Some other explanation is 468 Making Health Care Decisions: Appendix D needed to account for the physician who did not “recognize the phenomenon you are describing.” The answer may lie in the meaning refusal has for clinicians. Clinging to a (ofttimes de- served) self-image as individuals selflessly dedicated to the alleviation of suffering, many physicians find it difficult to acknowledge patients’ rejections of their ministrations. The im- plications of such rejections—that the doctor has not been acting in the patient's best interests, but perhaps in some selfish way, or that the patient knows better than the physician what is required for the patient's care—are often unacceptable to physi- cians. They may respond in a manner, therefore, that negates those implications. Anger, attribution of irrationality to the patient, and denial that the refusal has occurred all serve this purpose. A number of times during the course of this study, the investigators approached attending physicians to solicit their views of the reasons for their patients’ refusals, only to be told that no refusal had, in fact, occurred. This inevitably sent the investigators back to the nursing staff and/or the patient for confirmation of the event. The evidence was clear: two refusals were in progress. The patient was refusing treatment and the physician was refusing to acknowledge the existence of the patient’s act. It should be emphasized that this was not the ordinary, or even a common, response to treatment refusal. Most physicians acknowledged the presence of a refusal and proceeded to deal with it. But the tendency to deny refusal as a reality may account in part for the neglect of refusal as a clinical phe- nomenon and the disbelief with which accounts of the purpose of this study were met. If refusal is to be seen for what it is, however—both a reflection of difficulties in the doctor-patient relationship and a source of further difficulties—the first step must be to acknowledge the phenomenon as a reality, in fact a commonplace, of hospital-based medical care. Similarly, the possibility of taking appropriate preventive and reactive meas- ures depends on such recognition. Prevention of Treatment Refusal. There are always costs associated with refusal of treatment. Even if resolved rapidly and to the satisfaction of all parties, delay of treatment and diagnostic procedures means a prolonged hospital stay, with its attendant economic costs. In addition, refusal burdens the doc- tor-patient relationship with heavy emotional baggage. Patients may feel angry that such a measure was necessary to draw attention to their feelings, guilty about “hurting the doctor”— since many of them accurately sense the narcissistic wound physicians often suffer when a patient refuses—or fearful that their care-takers will retaliate by denying them needed care at some later point. Doctors may, in fact, respond as patients fear, labeling them as “difficult,” “gomers,” or “bitches” (all reac- Treatment Refusal 469 tions observed in this study), and seeking to avoid responsibil- ity for providing further care. And, of course, when a refusal is not resolved and the patient never receives the required medical care, an additional cost may be paid in human suffering and loss of life. The costs just enumerated suggest that refusal is a relatively inefficient way for patients to participate in their medical care. If refusals can be prevented, without ignoring legitimate patient concerns over what is being done to them, why, and by whom, that would appear to be the preferable course. Although the results of this study suggest that many of the causative influ- ences in refusal—for example, previous bad experiences with medical care or individual psychopathology—are beyond the power of physicians to change, the discovery that in most cases multiple factors interact to produce refusal suggests that inter- vention in those areas under physician (and hospital) control may have considerable impact on the process. Communication with patients is the most obvious area to which attention is directed. We are willing to conclude, based on this study and a companion study performed for the Presi- dent's Commission by Lidz, et al., that it was common in the hospitals we studied for physicians to fail to inform patients about the nature, purpose, and risks of planned procedures in a way that would enable them to make meaningful decisions. We further suspect that these hospitals are typical of most settings in the country. Attention to providing adequate information, which means not merely detailing what is to be done, but also eliciting patients’ fears and hesitations about the procedures and responding to them, would, we believe, substantially de- crease the rate of refusal of treatment. In fact, improving com- munication with patients is probably the single most effective preventive measure that can be taken. The identification of doctor-patient communication as a problematic area is not unique to this study. Other critiques and analyses of modern medical care, particularly in referral centers where doctors and patients are often strangers to each other, have come to the same conclusion. Yet, the costs of improving communication are frequently omitted from the analysis. The modern hospital is structured to facilitate the efficient applica- tion of technology to the diagnosis and treatment of medical disorders. As much as the faults in the doctor-patient relation- ship may be decried, they contribute to the efficiency of the system. If doctors were to take the time to provide adequate explanations to patients, much less to get to know patients well enough to intuit their fears and anticipate difficulties that might arise, other tasks they now perform would be neglected instead. In the end, more physicians would be needed to deliver care to the same number of patients. The resulting increase in costs, coming at a time when society is already looking for ways of 470 Making Health Care Decisions: Appendix D cutting back on the expense of hospital care, might well be unacceptable on a societal level. Some may claim that it takes the same amount of time to be an empathic and informative physician as it takes to be a cold and enigmatic one. This is almost certainly not the case. If we advocate changes in the way hospital care is delivered, we ought to be prepared to bear the costs those changes will bring. In determining the relative value of encouraging doctor- patient communication, however, one hidden cost of failing to act ought to be taken into account. There are undoubtedly a large number of cases in which patients should (in some objec- tive sense) refuse or question the need for treatment, but fail to do so. Patients who are silently suffering side effects of medica- tion (like the patient who kept taking her laxative until her diarrhea became too much to tolerate) or who are enduring unreasonable levels of anxiety as a result of unasked questions and unspoken fears are less obvious victims of the current system. Unless physicians take the initiative in changing the way hospital-based care is delivered, these patients will con- tinue to suffer. While neither the cost nor the extent of their suffering can be quantified, it should not be forgotten. Response to Refusal. Whether or not preventive measures are taken, it is unlikely that any phenomenon as prevalent as treatment refusal will rapidly become extinct. The data suggest some ways in which the response to refusal can be made more effective. When responses to refusal by nurses or physicians were unavailing, it was often because they were undertaken in an undifferentiated way. That is, an attempt was made to convince the patient to undergo the treatment or procedure without first determining the patient’s understanding of the treatment and reasons for refusing. The care-giver was thus often guilty of mouthing platitudes about the importance of treatment, or of addressing what he or she thought were the patient’s concerns, rather than confronting those concerns directly. Examples of such behavior included case example #7, in which the efforts of the physicians to talk the patient into having spinal anesthe- sia were focused on the possibility of long-term effects from delaying fixation of the fracture. The patient’s refusal, however, was grounded in his fear of another hypotensive episode while under anesthesia, a concern that was apparently never directly addressed. The reasons for this failure to come to grips with the issues underlying the patient's refusal are probably similar to the causes of other difficulties in doctor-patient communication. Time and an ability to relate emphatically are required to en- gage the patient in a discussion of his or her concerns. Both may be in short supply on a busy medical ward. But having failed to address the patient's concerns, the physician may then con- Treatment Refusal 471 clude that the refusal is irreversible and stop trying to under- stand it. The importance of individualizing the response to refusal cannot be overstated. Along with the failure to ascertain the real basis for the patient's refusal, we witnessed what might be termed a “false adherence” to the ideal of autonomy on the part of some physi- cians. Rather than undertaking an exploration of the basis for the patients refusal, or taking the time to explain what the treatment or procedure involved, some physicians were all too willing to grant their patients the “right” to refuse. They as- sumed both that their patients had an accurate conception of what they were refusing and what the effects might be, and that the patient had made a rational and irreversible decision. It was almost as if the doctor were saying, “If this patient doesn’t want the benefit of my services, that’s his right and it's fine with me. There's lots of work for me to do elsewhere.” The fallacy in this approach should be evident. Few pa- tients actually understood the consequences of their refusal, some acted on what might be considered “irrational” grounds, and, as the follow-up showed, few refusals were truly un- changeable. We have called this a “false” adherence to the ideal of autonomy because in these cases it often appeared as if the physician's anger over the patient's refusal was a more impor- tant motivating factor than any principled belief in patient self- determination. There are many circumstances in which patients’ refusals should be respected. But it is in no way an abrogation of pa- tients’ autonomy to provide information about treatments re- fused or about the consequences of refusal. In fact, a desire for such information often appeared to be the motivating factor behind refusals. When physicians responded by backing off from the original recommendation for treatment, patients were left bewildered. If refusals, which from the patients’ point of view, represented requests for more information or expressions of a desire to be talked into anxiety-laden procedures (such as surgery), led physicians to withdraw their recommendation for treatment, patients were left with the conclusion that the pro- cedures had not been important in the first place. They became even more determined to refuse. In addition, this then led them to wonder about the importance of other recommendations their doctors had made and served to lessen their overall level of confidence in those who were responsible for their care. In view of the disinclination of many physicians to explore with their patients in any depth the basis for refusals, one might expect that considerable use would have been made of the psychiatric consultation service for that purpose. That was rarely done in this study, unless the patient had a known psychiatric diagnosis. Physicians frequently resisted sug- gestions that psychiatric consults be requested by noting that 472 Making Health Care Decisions: Appendix D the patient was not “crazy,” and therefore psychiatry had little to offer. They were also frequently concerned that calling in someone to address the issue of refusal with the patient would have the effect of angering the patient, or of provoking further refusals or a request for discharge against medical advice. The investigators were impressed with how frequently they would be warned by house officers and attending physicians that a particular patient was “angry,” “difficult,” or likely would be uncooperative. In fact, only one patient who we were able to approach refused to talk about her refusal, and many patients seemed pleased to have an opportunity to do so. In large part, we suspect, the difference between the way these patients acted with their doctors and with us reflected the advantages of having someone outside the scope of the confron- tation talk with the patient. Patients were often pleased to talk to someone who was interested in their hospital experiences and could listen sympathetically as they described the events that led to their refusal. We conclude that it is incorrect to assume that intervention with the refusing patient will inevita- bly make things worse. Patients are likely to respond favorably to attempts to discover the difficulties underlying their refusals and to opportunities to express the feelings involved. Legal and Ethical Implications. The results of this study suggest that there is a wide divergence between legal theories concerning medical care and the reality of hospital-based prac- tice. The doctrine of informed consent, in particular, appears to suffer in the transition from judicial decision and statute book to the hospital ward. Informed consent doctrine envisions pa- tients as autonomous entities making rational decisions about their medical care. After accumulating information from the physician—for the doctrine mandates (in one of its most com- mon forms) the disclosure of the nature and purpose of the procedure, alternatives, and all risks that a reasonable person might find material —the patient, according to the doctrine, chooses whether to undergo the recommended procedure or to select one of the alternatives. There has been a great deal of discussion about the ability of patients to understand and thoughtfully employ intrinsically complicated medical data. For the refusers on the wards exam- ined in this study, such a dispute was academic at best. Often patients were not told what treatment or procedure had been ordered for them, much less asked to decide whether or not to accept it. The purpose of the procedure was frequently obscure and the risks commonly went unmentioned. Presentation of alternatives was extraordinarily rare. The main concern of the patients we interviewed was not to select the best treatment from those available, but to find out what was being selected for them and why. Treatment Refusal 473 (An important caveat must be inserted here. The investiga- tors generally did not have a chance to observe interactions between attending physicians and patients, although house staff-patient interactions were extensively observed. One would expect that discussions of major treatments and procedures, especially major surgical procedures, which were more often left to the attendings, might correspond more closely to the model envisioned by the doctrine of informed consent. How- ever, our conclusions are probably valid for the discussions we observed about diagnostic procedures, medications, and ad- junctive care, which were generally negotiated at the house staff level.) Thus, it would seem that the threat of malpractice suits for failure to obtain informed consent, though much discussed in medical journals, has been an insufficient stimulus to prompt the hospital-based physicians in our study to conform to the doctrine. The ethical implications of this finding are also significant, particularly since the evolution of informed consent law repre- sents the legalization of widely subscribed ethical imperatives. Also, although it is unclear if informed consent as a legal doctrine applied to many of the treatments and procedures that were refused (e.g., many elements of adjunctive care and non- invasive tests and procedures), the ethical principles underly- ing the doctrine are still relevant. The right of patient autonomy in decisionmaking (in the sense of patients having sufficient information to reach independent conclusions) is clearly more enthusiastically accepted in textbooks of medical ethics than on the medical ward. Conversely, despite the fashionable de- rogation of paternalism in medical care, decisionmaking by doctors for patients (presumably in the patients’ interests) ap- pears to be the bedrock upon which much hospital-based care is built. Overt coercion was, nonetheless, not observed except in the case of incompetents. A second area of legal regulation to which our findings speak is substitute decisionmaking for the incompetent. Most recent legal decisions in this area have held that before an incompetent person (for example, someone who is severely demented) can be treated, or before treatment can be withheld in life-threatening circumstances, a formal declaration of in- competency and appointment of a guardian is required. Despite the fact that many of the patients we observed were either temporarily or permanently incompetent, only one patient ac- tually went through this process. In most cases, family mem- bers were permitted to sign incompetent relatives into the hos- pital, to consent to treatment (including surgical procedures), and to play a significant role in the decision to terminate life support when the patient was near death. Although this role for 474 Making Health Care Decisions: Appendix D . family members was formerly widely accepted, it has been called sharply into question in a number of recent court cases, and it is currently of uncertain validity. Legal reacceptance of the role of families as proxies, al- though desirable in many respects, should be approached with caution. In this study, familial interests often diverged from patients’ interests, because of family members’ investments in the survival of their loved ones. When offered the opportunity to make decisions on behalf of their relatives, many family members declined to fill that role, apparently because of long- standing patterns of decisionmaking and dependency in the family. Finally, family members appeared to be almost always less well informed than most competent patients and, since the individual in the family who actually held the decisionmaking power was not always easy to identify, adequate education of family members for a decisionmaking role raises the prospect of an enormous investment of time and resources. Although these factors are most critical when families are being asked to make substituted decisions for incompetent patients, they apply equally as well to the participation of family members in the decisionmaking of competent patients. Even in the absence of familial consent for the treatment of incompetent patients, however, involuntary treatment of pa- tients deemed by the staff to be incompetent was a common practice. It should be noted that the staff in several of the hospitals studied appeared to have a uniform rule of thumb to determine incompetency, namely that a patient be disoriented and incoherent—a stricter standard than most courts would use. Nonetheless, once a patient fell into this category, he or she was often restrained (using arm and leg restraints and “Posey” vests), treated with sedative and antipsychotic medication to control disruptive or annoying behavior (e.g., loud screaming), and given whatever treatment short of surgery or intrusive diag- nostic procedures (generally classified as those that would re- quire anesthesia) the physicians deemed necessary. The differences between these practices and those in many psychiatric hospitals, which also often deal with incompetent patients, is quite striking. As a result of a number of court decisions and statutes, psychiatric facilities in many jurisdic- tions are required to obtain formal determinations of incompe- tency and substituted consents from legally authorized proxies (family members do not suffice) before involuntary treatment can take place. In addition, medication and restraints cannot be used merely because patients are loud or wandering from room to room, but only if patients endanger their own safety or the safety of other patients. The emergency exception to these re- quirements (i.e., doctors may act as necessary to preserve the Treatment Refusal 475 life of incompetents in emergencies and the law will presume consent) did not usually provide a justification in this study. Many of the cases of involuntary medication and restraint did not involve emergencies, and several went on for prolonged periods of time (up to several weeks), during which the ap- pointment of a guardian might have been sought. The advantages of this less formal model of decisionmak- ing for incompetents are evident in the hospital setting. A physician could hardly permit a disoriented patient to fall out of bed or to interfere with the care of other patients. It would seem cruel to deny admission or needed medical care to pa- tients during the pendency of guardianship proceedings. A certain measure of paternalism is probably justified in these cases, at least over the short term. Whether such actions should continue for a number of weeks or where major treatment deci- sions are concerned without taking steps to insure that patients’ interests are formally protected is less clear. Discussion of these issues might help physicians, especially those still in training, clarify both the necessity for paternalistic action and the appro- priate limits to paternalism. The judiciary will ultimately have to decide whether to endorse the informal decisionmaking pro- cesses currently used in the medical hospital, or the more formal processes now required in many psychiatric settings. A final comment on legal and ethical issues concerns what has come to be called “death with dignity.” Here too, despite the court rulings in some jurisdictions formalizing the decision process, the decision as to whether to terminate life support in these hospitals was made by the physicians, taking into ac- count the wishes of the family. As noted earlier, the widespread publicity given the concept of death with dignity, as well as the vociferous wishes of families, were often insufficient to prevent physicians from carrying out what they considered routine measures to combat complications in patients who were clearly terminal. One patient, for example, who was suffering from metasta- tic cancer and exhibited a repeated inability to sustain his blood pressure, came out of his stuporous state long enough to request that his doctors discontinue life-support efforts and instead administer sufficient morphine to end his life. Instead, the next time his blood pressure dropped, vigorous support measures were instituted. They were terminated only when it was clear that they had failed, and a medically oriented deci- sion was made at that point to allow the patient to die. Teaching and writing about “death with dignity” has clearly made an impact on the medical profession, but not enough of an impact to overcome in all cases the natural tendency of physicians to treat what can be treated. 476 Making Health Care Decisions: Appendix D Conclusion The questions posed earlier about the implications of treat- ment refusal can now be answered, at least in a tentative way, based on the data that have been presented. Refusal is indeed evidence of difficulty in the doctor-patient relationship, but it also reflects some positive efforts by patients to establish con- trol over their care. Although a variety of causative influences interact in a complex manner, most refusals, even if not reflec- tive of a real understanding of the situation, could not be said to be “irrational,” at least insofar as that term connotes a detach- ment from reality. It is not difficult to empathize with most patients who refuse treatment, even when one recognizes that they will be harmed by their actions. We believe that an examination of the seemingly mundane aspects of medical care—the practices and ethics of the routine case—can prove as useful to clinicians and theoreticians as a focus on the more dramatic issues. This study of the phe- nomenon of treatment refusal is a reflection of that belief. An examination of the causes of refusal points to significant diffi- culties in the modern doctor-patient relationship. Whether soci- ety will decide that the gains from repairing the flaws in the relationship are worth the ultimate cost remains to be seen. References 1. Fiore N: Fighting cancer - one patient's perspective. N ENGL |] MED 300:284-289, 1979. 2. Veatch RM: Models for ethical medicine in a revolutionary age. HASTINGS CENTER REP 2 (June): 5-7, 1972. 3. Brody DS: The patient's role in clinical decision making. ANN INT MED 93:718-722, 1980. 4. Blumgart HL: Caring for the patient. N ENGL |] MED 270:449-456, 1964. 5. Ingelfinger FJ: Arrogance. N ENGL |] MED 303:1507-1511, 1980. 6. Groves JE: Taking care of the hateful patient. N ENGL J] Mep 298:883-887, 1978. 7. Kris K: Psychiatric consultation in the management of patient ambivalence interfacing with the doctor-patient relationship. Am J PsycHIATRY 138:194-197, 1981. 8. Cousins N: ANATOMY OF AN ILLNESS. New York, W.W. Norton,1979. 9. Hastings AC, Fadiman J, Gordon JS (eds.): HEALTH FOR THE WHOLE PERSON. Boulder, Westview Press, 1980. 10. Boston Women's Health Book Collective: Our Bobiks, OUR SELVES. New York, Simon and Schuster, 1976. 11. Levin A: TALK Back 10 Your Doctor. Garden City, NY, Doubleday, 1975. 12. Illich I: MepicaL Nemesis. New York, Pantheon, 1976. 13. Meisel A: The expansion of liability for medical accidents: From negligence to strict liability by way of informed consent. NEBRASKA Law REVIEW 56:51-152, 1977. Treatment Refusal 477 14. Annas GJ, Glantz, LH, Katz BF: INFORMED CONSENT TO HuMAN EXPERIMENTATION: THE SUBJECT'S DILEMMA. Cambridge, MA, Ballinger, 1977. 15. Meisel A, Kabnick LD: Informed consent to medical treatment: An analysis of recent legislation. UNIVERSITY OF PITTSBURGH Law REVIEW 41:408-564, 1980. 16. Annas GJ: THE RiGHTs OF HosPITAL PATIENTS. New York, Avon Books, 1975. 17. In re President and Directors of Georgetown College, Inc., 331 F.2d 1000 (D.C.Cir.), cert. denied, 377 U.S. 978 (1964). 18. In re Estate of Brooks, 32 I1l.2d 361, 205 N.E.2d 435 (1965). 19. John F. Kennedy Memorial Hospital v. Heston, 58 N.J. 576, 279 A.2d 670 (1971). 20. Lane v. Candura, 376 N.E.2d 1232 (Mass. 1978). 21. In the Matter of Spring, 405 N.E.2d 115 (Mass. 1980). 22. In re Quinlan, 70 N.J. 10, 355 A.2d 647, cert. denied sub nom. Garger v. New Jersey, 429 U.S. 922 (1976). 23. Eichner v. Dillon, 52 N.Y.2d 263, 420 N.E.2d 64 (1981). 24. Superintendent of Belchertown State School v. Saikewicz, 373 Mass. 728, 370 N.E.2d 417 (1977). 25. Rogers v. Okin, 478 F. Supp 1342 (D.Mass. 1979), aff'd in part, 634 F.2d 250 (1st Cir. 1980), cert. granted, 49 U.S.L.W. 3779 (1981). 26. Rennie v. Klein, 4462 F. Supp. 1131 (D.N.]. 1978), 476 F. Supp. 1294 (D.N.]J. 1979), Nos. 79-2576, 79-2577 (3d Cir. July 9, 1981). 27. In re K.K.B., 609 P.2d 747 (Okla, Sup. Ct. 1977). 28. In re Yetter, 62 Pa. D.&C.2d 619 (1973). 29. In re Custody of a Minor, 379 N.E.2d 1053 (Mass. 1978). 30. Himmelhoch J, Davis N, Tucker G, et al.: Butting heads: Patients who refuse necessary procedures. PSYCHIATRY MED 1:241-249, 1970. 31. McCartney JR: Refusal of treatment: Suicide or competent choice? GEN Hosp PSYCHIATRY 4:338-343, 1979. 32. McCartney JR: Suicide versus right to refuse treatment in the chronically ill. PsycHOSOMATICS 19:548-551, 1978. 33. Bowes WA, Selgestad B: Fetal versus maternal rights: Medical and legal perspectives. OBSTET GYNECOL 58:209-214, 1981. 34, Leiberman JR, Mazor M, Chaim W, et al.: The fetal right to live. OBsTET GYNECOL 53:515-517, 1979. 35. Appelbaum PS, Gutheil TG. Drug refusal: A study of psychiatric inpatients. AM J] PSYCHIATRY 137:340-346, 1980. 36. Bosk CL: Occupational rituals in patient management. NEw EncL] MED 303:71-76, 1980. #U,S, GOVERNMENT PRINTING OFFICE : 1983 0-401-121/846 President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research Suite 555 2000 K Street, NW Washington, DC 20006 Postage and Fees Paid President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research Official Business Penalty For Private Use $300 U.S.MAIL B Ra 50 viliiiiin co2a78u47519