Office of Technology Assessment Congressional Board of the 99th Congress TED STEVENS, Alaska, Chairman MORRIS K. UDALL, Arizona, Vice Chairman Senate ORRIN G. HATCH Utah CHARLES McC. MATHIAS, JR. Maryland EDWARD M. KENNEDY Massachusetts ERNEST F. HOLLINGS South Carolina CLAIBORNE PELL Rhode Island WILLIAM J. PERRY, Chairman H&Q Technology Partners DAVID S. POTTER, Vice Chairman General Motors Corp. (Ret.) EARL BEISTLINE Consultant CHARLES A. BOWSHER General Accounting Office House GEORGE E. BROWN, JR. California JOHN D. DINGELL Michigan CLARENCE E. MILLER Ohio COOPER EVANS Iowa DON SUNDQUIST JOHN H. GIBBONS (Nonvoting) Advisory Council CLAIRE T. DEDRICK California Land Commission JAMES C. FLETCHER University of Pittsburgh S. DAVID FREEMAN Consultant JOSEPH E. ROSS Acting Director Congressional Research Service Director JOHN H. GIBBONS Tennessee MICHEL T. HALBOUTY Michel T. Halbouty Energy Co. CARL N. HODGES University of Arizona RACHEL McCULLOCH University of Wisconsin LEWIS THOMAS Memorial Sloan-Kettering Cancer Center The Technology Assessment Board approves the release of this report. The views expressed in this report are not necessarily those of the Board, OTA Advisory Council, or individual members thereof. PAYMENT FOR PHYSICIAN SERVICES Strategies for Medicare OTA Reports are the principal documentation of formal assessment projects. These projects are approved in advance by the Technology Assessment Board. At the conclusion of a project, the Board has the opportunity to review the report, but its release does not necessarily imply endorsement of the results by the Board or its individual members. CONGRESS OF THE UNITED STATES % !; Olllce of Technology Assessment ~ / Washington, D. C. 20510 0,.¢> 3>"' l"4>cFt'NQLOG"i r.~":> Recommended Citation: U.S. Congress, Office of Technology Assessment, Payment for Physician Services: Strategies for Medicare, OTA-H-294 (Washington, DC: U.S. Government Printing Office, February 1986). Library of Congress Catalog Card Number 85-600641 For sale by the Superintendent of Documents U.S. Government P:Hnting Office, Washington, DC 20402 Foreword Medicare coverage of physician services for elderly and disabled beneficiaries improves their financial access to medical care. But Medicare's payment methods have also fueled increases in expenditures for physician services, which are now one of the most rapidly growing parts of the Federal budget. The method of customary, prevailing, and reasonable charge payment is inherently inflationary and contains incentives for providers to use additional and more expensive services. To curtail continuing increases in expenditures for physician services, the Deficit Reduction Act of 1984 (Public Law 98-369) froze physician charges to Medicare beneficiaries for 15 months beginning July 1, 1984. That act also mandated OTA to examine alternative methods of paying for physician services in order to guide payment reform. The House Energy and Commerce Committee, the House Ways and Means Committee, and the Senate Finance Committee have jurisdiction over physician services under Medicare and that section of the act. The Senate Special Committee on Aging also requested OTA to study the effect of physician payment methods on the use of medical technology. In preparing this report, OTA staff drew on the expertise of members of the advisory panel, members of the OTA Health Program Advisory Committee, and experts in medicine, economics, insurance, industry, and health policy. Drafts of the final report were reviewed by the advisory panel, chaired by Dr. Sidney S. Lee; OTA's Health Program Advisory Committee, also chaired by Dr. Lee; and numerous individuals and organizations with expertise and interest in the area. We are grateful for their assistance. Key OTA staff involved in the analysis were Jane E. Sisk, Charles L. Betley, Polly M. Ehrenhaft, Peter McMenamin, Elaine J. Power, Gloria Ruby, Ellen S. Smith, and Kerry Britten Kemp. (!:116~ JOHN H. GIBBONS Director List of Related OT A Reports • Medicare's Prospective Payment System: Strategies for Evaluating Cost, Quality, and Medical Technology. OTA-H-262, October 1985, GPO stock #052-003-01010-1. • Technology and Aging in America. OTA-BA-264, June 1985, GPO stock #052-003-00970-6. • Federal Policies and the Medical Devices Industry. OTA-H-229, October 1984, GPO stock #052-003-00965-0. • Medical Technology and Costs of the Medicare Program. OTA-H-227, July 1984, GPO stock #052-003-00957-9. · • Medical Technology Under Proposals To Increase Competition in Health Care. OTA-H-190, October 1982, NTIS order #PB 83-164 046. • The Implications of Cost-Effectiveness Analysis of Medical Technology. OTA-H-126, August 1980, NTIS order #PB 80-216 864. • Assessing the Efficacy and Safety of Medical Technologies. OTA-H-75, September 1978, NTIS order #PB 286 929. • Technical Memoranda: -Update of Federal Activities Regarding the Use of Pneumococcal Vaccine. OTA-TM-H-23, May 1984. GPO stock #052-003-00955-2. -Diagnosis Related Groups (DRGs) and the Medicare Program: Implications for Medical Technology. OTA-TM-H-17, July 1983, GPO stock #052-003-00919-6. • Case Studies: -Nurse Practitioners, Physician Assistants, and Certified Nurse Midwives: Quality, Access, Cost, and Payment Issues. (forthcoming) -Nuclear Magnetic Resonance Imaging Technology: A Clinical, Industrial, and Policy Analysis. OTA-HCS-27, September 1984, GPO stock #052-003-00964-1. -Cost Effectiveness of Automated Multichannel Chemistry Analyzers. OTA-BP-H(9)-4, April 1981, NTIS order #PB 81-209 793. NOTE: Reports are available through the U.S. Government Printing Office, Superintendent of Documents, Washington, DC 20402, (202) 783-3238; and the National Technical Information Service, 5285 Port Royal Rd., Springfield, VA 22161, (703) 487-4650. OTA Project Staff-Payment for· Physician Services: Strategies for Medicare Roger C. Herdman, Assistant Director, OTA Health and Life Sciences Division Clyde J. Behney, Health Program Manager Jane E. Sisk, Project Director Charles L. Betley, Research Assistant Polly M. Ehrenhaft, Senior Analyst Peter McMenamin, Senior Analyst Elaine J. Power, Research Analyst Gloria Ruby, Senior Analyst Ellen S. Smith, Analyst Kerry Britten Kemp, Health and Life Sciences Division Editor Other Contributing Staff Becky Berka, Research Assistant Cynthia P. King, Analyst Virginia Cwalina, Administrative Assistant Beckie Erickson, P.C. Specialist/Word Processor1 Carol Ann Guntow, Secretary/Word Processor Specialist Diann G. Hohenthaner, P.C. Specialist/Word Processor Contractors Morris L. Barer, Robert G. Evans, and Roberta Labelle, University of British Columbia Alexander M. Capron, University of Southern California Morris F. Collen, Northern California Kaiser-Perrnanente Medical Care Program Louis P. Garrison, The Project HOPE Health Sciences Education Center Glenn T. Hammons, Robert H. Brook, and Joseph P. Newhouse, The Rand Corp. Lisa I. Iezzoni, Oren Grad, and Mark A. Moskowitz, Boston University David A. Juba, The Urban Institute Lois P. Myers, John M. Eisenberg, and Mark V. Pauly, University of Pennsylvania Michael A. Riddiough, Riddiough & Associates Jonathan A. Showstack, Eliseo J. Perez-Stable, Eric Sawitz, University of California, San Francisco James Vertrees, Dennis Tolley, and Kenneth Manton, La Jolla Management Corp. 'Until August 1985. Advisory Panel-Payment for Physician Services: Strategies for Medicare Sidney S. Lee, Chair President, Milbank Memorial Fund, New York, NY John R. Ball Associate Executive Vice President American College of Physicians Washington, DC Thomas L. Beauchamp Professor of Philosophy and Senior Research Scholar Kennedy Institute of Ethics Georgetown University Washington, DC Karen Davis Chair Department of Health Policy and Management School of Hygiene and Public Health Johns Hopkins University Baltimore, MD Richard C. Dever Fellow and Governor at Large for Florida American College of Surgeons Jacksonville, FL Joseph Eichenholz Assistant Vice President Affiliated Businesses Group CIGNA Corp. Hartford, CT Peter D. Fox Vice President Lewin & Associates Washington, DC Jack Hadley Director Center for Health Policy Studies Georgetown University Washington, DC Ronald E. Henderson Physician, private practice Birmingham, AL Jack A. Meyer Director Health Policy Studies American Enterprise Institute Washington, DC Janet B. Mitchell Vice President Health Economics Research Chestnut Hill, MA Vita R. Ostrander President American Association of Retired Persons Washington, DC Thomas 0. Pyle President and Chief Executive Officer Harvard Community Health Plan Boston, MA Uwe E. Reinhardt Professor Department of Economics Princeton University Princeton, NJ C. Burns Roehrig President American Society for Internal Medicine Boston, MA Jerald R. Schenken Vice Chairman Council on Legislation American Medical Association Omaha, NE Steven A. Schroeder Chief of Division of General Internal Medicine Department of Medicine University of California San Francisco, CA Jack K. Shelton Manager Employees' Insurance Department Ford Motor Co. Dearborne, MI Robert H. Taylor Executive Committee, Board of Directors American Academy of Family Physicians Spartanburg, SC B. Elizabeth Tunney Director, Legislation Retail, Wholesale, and Department Store Union, International New York, NY Sankey V. Williams Associate Professor Section of General Medicine Hospital of the University of Pennsylvania Philadelphia, PA Contents Chapter Page 1. Summary and Policy Options 3 2. Physician Payment Under the Medicare Program: Problems and Changing Context .... 0 •••• 0 •• 0 • 0 • 0 0 0 ••• 0 0 • 0 0 •••••• 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 39 3. Overview of Alternative Physician Payment Methods Under Medicare: A Framework for Evaluation 0 • 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 • 0 ••••• ~ • 0 • 0 0 0 0 0 0 0 0 • 0 0 0 0 0 81 4. Modifications to Customary, Prevailing, and Reasonable Charge Payment 0 0 97 5. 0 0 0 •• 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0. 0 0. 0 0 0 0 0 0 0 0 0121 Payment Based on Fee Schedules 6. 0 • 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 • 0 ••• 0 0 0 0 0 0 0 0 0 0 0 0 0 0 155 Payment for Packages of Services 7. 0 ••• 0 •• 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ••• 0 0 ••• 0 0 0 0 0179 Capitation Payment. Appendix Page A. Method of the Study .... 0 0 0 ••• 0 0 0 0 0 0 • 0 0 • 0 0 0 0 0 ••• 0 0 0 0 0 0 0 0 0 0 0 0 0 0 • 0 0 ••• 0 211 B. Acknowledgments and Health Program Advisory Committee 0 0 0 0 0 0 0 • 0 0 0 0 0 0 213 C. Medicare and Medicaid Payment for Physicians' Services .. 0 0 0 0 0 0 • 0 • 0 ••••• 219 D. Private Sector Approaches to Physician Payment ........ 0 •• 0 0 0 0 0 0 0 •• 0 0 0 0 236 References .. 0 251 0 • 0 0 0 0 •• 0 ••••••••••• 0 0 0 0 ••••• 0 0 0 0 ••• 0 • 0 0 0 0 0 0 0 0 ••••••••• 0 • 0 0 Glossary of Acronyms and Terms Glossary of Acronyms IMC -International Medical Centers, Inc. IPA -individual practice association MEl -Medicare Economic Index MRI -magnetic resonance imaging OTA -Office of Technology Assessment (U.S. Congress) PPO -preferred provider organization PRO -(utilization and quality control) peer re view organization ProPAC -·Prospective Payment Assessment Commission PSRO -professional standards review organization RVS -relative value scale TEFRA -Tax Equity .and Fiscal Responsibility Act of 1982 (Public Law 97-248) UCR -usual, customary, and reasonable Glossary of Terms Access: Potential and actual entry into the health care system. Actual charge (Medicare): The charge billed by a physician or other supplier of Medicare Part B medical services. Along with the provider's customary charge and the prevailing charge in the locality, the actual charge is used to determine approved charges. Allowed charge (Medicare): See approved charge. Ambulatory services: Medical services provided to patients who are not hospitalized. Ancillary services or technology: Medical technology or services used directly to support basic clinical services, including diagnostic radiology, radiation therapy, clinical laboratory, and other special services. Approved charge (Medicare): An individual charge determination made by a Medicare carrier on a covered Part B medical service or supply. In the absence of unusual medical circumstances, it is the lowest of: 1) the physician's or supplier's customary charge for that service, 2) the prevailing charge for similar services in the locality, 3) the actual charge made by the physician or supplier, and 4) the carrier's private business charge for a comparable service. Also called allowed charge or reasonable charge. AAFP AAPCC AARP ACIP ACP ACR ACS AHCCCS AMA ASC ASIM BC!BS CBO CFR -American Association of Family Phy sicians -average adjusted per capita cost -American Association of Retired Persons -Immunization Practices Advisory Com mittee (CDC) -American College of Physicians -adjusted community rate -American College of Surgeons -Arizona Health Care Cost Containment System -American Medical Association -ambulatory surgical center -American Society of Internal Medicine -Blue Cross and Blue Shield Association -Congressional Budget Office (U.S. Congress) -Code of Federal Regulations CHAMPUS-Civilian Health and Medical Program of CMP CPI CPR CPT-4 CT DHHS DRG ESRD ESWL FDA FR FTC GAO GMENAC GNP HCFA HCPCS HMO HRSA the Uniformed Services (Department of Defense) -competitive medical plan -Consumer Price Index -customary, prevailing, and reasonable -Current Procedural Terminology, 4th Edition -X-ray computed tomography -U.S. Department of Health and Human Services -diagnosis-related group -end-stage renal disease -extracorporeal shock wave lithotripsy -Food and Drug Administration (DHHS) -Federal Register -U.S. Federal Trade Commission -General Accounting Office (U.S. Congress) -Graduate Medical Education National Advisory Committee -Gross National Product -Health Care Financing Administration (DHHS) -HCFA's Common Procedure Coding System -health maintenance organization -Health Resources and Services Admin stration (Public Health Service, DHHS) ICD-9-CM -International Classification of Diseases, Assignment (Medicare): An agreement by a provider 9th Revision, Clinical Modification (physician or supplier) to accept a Medicare beneICU -intensive care unit ficiary's rights to benefits under Supplementary IOL -intraocular lens Medical Insurance (Part B), to bill the Medicare car rier rather than the patient, and to accept Medicare's approved charge paid by the carrier as payment in full (excluding the beneficiary's 20-percent coinsurance and the deductible). The provider may then bill the beneficiary only for the coinsurance and any applicable deductible. Average adjusted per capita cost (AAPCC): As defined under the Tax Equity and Fiscal Responsibility Act of 1982 (Public Law 97-248), the AAPCC is the estimated average per capita amount that Medicare would pay if covered services for Medicare competitive medical plan (CMP) members were furnished in local fee-for-service practices. The AAPCC formula consists of the product of three major components: 1) the U.S. per capita Medicare cost as projected to the current year; 2) an adjustment based on the historical relationship between national Medicare costs and Medicare per capita reimbursements in the local area that a CMP serves; and 3) an adjustment for the differences between persons who choose to enroll in a CMP and the Medicare population at large from which CMP enrollees are drawn. Billed charge (Medicare): See actual charge. Capitation payment: A method of paying for medical care by a prospective per capita payment that is independent of the number of services received. Carrier (Medicare): Organizations, typically Blue Shield plans or commercial insurance firms, under contract to the Health Care Financing Administration for administering Part B of the Medicare program. Their tasks include computing reasonable charges for physician services, making actual payments, determining whether claims are for covered services, denying claims for noncovered services, and denying claims for unnecessary use of services. Case mix: A measure of the mix of cases being treated by a particular health care provider that is intended to reflect the patients' different needs for resources. Case mix is generally established by estimating the relative frequency of various types of patients seen by the provider in question during a given time period and may be measured by factors such as diagnosis, severity of illness, utilization of services, and provider characteristics. Coinsurance: That percentage of covered hospital and medical expenses, after subtraction of any deductible, for which an insured person is responsible. Under Medicare Part B, after the annual deductible has been met, Medicare will generally pay 80 percent of approved charges for covered services and supplies; the remaining 20 percent is the coinsurance, which the beneficiary pays. Competitive medical plan (CMP): A health plan option available to Medicare beneficiaries that provides physicians' services, laboratory, X-ray, emergency, and preventive services, and inpatient hospital services, assumes risk for the provision of the required services and out of area coverage, and meets certain requirements to assure financial solvency. Such a plan is eligible to enter into a Medicare risk contract in return for a capitation payment under the Tax Equity and Fiscal Responsibility Act of 1982 (Public Law 98-248). Cost-sharing: That portion of the payment to a provider of health care services that is the initial liability of the patient and that may include deductibles, copayments, coinsurance, and under Medicare Part B, unassigned liability. Also, the general set of financial arrangements under which health care insurance is contingent on a purchaser's acceptance of the obligation to pay some portion of the reimbursements for those services. Current Procedural Terminology, 4th Edition (CPT4) Coding: A taxonomy of procedures performed by physicians that is used for recording and billing for services rendered. This taxonomy has been incorporated in the HCFA Common Procedure Coding system, which all Medicare carriers are now required to use. Customary charge (Medicare): In the absence of unusual medical circumstances, the maximum amount the a Medicare carrier will approve for payment for a particular service provided by a particular physician practice. The customary charge is computed by the carrier based on actual charge data for a specific service performed by one physician (practice or supplier) to his or her patients in general. Customary, prevailing, and reasonable (CPR) method (Medicare): The method used by Medicare carriers to determine the approved charge for a particular Part B service from a particular physician or supplier. Under this method, the approved charge is limited to the lowest of the physician's actual charge for the service, the physician's customary charge for the service, and charges by peer physicians or suppliers in the same locality. If necessary, prevailing charges are adjusted by the Medicare Economic Index. Deductible: An initial expense of a specified amount of approved charges for covered services within a given time period (e.g., $75 per year) payable by an insured before the insurer assumes liability for any additional costs of covered services. The Part B deductible is the portion of approved charges (for covered services each calendar year) for which a beneficiary is responsible before Medicare assumes liability. Diagnosis-related groups (DRGs): Entries in a taxonomy of types of hospitalizations based on groupings of diagnostic categories drawn from the International Classification of Diseases and modified by the presence of a surgical procedure, patient age, presence or absence of significant morbidities or complications, and other relevant criteria. DRGs have been mandated for use in establishing payment amounts for individual admissions under Medicare's prospective hospital payment system as required by the Social Security Amendments of 1983 (Public Law 98-21). Extracorporeal shock wave lithotripsy (ESWL): A technique for the disintegrating of upper urinary tract stones that uses shock waves generated outside a patient's body and does not require a surgical incision. Fee-for-service payment: A method of paying for medical care in which each service perforned by an individual provider can bear a related charge. Fee schedule: An exhaustive list of physician services in which each entry is associated with one specific monetary amount representing the approved payment for a given insurance plan. Fee screen: A limit used to determine an insurer's approved charge for a p·articular physician service, such as under Medicare the physician's customary charge or the locality prevailing charge for the service in question. Fee screen year: The calendar period during which a particular year's CPR limits are in effect. As of September 30, 1984, fee screen years run from October 1 through September 30 of the following calendar year, with fee screen year 1985, for example, beginning on October 1, 1984 and ending on September 30, 1985. Prior to the Deficit Reduction Act of 1984 (Public Law 98-369), fee screen years began on July 1 of a calendar year and continued through June' 30 of the next year. Fiscal intermediary: An organization that acts as an agent and purchaser of health care insurance or health care services for insureds. Health maintenance organization (HMO): A health care organization that acts as both insurer and provider of comprehensive but specified medical services. A defined set of physicians provides services to a voluntarily enrolled population for a prospective per capita amount (i.e., by capitation). Prepaid group practices and individual practice associations are types of HMOs. Individual practice association (IPA): A type of HMO whose physicians usually practice in private offices and are paid by the HMO on a fee-for-service basis. Members, however, pay the HMO for coverage through capitation payments. Inpatient services: Services provided to patients who are hospitalized. Intermediaries (Medicare): Organizations, typically Blue Cross plans or commercial insurance firms, under contract to the Health Care Financing Administration for administering Part A of the Medicare program. Their tasks include determining reasonable costs for covered items and services, making payments, and guarding against unnecessary use of covered services for Medicare Part A payments. In terrnediaries also make payments for horne health and outpatient hospital services covered under Part B. Locality (Medicare): For the purpose of making Medicare approved charge determinations, a locality is identified as a geographic area for which a carrier derives the prevailing charges for services. Usually, a locality is a political or economic subdivision of a State include a cross-section of the population with respect to economic and other characteristics. Magnetic reasonance imaging (MRI): An imaging technique based on the physical response of atomic nuclei to imposition of a forceful external magnetic field, and hence not requiring the ionizing radiation associated with X-ray technologies. Managerial technology: Technology used to facilitate and support the provision of health care services but not directly associated with patient care, including administration, transportation, and communication, both within and among health care facilities. Mandatory assignment: An alternative to the present system of Medicare assignment. Under a system of mandatory assignment, only those services for which a physician had agreed to accept the Medicare determination of approved charges as payment in full would be reimbursable. A beneficiary who received a service from a physician who did not agree to accept the Medicare approved charge as payment in full would not be able to be reimbursed for that service under a policy of mandatory assignment. Medical technology: The drugs, devices, and medical and surgical procedures used in medical care, and the organizational and support systems within which such care is provided. Medicare Economic Index (MEl): The index that the Medicare program uses to set limits on physicians' prevailing charges. The MEl is based on estimates of the costs of producing physician office services and a measure of increases in earning levels in the gep.eral economy, as specified by the Social Security Amendments of 1972 (Public Law 92-603). Medigap insurance: Private supplementary medical insurance covering out-of-pocket expenditures of Medicare beneficiaries such as deductibles and coinsurance, but typically not covering unassigned liability for physician services provided under Part B. Nonassigned liability: See unassigned liability. Nonparticipating physician (Medicare): A physician practice that has not elected to become a Medicare participating physician, i.e., one that has retained the right to accept assignment on a case-by-base basis. Compare participating physician. Nonprocedural service: A service, such as an office visit, that may involve but does not depend in a major way on a medical device. Opportunity cost: In economics, defined as the return available from the best alternative use of a particular resource, for example, the value of the other products that might otherwise have been produced by the resources used in the production of a particular good or service. Any single opportunity taken will have a cost in terms of an opportunity forgone. Packages of services: Groups of related physician services or functions that have either uniform content or expected therapeutic effect, or that involve sets of alternative, commonly performed but not required services complementary to a particular major physician service. Part A (Medicare): Medicare's Hospital Insurance program, which provides insurance benefits against the costs of hospital and related posthospital services for elderly and disabled beneficiaries. Part A, which is an entitlement program for those who are eligible, is available without payment of a premium, although the beneficiary is responsible for an initial deductible or copayment for some services. Those not automatically eligible for Part A may enroll in the program by paying a monthly premium. Part B (Medicare): Medicare's Supplementary Medical Insurance program, which provides insurance benefits for medically necessary physician services, hospital outpatient services, ambulatory physical therapy and speech pathology services, comprehensive rehabilitation facility services, and various other limited ambulatory services and supplies, such as prosthetic devices and durable medical equipment. Part B also covers home health services for those Medicare beneficiaries who have Part B coverage only. Enrollment in Part B is optional and requires payment of a monthly premium. The beneficiary is also responsible for a deductible and a coinsurance payment for most covered services. Participating physician (Medicare): A physician practice that has elected to provide all Medicare Part B services on an assigned basis for a year. In return for forgoing the right to bill for Part B services on a unassigned basis, the participating physician is listed in a directory of participating physicians available to beneficiary organizations and may receive greater increases in Medicare approved charges than nonparticipating physicians. Preferred provider organization (PPO): A form of health care delivery system in which an agreement is made between providers and purchasers that patients who seek medical care from the "preferred providers" will obtain benefits such as reduced cost sharing. In return for the potential increase in volume of patients, the preferred providers may agree to discount their charges or to submit to enhanced utilization review. Prepaid group practice: A type of HMO consisting of group practice that provides or arranges comprehensive covered services for enrollees, who pay by capitation. Prevailing charge (Medicare): In the absence of unusual medical circumstances, the maximum amount a Medicare carrier will approve for payment for a particular service provided by any physician practice within a particular peer group and locality. Generally, this amount is equal to the lowest charge in an array of customary charges that is high enough to include 75 percent of all the relevant customary charges. Procedural service: A service, such as endoscopy, that is dependent in a substantial way on the use of a medical device. Prospective payment: Payment for medical care on the basis of rates established in advance of the time period in which they apply. The unit of payment may vary from individual ·medical services to broader categories, such as hospital case, episode of illness, or person (capitation). Prospective Payment Assessment Commission (ProPAC): A commission established by the same law that created the DRG-based prospective payment system for Medicare (Public Law 98-21) to make recommendations to the Secretary of Health and Human Services on the annual update factor and on adjustments of DRG classifications and weights. Quality of care: The degree to which actions taken or not taken maximize the probability of beneficial health outcomes and minimize risk and other untoward outcomes, giveri the existing state of medical science and art. Quality assurance: Integrated programs that attempt to protect or raise quality of care by conducting assessments, taking action to correct problems found, and following up with corrective interventions. Reasonable charge (Medicare): See approved charge. Relative value scale (RVS): A list of all physician services containing a cardinal ranking of those services with respect to some conception of value, such that the difference between the numerical rankings for any two services is a measure of the difference in value between those services. Third-party payment: Payment by a private insurer or government program to a medical provider for care given to a patient. Unassigned liability: The difference, if any, between a physician's actual charge for a service on an unassigned claim and the Medicare approved charge for that service. Usual, customary, and reasonable charges (UCR): In private health insurance, the bases for reasonable charge reimbursement of physicians. This approach was developed in the early 1960s somewhat before the introduction of Medicare, and was adapted by Medicare as the model for CPR. "Usual" refers to the individual physician's fee profile, equivalent to Medicare's "customary" charge screen. "Customary," in this context, refers to a percentile of the pattern of charges made by physicians in a given locality (comparable to Medicare's "prevailing" charges). "Reasonable" is the lesser of the usual or customary screens. h44 I %. g 1 . J.. .. I J . . ; . .: :. t 1 . £& .!I LZ, .i J. AI . .lUll. d& , 2J) .. \1 .:: .,4) Jill L A Cha,pter 1 SQmmciry and PoliCy: OptipOS There's alw~ys ~h easy solution to ever)i problem ...... ne~t, pl~usP ble, and wrong .. \"j~ ' -H.L Mencken ·.· .. , J4. .. I ....•UL .. J.L ··~· I t)Ji .HI. .... . . j z.. .Ji .. .:;; . )II ... J .. .Contents Page Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 .>~Scope of the Study~.>....:.... : .............•.... :; . . . . . . . . . . . . . . . . . . . . s Physician Payment Under Medicare: Problems and Changing Content . . . . . . . 6 PolicyOptions: .. '·····•···· ........... ............:.... ··''···· .. 9 General Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Continuation of P·rese:ht Payment Adimgeri).ents ... : , ... : . . . . . .. . . . 18 Payment Based on Fee Schedules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Payment for Packages of Services ... •, .... '~-... : ........·; ... :.?; • • . • • • . . :::24 Capitation Payment ........... : . . . . ................ , . . . . . . . . . . . 28 ;,Contlusioh' . .. ~ ...' ', ...'.:'<,.-•• ~,~:: • • • • • • • • • • • ••• ,: ••• ,: ~~ • • • • • • • •• ,;::',:33 List of Figures :·f'igur~ No. Page J-1. Congressional Options for Medicare Payment of Physician · ·'and OtherServices .. , ... , ........•........·.... >... .......•... ·. 10 1-2. Alternative Methods of Medicare Payment for Seryices .Provided to a . Hypothetihl Patient~Presenting the Symptom;·of E~trenie Flank Pain ..... '1:.2s Chapter 1 Summary and Policy Options INTRODUCTION In an era of concern about Federal budget deficits, the growth and size of Medicare expenditures on physician services have made this sector an obvious target for constraining outlays. During the 1980s, Medicare expenditures for physician services have risen at an average rate of 16 percent per year and in fiscal year 1985 reached an estimated $19 billion. For 1985, Medicare's Supplementary Medicare Insurance (Part B) program, which includes physician expenditures,1 was estimated to be the fourth largest domestic program in the Federal budget, following Social Security, Medicare's Hospital Insurance (Part A) program, and Medicaid2 (401,553). Important policy concerns apart from rising Medicare expenditures are at issue. There are indications that medical care is not being provided efficiently-that more or fewer resources than appropriate are being used to manage medical conditions and that the benefits gained from additional services may not be worth their costs. Wide variations exist in the use of physician and hospital services that cannot be explained by differences among populations (568). Observers have concluded that some technologies, such as vaccines, have been underused (397). On the other hand, populations with lower use of hospitalization and associated physician services have suffered no apparent ill effects over time (65, 279,285). To some extent, the present situation stems from the fact that medicine is not an exact science. Alternatives exist for the management of many medical conditions, and physicians use their ex 'Medicare's Part B program covers physician, home health, and ambulatory services. Eligibility for Social Security benefits determines entitlement for coverage of Part A, which pays for hospital and related post-hospital services. People eligible for Part A and U.S. residents over age 65 may enroll in Part B. Enrollees pay a monthly premium, a deductible, and 20 percent of the charges allowed by Medicare (493). In fiscal year 1984, physicians' services accounted for 85 percent of Part B expenditures (553). 2Expenditures for national defense and net interest also exceeded those for Part Bin 1985 (470). pertise and judgment to determine the medical care for particular patients. Given the discretion that physicians exercise and the lack of definitive information available to clinicians concerning the efficacy and cost-effectiveness of medical technologies, it is perhaps not suprising that there are substantial variations in practice patterns. But past policies regarding health insurance coverage and payment methods have also played a major role in rising medical expenditures. In general, health insurance dulls the sensitivity of consumers, physicians, and other providers to the financial implications of using medical care (137). With existing levels of cost-sharing, elderly Medicare beneficiaries in general are likely to be sensitive to the cost of using medical care. But coverage and beneficiary cost-sharing vary across settings and technologies. Both elderly Medicare beneficiaries and their physicians are likely to be less sensitive to the cost of technologies that have more nearly complete Medicare coverage. Routine checkups are statutorily excluded from coverage, for example, while physician services for certain surgery performed in designated ambula- Photo credit: Merck, Sharp, & Dohme Pneumococcal vaccination, shown here, is the only preventive service covered by Medicare for all beneficiaries. 4 • Payment for Physician Services: Strategies for Medicare tory surgical centers or hospital outpatient settings is reimbursed at the rate of 100 percent of ap proved charges. Although Medicare's payment policies for ambulatory surgery are intended to encourage physicians and beneficiaries to use less costly settings, in some cases, program expendi tures for ambulatory surgery have exceeded the amount that Medicare would have paid for an in patient case (161).3 Overall, Medicare pays 45 percent of the medical expenses of its elderly beneficiaries; this includes 74 percent of their hospital expenses and 55 percent of their physician expenses, but a much smaller percentage of their other medical expenses (551). About 65 percent of elderly people outside of institutions have private insurance to supplement their Medicare coverage. Despite Medicare and other coverage, elderly people still bear substantial medical expenses; in 1984, elderly people spent an estimated 15 percent of their average income on out-of-pocket costs for health care, the same percentage as in 1966 before Medicare was fully implemented (495). Medicare's traditional payment methods for physician services, like those of other payers, have provided incentives for physicians to provide additional services, regardless of the additional benefit to be gained by beneficiaries. Medicare pays physicians and other Part B providers a fee for each service performed. This fee-for-service payment method places the financial risk for the care provided on the Medicare program and the beneficiary, not on the physician. With fee-for-service payment, physicians have an incentive to perform additional services, provided that the additional revenue they receive exceeds their costs.4 Because much uncertainty exists in medicine and physicians must exercise discretion in their clinical decisions, there is much room for them to recom 3ln October 1, 1985, Medicare began phasing in a prospective payment system for beneficiaries' inpatient care. Under this system of payment according to diagnosis-related groups (DRGs), Medicare pays a fixed amount based on diagnosis for the operating costs of beneficiaries' inpatient admissions. •"Cost" here refers to the physician's expenses of resources used in the course of providing a service. The physician's cost includes his or her "opportunity cost," that is, the payment that the physi cian could obtain for other activities, such as performing another service for another patient. In contrast to cost, price or approved charge refers to the revenue that a physician receives for a service. mend additional followup visits or procedures within the bounds of accepted medical practice. Medicare's payment rates are based on what physicians have charged in the past, a system that is inherently inflationary (262). Under Medicare's customary, prevailing, and reasonable (CPR) payment method, the Medicare approved charges is limited to the lowest of a physician's billed charge, the customary charge for that service based on that physician's prior billings, and the prevailing charge for that service based on comparable physicians' prior billings for the same service. An additional limit on the prevailing charge for a service is set by the Medicare Economic Index, which measures changes in practice expenses and general earnings. Depending on whether the physician "accepts assignment," either the Medicare administrative carrier"' or the beneficiary pays the physician. The beneficiary, after having met the annual deductible, is entitled to have Medicare pay 80 percent of the approved charge for a Part B service. If the physician accepts assignment, she or he accepts Medicare's approved charge as payment in full and may collect only the beneficiary's 20-percent coinsurance and any remaining deductible from the beneficiary. If the physician does not accept assignment, the beneficiary is liable for any difference between the physician's actual charge and Medicare's approved charge (the beneficiary's unassigned liability), plus the coinsurance and any deductible. Since October 1, 1984, physicians have been able to become Medicare "participating physicians" by agreeing to accept assignment for all Medicare claims for the next 12 months. From July 1, 1984 to October 1, 1985, the customary and prevailing charges of all physicians and the billed charges of "nonparticipating" physicians were frozen. In the absence of passage of Medicare's authorization for fiscal year 1986, Congress in De 'The terms approved charge, reasonable charge, and allowed charge are used interchangeably to connote the amount that Medi . care pays for a specific physician service. After the beneficiary has paid an annual deductible, Medicare pays 80 percent of the approved charge, and the beneficiary is responsible for 20 percent. 'To administer Part B, Medicare contracts with private organizations termed carriers, which are primarily insurance companies that also have private lines of business. Ch. 1-Summary and Policy Options • 5 cember 1985 temporarily extended the freeze to amine different ways of paying for physician servMarch 15, 1986. ices under Medicare. The act specified that OTA was to pay particular attention to the following Like the use of the service as the unit of paytopics: any inequities in relative payments by typement, the structure of relative fees under the CPR of service, locality, and specialty or in relativepayment method has not contained incentives for payments between procedural and nonproceduralthe efficient provision of medical care. Medicare, services; incentives for providers to accept assign as well as most private third-party payers, has hisment; the effects of alternative payment methodstorically paid higher rates for urban, specialist, and levels on the use of services; and possibleand inpatient services than for rural, generalist, methods to develop fee schedules. The House Enand ambulatory services. Medicare has also reergy and Commerce Committee, the House Wayswarded services dependent on equipment more and Means Committee, and the Senate Finance highly than historytaking and counseling (227,424). Committee have jurisdiction over Medicare Part The Deficit Reduction Act of 1984 (Public Law B and this section of the law. In addition, the Sen98-369), which established the participating phyate Special Committee on Aging requested OTA sician program and froze physician charges to to study the effect of physician payment methMedicare beneficiaries, also mandated OTA to ex-ods on the use of medical technology. SCOPE OF THE STUDY This report uses the term "physician services" The problems with Medicare's current system for services that are commonly provided by phyof paying for physician services are examined in sicians but are sometimes provided by other chapter 2, and chapter 3 presents a framework professionals or organizations.7 Clinical laborafor evaluating alternative methods of payment to tory tests, for example, may be performed in a deal with these problems. Subsequent chapters fophysician's office, an independent clinical laboracus on the analysis of specific payment alternatory, or a hospital laboratory. Similarly, refractives: modifications to Medicare's customary, pretion and fitting of corrective lenses may be provailing, and reasonable (CPR) charge payment vided by physicians, such as ophthalmologists, (ch. 4); payment based on fee schedules (ch. 5); or by other professionals, such as optometrists. payment for packages of services (ch. 6); and capiFurthermore, some of the alternative payment tation payment (ch. 7). methods that are considered in this report entail The remainder of this chapter summarizes thesepayments for the institutional as well as the phytopics and presents policy options for Congresssician portion of inpatient care. to address the problems identified and to pursue This report uses OTA's definition of medical strategies culminating in different payment retechnology: the drugs, devices, medical, and surforms. Appendix A describes the method of con gical procedures used in medical care, and the orducting the study; appendix B acknowledges the ganizational and supportive systems within which valuable assistance of several individuals; and apsuch care is provided. pendixes C and D present background information on topics related to but broader than Medi 7The Social Security Act defines "physicians' services" as profescare, Medicaid, and private sector approaches to sional services performed by doctors of medicine, osteopathy, denpaying for physician services. In addition to thetistry, podiatry, and optometry; and chiropractors (Section 1861 main report, a case study on extracorporeal shock (r)). In addition, health maintenance organizations and competitive medical plans may furnish the services of certain other health profeswave lithotripsy (ESWL) is being published in con sionals without the direct supervision of a physician: physician asnection with this assessment. sistants, nurse practitioners, and clinical psychologists (SO FR 1351, 42 CFR 417.416). 6 • Payment for Physician Services: Strategies for Medicare PHYSICIAN PAYMENT UNDER MEDICARE: PROBLEMS AND CHANGING CONTEXT The Medicare program is intended to help Differences in assignment rates affect beneficiaries'elderly and disabled people meet their medical exout-of-pocket costs. For the claims of elderly benpenses. Expressed as the missions of the program, eficiaries, assignment rates increase with the agethis goal entails promoting the delivery of qualof the beneficiary, but for disabled people, assignity health care services to beneficiaries, making ment rates have been highest for the youngest age those services accessible to them, and doing so group. Assignment rates also rise substantially forin a manner that is consistent with the costhigher bills. In general, the data are consistenteffective delivery of services within both Mediwith the hypothesis that physicians accept assign care and the general U.S. health care system ment more readily when there is a greater risk of (491,508). incurring a bad debt. Over the life of the program, per capita MediAlthough Medicare's Part B program is ana care expenditures for physician services have risen tional program funded through general revenues at roughly the same rate as increases for the and beneficiary premiums and deductibles are uniUnited States; however, since 1978 and especially form across the country, Medicare's payments onsince 1982, per capita Medicare expenditures for behalf of beneficiaries vary considerably. Acrossphysician services have risen more rapidly than the United States, there is more than a twofoldexpenditures for the Nation as a whole. Although variation by carrier jurisdiction8 in Medicare exgrowth slackened in 1984, total expenditures for penditures per beneficiary for physician and otherMedicare's Part B program are expected to con medical services (525). This variation depends ontinue to rise by almost 14 percent per year through the proportion of beneficiaries who exceed thefiscal year 1990 (401). Increases in the beneficiMedicare deductible and are thus eligible for reim ary population have accounted for only a minor bursement; that proportion, in turn, depends onpart of this growth. From 1976 to 1982, Medicare variations in health, service volume, physicians' expenditures for physician services for elderly peocharges, and the stringency with which Medicareple increased 18 percent per year-2 percent from carriers determine approved charges. enrollment increases, 10 percent from price increases, and 6 percent from increases in the numCompared to beneficiaries in States with high ber of services per enrollee (133). Claims per benapproved charges, beneficiaries in States with low eficiary have risen continuously throughout the approved charges have to receive more services history of Medicare, from 1.1 in 1967 to 7.9 in to meet the deductible and qualify for program 1984 (527). payments. On the other hand, for a given volume of services, beneficiaries in States with lowerThere is substantial variation in aspects of approved charges may have lower out-of-pocketMedicare payment, including assignment rates, expenses. Even within a national program, proannual expenditures per beneficiary, and relative vision of a uniform real level of benefits requiresrates paid for certain services. This variation, disthat Medicare pay different prices across jurisdiccussed further below, may be indicative of probtions to reflect different practice costs.lems regarding quality, access, and efficiency. But .substantial variation is to be expected within a Within States, variations generally reflect dis parities in payment levels between urban and ru national program serving over 30 million beneficiaries in thousands of local markets, and little ral areas. Under the Social Security Act, Medior no consensus exists regarding whether specific care carriers are given discretion in identifyinglocalities for payment purposes. Because of this variations signify problems. carrier discretion, the entire State is the localityAcross the United States, assignment rates varyfrom 17 percent for elderly people in South 'In fiscal year 1984, the 58 jurisdictions across the United StatesDakota to 87 percent in Rhode Island (296). were administered by 40 carriers (535). for 18 States, while Michigan has 2, Pennsylvania has 4, and Texas has 32localities. Across localities, the range in charges for specific services is often great. In 1980, the highest prevailing charge exceeded the lowest by 159 percent for cataract removal and by 536 percent for chest X-rays (494). Even after adjustment for cost-of-living differences, great variations continue to exist (SO), but the costs of operating practices of equivalent size and style are not available. To the extent that differences in approved charges exceed differences across local market areas, reducing the number of localities for charge determination is a reasonable goal. Carriers also differ across the United States in their use of physician specialty to determine approved charges for services. Four carriers make no distinction among physician specialties (473), while the carrier for Pennsylvania has had 58 different groups. For many services, prevailing charges are specialty specific regardless of carrier policy, because one specialty typically performs a certain procedure. Few cataract removals, for example, are performed by physicians who are not ophthalmologists. Specialty-specific determinations may have the most effect on approved charges for physician visits, which are performed by many different specialties and account for about half of all physician services provided to beneficiaries. The justification for recognizing higher approved charges for specialists compared to generalists is that specialists provide either higher quality or different services. Some evidence suggests that higher quality care is provided by physicians practicing in the area of medical care for which they were trained (so called "modal specialists") (370). The difficulty arises in determining in specific cases when services, mainly visits, performed by specialists and generalists constitute similar services and when a specialist or generalist is the modal specialist and deserves higher payment. There is no empirical literature to guide determinations for specific cases. Medicare payment for physician visits also varies by the site of service, with a hospital visit paid more than a nursing home visit, and a nursing home visit paid more than an office visit (494). In 1982, average prevailing charges were 11 to 32 percent higher for inpatient visits than for office Ch. 1-Summary and Policy Options • 7 visits. Since physicians do not pay hospitals for the use of their facilities, these differences suggest an incentive for physicians to favor the hospital as the site of care. However, the nomenclature of physician services may be misleading in this instance. A limited inpatient visit may differ from a limited office visit because inpatients are usually sicker and may require more physician time and skill. Large differences appear to exist in relative approved charges for procedural services such as endoscopy, which depend in a major way on the use of medical devices, and nonprocedural services such as office visits, which use medical devices only incidentally.9 One study reported that after adjustment for such factors as training, resource cost, and service complexity, physicians were paid as much as four to five times more per hour for inpatient surgery than for office visits (227). Within the office, the lack of additional payment for such primary-care services as historytaking or nutritional counseling contrasts sharply with the additional income that can be generated from, for example, providing laboratory tests, interpreting an electrocardiogram, or performing an endoscopy. In office practice, payment 'Although procedural services are often referred to as cognitive services, both procedural and nonprocedural services use cognitive skills. Photo credit: American College of Physicians, HEALTHSCOPE film series The lack of additional payment for primary-care services such as counseling and historytaking contrasts sharply with the additional payment for services, such as interpreting electrocardiograms, that depend greatly on medical devices. 8 • Payment for Physician Services: Strategies for Medicare rates are such that physicians might realize greater net incomes from performing an additional diagnostic test than from seeing an additional patient (424). The establishment and maintenance of high payment rates for equipment-embodied and surgical technologies may have contributed to payment differentials between procedural and nonprocedural services. Many technologies are priced high when new because they are complex and requre special skills to perform. Even if over the years the required physician time and other resources decline and the necessary skills become more commonplace, the initial price is maintained. Differentials in Medicare payment rates for certain services raise the concern that they may be affecting the quality of care received by beneficiaries and the cost of care paid by Medicare and beneficiaries. Differences in net revenue to providers would be most likely to influence medical decisions for which the medically and ethically correct choice is unclear (194). The comparison involves both the net revenues from services that are substitutes for a particular patient and the physician's opportunity costs of providing services to another patient. In the case of a beneficiary who has private supplementary insurance to cover cost-sharing, the additional cost to the patient of a diagnostic procedure, such as endoscopy, may be negligible. Since the test may provide useful information and requires little time, the increase in revenue to the physician of several hundred dollars may be a strong incentive to perform the test. Currently, beneficiaries may find it harder to obtain nonprocedural than procedural services. There is evidence that carriers have paid a lower percentage of billed charges for visits than for surgeries (247,294), that assignment rates have been lower for primary-care specialties than for surgical ones, and that beneficiary out-of-pocket payments have been a larger part of revenue associated with the Medicare program for primary-care physicians than for surgeons and radiologists (247). There is no indication, however, that beneficiaries' health has suffered from lack of access to primary-care services. Variations in payment rates also result from the application of the Medicare Economic Index. The effect of the index varies greatly, depending on the services and specialty. In 1980 in California, the index affected almost no payments for eye exams from ophthalmologists but affected almost all payments for basic anesthesiology services from anesthesiologists (187). On the other hand, by capping prevailing charges in urban areas, the index in effect prevented urban-rural differentials from increasing (359). The changing context of medical practice adds other considerations to an analysis of Medicare's payment policies. In recent years, physicians have felt under greater competitive pressure. In part, this sense may have come from the increases in physician supply, which has grown rapidly over the past decade and is expected to outstrip requirements for additional physician services for the rest of the century (544). The sense of greater competition may also have come from activities of employers to contain increases in health insurance premiums and of Federal and State governments to moderate increases in their health care expenditures. Perhaps in response to these changing circumstances, innovative practice arrangements are burgeoning, and physicians are increasingly entering organizational and payment systems, such as prepaid group practices, individual practice associations, and preferred provider organizations (PPOs), that differ from traditional fee-for-service solo practice in utilization controls, payment methods, and benefit design. Although these organizations usually exert more control over the availability and use of resources than physicians would experience in solo practices, physicians in these organizations gain greater predictability in patient load, income, and practice hours. As a result of prospective hospital payment systems that Medicare and several States have adopted in recent years, hospitals have new incentives to reduce inpatient operating costs. Cutbacks in lengths of stay appear to be affecting the inpatient services that physicians perform, but payment for services in the ambulatory settings including physi Ch. 1-Summary and Policy Options • 9 dans' offices has remained relatively unconate discounts. Although the determination of strained. approved charges might be considered a form of quantity discounting, one might expect greater re Greater Medicare expenditures can be expected ductions for services provided primarily to benas the increasing supply of physicians enables the eficiaries, such as cataract surgery. Medicare alsogrowing demand from more numerous and more lacks arrangements with PPOs, organizationselderly beneficiaries to be realized. Because of the which contract with physicians and sometimesincreasing supply of physicians, however, these hospitals to provide services at lower than usualproviders may be more willing to accept lower rates on the expectation that patient load will be prices for their services and lower increases in their greater. incomes. To the extent that competition would lead physicians to moderate their billed charges, Review of Medicare's payment of physician Medicare's present CPR system would permit the services raises questions regarding the quality, ac program and its beneficiaries to benefit from lower cessibility, and efficiency of beneficiaries' medicosts. But under CPR, Medicare could also excal care. It is clear that Medicare expenditures for perience increases in use and expenditures if phyphysician services are currently unpredictable and sicians chose to maintain their incomes in the face lie largely outside the control of the program and of greater competition by increasing the discreits beneficiaries. Using fee-for-service as the tionary use of services or if beneficiaries method of payment and CPR as the basis for determining approved charges has been associated demanded more services in response to lower with continual increases in claims per beneficiarycharges. and in recent years with more rapid expenditureRecent changes in legislation and regulation increases for Medicare than for the Nation as a have made participation in Medicare more attracwhole. Nor does the pattern of variations in aptive to risk-sharing health maintenance organizaproved charges among services appear consistenttions (HMOs) and other competitive medical with incentives for providers to deliver good qualplans (CMPs), and beneficiary enrollment in ity care in an efficient manner. There is also noHMOs mushroomed during 1985. Nevertheless, question that variations in payment levels have it appears that Medicare has not fully taken led to confusion among providers about the apadvantage of opportunities in the marketplace. proved charges that they may expect for a serv Despite the fact that beneficiaries account for a ice and among beneficiaries about their out-oflarge share of certain physician services, Medi pocket expenses. care uses a standard formula to determine ap proved charges and has not attempted to negoti- POLICY OPTIONS To address the problems identified with Mediment for packages of services on a wide scale. Uncare's current system of paying for physician servder the fourth strategy, Medicare would pay for ices, Congress could undertake four different all beneficiaries' medical care by capitation paystrategies, depending on the payment method that ment. In addition to the four strategies, a set of Congress ultimately wished to adopt for Medigeneral options addresses problems that are likely care (see figure 1-1). The first strategy would reto continue under all of the payment alternatives tain CPR as the mainstream payment method, but that continue payment for individual or packages continue other payment methods in specific cirof services, that is, for all of the alternatives excumstances, such as capitation payment for bencept capitation payment (see figure 1-1). eficiaries who elected to enroll in HMOs. A second strategy would replace the CPR payment The four sets of payment alternatives vary with approach with payment based on fee schedules. respect to the unit on which medical care is based. The third strategy could be adopted if Congress Two of the alternatives, modifications to CPR wished to explore the strategy of moving to pay-payment and payment based on fee schedules, - <::) • (!) ~ 0' ..., ~ c:;· ~- Figure 1·1.-Congressional Options for Medicare Payment of Physician and Other Services --:-· .\~-:-~;;J t;;:::-:-:-:--:-:~ f7;;:::;--:-;;;;;:::;:-:-:::~ r·-:-:-:-.:-:-;-::!lv,Op~j.liM:• •···l t;•~:crc.opljlflf,6!ltlop:SJJ;ccc 'I 1• ••lPdJi}bn'+1;~~'1; I Reduce fee 1 •·• . RedlJce .. · .: •·• Adopt; 1 I Mandate I .1 I. dlffe;~nf~als ;•.I tl~mbera:;~bd~s .l I voiu~e c.p~~~':j 1 ~:'!~~~~ment L-·L:i:j~~-"-r ~.,...L·· c~··1---_J I I I ij~ji~j"---------------;;:~::j--r:~--:::--::--::~--:::--::--::--:::--:::--:---- a Dashed lines indicate that the strategies below are possible with or without the adoption of these general options. SOURCE: Office of Technology Assessment, 1985. (/) (!) Ri·····. ~<1~~~1§~~;:[1 ~- I Es11i6Hsh• . . . I (!)1physictait aa 1 L' qo[!fwtG,P""'~1 I !"! (/) iii ---------.;;.;'~''" () ib CQ .ayment System . . . . . . . . . . . 54 Beneficiary and Provider Confusion . . . 55 . ,inefficiencies in\the Ii)elivery of . Medical Care . . . . . . . . . . . . . . . . . . . . . 56 ,.,yariations. in AnnuahExpenditures ·· per Beneficiary ..·: . . . . . . . . . . . . . . . . 56 Page Variations Related to Assignment 57 Perceived Payment Imbalances ....... . 59. Summary of Variations ............. . 71 The Changing Context of Physician Payment ................. . 72 Changes in Policiesof Hospital · P~yment ..·............. .' . . . . . . . . . 73 Changes in the Elderly Population,. . . . 73 Changes in Medical Providers .........> 7 4 Implications for Medicare Expend!'tures ................. : ... . 76. Conclusion .......................... . 77 list of'Tables Table No. .· J;age. 2~i. Medicare Part B Enro~llmerit, .Reimbursement Amounts, and Claim~·Volilme, Fisc~} Years 2.:.2. 1?.67-84 .. ~;· .. ·.:: ....:. : .. '.· ... ·, .... Percent Distribution of Medicare 41 Approyed Charges for Phy:sicians' Services, by Colli.bin~tion~ of Piace ~l}d Type of Ser:\rice, J983......... . 43 2-3. Components of Increases in Total Medicare Approyed C:harg(;!s for PhysiCian Servi&s per Aged Enrollee, 1967-83 ................ . 44 2-'l. Annuar Percenta~e lntreases in Medicare Pay:ments f()r Physicians' Services for 'Aged Beneficiaries, 1975-82 ........................ . 45 lJ.S. Physician Expenditures and Factors Accounting for Growth, 1965-83 . . . . . ; ; ............(.... 46 2-6. Trends in the Gross Income, E:Xpenses, Net Incom~;· and'·Real Net Income of Phy:sicians, 1970-84 . . . . . 47 Mean Physician Net Ihcome After Expenses Before Taxes, Selected Years, '1973,83 ...•... ;:L.. ·47 2-8. Gross Physician £arnings From Medieare, 1:981 ..'; •. . . . . :~j:. . . . . . . 48 :0·· Table No. Page 2-9. Percent of Self-Employed Physicians Reporting Specjfic Percentile Ranges of Patients With Medicare Coverage, 1984 ................. . 49 2-10. Medicare Reasonable Charge . Redu<;tions per Claim, January-March, 1985 .................... . 58 2-11. High and .~ow Prevailing Charges in Localities for Five Selected · . Procedures, Fe.e Screen Year 19~0 60 z'-12. Eldetly Population's Share of · Market for Selectedinpatient Surgical and Dragnostic and . . Therapeutic Procedures, 1983 .....<. 70 2-13. Distribution ofMed!care Approved Charges Acros~ CPR Limits, and >Specialty~md Type 6f Service ..... 2-14. Elderly PopulaJion in the United . States, Actual and ProjeCted, by Age Cohort, 19.70-2050 .......... . 74 2-15. Life Expectancy!at Birth and Age . . . 65, by Sex and Calendar Year ... 74 2:16. Medicare Enrollees Served and Their Reimbursement, by Age, 1982 . : .............'. ...·': 2-17. Active Physicians in the United . States and' Estilli.ated.Reql1irements};:· 1970-2000 ..................... . 75 list of Figures Figure No. . . .· > . . .... Pagf!. ·z-1. Percerif Distribution 6f Medicate Approved Charges for Physician . Services by~Ty:pe of Servie~, 1:983 .. •. 2-2. Percent Distribution of Medicar.e Approved Gharges fo? Physiciai\ · . Services by: Place of Service, 1983 .. , 2::3. Pe'rceni Gr6wtW1n U.'S. aiid ·:-· Medicare per Capita f'hysi<;ian Expenditure~~ 1968-83:~ .. :·:;:: . ...... . 1'1 Chapter 2 Physician Payment Under the Medicare Program: Problems and Changing Context INTRODUCTION The law establishing the Medicare program was enacted in 1965 as a means to enhance access of elderly people to hospital and physician services by providing insurance that would reduce the outof-pocket costs of such care. In this regard, the program has largely succeeded. This success, however, has come at an increasing cost to the Medicare program. Furthermore, elderly people have not been immune to increases over time in outof-pocket costs for Medicare premiums, deductibles,1 coinsurance, and "nonassigned" liability for covered services-not to mention the total costs for those health care services that are not covered by Medicare. Finally, there is some concern that the program does not provide equal financial protection to all beneficiaries. In particular, there are perceived imbalances by region, location within region, type of service, and other factors not related to eligibility. The Medicare program represents a major part of U.S. health insurance coverage, which has increased greatly over the past generation. Although health insurance has improved people's access to medical care, it has also fueled the use and cost of medical technology (129, 137). The nature of insurance coverage and the specific payment methods that have been used by Medicare and other third-party payers have dulled the sensitivity of consumers, physicians, and other providers to cost considerations. The result has often been inappropriate technology use and higher expenditures than warranted for the health benefits received (483). Until recently, increases in hospital expenditures under Part A of Medicare have attracted the most attention and concern because hospital expenditures have accounted for the largest share of total 'in constant dollar terms, there has been a decline in premiums and the deductible over time, but total real out-of-pocket costs for beneficiaries have increased. Medicare expenditures and have been growing at a high rate. However, the increase in Medicare hospital expenditures has slowed since fiscal year 1983; and in October 1983, Medicare began paying for inpatient operating costs by diagnosisrelated groups (DRGs). For fiscal year 1984, expenditures grew faster for physician and other services under Part B than under Part A or indeed for any other component of the Federal budget (401). As attention has turned to expenditures for Part B services, Medicare's method of paying for physician services according to customary, prevailing, and reasonable (CPR) charges has come under particular criticism. In fact, the inherent inflationary bias in the CPR approach has been demonstrated both theoretically (151) and empirically (189). This situation contrasts with the "financial" goals posed for the Medicare payment system of achieving at least predictable and preferably contained levels of beneficiary and program expenditures. Other developments in the medical care sector also affect Medicare's payment of physician services. Changes taking place in the supply of physicians and the organization of their practices may result in a more competitive market for physician services and a new environment for Medicare program payment policies. This chapter reviews the increases in Medicare expenditures for physician services along with other current issues in physician reimbursement in the Medicare program. It also identifies current developments outside of Medicare that may affect physician payment. The discussion in this chapter reviews the context for addressing both Medicare's physician reimbursement issues and the other general objectives of the Medicare program: promoting access of Medicare beneficiaries to health care services of an acceptably high quality delivered in a cost-effective manner. 40 • Payment for Physician Services: Strategies for Medicare EXPENDITURES FOR PHYSICIAN SERVICES In 1984, the Nation spent $75.4 billion on physician services ( 507). This was an increase of 9. 3 percent over the previous year, exceeding the rate of growth in expenditures on all health care services and supplies in general and the growth in hospital expenditures in particular. Medicare expenditures on physician services in 1984 were $14.6 billion, or 19.3 percent of the total. All Federal expenditures for physicians services in 1984 were $16.9 billion, compared to an estimated $200 million in 1965. As a proportion of all expenditures for physician services since that time, Federal expenditures have increased from 1.8 to 22.4 percent (165,507). Medicare Expenditures for Physician Services Method of Physician Payment Under Medicare The predominant method of physician payment under the Medicare program is fee for service. Although some Medicare funds for physician services are paid to hospitals and other institutions (e.g., health maintenance organizations (HMOs)) that may employ salaried physicians or retain physicians on other than a fee-for-service basis, such arrangements represent a very small fraction of the Medicare business. Of Part B incurred allowed charges for physician services in the year ending June 30, 1983, 96 percent originated with individual patient bills submitted on the sb:mdard physician claims forms for fee-for-service practice ( 553). 2 Reasonable or approved charges for those claims are determined through the CPR charge determination process, which is described in appendix C. Medicare's "approved charges" for any service are limited to the lowest of the physician's billed charge, the customary charge for tne service b·ased on that physician's prior billing~ to the Medicare carrier, or the prevailing charge for that service based on comparable physicians' prior bill 2 Comparable statistics are not available with respect t'o the volume of Part A funds used for physician reimbursement. Although much of this Part A funding will be used to pay salaried physicians, hospitals may bill carriers for services performed by sal~ried physicians. ings to the carrier for the same service as adjusted, if necessary, by the Medicare Economic Index (MEl). As a result, Medicare carriers (as most large private physician insurance programs) typically do not approve the full amount of a physician's charges for a service provided to a Medicare patient. In the first quarter of 1985, the average reduction due to the CPR process was 26.2 percent (535). For a bill with submitted charges of $100, therefore, approved charges would average $73.80. (The carrier would pay the physician 80 percent of the approved charges, or $59.04, less any unpaid patient deductibles.) Contrary to the conventional wisdom, not all physician claims are submitted at amounts that exceed the CPR limits. Through the end of calendar year 1984, 18.3 percent of all claims were submitted at or below the CPR limits (535). Physicians are paid for their services to Medicare beneficiaries either directly by the beneficiary or by a Medicare carrier, depending on whether the physician "accepts assignment." By statute, it is only the Medicare beneficiary who is entitled to be paid a reimbursement benefit. That benefit is equal to 80 percent of the approved charge for the service once the beneficiary has approved bills that exceed the annual deductible. Instead of being reimbursed directly, the beneficiary may elect to assign the benefit to the physician who provided the service. If the physician accepts assignment, he or she must accept the approved chargeas payment in full (and may bill the beneficiary for the 20-percent coinsurance and any remaining deductible). If assignment is not accepted, the physician's expected full payment is not bound by the approved charge, and the beneficiary is liable for any difference between the physician's actual charge and~the allowed charge, in addition to the coinsurance and deductible. Medicare's approved charge, however, is determined without regard to assignment. Prior to October 1, 1984, each physician was free to make assignment decisions on a case-bycase basis. 3 Passage of the Deficit Reduction Act 3 ln cases where a physician treated a patient who was eligible for both Medicare and Medicaid, accepting assignment was mandatory. And, in the case where a physician provided more than one service to a beneficiary on the same day, assignment would have to be ac Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 41 of 1984 (Public Law 98-369), however, introduced the concept of Medicare "participating physicians" along with a 15-month freeze on customary and prevailing charges for all physicians and a freeze on submitted charges by "non-par" physicians (i.e., physicians who did not elect to become participating physicians). A physician who elected to become a participating physician agreed to accept assignment for all Medicare claims for the next 12 months. In return, that physician would be listed in a directory of participating physicians available to beneficiaries, and would be allowed to increase billed charges. According to the provisions of the Deficit Reduction Act, participating physicians would receive higher approved charges in fee screen year4 1986, while the approved charges of the non-pars in fee screen year 1986 would not increase appreciably beyond the fee screen year 1984 levels.5 Although non-par physicians are not required to accept assignment on 100 percent of their claims, they may continue to accept assignment on a case-by-case basis. Participating physicians represent 29.8 percent of all physicians who receive payment under the Medicare program (518). In the first quarter of 1985, participating physicians submitted 36.1 percent of all physician claims to Medicare and 56.5 percent of all assigned claims. Participating physicians accounted for 34.9 percent of covered charges for physician services (537). Composition of and Growth in Medicare Expenditures for Physician Services In fiscal year 1984, Medicare carriers processed 229 million Part B claims (527), approximately 7 cepted on all of those services or none of the services. The physician in that case could not accept assignment for only some of the services. Beginning in fiscal year 1985, however, assignment could be accepted for laboratory services only without the requirement that assignment be accepted on all services if it was accepted on any service. 'A fee screen year is the calendar period during which a particular year's CPR limits are in effect. Prior to the Deficit Reduction' Act of 1984 (Public Law 98-369), fee screen years began on July 1 of a calendar year and continued through June 30 of the next year. As of Sept. 30, 1984, fee screen years run from Oct. 1 through Sept. 30 of the following year, with fee screen year 1985, for example, beginning on Oct. 1, 1984. •Because the freeze limits were based on the charges from the last 3 months of fee screen year 1984, it is conceivable that non-pars who had increased their fees between July 1, 1983, and Mar. 30, 1984, would receive increases in their customary charges in spite of the freeze. claims per enrollee. The average claim included charges for covered services of $128.74; average approved charges per claim were $97.61. Total claims volume has grown at an average annual rate of 12.6 percent since 1968, while annual growth in claims per enrollee has averaged 9.4 percent (see table 2-1). Eighty-five percent of Part B expenditures are for physician services, with the bulk of the remainder going to outpatient departments (553). As shown in figure 2-1, the expenditures are concentrated in the areas of medical care and surgery, at 37.3 and 33.7 percent, respectively, of total approved charges in 1983 (69). Diagnostic radiology and diagnostic laboratory services represented 8.4 and 8.0 percent, respectively, of total approved charges, with all other physician services combining to total 12.6 percent. Most of the expenditures for physician services are for services provided in the hospital. In 1983, the most recent year for which estimates are available, 61.9 percent of all approved charges were Table 2·1.-Medicare Part 8 Enrollment, Reimbursement Amounts, and Claims Volume, Fiscal Years 1967·84 (in millions) Number of Total Number Claims per Fiscal year beneficiaries dollars of claims beneficiary 1967 17.8a $ 664 19.7 1.1 1968 18.0 1,390 34.2 1.9 1969 18.8 1,645 39.3 2.1 1970 19.3 1,979 43.8 2.3 1971 19.7 2,035 49.1 2.5 1972 20.0 2,255 54.5 2.7 1973 20.4 2,391 58.5 2.9 1974 22.6 2,874 68.0 3.0 1975 23.3 3,765 81.4 3.5 1976 ...... 24.1 4,672 93.5 3.9 TQb ....... 1,269 1977 ...... 24.8 5,867 110.0 4.4 1978 ...... 25.6 6,852 122.1 4.8 1979 ...... 26.3 8,259 136.2 5.2 1980 ...... 26.9 10,144 154.5 5.7 1981 27.5 12,345 171.7 6.2 1982 28.0 14,806 188.3 6.7 1983 28.5 17,487 208.4 7.3 1984 29.0 19,473 229.0 7.9 aAfter 1977 enrollment is as of June 30, not the end of the fiscal year, Sept. 30. bTransition quarter. SOURCES: Enrollment, years ending June 30 and incurred reimbursement amounts: U.S. Federal Supplementary Medical Insurance Fund, Board of Trustees, "1985 Annual Report of the Board of Trustees of the Fed· eral Supplementary Medical Insurance Trust Fund," Washington, DC, Mar. 28, 1985. Claims volume: U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Data Management and Strategy, Division of Reports and Analysis, Com· piled Carrier Workload Reports, 1985. 42 • Payment for Physician Services: Strategies for Medicare Figure 2·1.-Percent Distribution of Medicare Figure 2·2.-Percent Distribution of Medicare Approved Charges for Physician Services by Type Approved Charges for Physician Services by Place of Service, 1983 of Service, 1983 Anesthesia Other Hospital (4.8%) outpatient (2.8%) (5.9%) SOURCE: I. Burney and G. Schieber, "Medicare Physicians· Services: The Composition of Spending and Assignment Rates," Health Care Financing Rev1ew, forthcommg. provided in an inpatient setting. Physicians' offices and outpatient departments were the second and third ranked sites, with 29.2 and 5.9 percent, respectively, of approved charges (see figure 2-2). In terms of the most significant place of service/type of service combinations, 27.5 percent of total approved charges were for surgical services in a hospital, 18.8 percent were for medical services in a hospital, and 15.5 percent were f~r medical services in a physician's office (see table 2-2). Internal medicine was the specialty that received the highest proportion of Medicare physician expenditures, accounting for 20.4 percent of total approved charges in 1981 (69). The medical specialties as a whole accounted for 28.5 percent of 1981 approved charges, and general and family practice combined accounted for an additional 11.5 percent. Surgical specialties accounted for 34.8 percent of total approved charges, with the services of general surgeons representing 9.6 percent of the total and those of ophthalmologists representing 8.2 percent. The distribution of specialists' charges by type of service is unr~markable, with general and family practice and' most SOURCE: I. Burney and G. Schieber, "Medicare Physicians' Services: The Com position of Spending and Assignment Rates," Health Care Financing ReVIew, forthcoming. medical_ specialties billing most of their charges for med1cal care and most surgical specialties billing most for surgery. With few exceptions, most specialties have higher total billings for services provided in the hos~ital than in an office (69). Two specialties, however, received more than SO percent of 1981 approved charges for services provided in their offices: otolaryngology (50.3 percent) and dermatology (91.1 percent). In spite of the CPR limits or, as some would have it, because of them, approved charges for physician services per aged Medicare enrollee increased by 591 percent between fiscal year 1968 and fiscal year 1983. Medicare Part B benefit payments totaled $1.4 billion during fiscal year 1968; 16 years later, benefit payments had increased to more th~n $19.5 billion, an increase of 1,400 percent (553). These increases were due to a variety of factors in addition to the changes in approved charges, including changes in enrollment, changes in physicians' billed prices, and changes in utilization. In order to explore these increases, one can ex Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 43 Table 2·2.-Percent Distribution of Medicare Approved Charges for Physicians' Services, by Combinations of Place and Type of Service, 1983 Type of service All types ............. Medical care .......... Surgery . .. . . . . . . . . . . . Consultation .......... Diagnostic radiology ... Diagnostic laboratory .. Radiation therapy ...... Anesthesia ........... Assistant-at-surgery .... Other medical ......... aopo~outpatient department. bsNF~Skilled nursing facility. *~Less than 0.05%. All Office Inpatient 100.0 29.2 61.9' 37.3 15.5 18.8 33.7 3.8 27.5 3.8 0.7 2.9 8.4 3.4 3.7 8.0 5.1 2.2 1.2 0.4 0.2 4.8 4.7 ' 1.8 1.8 0.9 0.4 0.1 Place of service Home OPDa 0.6 5.9 0.5 1.1 2.4 0.1 1.3 0.4 0.5 0.1 0.1 0.1 0.1 Lab SNFb Other 0.3 1.3 0.6 1.1 0.3 0.1 0.1 0.3 0.3 SOURCE: I. Burney and G. Schieber, "Medicare Physicians' Services: The Composition of Spending and Assignment Rates," Health Care Financing Review, forthcoming. amine the changes in Medicare expenditures by partitioning expenditures as follows: Total Medicare number of x per capita x average expenditures beneficiaries use physician enrolled prices From this it also follows that: Change in change in + change in + change in total enrollment utilization physician expenditures prices' Since the beginning of the Medicare program, enrollment has grown at an average annualized rate of 2.4 percent for the aged population and 3.0 percent in aggregate (553). (The relatively small disabled Medicare population grew at an annualized rate of 7 percent per year from 1974 through calendar year 1981, after which enrollment declined.) The annual increase in the enrollment of the aged population has been so nearly constant-just in excess of 2 percent-that yearto-year fluctuations in reimbursements are almost entirely derived from changes in utilization or physician prices. 'Although conceptually accurate, in practice it is difficult to com pletely distinguish changes in utilization from changes in price. For example, the most common measure of physician fee inflation is the Professional Services Index of the Medical Care Component of the Consumer Price Index. This index is computed by pricing a fixed market basket of physician procedures from a fixed cohort of roughly 700 physician practices. As a result, the index reflects neither changes in the mix of physician services available in the market nor changes in the mix of physician practices active in the market. Therefore, simply "deflating" physician expenditures with the index may not yield an entirely accurate estimate of changes in utilization. From June 30, 1967, to June 30, 1983, approved charges per aged enrollee increased 591 percent, or 11.5 percent per year. The increase in approved charges per disabled enrollee from June 30, 1974, to June 30, 1983 was 390 percent, or 18.3 percent per year. Further, as shown in table 2-3, through fiscal year 1983, the aggregate increases in allowed charges per enrollee had been accelerating. With only two exceptions, the year-to-year total increase in recognized charges per aged enrollee increased in every year between 1970 and 1983 (553). Through June 30, 1970, the increase was 4.0 percent; by 1974, it was 8.9 percent; in 1978, 13.3 percent; in 1980, 16.0 percent; and in 1983, charges per enrollee increased 20.6 percent. 7 From 1978 onward, Medicare's approved charges per enrollee have consistently increased faster than total per capita expenditures for physician services in the United States (see figure 2-3). Of the 1968 to 1983 annualized increase of 11.5 percent per year, 6.9 percent was due to price increases and 4.6 percent was due to residual factors that include changes in utilization. Although there are no consistent trends in either price changes or the residual factors analogous to the accelerating change in approved charges per enrollee, the rate of price increase rose substantially 7Comparable fiscal year 1984 estimates will not be available until the preparation of the Federal Supplementary Medical Insurance Trustees report for 1986. In fiscal year 1984, the aggregate reimbursements for aged Medicare beneficiaries increased 12 percent over the previous fiscal year, compared to increases in excess of 19 percent in each of the 5 preceding fiscal years (533). 44 • Payment for Physician Services: Strategies for Medicare Table 2·3.-Components of Increases in Total Medicare Approved Charges for Physician Services per Aged Enrollee, 1967·83 (in percent) Price change factors CPR fee screens Residual factors Year Total (ending June 30) CPia Cumulative Annual Net Gross Denials Net increase 1967 .............. 7.6 -2.6 •••• 0 ••••••••• 1968 5.9 -3.6 -0.6 5.3 10.8 -1.4 9.4 14.7 1969 .............. 6.2 -5.0 -1.5 4.7 2.9 -0.4 2.5 7.2 1970 .............. 6.7 -7.5 -2.8 3.9 3.2 -3.1 0.1 4.0 1971 .............. 7.5 -10.1 -3.0 4.5 3.6 -3.2 0.4 4.9 1972 . . . . . . . . . . . . . . 5.2 -11.2 -1:2 4.0 2.3 0.4 2.7 6.7 1973 .............. 2.6 -11.7 -0.5 2.1 5.7 -0.6 5.1 7.2 1974 .............. 5.0 -13.2 -1.6 3.4 6.1 -0.6 5.5 8.9 1975 .............. 12.8 -16.2 -3.6 9.2 3.8 -0.3 3.5 12.7 1976 .............. 11.4 -18.6 -2.9 8.5 2.9 0.1 3.0 11.5 1977 .............. 10.2 -19.5 -1.0 9.2 3.3 0.1 3.4 12.6 1978 .............. 8.9 -19.4 0.5 9.4 3.8 0.1 3.9 13.3 0 ••••••••••••• 1979 8.6 -20.0 -0.5 8.1 3.9 -0.3 3.6 11.7 1980 .............. 11.5 -22.1 -2.4 9.1 6.8 0.1 6.9 16.0 1981 .............. 11.1 -24.5 -2.8 8.3 7.1 0.7 7.8 16.1 0 ••••••••••••• 1982 9.9 -23.9 1.5 11.4 5.9 0.5 6.4 17.8 1983 .............. 8.2 -23.4 1.6 9.8 10.9 -0.1 10.8 20.6 "CPI ~Medical Care Component of the Consumer Price Index. SOURCE: U.S. Federal Supplementary Medical Insurance Trust Fund, Board of Trustees, "1985 Annual Report of the Board of Trustees of the Federal Supplementary Medical Insurance Trust Fund," Washington, DC, Mar. 28, 1985. Figure 2·3.-Percent Growth in U.S. and Medicare per Capita Physician Expenditures, 1968·83 «; Q) 21 >. VJ :::l 19 0·;; 17 ~ c. E 15 ,g 13 Q) OJ c: 11 "' ~ (.) 9 Q) OJ "' 7 c Q) (.)a; a. 5 1968 1971 1974 1977 1980 1983 Years ____ U.S. total ---Medicare aged only SOURCE: M. Freeland, Bureau of Data Management and Strategy, Health Care Financing Administration, U.S. Department of Health and Human Serv· ices, Baltimore, MD, personal communication, Apr. 4, 198.~; and U.S. Federal Supplementary Medical Insurance Trust Fund, Board of Trustees, "1985 Annual Report of the Board of Trustees of the Federal Supplementary Medical Insurance Trust Fund," Washington, DC, Mar. 28, 1985. after 1974, and the rate of increase in the residual factors also rose substantially after 1979. Increases in expenditures have not been uniform across physician specialties or the types of services that they provide. Data have been analyzed from the 5 percent Bill Summary Record sample to disaggregate changes by specialty and type of service over the period 1975 to 1982 (133). Some of these data are reproduced in table 2-4. (Over this time period, total physician expenditures for the aged increased by 18 percent: 2 percent from enrollment increases, 10 percent from increases in reimbursements per service, and 6 percent from increases in services per enrollee.) Specialists in cardiovascular disease saw their Medicare reimbursements rise by 26 percent in total, half from increases in service volume, half from increases in reimbursements per service. Ophthalmologists and general surgeons also enjoyed comparable increases-of 13 percent-in reimbursement per service, while many other specialties saw increases in reimbursements per service of 9 or 10 percent. Increases in the provision of services appeared to be much more variable across specialties. Pathologists' services increased by 21 percent over that time period, while general and family practitioners' services increased only 2 percent and those of general surgeons increased but 1 percent (133). U.S. Expenditures for Physician Services As indicated earlier, Medicare is only one player in the market for physician services, and Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 45 Table 2·4.-Annual Percentage Increases in Medicare Payments for Physicians' Services for Aged Beneficiaries, 1975-82 Total Reimbursements/ Specialty reimbursements Services service Cardiology ........................ . +26% +13% +13% Pathology ......................... . 25 21 4 Ophthalmology .................... . 22 9 13 Radiology ......................... . 22 13 9 Podiatry .......................... . 20 11 9 Dermatology ...................... . 20 10 10 Otology/laryngology/rhinology ....... . 17 7 10 Orthopedic surgery ................. . 17 7 10 Internal medicine .................. . 17 7 10 Urology ........................... . 14 5 9 General surgery .................... . 14 1 13 General practice/family practice ..... . 12 2 10 SOURCE: L. Etheridge and D. Juba, "Medicare Payments for Physicians' Services," Health Affairs 3(4):132-137, Winter 1984. Part B physician service payments in 1984 were 19.3 percent of all expenditures for physician services (507). As a result, trends in that larger physicians' market must be observed to understand both the source of some of Medicare's problems and the prospects for their resolution. Much like Medicare expenditures for physician services, expenditures for physician services in general are a function of the size of the population, per capita use of physician services, and price per service. Hence the change in physician expenditures is a function of changes in prices, changes in per capita use, and changes in the population. The Health Care Financing Administration (HCFA) has developed internal estimates to further partition price changes into those due to price changes in the general economy and price changes in physician services that differ from those in the general economy. These estimates are reported in table 2-5. In 1965, an estimated total of $8.5 billion was spent on physician services in the United States (165). By 1984, that expense had expanded more than eightfold to $75.4 billion, a rate of growth of 11.5 percent per year. The years of the largest growth in physician expenditures occurred in 1969 to 1971, prior to the imposition of the Economic Stabilization Program; in 1976 to 1977, arguably as a result of the malpractice crisis of 1976; and during the inflationary period of 1979 to 1981. In fact, just over half of the growth in physician expenditures since 1965 can be ascribed to inflation in the general economy as measured by the Gross National Product (GNP) deflator. Physician fee inflation has exceeded inflation in the general economy with the exception of the 1972 to 1974 period of the Economic Stabilization Program. Since 1965, the total excess has been 39 percent, averaging just less than 2 percent each year (152). The difference between the GNP deflator and the professional services index of the medical care component of the Consumer Price Index (CPI) reached a high of 6 percentage points between 1975 and 1976. Since 1980, this excess inflation has been accelerating as medical inflation has continued while general inflation has declined. Between 1982 and 1983, medical inflation was twice the rate of inflation in the general economy. Physician fee inflation in excess of general inflation contributed 15 percent of the total growth in physician expenditures since the beginning of the Medicare program (152). The rate of growth in the general population has been fairly constant since 1965, at approximately 1 percent annually. Per capita use of physician services has increased only slightly since 1965 and exhibited actual declines from 1980 through 1983. Together, growth in the use of physician services has represented just under 10 percent of the total growth in expenditures for physician services since 1965 (152). Finally, one-quarter of the growth of physician expenditures can be ascribed only to the residual category. In the framework of the National Health Accounts (165), this residual can be interpreted as either an increase in the intensity or complex 46 • Payment for Physician Services: Strategies for Medicare Table 2·5.-U.S. Physician Expenditures and Factors Accounting for Growth, 1965·83 Percent change from previous year Total dollars Total Physician fee Use of physician Change in Other Year (in billions) dollars GNP deflatora inflationb services per person population factors NAC NA NA NA 1965 8.5 NA NA 8.2 3.2 2.5 -3.8 1.2 5.1 1967 10.1 9.8 2.9 3.8 0.0 1966 9.2 0.9 2.1 1968 11.1 9.9 4.7 1.2 -1.8 1.1 4.7 1969 ...... 12.6 13.5 5.3 1.7 1.6 1.0 3.9 1970 ...... 14.3 13.5 5.5 2.1 7.9 1.1 -3.1 1971 15.9 11.2 5.2 1.9 4.9 1.3 -2.1 1972 ...... 17.2 8.2 4.4 -1.0 0.8 1.1 2.9 1973 19.1 11.0 5.8 -2.4 0.7 0.9 6.0 1974 21.2 11.0 8.7 0.3 -0.7 0.8 1.9 1975 24.9 17.5 9.8 3.0 1.7 1.0 1.9 1976 27.6 10.8 5.5 6.0 -1.4 1.0 -0.2 1977 31.9 15.6 6.1 3.4 -2.4 1.0 7.5 1978 35.8 12.2 7.5 0.9 -1.1 1.0 4.0 1979 40.2 12.3 8.8 0.5 -0.2 1.1 2.1 1980 46.8 16.4 9.7 1.3 0.8 1.2 3.5 54.8 17.1 10.1 1.3 -2.4 1.0 7.1 1981 1982 61.8 12.8 6.2 3.3 -1.9 1.0 4.1 1983 ...... 69.0 11.7 3.9 3.9 -0.6 1.0 3.5 1965·83 Total change •..•.......••.. 711.8 199.0 39.1 1.5 20.5 70.6 Average annual change ....•• 12.2 6.3 1.9 0.1 1.0 3.0 aThe GNP deflator is a measure of the inflation in the general economy. bphysician fee inflation is measured here by the Physicians' Services Component of the Consumer Price Index minus the GNP deflator. cNA~Not available. SOURCE: U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Data Management and Strategy, Office of Financial and Actuarial Analysis, Division of National Cost Estimates, unpublished data, 1985; and R.M. Gibson, K.R. Levit, H. Lazenby, et al., "National Health Expendi· lures, 1983," Health Cam Financing Review 6(2):1-30, Winter 1984. ity of the average physician service-possibly due to technological change-or an increase in the fee for the average service that is not accounted for in the physicians' services price index (that estimates changes in prices with respect to a fixed market basket of services). In fact, over the period from 1968 to 1983, the increases in per capita expenditures for physician services for the Medicare program have risen at about the same rate as per capita increases in the country as a whole (see figure 2-3). Over that time period, the United States as a whole has averaged increases of 11.6 percent. The comparable statistic for the Medicare program is 11.5 percent. Through 1977, the Medicare increase was less than that of the United States as a whole in 8 out of 10 years. Only since 1978 has the Medicare increase consistently exceeded the aggregate increase. Physician Incomes Payments for physicians' services can also be considered as income to physicians. Those in comes have also been increasing. For example, average gross professional revenues more than doubled from $81,800 in 1973 to $192,200 in 1983 (391). Physicians' average net income also rose over that decade, but in constant dollar terms, it was nearly constant. Average real net physician income in 1984 was 4 percent lower than in 1970 (see table 2-6). However, this pattern of stable or declining real income was common to many occupations during the period of the 1970s with its high inflation rates. During the same period, earnings in constant dollars of workers in private nonagricultural industries fell 9 percent (500). Income data indicate that there have been substantial differences among physicians by specialty in both the level of income and the growth of income level. Hospital-based specialists in anesthesiology and radiology have had both the highest and the most rapidly increasing net incomes. General practitioners have had the lowest net incomes, on average. Net incomes for general practitioners actually declined between 1981 and 1983 by 2.6 percent a year (see table 2-7). Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 47 Table 2·6.-Trends in the Gross Income, Expenses, Net Income, and Real Net Income of Physicians, 1970-84 Average Average Average Median Real Year gross income expenses net income net income net incomea 1970 ....................... $ 66,100 $24,300 $ 41,800 NAb $41,800 1971 ....................... 74,200 28,900 45,300 NA 43,400 1972 ....................... 78,500 31,300 47,200 NA 43,900 1973 ....................... 80,800 32,200 48,600 NA 42,700 1974 ....................... 86,000 34,000 52,000 NA 41,200 1975 ....................... 94,900 38,500 56,400 $54,000 40,800 1977 ....................... 106,700 45,500 61,200 56,300 39,200 1978 ....................... 111,900 48,400 65,500 60,000 39,000 1979 ....................... 131,300 52,900 78,400 73,200 41,600 1981 ....................... 167,000 74,000 93,000 78,000 39,400 1982 ....................... 177,900 78,400 99,500 85,000 40,100 1983 ....................... 192,200 85,900 106,300 90,000 41,600 1984 ....................... 201,000 92,600 108,400 92,000 40,200 aAverage net income in 1970 dollars. bNA~Not available. SOURCE: American Medical Association, Profile of Medical Practice 1981 (Chicago, IL: AMA, 1982); G.L. Glandon and J.L. Werner, "Physicians' Practice ExperienceDuring the Decade of the 1970s," J.A.M.A. 244(22): 2514-2518, Dec. 5, 1980; "Average Net Income and Expenses of Physicians-1981," SMS Report 1:5, J,une 1982; and R.A. Reynolds and R.L. Ohsfeldt (eds.), Socioeconomic Aspects of Medical Practice 1984 (Chicago, IL: American Medical Association, 1984). Table 2·7.-Mean Physician Net Income After Expenses Before Taxes, Selected Years, 1973-83 (in sooo)a 1973 1974 1975 1977 1978 1979 1981b 1982b 1983b All physicians ............... $48.6 $52.0 $56.4 $61.2 $65.5 $78.4 $ 93.0 $ 99.5 $106.3 Specialty: General/family practice ..... 41.9 44.7 45.4 51.1 54.6 62.0 72.2 71.9 68.5 Internal medicine .......... 47.8 51.4 57.0 61.5 63.8 76.2 85.1 86.8 93.3 Su~e~ ................... 57.4 60.5 68.2 74.0 82.6 96.0 118.6 130.5 145.5 Pediatrics ................. 41.1 42.1 44.3 48.2 51.2 60.4 65.1 70.3 70.7 Obstetrics/gynecology ...... 55.4 61.7 63.3 69.9 70.3 91.8 110.8 115.8 119.9 Radiology ................. 59.5 63.8 75.2 76.7 81.5 98.0 116.9 136.8 148.0 Psychiatry ................ 38.4 41.3 44.8 48.2 50.2 62.6 70.6 76.5 80.0 Anesthesiology ............ 48.1 54.4 57.1 65.5 74.2 91.4 118.6 131.4 144.7 Census division: New England .............. 44.2 46.3 47.2 53.1 54.9 66.6 85.0 82.2 84.5 Middle Atlantic ............ 43.8 47.7 53.2 55.9 59.1 73.2 85.6 91 "1 98.6 East North Central ......... 50.5 54.2 59.9 62.7 69.9 81.2 100.9 106.2 114.3 West North Central . . . . . . . . 51.5 53.6 56.6 61.1 70.2 79.4 87.4 106.5 110.5 South Atlantic ............. 50.3 54.4 58.2 61.8 64.9 79.8 92.6 97.9 106.7 East South Central ......... 53.3 58.4 65.5 68.2 79.7 87.0 97.5 106.8 114.9 West South Central ........ 52.8 57.7 61.4 67.9 70.9 85.8 101.6 118.7 124.4 Mountain ................. 47.4 49.5 54.7 57.5 61.8 73.5 92.6 95.8 91.4 Pacific ................... 48.1 50.9 54.8 63.6 64.9 78.6 91.7 92.9 103.1 Type of practice: Solo ...................... 45.3 48.5 51.6 56.3 61.3 75.8 88.4 93.4 100.0 Non-solo .................. 52.8 55.6 61.1 68.3 69.9 80.7 96.6 104.0 111.3 Location: Nonmetropolitan ........... 46.9 48.5 50.2 56.7 64.8 74.1 87.1 86.9 87.2 Metropolitan Less than 1,000,000 ...... 50.3 53.7 58.8 63.2 67.4 78.8 99.6 103.9 111.0 1,000,000 and over ....... 47.5 51.5 55.6 60.6 63.9 78.8 90.2 98.4 106.3 Physician age: Less than 36 years ......... 32.8 40.6 43.7 49.6 49.0 64.3 62.5 73.3 77.0 36-45 years ................ 51.9 57.1 62.9 69.9 70.1 87.5 98.1 108.2 110.2 46-55 years ................ 55.0 58.9 62.3 67.7 76.2 87.1 110.8 116.5 133.6 56-65 years ................ 48.3 49.3 54.1 58.7 65.3 75.9 95.6 99.5 103.1 66 or more years ........... 31.9 34.0 35.0 36.8 44.4 54.9 68.3 64.3 71.9 aData, other than in the specialty breakdown, are based on responses from physicians in all specialties. bcaution should be observed in comparing results for 1981-83 with results for previous years because of changes in methodology made in the transition from the periodic surveys of physicians to the socioeconomic monitoring system. Results for 1981 and 1982 in the location breakdown reflect corrections of previously reported results. SOURCE: R.A. Reynolds and R.L. Ohsfeldt (eds.), Socioeconomic Aspects of Medical Practice 1984 (Chicago, IL: American Medical Association, 1984). 48 • Payment for Physician Services: Strategies for Medicare The variations in income parallel the variation in return on training among specialties. Dresh assessed the net capital value of returns to physician training for various specialties compared to general practice using a measure of lifetime physician earnings (113). He found that, except for pediatricians, psychiatrists and allergists/dermatologists, the adjusted lifetime earnings of medical and surgical specialists were much higher than those of general practitioners (113). There is also a marked difference in annual earnings among specialties. Several studies confirm this finding and indicate that physician earnings for some specialties are higher than those for other specialties even when allowances are made for the opportunity costs of education and capital spent on education and offices (46). Medicare's contribution to physicians' income also varies by specialty (353) (see table 2-8). A report of a recent survey of physicians indicated that only 8.4 percent of self-employed physicians had no Medicare patients in 1984 (406a). Those physicians who reported providing care to some Medicare beneficiaries indicated that 31.3 percent of their patients had Medicare as the primary source of insurance coverage (see table 2-9). What is not known is whether, and if so, to what extent, specialty differentials contribute to the differences in Medicare payment among specialty groups. Numerous other variables, such as volume of Medicare services provided, relative payments for procedural services vs. nonprocedural services, and the different mix of services provided by different medical disciplines are also contributory factors. Physician Control Over Expenditures for Physician Services Although physician expenditures represent 22 percent of all health expenditures (165), physicians have considerable influence on the use of a wide variety of nonphysician services. Blumberg estimated the fraction of all health care services under physician control and the relative cost of those services (46). By taking the product of those two factors for each type of service and summing over all services, he estimated physician control for 1976 as 69.8 percent of total costs. With respect Table 2·8.-Gross Physician Earnings From Medicare, 1981 Medicare Percent of total Specialty income gross income Anesthesiology ............ . $32,790 22 Family practice ............ . 21,220 15 General practice ........... . 21 '170 18 General surgery ............ . 43,750 25 Internal medicine .......... . 39,630 29 Neurology ................ . 37,390 24 Neurosurgery .............. . 37,310 18 Obstetrics/gynecology ...... . 8,530 • 5 Ophthalmology ............ . 49,010 24 Orthopedic surgery ......... . 43,220 17 Pathology ................. . 28,000 21 Pediatrics ................. . 1,170 1 Plastic surgery ............ . 18,780 12 Psychiatry ................ . 6,370 6 Radiology ................. . 49,730 28 Thoracic surgery ........... . 72,420 35 All surgical specialties . . . . . . 38,910 20 All nonsurgical specialties . . . 24,660 17 All MDs . . . . . . . . . . . . . . . . . . . 27,490 17 SOURCE: A. Owens, "How Much of Your Money Comes From Third Parties?" Medical Economics 60:254·263, Apr. 4, 1983. to individual services, physician control ranged from 91 percent of the cost of hospital expenditures to 20 percent of the cost of "other professional services." Physician control was assessed by estimating the proportion of services that patients received on the direction of their physicians. In aggregate, physician control over ambulatory care was estimated as 61.5 percent, ranging from 45 percent in pediatrics to 81 percent in psychiatry. Physician control over all physician services in all sites was estimated to be 76 percent. Another perspective on this question comes from recent work on the potential use of DRGs for physician reimbursement purposes (571). If physician charges represent 20 percent of all health care bills but physicians are responsible for 70 percent of all charges, one would infer that physicians order services of roughly 2.5 times the value of their own services. Based on all the physician approved charge data within 2 months on either side of a hospital stay for Medicare beneficiaries in Florida in 1981, West et al. estimated that physicians as a whole ordered hospital services 1.73 times the value of their own services (571). With respect to only those physician services provided during the hospital stay, the Florida statistic would be 2.2. In South Carolina for the same year, physician charges during a stay and within 2 months Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 49 Table 2·9.-Percent of Self-Employed Physicians Reporting Specific Percentile Ranges of Patients With Medicare Coverage, 1984 Percent of self-employed physicians Percent of patients with Medicare coverage 0 ............. . <10% ............ . <20 .............. . <30 .............. . <40 .............. . <50 .............. . <60 .............. . <70 .............. . <80 .............. . Physicians with All some Medicare physicians patients 8.4% 0 30.0 23.5% 44.2 39.0 60.8 57.2 75.8 73.5 86.1 84.7 92.0 91.1 95.3 94.8 98.6 98.4 asynthesis of reported percentage of physicians without Medicare patients and estimated number of physicians with specific Medicare percentages. These per· centages were not directly combined in the source report. SOURCE: M.L. Rosenbach, S. Hurdle, and J. Cromwell, An Analysis of Medicare's Physician Participation Agreement Program (Chestnut Hill, MA: Health Economics Research Center, Oct. 29, 1985). of either side of a Medicare hospitalization were accompanied by 2.7 times that amount in hospital services. Hospital charges were 3.3 times the value of Medicare physician allowed charges for strictly inhospital physician services. When the physician charges were further disaggregated to identify the physician practice that alone was responsible for the largest fraction of physician charges, each "lead" practice was responsible for 3.8 times the value of own charges in Florida and 4.8 times the value of own charges in South Carolina. Whether physicians control 70 percent or more of additional health care services, the potential spillover effect of physician payment reform on those additional health care dollars heightens the importance of any reform. PHYSICIAN PAYMENT CHANGES AND PRACTICE CHOICES With any change in the method of physician payment adopted by Medicare, one can expect responses by physicians in practice. The variety of choices available to physicians in response to payment changes includes both entrepreneurial and clinical decisions. Net payment to the physician, however, is only one factor in the physician's decisionmaking process. A patient's health status, presenting complaints, income, health insurance coverage, and health insurer's utilization controls, in addition to the physician's experience, practice style, repertoire of skills, and available equipment may be equally if not more important in influencing both the clinical and business decisions of the physician. These decisions include choices among particular physician services, choices with respect to the volume of services provided to Medicare beneficiaries, and choices with respect to physician participation in the Medicare program on an assigned basis. Physicians as Entrepreneurs: Accepting Assignment Two basic entrepreneurial decisions that physicians must make with respect to the Medicare program are: 1) whether to accept Medicare beneficiaries as patients and bill the program for services provided to those patients, and 2) whether to accept assignment.8 The factors that influence these decisions have been studied, and some significant factors identified, in studies using an economic model of the physician as entrepreneur based on the assumption that the physician is a discriminating monopolist9 (184,188,317,357,402). There is little question about the effects of changes in approved charges on physician participation in the Medicare program. The higher the ratio of approved charges to billed charges, the more likely a physician is to accept assignment for Medicare patients. The higher that ratio, the more services will be provided to Medicare patients per capita and the greater will be the number of individual Medicare patients treated 'Since the passage of the Deficit Reduction Act of 1984, all physicians who provide services to Medicare beneficiaries have been asked to make an annual election either: 1) to become a Medicare participating physician and accept assignment for all Medicare claims, or 2) to retain the option of accepting or rejecting assignment on a case-by-case basis. •Strictly defined, a monopolist is the only seller of a particular good or service in a particular market. A seller who can influence (raise) the final market price due to control over a substantial por tion of the total volume of a particular commodity or service has a substantial degree of monopoly power. A monopolist who can maintain different prices for different consumers is a discriminating monopolist. 50 • Payment for Physician Services: Strategies for Medicare by that physician. The estimated relationships have been consistent and positive. The estimates of physician responsiveness to Medicare's relative allowances have clustered around the value of 0.7 (293), implying that for a 10-percent increase in the ratio of approved to billed charges, there would be an increase of 7 percent in assignment.· With respect to the influence of approved charges on the decision to become a Medicare participating physician, one would expect participating physicians to be comparable to physicians who exhibited high assignment rates prior to the participating physician program (317). Early evidence suggests that those physicians who did elect to "participate" had relatively high assignment rates prior to the initiation of that program (15). An initial study of participating physicians showed, in fact, that previous assignment rates and the percentage of the usual fee reimbursed by Medicare were the most important economic variables associated with the decision to participate (94). In all likelihood, modification of CPR or conversion of Medicare physician payments from CPR to some other system would result in increases in approved charges for some physicians for some services and might result in decreases for others. Therefore, one would expect a decrease in the probability of assignment being accepted in those instances where approved charges were reduced and an increase where approved charges were raised. (Similarly, one would expect an increase in the probability that a physician would become a "participating physician" if his or her allowed charges had been increased.) The individual effects on specific physicians would depend on their approved charges in aggregate under CPR relative to those of their peers. The financial effects of changes in physicians' allowed charges on individual beneficiaries would depend on the constellation of physicians providing services and the individual services provided in treatment. Given both increases and decreases in approved charges, one would expect both decreases and increases, respectively, in nonassigned liabilities and increases and decreases, respectively, jn beneficiary cost-sharing liabilities. For example, a beneficiary whose physician experiences an increase in approved charges would be more likely than otherwise to have that physician accept assignment, thereby reducing the expected nonassigned liability. At the same time, however, that beneficiary would face an increase in coinsurance liability equal to 20 percent of the increase in the allowed charge. The net financial effect on any one beneficiary would depend on his or her physicians' combined assignment/participation behaviors and changes in allowed charges. Changes in the quality of care received by beneficiaries can also be expected to accompany the financial changes occasioned by physician decisions on assignment. In response to any net increases in out-of-pocket liabilities, Medicare beneficiaries may choose to forgo the use of some physician services. For some beneficiaries, such a change might actually result in an increase in quality through the reduction in the probability of receipt of some physician services that are inappropriate to the patient's condition. On the other hand, the provision of otherwise appropriate services might also be reduced, and the probability of receipt of appropriate services that are not currently being provided might also decline. For poor patients, a reduction in care would be likely to have an adverse effect on their health (194). Quality of care might also be affected if patient choice among physicians were to be restricted because of physician decisions about assignment. If it were the case that those physicians who provided relatively high quality experienced the greatest reductions in allowed charges, the beneficiaries' out-of-pocket costs for securing access to those physicians would be expected to exhibit greater than average increases. If those physicians' patients switched to other sources of care, quality might decline. However, there is no evidence associating physician quality and the level of allowed charges. Further, there is some evidence that patients will not switch providers in response to changes in out-of-pocket costs (288). A Medicare physician payment reform that is designed to reduce Medicare program expenditures probably will result in increased beneficiary liabilities as long as the case-by-case assignment choice remains an option for physicians and as long as there exists a private market for physi Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 51 dans' services. A net decrease in average allowed charges can be expected to lead to reductions in assignment by nonparticipating physicians and reductions in the numbers of physicians who elect to become participating physicians. The participating physician concept has modified physicians' options with respect to assignment. One can infer, however, that the participating physician decision is analogous to the case-by-case assignment decision (15), an inference that appears to be confirmed in a study using American Medical Association survey data (94). Data from the Medicare carriers indicate that assignment rates for all claims have increased under the participating physician program (537). A study based on physician survey data from five specialties suggested that net assignment rates for physician visits would decline with respect to a possible "all or none"10 assignment system (317). The Effect of Relative Prices on Technology Choices If one assumes that a physician has made the decision to accept Medicare patients on either an assigned or unassigned basis, in effect, an array of relative expected payments available for specific services is established in advance. That is, the physician can know that he or she may expect to receive $16 from Medicare and $4 from the patient for an assigned office visit, for example, or $48 from Medicare and $52 from the pa tient for an unassigned sigmoidoscopy. At that point, one might begin to ask about the impact of such price differences. In theory, in addition to factors specific to the patient, the patient's health insurance, the patient's physician, and the physician's practice, relative prices can influence physicians' clinical decisions. Specifically, one can identify two types of clinical choices where relative prices may make a difference: 1) choices between two services that are substitutes for one another, and 2) choices among services that may be complements. 10Under an "all or none" system, a physician would have to choose between accepting assignment on all Medicare claims versus not being able to accept assignment on any Medicare claims. The current system might be described as "all or some." 56-119 0 -86 -3 QL 3 The choice between substitutes is usually illustrated by the classic distinction between medical and surgical treatments for a particular disease. In fact, there may actually be more than two treatments that can be substitutes as in the case of (surgical) open nephrolithotomy, (catheter-based) percutaneous nephrolithotomy, and extracorporeal shock wave lithotripsy (ESWL) for the treatment of kidney stones. An alternative type of substitution may occur if the physician has the option of delegating the performance of a particular diagnostic test, for example, to an assistant. Similarly, the physician may substitute time spent in performing specialized diagnostic tests, such as endoscopies, for his or her own direct patient contact time. The distinction between substitutes and complements may become blurred at this point, and discerning acttial choices from billing records becomes especially difficult. Although it may involve a complementary service, when physicians choose to perform endoscopies or other diagnostic tests by themselves, they are substituting time with one patient for time that might have been spent with another patient. When would relative price make a difference? Where there are clear medical indications of the advantage of one technology over another or clear contraindications against one choice, price may not matter much to the physician. Differences in net revenues to the physician would be more likely to influence medical decisions for which the medically and ethically correct decision is unclear (194). The relevant comparison with respect to net revenues involves not only the net revenues anticipated from the particular services that may be substitutes for the patient in question, but the opportunity costs of providing services to another patient. For example, although a physician might receive a greater net revenue from Medicare for providing a limitedn office visit rather than a brief "The manual of Current Procedural Terminology, 4th ed. (CPT-4) defines a limited service as follows, "a limited level of service is one pertaining to the evaluation of a circumscribed acute illness or to the periodic re-evaluation of a problem including an interval his tory and examination, the review of effectiveness of past medical management, the ordering and evaluation of appropriate diagnos tic tests, the adjustment of therapeutic management as indicated, and the discussion of findings and/or medical management" (85). Brief involves "a level of service pertaining to the evaluation and treatment of a condition requiring only an abbreviated history and examination." 52 • Payment for Physician Services: Strategies for Medicare office visit to a Medicare beneficiary, that physi cian might choose to provide the brief visit if still greater net revenues were available for provid ing services to patients with insurance that pro vide higher payments than Medicare. The effects of relative prices on treatment choices, however, have not been studied empiri cally. This situation is partly a result of data limi tations, but it also derives from the difficulty in empirically framing the question for analysis. Finding specific examples where a sufficient num ber of individual physicians face a particular choice among substitute services involving comparable patients but with differing relative payment levels is not easy. Several authors have speculated on the potential effects of relative revenue differences involving services that may be viewed as complements to office visits (4,424). Schroeder and Showstack note that the per physician net incomes of a group practice that performed eight specific diagnostic tests in the practice would be nearly three times greater than that of a physician in a solo practice in which virtually no diagnostic testing was performed by the physician in the office even though each physician in the group practice might see fewer patients than the solo physician (424). In the case of a Medicare beneficiary whose coinsur ance and deductible were covered by a private supplemental insurance policy, the additional cost to the patient of an endoscopic exam might be negligible, but the increase in revenue to the physician who complements the patient's treatment with that exam may be several hundred dollars. Because the information provided by the test may be useful and the time required to perform the ~est is relatively short (423), the incentive to perform the test may be nearly irresistible. Is There Too Much Service With Fee for Service? One issue that should be addressed at this point involves the incentives faced by physicians under a fee-for-service payment system. In one sense, fee-for-service incentives are volume increasing because the physician can receive an extra payment for each extra service performed and billed. The gross price per service alone, however, will not establish a positive incentive. Any incentive will depend on the available revenue per service net of cost. If, for example, a physician can spend the same amount of time to administer an injection of pneumococcal vaccine or prescribe a drug, and if the additional cost to the physician for the vaccine is $5 with no appreciable additional cost for a prescription pad, a positive incentive exists to prescribe rather than to inject even if the payment for the injection is $10 and that of the visit with a prescription is $6. One might argue, however, that physicians behave as if virtually all their costs were fixed; 12 hence gross payment levels do indicate incentives. Physician obligations for employee salaries, space and equipment, insurance, and transportation may be considered by physicians to be fixed annually, and those obligations may represent 70 to 90 percent of all office expenses (355,391). Further, since total physician office costs are approximately 40 percent of gross professional revenues, most services may appear to yield profits and hence embody an incentive to provide more. This argument assumes, however, that physician time is of no value in and of itself. When the alternative revenues that a physician might generate with his or her time, the opportunity costs, are included in the calculation of costs, incentives for greater amounts of service exist only when net payments exceed those opportunity costs. From this perspective, any incentives would be a function of the level of payment in addition to the method throughwhich that payment level is derived. In one sense, Medicare payments may be too high for some services. They may be too high in general. For example, nearly one claim in five is paid at or below the level of the physicians' customary charges (535). Because a physician will be paid the lower of the billed charge or the customary (or prevailing) charge, one would expect billed charges to exceed customary charges if approved charges were consistently too low. Lower approved charges within the context of a fee-for "Fixed costs are those costs that do not vary as output varies. Most overhead costs can be considered fixed costs. Although a policy for professional liability insurance may subsequently be canceled, a physician's expenditures for such insurance are fixed at rates estab lished annually. Some office overhead expenditures, e.g., heat or electricity, however, are not fixed, since they will cease to be incurred if the physician does not have his or her office in operation. Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 53 service, therefore, might retain beneficiary access to physician services in addition to reducing some inappropriate incentives to provide too much service. Volume Responses and Induced Demand The additional possibilities of providing physician services that in some sense are complements to treatment lead to the issue of changes in the aggregate volume of physician services in response to changes in allowed charges. In a market economy, most suppliers would respond to a decrease in prices paid for their goods and services by reducing the quantity they were willing to sell. One might believe, therefore, that reductions in approved charges for physicians services would lead to reductions in Medicare expenditures for those services, even if there were no reduction in the volume of services provided. In fact, one might believe that a reduction in approved charges by Medicare would lead to a reduction in the number of services provided by physicians to beneficiaries, reducing expenditures by an even greater amount. However, in response to changes in approved charges, beneficiaries and physicians may appear to change their behaviors in ways that increase service volumes. For example, in response to a decrease in approved charges some patients might want to increase their purchases of physician services. If, in addition, physicians can control service volumes, an alternative approach to payment·reform based solely on reductions in allowed charges may be needed to control increases in expenditures. The question of whether physicians in particular can influence the use of their services and hence frustrate cost control efforts based solely on con trols on fees is one of a number of issues included under the topic of iatrogenic-or supplier-induced demand. The possibility that physicians might induce demand for their services has been the subject of empirical studies since as early as the late 1960s (389). In particular, studies of cases where public health insurance programs have reduced or frozen physician fees have suggested that such efforts have not controlled costs (158). Unfortunately, none of the studies unequivocally proves or disproves the magnitude or even the existence of induced demand. For example, a study of California physicians' billings to Medicare during the period of the Economic Stabilization Program, found that in spite of the fee freeze overall costs rose substantially because there was an increase in the volume of services provided to Medicare beneficiaries (215). This result is often cited as evidence that even if price controls do control price, they do not control expenditures. However, an alternative explanation for the increase in Medicare volume during the Economic Stabilization Program can be found in the well-established positive relation between Medicare participation and allowed charges. The Economic Stabilization Program was instituted in August 1971, 1 month after Medicare approved charges had been increased for fee screen year 1972 and 5 months prior to the (January) period typically exhibiting the largest increases in physicians charges. As a result, Medicare approved charges relative to private market payments were frozen at a level typically higher than that of any other time of the year. Given that physician participation in the Medicare program has been found to be positively related to the ratio of approved charges to billed charges, one would have expected physicians to increase the volume of services provided to Medicare patients. If not resolved, the current debate on this issue only simmers. There appears to be some physician volume response to reductions in physician prices. Quebec's experience indiCates that fee controls can be effective in reducing the rate of growth in physician expenditures, in spite of some volume increases (28). If the volume response is perceived as potentially vitiating the desired effect of physician payment reform, an initial step might be to monitor volume changes to ascertain the need, if any, for additional controls. Cost-Shifting One other potential physician response to reductions in Medicare approved charges is that physicians might raise their non-Medicare charges, a form of "cost-shifting." However, if non-Medi 54 • Payment for Physician Services: Strategies for Medicare care patients are responsive to price, i.e., if their willingness to purchase physician s~rvices is reduced when prices increase, rational physicians would reduce their charges to non-Medicare patients rather than increase them in response to reductions in approved charges (188,357). It may well be the case that physicians may choose not to participate in the Medicare program if payments are reduced to levels significantly below those of the non-Medicare market, but there is no theoretical or empirical evidence for physicians' cost shifting. It is possible that some physicians might appear to provide greater quantities of service to their non-Medicare patients as an additional response to reductions in Medicare approved charges. This might also be perceived as "cost-shifting." However, other things being equal, if non-Medicare patients are responsive to price, there should be no net increase in physician service volumes to those patients unless there is a decrease in average fees charged to those patients. Alternatively, if some physicians elect to serve fewer Medicare patients in response to a decrease in Medicare approved charges, there might be a reduction in appointment delays or office waiting times. This de crease in the "time price" might then be followed by an increase in demand for services by those physicians' non-Medicare patients. ISSUES WITH RESPECT TO MEDICARE'S PHYSICIAN PAYMENT SYSTEM Then HCFA Administrator Leonard Schaefer enunciated the basic missions of HCFA in 1979 (508): • to promote the timely, cost-effective delivery of appropriate, quality health care services to its beneficiaries; • to make beneficiaries aware of the services for which they are eligible, and to make those services accessible to them in the most effective manner; and • to ensure that its policies and actions promote efficiency and quality within the total health delivery system which serves all Americans. This mission statement can provide the basic starting point in examining whether Medicare's physician payment systems foster or hinder the achievement of those objectives. As might be expected, however, it is easier to raise the issues than to resolve them. As reviewed above, Medicare expenditures for physician services have continued to increase, but until 1982, Medicare increases were roughly in concert with those observed in all U.S. expenditures for physician services. There are perceived excessive variations in such aspects of Medicare physician services as payment levels, assignment rates (and hence effective financial coverage), and utilization of services. Such variations are consistent with problems in quality, access, cost and/or efficiency, but one would also expect to observe even some substantial variations in a national program serving more than 30 million beneficiaries in thousands of local markets. Although many observers conclude that the variations are too great not to reflect a particular problem of interest, there is little or no consensus about whether the variations signify actual problems. Although the many perceived variations in the Part B program may not unequivocally indicate the presence of problems, there seems to be no question that such variations have led to confusion on the part of both the beneficiaries and the providers. In addition, health insurance coverage, which insulates patients and providers from health care costs, and the design of the Medicare benefit package itself do not provide incentives for efficiency. Confusion and inefficiency are the first issues reviewed below. Following that discussion, the magnitudes of the potential problem variations are addressed. Potential problems include: • variations in annual expenditures per beneficiary; • variations related to assignment; and • payment level variations with respect to geographic areas, physician specialties, and place, type, and vintage of service. Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 55 Some variations that might be expected, such as those due to quantity discounts, for Medicare as a large payer, are not evident, and some variations are either exacerbated or moderated by the MEL These potential problem areas are reviewed in light of what is known about plausible levels of variations in payments and expenditures that might be expected in the Medicare program. Beneficiary and Provider Confusion The CPR method of determining approved charges, the possibility of case-by-case assignment choice by physicians, and even the existence of the Part B deductible itself all contribute to confusion about payments among both beneficiaries and providers. An Administrator of HCFA once noted, "We get something like 9 million letters a year on reimbursement, most simply wanting to know how the payment was arrived at" (98). Even with the Medicare participating physician program initiated under the Deficit Reduction Act of 1984 (Public Law 98-369), it is rarely straightforward for a Medicare beneficiary to establish in advance his or her out-of-pocket liability for covered Part B services. Because CPR in effect establishes a separate fee schedule for each physician practice, the approved charge for a specific procedure may vary widely within a given locality. Hence, there will be variations in beneficiary coinsurance liability for a specific service regardless of assignment. Further, some physicians may not be able to recall their Medicare approved charges when they recommend to the patient that a specific service be rendered, and when they refer a patient to another physician for a specialized service, they may not know all of the services that that physician may render, much less the charges for those services. In that case those physicians may not be able to tell patients what their expected out-of-pocket cost will be. Finally, for some infrequently performed procedures rendered near the beginning of a fee screen year, an approved charge for the procedure may not have been calculated for the physician practice. Neither the beneficiary nor the physician would then know the level of the approved charge until after a bill for the service had been submitted. Under Medicare's participating physician program, a significant proportion of practices have agreed to accept assignment on all claims. There are also directories available at the offices of the carriers that identify those practices. Unfortunately, because some physicians may have more than one practice and may not have elected to "participate" in each practice, the directories are not a perfect guide to 100-percent assigned practices (231). With respect to the providers, there are many situations in which a physician will not know how much he or she will be paid for treating a Medicare beneficiary. Prior to the implementation of the freeze, at the beginning of a fee screen year a physician would be likely to learn of that year's allowed charges only as reimbursements were received for services rendered in the new fee screen year. A physician could request information on those new approved charges from the carrier, but there was no organized information dissemination of approved charges to physicians from the carriers. However, even where the approved charges are known, those charges are not reimbursed by the carriers for patients who have not yet satisfied the Medicare deductible. Further, the deductible is assessed as of date of payment, not date of service. If a Medicare beneficiary received $75 of physician services on a nonassigned basis in January but did not file those claims until after receiving $75 of assigned physician services in March, the assigned services would be applied to the deductible and hence would not result in a payment from the carrier. Even though the patient may have indicated that he or she had already met the deductible, the physician who accepted assignment in this example would have to collect those charges from the beneficiary. Finally, provisions regarding elderly beneficiaries who are employed may also lead to uncertainty for providers. Medicare is not the primary payer for aged beneficiaries below age 70 who are covered by employer-offered health insurance plans. As a result, a physician who treats such a Medicare patient may find that the charge approved by the patient's insurer is not the same as the Medicare allowed charge. In addition, if the physician had accepted assignment and submitted a bill to the Medicare carrier, Medicare might 56 • Payment for Physician Services: Strategies for Medicare deny payment of the bill unless it had been first presented to the third-party carrier of the patient's employer. Inefficiencies in the Delivery of Medical Care In addition to improving people's ability to obtain medical care, health insurance affects people's decisions about using services and providers' decisions about purchasing and using technologies (343). Compared to those who pay for care out of pocket, cost is less of a deterrent to insured persons' decisions to seek care and to choose costly providers and technologies. Similarly, consideration of insured patients' finances is less of a concern to physicians and other providers who buy and use medical technologies. Thus, one would expect to observe Medicare beneficiaries demanding greater volumes of covered physician services at any price level than would be strictly cost ef fective. Further, this is more likely to be the case for those beneficiaries who: 1) receive services under capitation without copayment, or 2) have obtained supplemental insurance that pays for the Medicare deductible and coinsurance. Under feefor-service payment, in those cases where net revenues are increased by the increased provision of care, providers' financial incentives reinforce the beneficiaries' enhanced demand for services. Under capitation, net revenues are diminished by the increased provision of care; hence, the incentives regarding use for beneficiaries conflict with the incentives for those who receive the capitation payments. The Medicare benefit package may also contribute to inefficiency in the provision of physician services. Although providers render much preventive care to Medicare beneficiaries in the course of visits, many preventive services, such as physical examinations and influenza vaccinations, and some rehabilitative services, such as hearing aids, are not officially included under Medicare coverage. Exceptions are pneumococcal vaccination, which is covered for all beneficiaries, and hepatitis B vaccine, which is covered for end-stage renal disease patients and other categories of beneficiaries at high or intermediate risk of contracting hepatitis B (89). Beneficiaries are liable for the total charges of services not covered by Medicare, and might be expected to use less of such services than might be recommended on strictly medical or cost-effective grounds. On the other hand, legislation (the Omnibus Reconciliation Act of 1980, Public Law 96-499) has eliminated beneficiary cost-sharing in certain cases to encourage the use of less costly alternatives. For example, beneficiaries are liable for no deductibles or coinsurance for certain surgery performed in ambulatory surgical centers. The result from these design decisions would be expected to be inefficiency (higher cost for a given level of quality) in the provision of particular technologies and in the combination of technologies used for a given medical condition. For example, duplicative laboratory tests may be performed, diagnostic and therapeutic equipment and facilities may be used far below capacity, and the more expensive and hazardous hospital setting may be used when ambulatory care would be just as effective ( 483). Variations in Annual Expenditures per Beneficiary There is more than a twofold variation in reim bursements per Medicare enrollee across the 50 States. In 1982, for example, Medicare reimbursements for physician and other Part B medical services on behalf of aged beneficiaries in the United States averaged $517.93 per enrollee. In Nevada, however, the average was $842.29, while in West Virginia, the average was $305.15 per enrollee. The western census region as a whole-(l:veraged $654.40, nearly 40 percent greater than the'sotithern census region at $468 (525). Although not necessarily indicative of problems, there are also variations by age, gender, and race. As might be expected, Medicare reimbursements per enrollee increase as the age of the enrollee cohort increases. In 1978, average reimbursements per enrollee for physician services for aged Medicare eligibles averaged $197. For the age 65 to 69 cohort, the average was $152; for those aged 85 and over, the average was $259. During that same year, reimbursements on behalf of male beneficiaries were $214 compared to $186 for fe Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 57 males. This disparity widened over the 1975 to 1978 period. Average reimbursements for white beneficiaries were $201 in 1978, compared to an average of $153 for nonwhite beneficiaries. (There was no obvious trend in this relation over the 1975 to 1978 time period (297).) Part of the regional variation in Medicare expenditures per beneficiary is due to the variation in the proportion of beneficiaries who exceed the Medicare deductible-and who are thus eligible for reimbursementsY In 1982, 65.6 percent of both the aged and the disabled populations exceeded the deductible. In Rhode Island, however, 78 percent of the aged beneficiaries exceeded the deductible, while only 54.7 percent of Kentucky's aged beneficiaries exceeded the deductible. (There are no marked disparities in any State between aged and disabled Medicare populations in terms of the percentages exceeding the deductible.) One other factor that contributes to the variation in expenditures per beneficiary involves differences in the apparent relative stringency of the reasonable charge process. In the first quarter of 1985, 17.4 percent of physician claims were submitted with charges at or below the effective approved charge limits. Of the remaining 82.6 p~rcent of claims, the average reduction per cla1m was $32.84. As a result, Medicare's approved charges in aggregate were 74.5 percent of the total submitted charges. The differences in these statistics among carriers are striking. For example, Maryland Blue Shield reduced only 48.7 percent of claims, while 91.6 percent of Hawaii's claims were subject to reductions by its carrier, Aetna. The average reduction per claim for that time period was $19.98 in Vermont, but $44.49 in Nevada. Finally, allowed charges were 80.8 percent of total covered charges in Kentucky, but only 66.2 percent of total covered charges in Rhode Island (535). Thus, a Kentucky beneficiary with a nonassigned claim for $100 might expect to have to pay $33.36 out-of-pocket charges, while a Rhode Island beneficiary with a comparable claim might have to pay $47.04, 41 percent more. 13This in turn is due to variations in patient health and in both the level of allowed charges and service volume. Patients who do not initiate visits to physicians or patients with either very inexpensive doctors and/or medically very conservative doctors may not exceed the deductible. Variations Related to Assignment There is substantial variation in assignment rates across the United States. Assignment rates nationally declined from 1969 through 1977, but they have increased since that time reaching 59 percent of claims and 59.6 percent of charges in 1984.14 In 1982, when assignment was accepted on 51.8 percent of charges for the aged, assignment rates for elderly people increased as the age of the cohort increased, ranging from 47.3 percent for the age 65 to 69 cohort to 61.1 percent for those aged 85 or above. Assignment rates among female beneficiaries exceeded those of males, 52.6 percent compared to 50.6, and nonwhite beneficiaries exhibited higher assignment rates than whites, 79.9 percent compared to 49.3. Assignment rates for disabled beneficiaries exceeded those for aged beneficiaries. Within the disabled population, rates for females were greater than for males, and rates for nonwhites were greater than those of whites. Assignment rates for the youngest cohorts of disabled beneficiaries were the greatest at 88.7 percent for those younger than age 25 compared to 66.6 percent for those aged 45 to 64. Across the States in 1982, the rates for the aged ranged from 87 percent in Rhode Island to a low of 17 percent in South Dakota. Assignment rates for most major physician specialties ranged from 40 for otolaryngologists to 54.7 percent for cardiologists with most specialties at less than 50 percent (296). In fiscal year 1985, the first year of the participating physician program, 29.8 percent of the physician practices that provided services to Medicare beneficiaries elected to participate. Across the States, the percentage of participating physician practices ranged from a high of 53.9 percent in Alabama to a low of 5.6 percent of the practices in South Dakota. With respect to physician specialties with substantial Medicare volumes, 21.1 percent of anesthesiology practices elected to participate compared to 50.8 percent of the nephrologists. Of 17 distinct physician specialties reported by HCFA, 11 exhibited participation rates between 25 and 35 percent (518). (Early tabulations "Unless otherwise noted, these and subsequent statistics on assignment include the mandatorily assigned claims of beneficiaries who are entitled to both Medicare and Medicaid coverage. 58 • Payment for Physician Services: Strategies for Medicare from the second year of the program indicate that 15.3 percent of the participating practices of physicians, osteopaths, and limited license practitioners from fiscal year 1985 did not continue their participation into fiscal year 1986, although 13,718 new agreements were submitted. As a result, the aggregate participation rate for physicians, osteopaths, and limited license practitioners dropped from 30.4 to 28.4 percent (521).) Because of the assignment option, differences in the proportion of total charges that are approved yield differences in beneficiaries' out-ofpocket liabilities for covered services. In the first quarter of 1985, 81.6 percent of assigned claims were subject to reductions averaging $32.48 per claim, as a result of which 73.5 percent of total covered charges were allowed. Of nonassigned claims, 84.7 percent were subject to reductions, which averaged $32.84 per claim, yielding approved charges equal on average to 74.5 percent of total covered charges (535). Therefore, for claims that were subject to CPR reductions, expected beneficiary out-of-pocket cost was $18.02 for the average assigned claim. Adding an average coinsurance of $19.19 to the nonassigned liability of $32.84, the expected beneficiary out-ofpocket on an unassigned claim was $52.03 (see table 2-10). Historically the differences in the statistics between assigned and nonassigned claims have been small. Although a slightly higher percentage of nonassigned claims have been subject to reductions, the reductions on assigned claims have been somewhat greater both in absolute and percentage terms. Within the class of nonassigned claims, however, differences in the effective stringency of the reasonable charge process across carriers directly lead to differences in beneficiary liability. In dollar terms, the average reduction on nonassigned claims is exactly equal to beneficiary average nonassigned liability per nonassigned claim. Although the average for the country was $32.84 per claim in the first quarter of 1985, in Maine, the average nonassigned liability per claim was $17.37, while in the Mil111eapolis, Minnesota, region served by Travelers, this liability was $56.38 per claim (535). One might expect that beneficiaries would appear to react to these variations in nonassigned liability with more searching for doctors who accept assignment in those areas where average nonassigned claims were relatively expensive in terms of beneficiary out-of-pocket costs. This does not appear to be the case, however. There is little obvious positive correlation between assignment rates and average nonassigned liability per claim. In fact, in the carrier data reported to HCFA one can observe a slight negative correlationY Some evidence consistent with searching for assignment can be seen in the variations in assign 15ln the first quarter of 1982, the correlation between average nonassigned liability per claim and the assignment rate by carrier was -0.26. Possibly due to the increase in assignment rates concomitant with the participating physician program, the negative correlation between average nonassigned liability per claim and the assignment rate by carrier has been reduced. In the first quarter of 1985, this correlation was -0.12. Table 2·10.-Medicare Reasonable Charge Reductions per Claim, January-March, 1985 Type of claim Assigned Unassigned Average billed charge ...... . $122.35 $128.93 Percent of claims reduced .. . 81.6% 84.7% Percentage reduction ....... . 26.5% 25.5% Average CPR reduction ...... Average approved charge .... Medicare payment .......... Beneficiary coinsurance ..... Nonassigned liability ....... . (26.5% x $122.35) ($122.35 -32.48) (80% X $89.97) (20% X $89.97) $ 32.48 $ 89.87 $ 71.90 $ 17.97 $ 0.00 (25.5% X $128.93) ($128.93 -$32.84) (80% X $96.09) (20% X $96.09) $ 32.84 $ 96.09 $ 76.87 $ 19.22 $ 32.84 Total beneficiary cost ...... . $ 17.97 ($19.22 + $32.84) $ 52.06 SOURCE: U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Quality Control, Carrier Reasonable Charge and Denial Activity Report January-March 1985 (Washington, DC: U.S. Government Print· ing Office, 1985). Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 59 ment rates by annual charges per user. In 1978, assignment was accepted on 44.6 percent of total physician charges. For those aged patients with charges less than $100, the assignment rate was 30.3 percent. For those with charges between $100 and $149, the rate was 27.9 percent. For those with annual charges in excess of $149, however, as annual charges increased, assignment rates increased. The maximum average assignment rates were 52 percent for those beneficiaries with annual charges in excess of $2,500. This general pattern is consistent by specialty and holds for both the aged and disabled populations (297). These data are also consistent with the "big bill" hypothesis that physicians accept assignment more readily for services with high charges, accepting the Medicare approved charge with certainty rather than risk the possibility of incurring a relatively large bad debt. In 1981, with respect to those "big bills" with submitted charges in excess of $200, assigned charges were 52.9 percent of total submitted charges. For bills of $200 or less, assigned charges were 47.6 percent of the total (69). The assignment option probably exists in Medi care because Blue Shield Plans in the mid-1960s had participating physician options under which, for some patients, a physician would agree to accept as payment in full a fee that was below his or her usual charge for a particular service. A beneficiary who receives an assigned service from a particular physician has therefore received a discount from that physician's otherwise standard fee. From this perspective there is no correct or best level of assignment. A beneficiary who received all physician services on assignment is not necessarily better off than another beneficiary who received the same services from another physician with no services provided on assignment. The out-of-pocket costs of the first beneficiary could be higher than those of the second. Other things being equal, assignment can mean reduced out-of-pocket liabilities and hence reduced financial barriers to care. Assignment rates may thus be interpreted as imperfect indicators of beneficiary access to care, with higher assignment rates presumed to reflect better access. There is no evidence, however, that there are particular groups of Medicare beneficiaries who could not obtain access to needed physician services included in the Medicare benefit package. Perceived Payment Imbalances In addition to aggregate and per capita variations in expenditures and out-of-pocket liabilities, there are also significant variations in approved charges for specific services, i.e., individual fees. There are variations across States, across geographic areas within States, and across the physician specialties regardless of State. Variations have also been observed by site of service, by the type of service, by the vintage of the service, and by the apparent relative effective stringency of the MEI. By design, the legislation establishing the Medicare program did not mandate a national fee schedule for physician services, and the CPR system was at least partly adopted to allow recognition of local differences in charge levels. The observed ranges in approved charges, however, suggest to some that there is excessive variation in charges. Further, the variation in charges is not random. As a result, the incentive effects of Medicare's physician reimbursements may not be in concert with other public policy objectives. Geographic Variations For Medicare Part B payment purposes, the country is currently divided into 240 localities. In 18 States, the entire State is a locality; in the remaining States, there are two or more locali ties. Texas has the greatest number of localities with 32, California has 28, and Illinois has 16 (515). Although multiple localities may be identified and used to partition physician claims for the purposes of establishing prevailing charges and determining approved charges, it should be noted that not every procedure will have a prevailing charge established on a locality-specific basis. Relatively low-volume procedures may have a state-or carrier-wide prevailing charge even in some States with more than one locality. Medicare carriers were given the responsibility to identify localities in the original Medicare legislation. The Social Security Act, however, was permissive in that it did not require that a carrier identify two or more localities. The original guidelines indicated that localities were to embody substantial, relatively homogeneous, but not necessarily contiguous geographic areas. Homogeneous 60 • Payment for Physician Services: Strategies for Medicare but relatively small jurisdictions, such as particuexhibited lower rather than higher fees as might lar neighborhoods, were not to be identified as have been anticipated. distinct localities for payment purposes. Subse quent instructions to the carriers required HCFA Intrastate Variations.-Within those States permission to change the number of localities where carriers had established more than one locality, variations in prevailing charges between within any carrier jurisdiction. In recent years,many carriers have consolidated two or more lothe highest charge and lowest charge localities calities. have been commonly observed to exceed 50 percent (494). In general this reflects urban/rural payInterarea Variations.-The range across localment level disparities. A 1976 study showed thatities in charges for specific services is substantial. Medicare payment levels in urban areas exceededData from HCFA for fee screen year 1980 reveal those in rural areas by 23 percent (421). After adthat the highest prevailing charge for a brief foljustment for cost-of-living differences, Medicarelowup hospital visit exceeded the lowest by 373 prevailing charges in the largest standard metropercent. For extraction of lens the "excess" was politan statistical areas in 1975 averaged 17 per159 percent. For electrosection of the prostate, cent above the national average, while those inhysterectomy, and single view chest X-rays, re the counties with the lowest populations averagedspectively, comparable differences were estimated 8 percent below (494). If payment level differencesas 197, 143, and 536 percent (see table 2-11). exceed differences in physicians' costs of practice,these urban/rural disparities under Medicare will Four-, five-, and six-fold differences in prevailtend to discourage physicians from locating in ruing charges in 1980 were not aberrations. Data ral areas. To the extent that the Federal Governfrom fee screen years 1976 through 1980 show ment has a policy of trying to enhance access ofthose differences as relatively constant over time. residents of rural areas to physician services,A study of Medicare prevailing charge data from 1975 for a selection of surgeries also showed the Medicare's physician reimbursement policy in thisregard is at variance with national policy. same range of results (50). In addition, that re view examined whether those variations could be How Much Geographic Variation Is Enough? explained by differences in cost of living, malpracThere are arguments on both sides of the ques tice premiums, quality of care, or relative physition of whether to have identical or varying fees cian shortages. The findings of the study were that for the same service in different jurisdictions. Two cost-of-living adjusted fees still showed three-fold arguments for identical payment levels across variation; that neither quality differences nor maljurisdictions, such as might be produced through practice expense differences could explain the varthe use of fee schedules, involve: 1) the potential iation; and that relative physician shortage areas inappropriateness of different effective benefit Table 2·11.-High and Low Prevailing Charges in Localities for Five Selected Procedures, Fee Screen Year 1980 Locality prevailing charges Procedure High Low Range Ratio 1. Brief followup hospital visit by aninternist $ 33.10 $ 7.00 $ 26.10 4.73:1 •••••••••••••••• 0 2. Extraction of lens by anophthalmologist .................... 1,390.70 536.50 854.20 2.59:1 3. Electrosection of prostate by aurologist ........................... 1,410.40 475.25 935.15 2.97:1 4. Hysterectomy by an obstetrician/genecologist 1,305.20 536.50 768.70 2.43:1 ••••••••••••••• 0. 0 ••••• 5. Chest X-ray single view by a radiologist ••••••••••••••••• 0 35.00 5.50 29.50 6.36:1 SOURCES: U.S. Department of Health and Human Services, Health Care Financing Administration, "Medicare Part B Charges, Overview and Trends, Fee Screen Years, 1976-1980," Washington, DC, Feb. 3, 1982; and U.S. Congress, Senate Com mittee on Finance and House Committee on Ways and Means, Background Data on Physician Reimbursement Un der Medicare, S. Prt. 98-153 (Washington, DC: U.S. Government Printing Office, October 1983). Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 61 levels for a program supported by national taxes and beneficiary premiums that do not vary by jurisdiction, and 2) the implicit incentives in varying payment levels that make the Medicare program nonneutral with respect to physician location choice. (In particular, it is alleged that Medicare payments encourage new physicians to locate in urban areas while the explicit policy of the Department of Health and Human Services has been to encourage physicians to locate in "underserved," predominantly rural areas.) On the other hand, two arguments for varying payment levels involve: 1) variations in the costs of physician practices across jurisdictions·, and 2) variations in what may be market-determined prices for physician services across those areas. National Equity.-The Medicare Part B program is a national program. It is funded primarily (about 75 percent) through general revenues collected largely on the basis of ability to paynot State of residence. Beneficiary premiums for enrolling in Part B do not vary across the country, nor is there variation in the deductible that must be met prior to receiving reimbursements for approved charges. Part B enrollees are eligible for benefits regardless of their State of residence or the State in which they may receive physician services. Medicare payments on behalf of beneficiaries, however, do vary considerably. Variations in payment levels across States and within some States contribute to the variations in benefit pay ments both directly and indirectly. In particular, compared to beneficiaries who receive physician services in States with relatively high approved charges, those in States with low approved charges have to receive more physician services in order to meet the deductible and hence qualify for reim bursement. In effect, beneficiaries who are in some sense in equally poor health may not have equal financial access to care through the Medicare pro gram. Of course, beneficiaries in lower cost areas are likely to have lower cost-sharing for a given set of physician services. Location Incentives.-In general, Medicare allowed charges for physician services are highest in urban or suburban areas and lowest in rural parts of the country (471). However, it is exactly in those rural areas that beneficiaries may experience the most difficulty finding a source of medical care because physicians have not elected to establish practices in those locations. Recent evidence has suggested that increasingly fewer rural areas are without a specialist physician (344), but there is still enough of a perceived problem of unequal access to care that Federal policy remains committed to rectifying shortages in "underserved" areas. Therefore, the incentive effects of Medicare physician payment policies are in conflict with other Federal policies with respect to the encouragement of rural practice locations. The importance of this conflict, however, is far from clear. Based on 1981 revenue estimates, Medicare payments to physicians represent only 17 percent of all their gross professional revenues, ranging from a low of 1 percent for pediatricians to a high of 35 percent for thoracic surgeons (353). In this regard, one might want to design a system that was "location neutral" only to certain specialties, not necessarily including those specializing in tertiary care. However, to the extent that current levels of Medicare approved charges approximate those of the private market in individual localities for any specialty, spectacular increases in Medicare allowances would be required to reduce any aggregate location incentives due to differences in physician prices. Further, the empirical evidence shows that, other things being equal, the link between market-specific physician payment levels and location choice is weak (438). Thus, even if Medicare prices were adjusted to become location neutral, there would be little, if any, effect on local physician shortages. Differences in Practice Costs. -Although the possibility of varying cost levels provides a plausible argument for varying payment levels, the data on the degree of cost variation are equivocal. Average physician expenditures for practice inputs consistently have been the highest in the West South Central census division and lowest in the Middle Atlantic census division (390). In addition, average reported professional expenses have been highest in nonmetropolitan areas, and lowest in the largest of the metropolitan areas (390). Unfortunately, these gross differences in expenditures may mask both differences in practice volumes and differences in physicians' purchasing decisions as a result of their rational entre 62 • Payment for Physician Services: Strategies for Medicare preneurial responses to differences in price. For of allowed charges to submitted charges, by carexample, other things being equal, physicians who rier has been accepted as a measure of the degreepractice in an areas with relatively low commerto which Medicare payment levels match privatecial rent levels may choose to have larger offices market levels (215). In the first quarter of 1985,or more patient examining rooms. As a result, the range in this ratio across carriers went from those physicians' expenses for office space might a low of 66.2 percent in Rhode Island to a high be greater than, equal to, or less than those of their of 81.8 percent in Kentucky. For all but six carcounterparts in the higher rent districts. There are riers, the ratio of approved charges to billedno available data on the variations in the costs charges was between 70 and 80 percent. The naof operating physician practices of equivalent size, tional average was 73.8 percent (535). Reducing amenity levels, or style. the interstate variations in Medicare absolute ap The available evidence does suggest that curproved charge levels by paying the national averrent Medicare variations in payment levels exceed age would tend to increase the variation in Medi plausible differences in the costs of living, if not care's "comparability" to private market physician costs of an "equivalent" practice. As a result, some prices. In particular, it would reduce Medicare's consideration might be given during implemencomparability in such States as New York, Pennsylvania, Florida, Michigan, and Texas, all of tation of any Medicare physician payment reformto assessing the extent of some of the existing varwhich are currently below average in terms of theratio of approved charges to billed. The initial im iations to verify whether any remaining variationcan be justified. pact of such a policy might well be to reduce beneficiary access to care in States with above-aver Physician Opportunity Costs. -Although prac age ratios. One might argue, however, that if tice cost differences are important, differences in Medicare reduced its payment levels in States withphysicians' opportunity costs need to be considrelatively high physician prices, the private mar ered in establishing a Medicare payment policy ket would follow, thus bringing Medicare's chargesthat must also foster the goal of assuring benefiback into line and ameliorating any adverse imciary access to care. Various authors have found pact on physician participation in Medicare. that physicians' decisions with respect to Medicare program participation-either generally or Are Geographic Differentials Necessary?specifically on an assigned basis-are influenced Having examined the pros and cons with respectby the level of Medicare approved charges relato uniform payment levels, what can one contive to the physicians' billed charges (190,317,357). clude? First, a national fee determination procIf physicians' private pay patients (and their iness cannot be dismissed as a possibility for imsurers) are willing to pay relatively high fees reprovement. Providing uniform national benefitsgardless of-or even in excess of-differences in to Medicare enrollees is not an unreasonable goal.practice costs, constraining variations in MediFurther, the evidence on the correlation betweencare allowances to the levels of practice cost differpayment levels and the percentage of beneficiariesences may result-in high fee areas-in fewer phymeeting the deductible is consistent with effecsicians' either accepting Medicare patients or tively nonuniform insurance coverage under Mediaccepting assignment when they do see Medicare care Part n. (A Medicare beneficiary in Califorpatients. · nia need not be so sick as one in Oklahoma inorder to meet the deductible and hence qualifyIn fact, data from the Medicare carriers sugto receive additional reimbursements.) Eliminatgest that there is less variation in the degree to ing geographic differentials would certainly reduce which approved charges match private pay prices some of the administrative complexities of thethan in the absolute approved charges themselves. program, and fee schedules by jurisdiction could Physician submitted charges to Medicare have eliminate any confusion among beneficiaries orbeen found to be a good estimator of private marphysicians about what amounts Medicare willket prices even if most insurers rarely pay 100 perpay. (Fee schedules by jurisdiction could elimi cent of submitted charges. As a result, the ratio nate all variation in approved charges within a Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 63 jurisdiction.) To a certain extent, a national fee schedule might also make resistance to physician price increases easier for private market payers in relatively high cost States. Further, if there is significant competition between private insurers in the affected jurisdictions, the imposition of a national fee schedule would not necessarily contribute to market-wide price increases for physician services in the relatively low cost States even though the physicians in those States might increase submitted charges in response to the increase in Medicare payments. On the other hand, neither a national fee schedule nor even a set of 53 statewide fee schedules is a requirementY Although it is a national program, Medicare must operate in local markets across which the costs of operating a physician practice-including the opportunity costs of the physicians' own time-are not uniform. In fact, given that the market prices of specific physician services differ across the country, the provision (through those markets) of a uniform real level of benefits to Part B enrollees would require that different prices be paid in different jurisdictions. (In fact, a substantial factor in the origin of the "usual, customary, and prevailing" within Blue Shield Plans was the demand from national purchasers, such as the automobile manufacturers, for consistent paid-in-full benefits for members in all parts of the country (122,312).) Paying the same price for a particular service in all parts of the country would certainly imply large interregional transfers of funds within the Medicare program, and one would expect significant changes in beneficiaries' access to assigned services. What would be useful, if not required, would be an explicit effort to monitor the continued justification both for maintaining the level of differences in approved charges among jurisdictions and even for maintaining separate locality jurisdictions. Because program administration is eased and provider and beneficiary understanding can be improved when there are fewer rather than more localities in any State, reducing the number of localities to only those with reasonable justification is a plausible goal. There may be some negative spillover effects on assignment "Fifty States plus the District of Columbia, Guam, and Puerto Rico. rates concomitant with locality consolidation that should be weighed in advance against anticipated benefits.17 Specialty Variations One other source of variation in approved charges per unit of service is the use of specialtyspecific groupings of physician practices in implementing the CPR fee determination process. In the 1984 fee screen year, Medicare carriers established prevailing charge limits by specialty in all areas of the country except Florida, the area of Kansas served by Blue Shield of Kansas, North Dakota, South Dakota, and the area of New York served by Blue Shield of Western New York (471). All of the other carriers have established that there may be some services for which approved charges may be influenced by the specialty of the physician who performed the service. This approach may take the form of two prevailing charge screens, one for "generalists" and the other for "specialists." Alternatively, separate prevailing charge screens may be established for each of several sets of specialties. South Carolina, for example, has 33 prevailing charge screens, and Pennsylvania has 58 different groups. Although each physician's customary charge for a particular procedure is established solely with respect to his or her own submitted charges for that procedure, in jurisdictions that recognize more than two specialty distinctions, two or more specialty-specific prevailing charges might be established for the procedure. As a result, two physicians of different spedalties with identical customary charges might have different approved charges for the same procedure. In fact, because of physician specialization, most of the roughly 7,000 physician procedures will have specialty-specific prevailing charges whether or not the carrier in question recognizes specialty distinctions. For example, relatively few cataract operations are performed by physicians who are not ophthalmologists. As a result, the "Colorado consolidated its localites into a single locality in 1976. Assignment rates declined in each area of the State following the consolidation, consistent with the declining trend in assignment rates observed at that time in all parts of the country. Assignment rates declined the most, however, in those areas where prevailing charges were reduced as a result of the consolidation (394). 64 • Payment for Physician Services: Strategies for Medicare distribution of ophthalmologists' customary charges physicians in the low charge specialty that werewill determine the prevailing charge for this type above the 75th percentile in that specialty becauseof procedure even where all physicians' charges all other physicians in that specialty had been unfor the procedure are combined to determine a affected by the initial prevailing charge and wouldsingle prevailing charge. remain unaffected by the new one. If the servicevolume of the lower charge specialty were insig Specialty-specific prevailing charge screens that nificant compared to the other, the partition may make a difference involve those services that would have little effect on the higher volume, are performed by physicians of many different higher charge specialty, while reducing the ap specialties. The most prominent of such services proved charges of only those physicians in the low are visits (which account for 57 percent of all phy volume, low charge specialty which were above sician services provided to Medicare beneficiaries the 75th percentile in that specialty. Finally, if the and 33 percent of total approved charges (247)). service volume of the higher charge specialty were Among the prevailing charges in fee screen year insignificant compared to the other, the partition 1982 for a selection of 16 specific types of office might have a slightly negative effect on the ap or hospital visits, differences between general proved charges of the higher volume, lower practitioners and internists were observed of up charge specialty above the 75th percentile in that to 53 percent (494). Further, the prevailing charge group, while raising the approved charges of most for the general practitioners was lower than that of the physicians in the low volume, high charge of the internists in 15 of the 16 cases. specialty.Effects of Maintaining Separate Specialty ChargeScreens.-Under the CPR system, however, the For the most part, maintaining separate prevailprevailing charges set only a maximum on the aping screens for different specialties permits higherproved charge for a particular set of physicians; approved charges for the highest priced of theeach physician's customary charge also establishes higher priced specialties and reduces the approveda unique limit that may be the effective constraint charges of the highest priced of the lower pricedon the approved charge. Because the approved specialties. The effects on the beneficiaries ofcharge for a service from a particular physician maintaining separate specialty distinctions are notwill never exceed his or her customary charge for unequivocal. The out-of-pocket costs of a benethat service, the major effects of establishing sepficiary who receives service from one of the relaarate charge screens by specialty involve primarily tively low priced physicians in either of two spethose physicians in each specialty whose customcialties would be virtually unaffected by creatingary charges are high relative to their peers. The separate prevailing charges. The patient whoreactual effects of any specialty consolidation or ceives service from a high priced doctor in thepartition would depend on the relative volumes lower priced specialty will face reduced coinsurof service for the specialties in question and the ance but possibly a higher amount of nonassigneddegree of overlap in the distribution of customliability. Similarly, the patient who receives servary charges among those specialties. ice from a high priced doctor in the higher pricedspecialty will face increased coinsurance but a For example, if one ignores for the moment the somewhat lower level of unassigned liability. effects of the MEl, the results of creating two spe cialty screens where formerly there was one might Considerable attention has been given to com be as follows: If the service volumes of the two parisons of the prevailing charges of general prac specialties were comparable but the customary titioners and "specialists" due to the availability charges of one specialty were no higher than the of the Medicare prevailing charge directories 50th percentile of the other, the partition would (513). Although there is some concern that such raise the approved charges of only those physidifferentials may encourage specialization, therecians in the higher charge specialty whose customis no evidence that fee differentials in and of themary charges were above the 50th percentile. It selves have much influence on specialty choicewould lower the approved charges of only those (438), and no one has seriously suggested that the Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 65 relatively low fees paid by Medicare are solely responsible for the declining numbers of general practitioners. There is some question about the appropriateness of allowing individual physicians to declare themselves specialists and take advantage of higher prevailing charges (475). Recent analyses of the distributions of approved charges for individual procedures have found that, compared to surgeries, a much greater proportion of physician visits have approved charges equal to the prevailing charge (247,294). Where the MEl (or market competition) has compressed the distribution of approved charges within specialties, changes in specialty distinctions can have more dramatic effects. }uba estimated that if a fee schedule had been adopted based on average approved charges in South Carolina in 1983 that did not recognize specialty differentials, Medicare revenues for office visits for general practitioners would have increased 19.6 percent. Family practitioners would have observed an increase of 16.6 percent, and internists would have observed a decrease of 16.5 percent (247). These findings suggest a difference between fee schedules and CPR. Because of the presumption that fee schedule amounts will provide a limit for all physician payments-not just payments for the physicians with the highest fees relative to their peers, specialty distinctions may have more significant financial implications under a fee schedule than under CPR. Different Fees for Different Physicians.-There are no data with respect to the number of distinct specialties that have billed Medicare carriers for specific physician procedures. Office visits and hospital visits-which account for 33 percent of Medicare approved charges (247), however, are provided and billed by most of the medical specialties and subspecialties. It is commonly accepted that most surgeries are primarily specialty specific, but here, too, there may be instances where some fraction of particular surgeries may be performed by physicians outside of the spe cialty considered most likely or most appropri ate to perform that procedure ("modal" special ists). How does one determine the "right price" in these instances, and should that price be the same as is paid to the modal specialists? The common justifications for recogmzmg higher approved charges for specialists compared to general practitioners involved either higher quality or qualitatively different services provided by specialists even though the procedures (such as visits) may have the same label. Office visits of internists, for example, have been found to be 46 percent longer than visits to general and family practitioners (548). Although physician time is important, time alone may not fully describe the differences in professional effort that may be involved or the resources of knowledge and skill that may be brought to bear by the physicians in question. In order to account for such differences between physician services, various observers have introduced the concept of skill, complexity, urgency, intensity, stress, and severity. Although the concepts differ from one another, they are all interrelated with respect to the utilization of physicians' personal resources. Basic skills involve the clinical judgment needed to diagnose and choose appropriate therapeutic procedures. Complexity reflects the technical skills needed to perform the procedure. A patient's severity and the urgency of his or her medical situation will influence both the intensity of the physical or mental effort required of the physician and the stress due to the potential risk of the procedure in question. Previous studies have found a fair degree of consensus among physicians with respect to these types of complexity rankings across individual physician services (225,226,227,422). There is no empirical literature on whether such differences are evident with respect to a set of specific procedures performed by physicians of different specialties. Physician Opportunity Costs.-One might argue that physicians in different specialties elect to invest different amounts of time in specialty training, and that payment differentials should merely reflect such differences. Various authors have used a "returns to training" adjustment to account for differences in physicians' incomes and, notably, differences between the costs of various physician services (227)Y 181£ anything, these studies of income differences have tended to suggest that payment levels to specialists more than compensate those physicians for their additional investments in training (72,113). 66 • Payment for Physician Services: Strategies for Medicare When applied to specific physician services, however, this argument involves a potential dou ble adjustment for differences across specialties in physician opportunity costs. The problem is as follows: Although physicians do make an in vestment of time and money in obtaining specialty training beyond the "intern" level, part of there turn on that investment is the "specific training" (31) skills that allow the performance of relatively complex-and more highly paid-services. In the ory, services that involve primarily the "general training" skills that all physicians acquire will not warrant additional payment. If, for example, a gastroenterologist perceives the opportunity costs of the professional time devoted to an office visit in terms of the payments available for perform ing an endoscopy, it may be rational for that phy sician to bill accordingly. However, unless it can be shown that beneficiaries' access to compara ble specialists' services suffers or that the Medi care program can make operational a valid option dernand19 for that physician's more specialized skills, it may not be rational for Medicare to pay higher approved charges for that visit. Practice Status.-Because the arguments for the use of board certification as a basis for payment differentials are essentially a refinement of the general specialty differential arguments, these arguments will not be repeated in this section. The arguments with respect to higher payment levels for teaching physicians do involve a different perspective. In particular, teaching physicians may provide an adjunct service-teaching of new physicians-at the same time that they provide strictly medical services to Medicare beneficiaries. In addition, some may argue that because the opportunity costs of a teaching practice are high, higher payments than otherwise available for comparable services will be necessary to retain highly qualified physicians in the role of teachers. With respect to the first argument, Medicare has recognized a share in hospitals' direct and indirect education and training expenses of health professionals even under the prospective payment system. This situation might tend to legitimize the argument for higher payments for such physi 19An option demand would involve a payment for a service that although it may not be used by the purchaser is valued for its existence as an option. dans. Alternatively, one can argue that Medicare or other governmental contributions for such expenses should be made explicitly. Payments ernbodied in allowed charge differentials for teaching physicians could contribute to inequitable variations in beneficiary liability just as much as any locality or specialty differential. With respect to the question of the opportunity costs of teaching, one might want to examine evidence that the quality of the teaching staffs in the country have suffered due to relatively low payments available under Medicare before proceeding to raise payments in that regard. Variations by Site of Service Comparable though not necessarily identical services may be observed to have both differing customary charges and prevailing charges for a single physician practice depending on the site of service. In the HCFA Common Procedure Coding System (HCPCS) and most other physician service taxonomies, different procedure codes are assigned to physician encounters-visits-according to where they occur. Thus, one can have a lirnited20 (subsequent) visit in the physician's office (CPT-4 code 90050); in the patient's horne (90150); in the hospital (90250); in a skilled nursing facility, intermediate care facility, or other long-term care facility (90350); in a nursing horne, boarding horne, domiciliary, or other custodial care facility (90450); in an emergency department (90550); or in a critical care unit (99172). (A limited visit may also be provided as a consultation (90641) or for the purpose of issuing a second opinion (90650) without regard to site (85).) Average prevailing charges across the country (for fee screen year 1982) exhibited the pattern that for a given category of visit (such as limited or intermediate) a hospital visit commanded a higher allowed charge than a nursing horne visit, which in turn was higher than an office visit (494). Average prevailing charges in that year ranged from 11 to 32 percent higher for the inpatient visits compared to office visits. If one assumed that the medical content of the visits was comparable, this pattern implies an incentive to favor the hospital 20The definition of a "limited" service is provided in footnote 11 to this chapter. Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 67 as a site of service where additional physician practice costs, if any, for hospital visits were less than 11 percent higher than comparable office costs. (One might also note that the regulations promulgated with respect to physician services performed in an ambulatory surgical center provide for physicians to be paid 100 percent of approved charges on those services whether provided in an ambulatory surgery center or hospital outpatient department, if they accept assignment. Although accepting assignment in this case may lead to a reduction in a physician's bad debts, that physician's total expected revenues may still be greater when assignment is not accepted for those services. Further, whether the physician is paid more for ambulatory surgical services than for the same services provided to inpatients depends on the relevant array of customary and prevailing charges. A physician's total expected revenue may remain higher for hospitalized patients.) The arguments with respect to site differentials revolve around two questions. The first involves the issue of whether existing differentials inappropriately influence the site of care, particularly when in-hospital payments exceed those for services that might otherwise be provided in a physician's office. Second, for services provided outside a physician's office, there is the perception that some of the practice costs are not paid by that physician, and hence payment to the physician should be lower. For the most part, the first issue arises for separately billed physician visits, not for surgeries or interpretations. Prevailing charges for office visits have been shown to be lower than those for hospital visits of ostensibly the same variety (494). With respect to the second issue, it is argued that since physician's overhead costs account for roughly 40 percent of gross professional revenues, payments for services provided in outpatient departments, for example, should be limited to 60 percent of payments allowed for comparable services provided in the physician's office. The first issue regarding payments for inpatient visits may be a case where the nomenclature of physician services may be misleading. A limited hospital visit may be very different from a limited office visit even though both are described as limited visits. On the whole, patients seen in hospitals are sicker than those who are ambulatory. And on average their verified medical complaints may require more physician attention than the reported symptoms of their ambulatory counterparts. If there are some circumstances for which physician hospital visits are warranted to be cursory, that situation may argue for a single, per admission hospital care payment rather than daily visit payments, not necessarily for reducing daily payment rates to the level of office visit payments. With respect to both issues, differentials in payment may be compared to differentials in costs, including both variable costs with respect to the site of treatment and fixed costs of the physician's office or primary place of practice. Although physicians do have the use of highly qualified technical personnel in outpatient departments or emergency rooms, those persons, for the most part, are not substituting for similarly trained individuals in the physician's office. Physicians' costs for providing such services outside of the office may be lower than they would be otherwise, but most physicians bill only for a professional component for such care; they do not bill for the cost of services provided by the institution. Most physician's office employees are bookkeepers, receptionists, or secretaries. In effect, their compensation is a fixed cost to the physician that is unaffected by the amount of professional time spent on practice outside the office. Variable costs with respect to site may be limited to drugs and supplies, which represent 4 percent of physicians' gross revenues (117). Differences Among Procedural and Nonprocedural Services In addition to the obvious issue about the justification for establishing different approved charges for what appear to be identical services, there have also been questions raised about the appropriateness of apparently large differences in relative approved charges for different services. In particular, there is some concern that "procedural" services are overvalued compared to "nonprocedural" services. One HCFA study found that even after adjusting for differences in complexity, physicians were reimbursed as much as four 68 • Payment for Physician Services: Strategies for Medicare to five times more per hour for inpatient surgery than for office visits (227). Even within the office setting, the lack of additional reimbursement for such primary care services as history taking or nutritional counseling provided during a visit is in sharp contrast to the additional fees that can be generated by ordering and/or interpreting an electrocardiogram (EKG), performing an endoscopy, or providing laboratory tests. To the extent that physicians respond to relatively lower reimbursements for nonprocedural services, fewer of these services will be provided to Medicare beneficiaries with a possible increase in the subsequent demand for more expensive curative or ameliorative services. To the extent that net revenues from procedural services exceed those of the nonprocedural services, there may be a financial incentive to provide more of such services than would be appropriate on strictly clinical grounds. There has been a great deal of recent interest in identifying whether the extent of the differences observed between payments for procedural services and those for less technical nonprocedural services are warranted (16,17,103,136,195). There are a variety of reasons why the actual payment rates for specific services might differ from one another on either an absolute basis or as expressed in payments per unit of time. Such differences in payment may be due to differences in patient characteristics including health status differences, differences in the physical and mental demands on the physician occasioned by the service and/or circumstances in question, and differences in the length of training invested by individual physicians. The question remains whether the present physician payment systems-in which Medicare is only a subsystem-overcompensate for some of those differences. In particular, since the beneficiaries of the perceived overcompensation are also physicians in specialties that have relatively high estimated net incomes, namely, surgery, there is an issue of whether these perceived imbalances should be redressed concomitant with the initial implementation of any physician payment reform.21 Were overcompensation to be verified, the time of conversion to a new payment system might be 21A specific proposed remedy is the development of a resource cost-based relative value scale, which is reviewed in ch. 5. an opportune one. Any major modification of Medicare's physician payment system is likely to embody some years of conversion, much as the recent implementation of the prospective payment system for hospital payments. If there is a problem that needs correcting, delay until after the conversion might simply make a subsequent correction that much harder to implement. The question remains, however, how to identify whether there is a problem. The arguments and evidence on procedural! nonprocedural imbalances are as follows: physician payments for nonprocedural services, i.e., visits, are low compared to surgeries in terms of payment per unit of time spent with patients (227); specialties in which the bulk of practice involves procedural services receive higher net incomes than those specialties more heavily concentrated in nonprocedural services (35422 ); estimated rates of return to training are higher in medicine than in other learned professions and within medicine, higher in those specialties in which the procedural services are concentrated (72,113); patients' health would be improved if they received more primary/preventive/nonprocedural services, which in turn would be more available if those services were paid higher fees (336). Although the price and income differences may be evidence of imbalances, it is not clear that those differences alone are evidence of a problem, much less a problem to be redressed by Medicare. In a market economy, one would expect periodic imbalances between supply and demand and reductions over time in those imbalances as physicians, in this case, responded to just those market signals that are being produced. In a simple world, one would expect that more physicians in training would enter the surgical specialties and that more students in general would enter the profession of medicine because the returns to medicine exceed those of other learned professions. Longrun trends in medicine are consistent with such "corrections." There are more physicians per capita, but fewer physicians not pursuing a specialty. 22The author reports an increase in internists' net incomes relative to changes in the Consumer Price Index, and states, 'This suggests that [internists] have begun to succeed in their long-standing battle to reduce the third-party reimbursement gap between cognitive and procedural services." Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 69 Many observers would be unwilling to wait for the long run to arrive. They would assert that although some differences in payment are expected, the observed differences represent actual discrepancies in payment policy that are either not right in and of themselves or not right in that these payment differences lead to incentives that may inappropriately influence medical decisionmaking. With respect to the "correctness" of fee (or income) differences, there is no consensus. For example, in fee screen year 1984, the average prevailing charge for a cataract extraction was $981.77, nearly 50 times higher than the average prevailing charge for a limited office visit (532). During that same year, the median income of ophthalmologists was estimated to be $150,000, compared to a median income of $89;660 for internists (354). The face validity (or lack of validity) of such payment/income differences, however, rests primarily in the eye of the beholder. Whether ophthalmologists or internists or both are paid too much or too little is an open question. There is potential consensus with respect to whether such payment differences either inappropriately influence medical decisionmaking or threaten beneficiaries' access to care. If the relative approved charges of procedural services were so high as to lead to the provision of services of zero or negative benefits to patients, many physicians would agree that those prices were too high.23 On the other hand, if the relative approved charges of nonprocedural services were so low that physicians providing such services refused to see Medicare patients and if, as a result, those patients' health deteriorated, many would agree that those prices were too low. Verifying either of these states of the world in the current state, however, has proved elusive. There is a host of literature on variations in the use of hospital services and individual surgical services (272). None of these studies has identified a correlation between levels of use and levels of fees. Various observers have claimed to identify specific surgeries that may have been provided "Economists would argue that the price of a particular procedure was too high if services were provided at a price that exceeded the value of the change in health status expected to result from a par ticular procedure. in excess, but there is no indication that this surgical excess has been associated with excessive reimbursement rates. At the same time, there are no studies indicating that any particular groups of Medicare patients have not had access to needed health care services due to low reimbursement rates. Further, recent empirical evidence on the lack of dramatic effects for those who have forgone primary/preventive care (343,348) suggests that the health improvement argument for raising nonprocedural fees may be overstated. Where additional arguments might be made and where sufficient evidence may yet be developed involves differences in beneficiary access to specific types of health care services in terms of differential out-of-pocket liabilities with respect to different types of physician services. There is some evidence that reasonable charge reductions by carriers are relatively higher for visits than for surgeries (247,294), that assignment rates prior to 1984 were somewhat lower for primary care specialties than for surgical ones (247), and that beneficiary out-of-pocket expenses, if collected, were a larger part of total Medicare billings by the primary care doctors than for surgeons and radiologists (247). This situation may suggest that it is harder for beneficiaries to secure nonprocedural services. New vs. Old Finally, one other pattern observed among approved charges is that services of newer vintage or those that are provided by physicians of newer vintage have higher approved charges than those of older vintages. Specifically, new physician practices appear to have higher customary charges than more established ones, and newly introduced physician procedures have higher customary and prevailing charges than those procedures that have been commonly accepted for a longer time. With respect to a carrier's assessing claims from new physicians, there is no claims experience from which to compute a customary charge. Carrier rules have therefore been established to assign a customary charge in such cases by default. This default customary charge is equal to the 50th percentile of the distribution of comparable customary charges for the procedure in the relevent locality. As a result, new physicians can have ap 70 • Payment for Physician Services: Strategies for Medicare proved charges that, by definition, may exceed those of half of their more established colleagues. Prior to the freeze on submitted charges imposed by the Deficit Reduction Act of 1984 those more established colleagues were not prohibited from raising their own charges to retain a relative allowed charge position more in keeping with their experience.24 The question of the appropriateness of relatively higher approved charges for "new" procedures is more subtle. Very often a new procedure will require the acquisition of new skills or new equipment. The extra care required to execute the new procedure may require more time or place more stress on the physician performing the procedure for the first time. The relative value of the physician services involved in performing the procedure may be relatively high, and initial approved charges will reflect this. Although over time one would expect this relative value to decline as performance of the procedure becomes more routine, both the perceived relative value and the submitted charges of the procedure tend to become embedded in the structure of relative charges within a particular specialty. Since there is no periodic "zero-based" reevaluation of charges for specific procedures, the structure of approved charges simply drifts upward over time. (This is consistent with and may exacerbate the perceived imbalances in approved charges between procedural and nonprocedural services.) For example, coronary artery bypass surgery has been cited as a procedure that when first introduced required extraordinary expertise and enormous amounts of time. Initially, 3 or 4 procedures per week were a heavy workload, but today, some surgeons perform 3 to 4 procedures in 1 day. Furthermore, many of the surgeons' earlier tasks are now carried out by other professionals who bill independently from the surgeon. Surgeons' fees have not dropped but have increased more than the rate of inflation (403). Both cata 24A common misperception among physicians, however, was that Medicare "locked" them into a set of fee screens, over which sue~ cessively newer cohorts of physicians would leapfrog, leaving established physicians financially behind. Patients' expectations, if not their potential responsiveness to price changes may have inhibited physicians from raising charges as much as desired, but this outcome was not a function of Medicare regulations. ract surgery and blood chemistry tests, in particular, have also been cited as examples of this phenomenon ( 46). · Lack of Variations Due to Quantity Discounts Although Medicare reimbursements may account for only 17 percent of physicians' gross professional revenues, there may be some services for which Medicare revenues represent the bulk of all purchases. One might expect Medicare to get a better bargain in purchases of those services compared to physician services that are little used by Medicare beneficiaries. Table 2-12 indicates the proportion of specific inpatient services that are provided to elderly persons-who can be presumed to be Medicare patients. Although the reasonable charge reductions inherent in the cur- Table 2-12.-Eiderly Population's Share of Market for Selected Inpatient Surgical and Diagnostic and Therapeutic Procedures, 1983 Market share represented by population 65 and Procedure over (percent) Inpatient surgical procedures: Insertion of prosthetic lens .............. . 82.8% Extraction of lens ...................... . 79.5 Pacemakera ........................... . 79.4 Prostatectomy ......................... . 76.8 Arthroplasty and replacement of hip ...... . 74.2 Partial gastrectomy/resection of intestine .. 59.7 Dilation of urethra ...................... . 42.9 Mastectomy ........................... . 37.9 Direct heart revascularization ............ . 35.1 Open heart surgery ..................... . 32.0 Open reduction of fracture .............. . 31.4 Arthroplasty and replacement of knee .... . 27.9 Repair of inguinal hernia ................ . 27.5 Skin graft (except mouth or lip) .......... . 26.0 Inpatient diagnostic and therapeutic procedures: Endoscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.4% Radioisotope scan . . . . . . . . . . . . . . . . . . . . . . 47.0 Bronchoscopy . . . . . . . . . . . . . . . . . . . . . . . . . . 44.5 Computerized axial tomography (CAT) . . . . . 42.8 Esophagoscopy and gastroscopy. . . . . . . . . . 38.6 Arteriography and angiocardiography . . . . . . 38.3 Diagnostic ultrasound . . . . . . . . . . . . . . . . . . . 35.8 Pyelogramb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33.8 Cardiac catheterization . . . . . . . . . . . . . . . . . . 27.2 alncludes insertion, replacement, removal, and repair of pacemakers. bAn X-ray highlighting the kidney and urinary tract. Data source: Table 7, Advance Data, Sept. 28, 1984, No. 101. Vital and Health Statistics, National Center for Health Statistics, Department of Health and Human Services, Public Health Serivce. SOURCE: I. Burney and G. Schieber, "Medicare Physicians' Services: The Com· position of Spending and Assignment Rates," Health Care Financing Review, forthcoming. Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 71 rent CPR system may be considered to be a form of quantity discounting in and of themselves, one might expect that greater reductions would be observed for those services primarily provided to Medicare beneficiaries. In fact, this does not appear to be the case. In 1983 in South Carolina, for example, Medicare approved charges were roughly 75 percent of billed charges for all services, but for cataract extractions-the most common Medicare surgery in that State-approved charges were 90 percent of billed charges (294). Uneven Effects of the Medicare Economic Index (MEl) In 1972, in response to concerns that increases in physician fees under Part B were the cause of rather than the result of medical inflation, Congress mandated that an additional fee limit-an economic index-be included in the reasonable charge determination process. This index was to reflect changes in physicians' operating expenses and changes in general earnings levels and was to be used as a cap on prevailing charges.25 Prior to the imposition of the index, the maximum reasonable charge allowed by the carriers was equal to the "prevailing charge." The prevailing charge for any service was computed by the Part B carriers as the lowest customary charge that was no less than 75 percent of all customary charges when weighted by the volume of services billed. With the advent of the MEl, the value of the maximum reasonable charge was established as the "adjusted" prevailing charge, which was the lesser of: 1) the unadjusted prevailing, i.e., the 75th percentile; or 2) the product of the prevailing charge from fee screen year 1973 multiplied by the value of the MEl (117). Although some observers contend that the effect of the MEl has been to create de facto fee schedules, the actual effects are much less certain. The MEl has been constraining, and in the early 1980s, it appeared to be becoming more constrain "Inputs to the MEI are of two types, one reflecting increases in physician practice costs and the other reflecting increases in general earning levels. Of the first type, there are six practice costs measures: wages and salaries, office space, drugs and supplies, automobile expenses, professional liability insurance premiums, and all other practice expenses. General earnings levels measures included in the MEl are average weekly earnings of nonagricultural production and nonsupervisory workers and changes in productivity (117). ing over time. In fact, however, in the Medicare Directory ofPrevailing Charges the number of entries that indicate those prevailing charges that were due to the MEl declined each year from 1981 to 1984 (532). In addition, a study of fee screen year 1980 data from California for a selection of physician procedures found that the percentage of customary charges that might be directly affected by the MEl ranged from 24.5 percent of eye exams from ophthalmologists to 99.7 percent of basic anesthesiology services from anesthesiologists (187). Basically, this range goes from no effect to total effect. Further, an analysis of calendar year 1983 carrier data from the State of South Carolina showed that 43.2 percent of approved charges were established at the level of the adjusted prevailing charge (see table 2-13). Because the adjusted prevailing is the lower of the MEl cap or the actual 75th percentile, 43.2 percent must be considered an upper bound estimate of the impact of the MEl in that State (247). Finally, although some have alleged that the MEl has unfairly prevented reimbursements from rising in rural areas (415), California data show instances where in capping prevailing charges in urban areas, the MEl, in effect, prevented urban/rural disparities from increasing (359). Until recently, in performing the reasonable charge reduction process, carriers did not commonly record the specific limit-actual charge, customary, adjusted prevailing, or unadjusted prevailing-used to determine the approved charge for a specific claim. Because of this lack of data on the specific reasons for reasonable charge reductions and amounts of reductions by type of limit, there has been no definitive analysis of the impacts of the MEL Its inclusion in the reasonable charge process does make the process somewhat more cumbersome and potentially more confusing to providers, if not to the beneficiaries. Further, because by constraining some reimbursements but not others the MEl can lead to either increased or decreased payment differentials, the MEl also contributes to variations in payment levels across specialties and geographic areas. Summary of Variations The review of issues with respect to Medicare's physician payment system began with an indica 72 • Payment for Physician Services: Strategies for Medicare Table 2·13.-Distribution of Medicare Approved Charges Across CPR Limits by Specialty and Type of Service (South Carolina, 1983) CPR limit Billed Customary Prevailinga charge charge charge Otherb Specialty: All ........................ 15.7 38.7 43.2 2.3 General practice ............ 20.7 20.5 56.8 2.0 Family practice ............. 15.2 22.9 59.0 2.9 Internal medicine ........... 17.4 28.5 51.2 2.9 General surgery ............. 20.8 49.7 28.4 1.1 Orthopedic surgery .......... 13.1 38.1 48.1 0.7 Ophthalmology ............. 12.8 72.3 14.6 0.3 Radiology .................. 12.9 40.4 39.7 7.0 Type of service: Office visits ................ 12.4 19.7 65.2 2.8 Hospital visits .............. 11.4 22.4 64.6 1.7 Other medicine ............. 27.0 34.7 37.0 1.2 Surgery .................... 15.4 53.4 30.3 0.9 Radiology .................. 13.0 41.2 38.9 6.9 aAdjust~d prevailing charge. SOURCE: D. Juba, "Analysis of Issues Relating to Implementing a Medicare Physician Fee Schedule," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, November 1985. bAny amounts not equal to either the billed, customary, or prevailing limits. tion of beneficiary and provider confusion. In the Medicare is a national program with roughlysections that followed some of the other sources 30 million beneficiaries receiving physician serv of beneficiary and provider confusion were themices in thousands of communities in the United selves illustrated as issues. In the Medicare proStates and abroad. Some of the variability in the gram, one can observe variations in annual exprogram should be expected and much of the penditures, variations in the proportions of benvariability is desirable. eficiaries who meet the deductible, and variations What has not been included in the Medicarein assignment rates. In addition, approved charges program is an organized and timely review offor a particular service will vary by geographic Medicare's experiences to identify potential disarea, specialty of the physician, place of service, parities across the many dimensions of the protype of service, and "cohort" of either the servgram and to verify or refute the existence of suchice or of the physician performing the service. problems. Time and again one finds, 'There areThere are also variations in use of physician servno data." Although this may be taken to implyices across the country, and these variations have that there are no problems, in fact, it is more likelynot been found to correlate with variations in apto betoken the lack of solutions for the problemsproved charges. Finally, there do not appear to that do become evident. be variations in approved charges by quantity of service provided to Medicare beneficiaries, but there are unpredictable and uncertain variations in approved charges due to the MEl. THE CHANGING CONTEXT OF PHYSICIAN PAYMENT Changes in Medicare payment policies are beoutside of Medicare are likely to affect program ing discussed in a context that is itself in flux. From expenditures independently of changes in payment both the beneficiary and the provider sides of policies. The remainder of this chapter examines health care delivery, developments taking place the implications for future Medicare expenditures Ch. 2-Physician Payment Under the Medicare Program: Problems and Changing Context • 73 of changes in payment policies regarding hospitals; results from certain trends in the demographics of the elderly population, who make up 97 percent of Medicare beneficiaries (563); and developments with respect to the number and practice arrangements of physicians. Changes in Policies of Hospital Payment In October 1983, Medicare began paying for the operating costs of beneficiaries' inpatient care on the basis of DRGs. Until that time, Medicare reimbursed hospitals on the basis of the estimated costs that they incurred for Medicare patients. :This payment method encouraged the adoption and use of expensive technology rather than the efficient diagnosis and treatment of medical conditions. Beginning with the hospital payment reforms introduced in the Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA) (Public Law 97-248), the link between costs and Medicare payment levels was reduced. Under the new prospective payment system, Medicare pays a fixed amount based on diagnosis for the operating costs associated with inpatient admissions. Within each diagnostic category, the hospital has an incentive to use resources judiciously, including staff and equipment, and to reduce the length of stay. Incentives remain, however, to increase the number of admissions. During the first year of Medicare's DRG-based hospital payment system, lengths of stay for elderly people fell much more than the secular trend even though the prospective payment system was applied to a relatively small fraction of the hospitals in that year (489). Whereas the length of stay for people over age 65 had been falling by 1.9 percent per year, the length of stay during fiscal year 1984 fell 10.2 percent. Contrary to expectations, Medicare hospital admissions also declined during fiscal year 1984. The rate of increase in Part B expenditures fell substantially during the first year of the prospective payment system. During fiscal year 1984, Part B Medicare payments rose only 12 percent, in contrast to increases exceeding 19 percent in each of the 5 preceding fiscal years. This reduction is consistent with the likelihood that expenditures for physician hospital visits and consultations would be lower for patients with shorter lengths of stay. It is unlikely, however, that shorter lengths of stay accounted for all or even most of this reduction in Part B increases. Since hospital visits and consultations account for about 20 percent of Part B expenditures (68) and lengths of stay fell 10.2 percent, one might expect the growth in total Part B expenditures to have fallen about 2 percent because of DRG payment. Other factors, such as declines in price increases, may help to explain the overall decline. Changes in the Elderly Population The aging of the U.S. population is a long-term trend that is projected to continue into the next century. From 1970 to 1980, the cohort of people 65 years or older grew from 9.8 to 11.3 percent of the population. This cohort will account for 13.1 percent of the population in the year 2000 and 21.7 percent by 2050 (see table 2-14). Within the elderly population, the age structure is also changing. Those age 75 and older comprised 4.4 percent of the population in 1980, but will reach 6.5 percent by the year 2000. The growth of the elderly population stems mainly from previous changes in birth rates. Current increases in the 65 to 74 age group reflect higher birth rates after World War I. The size of this age group is projected to fall slightly by the year 2000 because of lower birth rates during the Depression and then to rise sharply as the baby boom of World War II reaches age 65 (498). Increases in life expectancy, although less important in explaining changes in the elderly population, have been substantial. Awoman of age 65 could expect about 17 more years of life in 1970, but will be likely to live almost 21 additional years in the year 2000 (see table 2-15). The ageadjusted death rate for people age 65 and older fell 22 percent from 1970 to 1982, with a much faster decline for women than for men (550,563). A pattern of higher use and expenditures can also be observed among the older age groups within the elderly population. As previously noted 74 • Payment for Physician Services: Strategies for Medicare Table 2·14.-Eiderly Population in the United States, Actual and Projected, by Age Cohort, 1970·20508 (numbers in thousands) 65 to 7 4 years 75 to 84 years ~85 years ~65 years Total population Year all ages Number Percent Number Percent Number Percent Number Percent 1970 203,302 12,447 6.1 6,124 3.0 1,409 0.7 19,980 9.8 1980 226,505 15,578 6.9 7,727 3.4 2,240 1.0 25,544 11.3 1990 249,731 18,054 7.2 10,284 4.1 3,461 1.4 31,799 12.7 2000 267,990 17,693 6.6 12,207 4.6 5,136 1.9 35,036 13.1 2010 283,141 20,279 7.2 12,172 4.3 6,818 2.4 39,269 13.9 2020 296,339 29,769 10.0 14,280 4.8 7,337 2.5 51,386 17.3 2030 304,339 34,416 11.3 21,128 6.9 8,801 2.9 64,345 21.1 2040 307,952 29,168 9.5 24,529 8.0 12,946 4.2 66,643 21.6 2050 308,856 30,022 9.7 20,976 6.8 16,063 5.2 67,061 21.7 aProjections are middle series. SOURCE: U.S. Department of Commerce, Bureau of the Census, Decennial Censuses of Population 1900-1980 and Projection of the Population of the United States: 1982 to 2050 (Advance Report), Current Population Reports, Series P-25, No. 922, October 1982; as cited in U.S. Congress, Senate Special Committee on Aging, and the American Association of Retired Persons, Aging America: Trends and Projections, (Washington, DC: U.S. Government Printing Office, Second Printing, 1984). Table 2·15.-Life Expectancy at Birth and Age 65, by Sex and Calendar Year Male Female Year At birth At age 65 At birth At age 65 1970 67.05 13.14 74.80 17.12 1980 69.85 14.02 77.53 18.35 1990 72.29 15.11 79.85 19.92 2000 73.42 15.71 81.05 20.81 2010 73.93 16.08 81.62 21.27 2020 74.42 16.45 82.18 21.73 SOURCE: U.S. Department of Health and Human Services, Social Security Ad· ministration, Office of the Actuary, September 1982, as cited in U.S. Congress, Senate Special Committee on Aging and the American As· sociation of Retired Persons, Aging America: Trends and Projections (Washington, DC: U.S. Government Printing Office, Second Printing, 1984). those aged 75 and older are more likely to have Medicare reimbursements and to have higher reimbursements per person served (see table 2-16). Older people have higher expenditures at least partly because they have higher death rates; people during the last year of life have had MedicIl,eumococc~l;,Vaccination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 ··:·~~J:rlical Lab~~~tory Testin8 ..· :·. . . . .. .. . . . . . . . . ........ ·88 (Jataract Surgery ...•. ; .. , . . . . . . . . . . . . . . . . . . . :;· 89 Magnetic Resonance Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Extracorpore(ll Shock Wave Lithotripsy ........... , . . . . . . . . . . . . . . . . . . . . . 93 ·+~·6~~· Table No. , Page 3-1. Medicare Payment and Beneficiary Liability for Cataract Surgery .:.>With lntra<;>~lar Le1;1s Implantation in Four Sites•.ofCare . . ·. I I I r,•. Chapter 3 Overview of Alternative Physician PaymentMethods Under Medicare:A Framework for Evaluation INTRODUCTION Because of shortcomings in the present system, This chapter briefly outlines the sets of paymentCongress and other policymakers are consideralternatives and variations among them. The secing alternative arrangements to pay for physician ond section of the chapter discusses the dimenservices under the Medicare program. This report sions across which the alternatives are examinedanalyzes four sets of alternatives: in chapters 4 (modifications to CPR), 5 (fee sched • modifications to the present system of payules), 6 (packaging), and 7 (capitation); and introduces certain issues common to several pay ment according to customary, prevailing, and reasonable (CPR) charges; ment alternatives. The chapter concludes with • payment based on a fee schedule, with prebrief descriptions of five medical technologies: determined payment rates that would be the pneumococcal vaccination, clinical laboratory same for similar services; testing, cataract surgery, magnetic resonance im • global payment for packages of related servaging, and extracorporeal shock wave lithotripsy. ices; and These technologies are used in subsequent chap • capitation payment, under which a predeterters to illustrate the effect of the various physimined amount would be paid for a benecian payment alternatives on specific technologies. ficiary's care during a time period. ALTERNATIVES FOR PAYMENT OF PHYSICIAN SERVICES The sets of payment alternatives considered in cian services" refers to services that are commonlythis report vary according to the unit by which provided by physicians but are sometimes promedical care is paid. Modifications in the CPR vided by other professionals or organizations. Ansystem and payment based on a fee schedule example is clinical laboratory tests, which maywould continue the service as the unit of payment; be performed in a physician's office, an independ the packaging approach would base payment on ent clinical laboratory, or a hospital laboratory.units that could range from ambulatory visits Similarly, optometrists provide some services, through therapeutic procedures to medical consuch as refraction and fitting of corrective lenses, ditions; and capitation (per capita) payment that may be provided by ophthalmologists. would pay according to the number of beneficiaries. The alternatives also vary according to the The alternatives discussed in this report conscope of medical services, the recipient of paycern how the Medicare program could pay for ment, and the methods of setting the payment physician services rather than how physicians relevel. ceive payment for their work. For example, Medicare might pay a health maintenance organizaThe core interest of this report is physician servtion (HMO) or hospital a per capita amount forices, although some alternatives include payments providing physician services to beneficiaries, butfor ancillary services and inpatient care. "Physi-the organization in turn could pay physicians on 81 82 • Payment for Physician Services: Strategies for Medicare a different basis, such as salary, fee-for-service, or some combination. Modifications in Payment Based on Customary, Prevailing, and Reasonable (CPR) Charges Under this set of alternatives, Medicare would continue to base its payment for physician services on reasonable charges calculated from the customary and prevailing charges billed by physicians and other providers (see app. C). The scope of services included in the payment and the recipients of payment would not change. What would change is the method of calculating reasonable rates. One set of changes would limit the rates paid for physician services, with the intention of reducing the increase in Medicare expenditures. Some of the options would apply to all physician services, such as lowering the percentile of prevailing charges used to determine al lowable charges. Medicare could also contract with preferred provider organizations to care for beneficiaries at discounted rates. Other options would apply more selectively to services with perceived payment imbalances relative to others. Rates for procedural services could be lowered relative to nonprocedural services, specialists and generalists could be paid the same rates for similar services, or geographic differences in rates could be reduced. Changing relative payment levels for certain services would also be possible under payment based on fee schedule. Payment Based on Fee Schedules Like the previous set of alternatives, payment by fee schedules would retain the scope of serv ices and recipients of payment of the present sys tem, but it would alter the method of calculating rates paid by the Medicare program. The sched ule of fees would be set in advance of the time period in which they were to apply, with similar rates set for services considered to be similar. Fees could be set on the basis of average charges billed in previous years, the cost of providing the serv ices, or rates negotiated with providers. Fee sched ules could also incorporate any changes desired in the relative prices paid for different services de pending on their location or content. Once developed, a fee schedule could be used in different ways. Medicare could treat the scheduled fee as the maximum allowable charge, but pay physicians a lower amount if they billed less. Or, Medicare could pay all physicians the applicable scheduled fee regardless of what was billed. In addition, Medicare could either require providers to accept the scheduled fee as full payment, or could pay beneficiaries the scheduled fee and permit providers to bill beneficiaries for additional amounts. Payment for Packages of Services This set of alternatives would package related services and pay for them as a unit. In comparison with the present system, changes could occur in the scope of services, recipient of payment, and method of calculating rates. Calculation of rates for most of the packages would require consideration of variation in resource use among patients and potential financial risk to the physician or other recipient of payment. The scope of services included in a package could range from a visit under an ambulatory visit package to all physician, ancillary, and possibly facility services under a total episode of care for a particular illness. Collapsing procedure codes would reduce the number of billing codes for services that have little distinction, such as 'brief" and "limited" office visits. The codes would be redefined as a single more comprehensive one (in this example, a short visit). A more diverse package is the ambulatory-visit package, in which an ambulatory visit to a physician and all ancillary services associated with that visit would be paid at a single rate. The rate could vary depending on the patient's diagnosis or reason for the visit. A third alternative, the special-procedure package, would pay a single rate for all physician services (including anesthesiologists and consultants) associated with a single procedure, such as cataract surgery. A variation of this alternative could include ancillaries and facility expenses for ambulatory procedures as well. Other possible variations of packages are packages for an ambulatory episode of care, an inpatient episode of care, or a total episode of care, which would include both ambulatory and in Ch. 3-0verview of Alternative Physician Payment Methods Under Medicare: A Framework for Evaluation • 83 patient services. Payment for inpatient physician services for the inpatient-episode-of-care package, for example, could be made according to diagnosis-related groups (DRGs) or other case-mix classifications. Such payments, which would include the services of attending physicians, anesthesiologists, and consultants, could be made to the attending physician, the medical staff, or the hospital. The goal of this set of alternatives is to contain Medicare expenditures by giving providers financial incentives for the more judicious use of resources, whether they be ancillary services, consultants, or facilities. The intention of payment for a package such as cataract surgery is that the attending physician consider cost more heavily that at present when ordering ancillaries, seeking consultations, or choosing the site for the surgery. Capitation Payment Under this set of alternatives, Medicare would pay a fixed amount set in advance and independent of the actual use of services for care to be provided beneficiaries during a certain time period. Although Medicare beneficiaries currently have the option of enrolling in HMOs paid on a capitation basis, capitation payment for all beneficiaries would entail changes in the recipient of payment, scope of services, and method of calculating rates. The recipient of capitation payment could be a risk-sharing plan, such as a traditional HMO. Alternatively, the payment could be made to geographic fiscal intermediaries, which would receive payments for all beneficiaries in that region. In both cases, beneficiaries would continue to have as one option continuation of present coverage, cost-sharing, and receipt of care from providers paid fees for services. This report considers two variations in the scope of services: 1) all acute and chronic care that lies outside of Medicare's payment system for hospitals facility expenses, and 2) all ambulatory and inpatient services. Calculation of capitation rates would require attention to differences in medical expenditures among beneficiaries and the potential financial risk to therecipients of payment. Payment of a fixed amount per beneficiary is intended to curb expenditures by giving providers a financial incentive to use the most cost-effective level and mix of medical professionals, sites of care, and other resources in managing patient care. Payment by capitation does not necessarily imply that individual physicians receive payment on a per capita basis, however. If the carrier received the capitation payment, for example, it could still pay physicians by fee for service or some other method. DIMENSIONS FOR EVALUATING PAYMENT ALTERNATIVES The payment alternatives discussed in this report are evaluated across five dimensions: • quality of care, • access to care, • cost, • technological change, and • administrative feasibility. These dimensions emanate from the goals of the Medicare program and concerns about its present shortcomings. The Medicare program was intended to help elderly and disabled people who needed assistance in meeting medical expenses (491). Concern with access to good quality care 56-119 0 -86 ·-4 : QL 3 was evident from the start of the program in requirements that providers had to meet in order to participate (487). Later amendments to the Social Security Act added utilization review and quality assurance, first by professional standards review organizations (Public Law 92-603) and later by utilization and quality control peer review organizations (Public Law 98-21). Cost is now a primary issue because current interest in reform of Medicare physician payment has been aroused by ever-rising program expenditures. These three dimensions-quality, access, and cost -are ones by which the medical care system in general and programs in particular are typically evaluated. 84 • Payment for Physician Services: Strategies for Medicare Technological change merits attention as a sepreports and commissions have concluded thatarate dimension because of the great influence that much medical technology has been used with inthe Medicare program has on this activity as part sufficient evidence of its efficacy. In addition,of Medicare's impact on the financing and delivsome technologies, such as diagnostic tests or hysery of medical care throughout the United States. terectomy, have been used when they provide litIn addition, administrative feasibility is examined tle or no additional health benefit and may evenseparately from cost and quality, to which it reharm the patient, while other technologies, suchlates, because it pertains to the ease with which as vaccinations and hypertension monitoring,changes in physician payment could be implecould greatly improve health if used more extenmented. sively (481,482). Studies to evaluate quality of care have often Quality of Care had difficulty measuring and evaluating outcome, Quality of care, given the existing state of medespecially the effect on the patient, because infor ical science and art, is the degree to which actions mation was not available or because a person's taken or not taken maximize the probability of health status depended on factors other than med beneficial health outcomes (health improvements) ical care. Patientoutcome is also difficult to evalu and minimize risk and other untoward outcomes. ate because it may change over time; whether a Health improvements include changes in the level patient's health status is judged to be improved of physical, psychological, and social functionmay depend on when it is measured. Therefore, ing (108). many studies have used the process of care, what a provider does for a patient, and the structure Quality is a multidimensional concept that deof care, the characteristics of facilities or propends on both technical and interpersonal aspects viders, as proxies to evaluate the quality of care. of medical care. Technical care entails the appliHowever, specific process measures, such as thecation of science and technology and encompasses use of a certain test, and structure measures, suchthe preventive, diagnostic, and therapeutic proas board certification of a specialist, are validcedures performed for a person's medical condiproxies only if they are associated with bettertion. Interpersonal aspects or the art of care con quality care.cerns the manner of the provider in delivering careand communicating with the patient (63,108). Access to Care Unlike recent definitions (108,194), this conception of quality does not include clinical efficiency, Access is the ease with which a beneficiary can that is, "the ability of the physician to arrive at obtain medical care. Access depends in part on a favorable solution to the patient's problem while the ability of people to overcome financial, spatial, psychological, or social obstacles to obtain consuming the minimum amount of resources necessary" (61). Consideration of what care is approcare. It also depends on the accessibility of themedical care system to people, which in turn de priate for a person's medical condition certainly entails weighing the implications for the use of pends on the characteristics of the organizations resources and their costs against the net health and individuals that provide care. benefits that are expected. However, this report Access is related to both quality and cost. Theconsiders net health benefits, costs, and efficiency ease with which people are able to obtain medias different concepts. The approach in this report cal care affects the kinds of services that they repermits examination of the multiple effects of a ceive and hence affects quality. The extent of papayment alternative and identification of the tient cost-sharing when services are performed istradeoffs that may be needed among cost containpart of financial access and directly affects the im ment and added health benefits. plications of a particular payment alternative forbeneficiaries' cost. This approach is also consistent with the presentsituation, in which deficiencies exist on clinical Despite its close relationship to these otheras well as efficiency grounds (483). Numerous dimensions, access is considered separately here Ch. 3-0verview of Alternative Physician Payment Methods Under Medicare: A Framework for Evaluation • 85 because of its importance to equity. Not only do physician payment alternatives vary in the financial and bureaucratic barriers to obtaining care that they present, but these barriers may well impede access for some groups, such as poor and frail people, more than others. Separate consideration of access will highlight any such problems for equitable treatment of beneficiaries. The extent to which physicians accept assignment for Medicare claims is closely related to patients' access to care. If a physician does not accept assignment for a service, he or she can bill that patient for an amount greater than the Medicare-determined allowed charge for the service. This additional amount could impede access to care for patients to whom it presents a financial barrier. However, as discussed in chapter 2, access and assignment are not synonomous. The relationship between the level of assignment and degree of access is not clear because it is not known whether the current rate of assignment represents a real barrier to many patients' ease in obtaining physician services. Nevertheless, it is reasonable to assume that an increase in assignment rates will improve access for at least some patients, and a decrease in assignment rates will reduce access. Thus, changes in assignment rates may be interpreted as changes in the accessibility of the medical care system, even though the current degree of accessibility has not been quantified. Costs and Efficiency In subsequent chapters, the implications of physician payment alternatives for medical care expenditures are considered from several perspectives. One is that of the Medicare program. As documented in chapter 2, expenditures on physician services have been rising by as much as 20 percent per year, a particularly disturbing trend in times of growing budget deficits and a particularly noticeable one in light of recent declines in the growth of hospital expenditures. Another perspective is that of Medicare beneficiaries. The financial implications of a physician payment alternative for beneficiaries may well differ in magnitude and direction from the implications for the Medicare program. Beneficiaries' costs now consist of premium payments for Part B coverage, a deductible amount, coinsurance for certain assigned services (see app. C), and, if the physician's charge exceeds the Medicare approved charge, any balance that the physician bills the patient for unassigned services. Under the current system, in which fees are paid for services performed and physicians have the option of taking assignment, a beneficiary's out-of-pocket costs depend on the volume of services used, the prices charged, and physicians' decisions about accepting assignment and billing beneficiaries above the approved charges. Costs may also be considered from the perspective of society as a whole. It is possible that Medicare's payment policies or physicians' billing and practice patterns may shift costs from Medicare or beneficiaries to (or from) other payers, such as employers who buy health insurance or State and local governments that are responsible for the medical care of indigent people. Of course, expenditures for medical care constitute income from the perspective of physicians and other medical providers. The payment alternative chosen affects whether these groups gain or lose income compared to the present situation. The level of costs matters to policymakers who are concerned about Medicare's budget and to beneficiaries who are living on fixed incomes. An issue in addition to the level of costs, however, is the efficiency with which resources are used to deliver medical care. There are two types of efficiency. Productive efficiency describes the performance of a service or delivery of medical care of a given quality with the least expenditure of resources. Allocative efficiency concerns not only whether care is provided as cheaply as possible given its quality and quantity, but also whether the costs expended for the additional care are worth the expected benefits to be gained. Efficiency rather than the level of costs addresses whether resources are being used appropriately in medical care or whether more benefit could be gained from applying them to different uses in medicine or elsewhere in society. A major shortcoming of present physician payment has been the lack of cost consciousness and financial incentives for efficiency among providers, patients, and payers (129,367). As a result, indi 86 • Payment for Physician Services: Strategies for Medicare vidual services have often been performed inefficiently, such as using unnecessary consultants and assistants for surgical cases and ordering duplicative laboratory tests to diagnose myocardial infarctions. Inefficiency also exists in the treatment of medical conditions with an inefficient mix of services, such as performing surgery for a cardiac condition that could be treated medically with a better or equal outcome or treating a case of influenza that could have been prevented with prior immunization. Technological Change Since Medicare pays 17 percent of the income of physicians as a group and as much as 35 percent for some specialties, such as thoracic surgery (353), that care primarily for elderly people, how Medicare pays for physician services and associated medical care can exert substantial leverage over prices and uses of medical technologies throughout society. The adoption of Medicare's payment methods by other payers reinforces these direct effects. Through its influence on the market for medical care, Medicare in turn shapes the market for medical devices and other technologies and affects the direction and extent of medical innovation (487). Until recent changes in Federal and State payment for inpatients, payment policies encouraged manufacturers to develop and market sophisticated products that increased quality of care and that were directed to acute hospital care. Technological development has slighted cost-saving devices, since potential purchasers had little incentive to adopt them, and preventive and rehabilitative devices, which have been much less likely to be covered by Medicare and other insurance. With the greater payment limitations on inpatient care and clinical laboratories, market incentives are now fostering the development of devices for ambulatory settings, especially for physicians' offices. State certificate-of-need laws, which regulate the purchase of expensive equipment and construction of facilities, contain similar incentives since they have applied to hospitals and certain other facilities, such as dialysis centers, but rarely to physicians' offices. As of Aprill985, only 13 States and the District of Columbia had certificate-of-need laws that applied to some or all major equipment acquired by nonhospital ambulatory care facilities and one State (Maryland) required that costly technologies in all settings be licensed (11). 1 Administrative Feasibility Although all of the physician payment alternatives considered in this report are feasible to administer, they all require some changes in administration, especially for the Medicare contractor or carrier (see app. D). These changes range from different methods of determining Medicare's approved charges and different coding procedures to negotiating with providers and assuming financial risk for utilization. Consideration of these differences will highlight changes necessary to implement the alternatives. 'The 13 States are Colorado, Connecticut, Iowa, Missouri, Montana, New Hampshire, North Dakota, Rhode Island, Utah, Virginia, West Virginia, Wisconsin, and Wyoming (11). SPECIFIC MEDICAL TECHNOLOGIES FOR SUBSEQUENT ANALYSIS In order to provide some concrete examples of the way in which different physician payment methods might affect medical technologies, subsequent chapters on specific payment alternatives will consider the implications of payment reform on five technologies: • pneumococcal vaccination, • clinical laboratory testing, • cataract surgery, • magnetic resonance imaging (MRI), and • extracorporeal shock wave lithotripsy (ESWL). These five examples can illustrate potential payment effects on a diverse set of technologies. Pneumococcal vaccination is a preventive technology that is low in cost and underused by the Medicare population (485). Clinical laboratory testing Ch. 3-0verview of Alternative Physician Payment Methods Under Medicare: A Framework for Evaluation • 87 is a diagnostic technology, consisting of many high volume, relatively low-cost procedures. Cataract surgery is a well-established therapeutic technology also performed at high volume on Medicare patients. Finally, MRI and ESWL are diagnostic and therapeutic technologies, respectively, that are new, expensive to purchase, and undergoing rapid technological change. All five technologies can be provided in more than one setting: hospital outpatient departments, inpatient settings, freestanding ambulatory facilities, or physicians' offices. Thus, these examples can illustrate the ways in which alternative physician payment methods might affect the site of care. Pneumococcal Vaccination The vaccine to protect against pneumococcal pneumomia, which represents about 14 percent of all pneumonias, is the only preventive technology that is part of Medicare's benefits for all beneficiaries. Covered by Medicare since July 1, 1981 (Public Law 96-611), the vaccine is indicated for use among persons with certain chronic illnesses, who are at a higher than average risk of contracting pneumococcal infection. The Centers for Disease Control's Immunization Practices Advisory Committee also recommends that all older persons, particularly those over age 65, receive the vaccine even if they are otherwise healthy (386). Nevertheless, estimates based on vaccine sales and physician surveys suggest that only 10 to 25 per cent of elderly people have been vaccinated (397, 485). Pneumococcal vaccination is a relatively stable technology. Since its approval by the Food and Drug Administration (FDA) in 1977, the vaccine has undergone one major change: In 1983, FDA approved a vaccine with capsular polysaccharides of 23 of the 83 pneumococcal types, an increase from the previous vaccine with 14 types (545). The 23-valent vaccine provides coverage against types that cause 90 percent of pneumococcal bacteremia and is marketed by two manufacturers (485). A single injection probably provides effective coverage for at least 3 to 8 years in elderly adults (485). The vaccine is an inexpensive technology as well, though the average Medicare allowed charge is probably lower than the average physician's charge for administering a dose. The average charge per dose was estimated at $11.37 in 1978, $4.90 for the vaccine and $6.47 for the physician's fee to administer it (485). In 1983, the average price for the vaccine had actually decreased to $4.43. If physicians' charges had increased at the same rate as the Consumer Price Index over those 5 years (57. 9 percent), the average physician's injection fee would have increased to $10.22. Total average charge for the vaccine in 1983 was therefore approximately $14.65 per person. Medicare approved charges vary by carrier and geographic region; in 1985, the approved charges of four carriers ranged from $7.00 to $11.10 (397), which assumes a low approved charge for injection. One Florida Medicare carrier, for example, reimbursed for pneumococcal vaccination at a rate of $5.50 for the vaccine itself plus $2.00 for the injection fee (105). It has been estimated that pneumococcal vaccination for a person age 65 or older could provide on average an additional 0.5 day of healthy life for about $8.00, or a rate of about $6,000 to gain a year of healthy life (485). The cost to the Medicare program was higher, estimated at about $8,000 per year of healthy life gained because Medicare does not pay for the total medical expenditures of program beneficiaries and therefore reaps only part of the savings in treatment costs due to a reduction in pneumococcal pneumonia. Medicare pays 100 percent of the allowed charge for pneumococcal vaccination; beneficiaries are liable for neither deductible nor coinsurance. They are, however, liable for any charges in excess of the allowed charge if the physician does not accept assignment. Since pneumococcal vaccination is a Part B service, hospitals can bill Medicare for the vaccine separately from inpatient facility services, which are paid according to ORCs (485). The use of preventive technologies for adults, such as pneumococcal vaccination, has characteristically been low, even among the patients of physicians who support their use (363). Neither adults nor the clinicians who care for them have been attuned to prevention in the way that parents and pediatricians have been for children. Although the extent to which financial incentives can affect physicians' decisions to use preventive technologies including vaccines is unknown, pneuma 88 • Payment for Physician Services: Strategies for Medicare coccal vaccination has faced special barriers. Uncertainty surrounded efficacy when the vaccine was first marketed in 1978. Although the Immunization Practices Advisory Committee strengthened its recommendations in 1984, the initial situation may have discouraged clinicians from recommending its use. In addition, people are unlikely to feel threatened by pneumococcal pneumonia because public awareness of the disease is low. Nor is it clear that clinicians perceive that elderly people are at higher risk from the disease (485). In September 1985, the Health Care Financing Administration awarded two demonstration projects to organizations that will offer packages of preventive services to Medicare beneficiaries and assess the cost-effectiveness of these services over a 6-year period. Payment for the package is limited to $100 per year. The package to be offered by the University of North Carolina includes both pneumococcal and influenza vaccinations (441). Clinical Laboratory Testing Clinical laboratory testing is of interest because it is an example of a technology that has low perunit but high aggregate costs (145) and may at times be overused or inappropriately used. In addition, it is a technology that is undergoing rapid and significant change. Laboratory tests are ordered by physicians for a wide variety of reasons. Test results may be used to assist in diagnosis, as with fecal tests to detect colon cancer; to establish clinical baseline values, as with tests of blood components; to monitor therapy, as with tests for drug levels in the blood that can indicate whether a patient is adhering to a prescribed drug regimen; or simply to reassure patients that a disease is absent or under control. An increase in "defensive medicine" may also play a role in physicians' decisions to order tests (284). Total revenues for clinical testing services in the United States have been estimated at $20 billion, making it a highly important component of the health care market (159). Most clinical laboratories today are highly automated, and current technological trends are to make them more so. The increasing automation combined with smaller equipment and a variety of diagnostic test kits has made the performance of most routine tests practical for group practices and even for individual physicians' offices. Advances in biotechnology have supported rapid change in testing methods through the use of monoclonal antibodies and other technologies to enable rapid, simple, and accurate in vitro diagnostic testing (484), and more dramatic changes are imminent. Historically, most testing has been done in hospitals, and about half of it still is (159). Independent and reference laboratories perform about a quarter of clinical tests. The most significant change in site of testing, however, is the return toward testing in physicians' offices, which accounts for the remaining 25 percent of clinical laboratory tests. Approximately 50 to 60 percent of all office-based physicians conduct some clinical laboratory tests in their offices, drawing approximately $5 billion in clinical testing revenues (159). Many of these physicians are in group practices, a target market for new technologies such as a recently developed blood analyzer (114). Some tests, such as those that indicate the possible presence of colon cancer or diabetes, have even been developed for home use by patients. Payment for clinical laboratory testing has been as dynamic an area as changes in the technology. Before July 1984, physicians could bill Medicare for the laboratory services they ordered, regardless of whether the tests were actually performed in the physician's office or in an outside laboratory (332). If a physician's claim indicated that the test was performed in the physician's office, Medicare paid physicians 80 percent of the reasonable charge (less any beneficiary deductible) (487). If the test was performed outside the physician's office, Medicare would pay the physician laboratory's approved charge plus a $3 handling fee. The physician would then pay the laboratory. If the physician did not accept assignment, the beneficiary in either case would be liable for all physician charges above the Medicare reasonable charge. Thus, the total payment to the physician for the test could be considerably higher than the laboratory's charge. Under this system, the physician might reap a financial reward for ordering the test even though it was actually performed elsewhere. Ch. 3-0verview of Alternative Physician Payment Methods Under Medicare: A Framework for Evaluation • 89 fJnoro credtr: American College of Physicians, HEAL THSCOPE film series Some devices for cl.inical laboratory testing, such as this blood glucose monitor for diabetic patients, have been developed for home use. More recent changes in the law have eliminated this financial reward to physicians who act as intermediaries, increasing the incentives for physicians to perform tests themselves (487). The Deficit Reduction Act of 1984 (Public Law 98-369) prohibited physicians from billing for laboratory services unless they are performed in a physician's office. It also established Medicare maximum payment levels for laboratory services, for a 3-year period beginning in 1984, at a fixed percent of the prevailing fee levels for each service (60 percent for physicians' offices, independent laboratories, and hospital laboratories serving nonhospital patients; 62 percent for hospital laboratory services to hospital outpatients). These fee levels are adjusted annually according to the Consumer Price Index; the maximum increase in payments for laboratory services prpvided from July 1985 through June 1986 has been set at 4.1 percent ~51). In 1987, a national fee schedule, presumably based on a method other than prevailing charge levels, will be developed for tests performed in physicians' offices and independent laboratories (487). Hospital laboratories, however, will revert to cost-based payment (as before 1984) unless an alternative payment mechanism is devised. The Deficit Reduction Act also changed arrangements regarding assignment for tests in physician's offices and independent laboratories. Independent laboratories and hospital laboratories serving outpatients must accept assignment, but Medicare will pay 100 percent of the fee schedule, thereby waiving coinsurance and deductible requirements for tests in these settings. Physicians who conduct their own tests may choose to accept or decline assignment, but if they accept, Medicare will again pay 100 percent of the fee schedule, waiving coinsurance and deductible. If they decline assignment, of course, the beneficiary is liable for both the deductible and a coinsurance equal to 20 percent of the Medicare-approved rate for the tests, plus any excess about the fee schedule amount. If the physician does not actually perform the test, Medicare payment to the physician is limited to a maximum payment of $3 for specimen collection, handling, and test interpretation. Hospital laboratory services to nonhospital patients are considered to be identical to independent laboratory services, and assignment is mandatory. For services to the hospital's own outpatients, the hospital is constrained by its Medicare provider agreement to accept Medicare payment as payment in full, effectively mandating "assignment" in these cases as well. In both cases, Medicare pays 100 percent of the fee schedule rate, so no beneficiary deductible or coinsurance is necessary (88). Cataract Surgery As one of the most frequent surgical procedures performed on the elderly (69,468), the removal of cataracts-a clouding of the lens of the eyereceives considerable attention from the Medicare program. The practice of cataract surgery has undergone major changes in the past few years. Once a major hospital procedure that involved a long 90 • Payment for Physician Services: Strategies for Medicare stay and post-surgical VISion correction with heavy spectacles, cataract removal is now a delicate but streamlined procedure that is commonly performed on ambulatory patients (161). In about 85 percent of cases, it now also includes the implantation of a prosthetic intraocular lens (IOL) to replace the natural one extracted from the eye (385). By comparison, in 1980 fewer than half of cataract extractions included an implantable lens (385). Medicare is the foremost payer of cataract surgery; persons over 65, most of whom are covered by Medicare, account for nearly 83 percent of inpatient cataract extractions (69). Concern has been expressed that in a few cases, this procedure is performed unnecessarily in patients whose cata-· racts did not yet impede their everyday activities (479). Cataract surgery is a particularly interesting procedure because of the wide variety of settings in which it can be performed under Medicare. These include hospital inpatient settings, hospital outpatient departments, 2 ambulatory surgical centers (ASCs), and other ambulatory settings not certified by Medicare as ASCs. (These settings are often referred to for Medicare purposes as "physicians' offices," although they may look nothing like the traditional office of a physician in solo 'Hospital outpatient departments can choose if they wish to be certified and treated for payment purposes as ambulatory surgical centers (ASCs). However, once this choice is made the outpatient department is subject to all of the constraints and payment methods imposed on ASCs (47 FR 34082). practice.) Table 3-1 summarizes Medicare reimbursement for cataract surgery in various settings. Reimbursement for costs associated with cataract surgery fall into three categories. First are the facility costs, which include surgical equipment, routine medical supplies, and nonphysician staff. Second are the professional costs for ophthalmic surgeons and surgical assistants. Third is the cost of the IOL, which is reimbursed as a prosthetic device. In certified ambulatory settings, these three components are reimbursed separately under Part B. For hospital inpatients, the facility and IOL costs are reimbursed under the Part A DRG rate; only professional fees are reimbursed under Part B. In noncertified ambulatory settings, Medicare Part Bpays the approved portion of the physicians' professional charge and the charge for the IOL. Medicare will not make any additional payment for the technical (facility and equipment) charges of physicians performing cataract surgery in this setting.3 Medicare hospital payment incentives and utilization controls have encouraged the trend toward ambulatory rather than inpatient cataract surgery. Hospitals are now paid a fixed rate for all services associated with the procedure when it is performed on inpatients, giving hospitals an incentive in many cases to provide it to ambulatory patients instead, for whom costs in most 'Medicare will pay a technical fee to the physician only for certain services, such as radiology, that are "incident to" treatment and have high equipment costs (202). Table 3·1.-Medicare Payment and Beneficiary Liability for Cataract Surgery With Intraocular Lens (IOL) Implantation in Four Sites of Care Site of care Ambulatory Medicare payment Hospital inpatient Hospital outpatienta surgical center Other ambulatory Facility payment ......100% DRG rate 80% reasonable cost 100%b class 4 rate 0 Physician paymentc ....80% approved charge 100% approved charge 100% approved charge 80% approved charge Intraocular lens (IOL) payment ............Included in DRG rate 80% approved charge 80% approved charge 0 Beneficiary liabilityc ...20% physician's 20% facility cost + 20% IOL charge 20% physician's charge charge 20% IOL charge + 20% IOL charge aExcept for those hospital outpatient departments that have chosen to be certified as ambulatory surgical centers. b150% if intraocular lens implanted. Cif physician accepts assignment. If not, beneficiary is liable for all charges over Medicare's approved charge. SOURCE: 47 FA 34082; 47 FA 34099. Ch. 3-0verview of Alternative Physician Payment Methods Under Medicare: A Framework for Evaluation • 91 cases are still reimbursed as incurred. Inpatient cataract surgery is also being monitored by many utilization and quality control peer review organizations, the prospective payment system's utilization control mechanism, which is intended to prevent hospital admissions of low-risk cataract patients that would otherwise be profitable for the hospital. Unlike hospital outpatient departments, which are paid according to their costs, ASCs are paid according to a fixed rate schedule (47 FR 34082). The facility cost portion of the cataract surgery procedure in an ASC is reimbursed at a single percase (class 4)4 rate under Medicare Part B. Beneficiaries receiving cataract surgery services in this setting are subject to neither deductibles nor copayments. At present, Medicare physician payment incentives for cataract surgery also tend to reinforce the trend toward ambulatory surgery. Since beneficiaries who undergo cataract surgery in hospital outpatient or freestanding ambulatory surgical settings are not liable for any copayment (47 FR 34082), physicians who accept Medicare assignment have a more assured reimbursement if they perform the procedure in these settings. Beneficiaries probably pay the least when they undergo cataract surgery performed in certified ASCs by physicians accepting assignment; in this setting beneficiaries are liable only for a portion of the charge for the IOL. In all other settings, beneficiaries are responsible for the Medicare Part B deductible and at least a 2Q-percent coinsurance of the physician's charge (in hospital inpatient and noncertified ambulatory settings) or the facility and IOL costs (in non-ASC hospital outpatient sites). Magnetic Resonance Imaging (MRI) MRI has gained attention as a potentially powerful new tool to complement the current diagnostic imaging armamentarium. Two characteristics make it particularly attractive. First, it uses electromagnetic fields instead of ionizing radiation to 'Class 4 has the highest payment level of the four ASC rate categories. If the cataract extraction includes implantation of an IOL, the ASC received 150 percent of the class 4 rate. Photo credit: Georgetown Medical Bulletin Medicare payment incentives reinforce the trend for cataract surgery, shown here, to be performed in ambulatory settings. produce images, so it lacks the ionizing radiation dangers of traditional X-ray and X-ray computed tomography (CT) scanning. Second, MR images are not distorted by signals from bone, a problem with conventional X-rays. But the powerful magnets that make MRI a novel and promising technology come at great expense. The cost of MRI, the logistical problems involved in providing it, and the uncertainty about the scope of its future applications have acted to slow its diffusion (234,449). Although MRI holds tremendous potential to advance diagnostic science and to replace other riskier modalities, it is largely unclear what the clinical role of MR imaging will or ought to be (234). At present, there are special indications for MRI only for anatomic areas that have never been adequately imaged by conventional modalities. For example, MRI is the modality of choice for scans of the posterior fossa region of the skull and the cervical spine and is a promising modality for imaging the pelvis, where the absence of ionizing radiation is particularly important (234). In the near future, most clinicians are likely to view MRI as a complement rather than as a substitute for X-ray CT or other diagnostic technologies. It is possible for the use of MRI to skyrocket as its uses become better defined. The central nervous system, as the most heavily explored area to date, offers the greatest potential for extensive 92 • Payment for Physician Services: Strategies for Medicare MRI use. Some researchers already consider MRI the modality of choice for initial screening of suspected brain disease. MRI's well-documented ability to delineate the plaques of multiple sclerosis may lead to its use for nonspecific complaints, mostly for patients younger than most Medicare beneficiaries. If MR technology improves as expected in cardiac and tumor imaging, the potential for widespread applicability in a Medicare population would also grow. At present, however, much of the clinical experiences is anecdotal, not from controlled trials (234). MRI at this point in its development is a classic case of diagnostic methods' outstripping therapeutic options. Obtaining a definitive diagnosis may be a desirable outcome in itself, but therapeutic limitations may make it unlikely that diagnosis will change the course of a patient's illness. For cerebellar and brainstem infarctions, for example, which by virtue of their location in the posterior fossa offer indications for MRI, little can currently be done to alter the prognosis for most patients (234). MRI has the potential to be useful in certain diseases such as some tumors for which treatments have been more successful, and greater knowledge about disease processes may ultimately lead to therapeutic advances. However, because the value of MRI in altering therapy or improving quality of care has not been adequately studied, it is difficult to ascribe an appropriate position for MRI in the provision of good quality medical care (234). Another major source of uncertainty to MRI purchasers concerns technical developments ( 449). Prospective buyers must choose among MRI systems with different types of magnets (resistive, permanent, and superconducting) and different magnet strengths, and considerable debate surrounds the relative efficacy and cost effectiveness of the different systems. The costs of equipment and site preparation range from about $1.7 million to $2.4 million, depending on the type and field strength of the magnet (447). An additional complicating factor is that magnetic resonance is also used to perform MR spectroscopy, which indicates relative concentrations of different compounds in tissues or organs. MR spectroscopy requires high field strengths and, although it has great promise, it is still in a research phase and its clinical importance is unclear (449). Providers do not want to purchase an unnecessarily expensive imager, but neither do they want to purchase a (still costly) less expensive device that will be outmoded in a few years. Nonetheless, a variety of physicians, including radiologists, neurosurgeons, neurologists, and cardiologists, envisage MRI as an important future component of their practice and are learning to perform it (234). Governmental policies have most likely slowed the diffusion of MRI and affected its distribution. By the end of 1984, 4 years after MRI's introduction into the United States, 108 MRI units were installed in the United States, 39 percent in ambulatory settings (449). MRI diffusion has been occurring during a period when payment for inpatient services has been undergoing great change. Medicare's payment of operating expenses by DRGs has constrained its payments to hospitals and given hospitals a financial disincentive to use technologies such as MRI that are likely to increase the cost of caring for patients. Although capital expenses connected with the purchase and installation of equipment have continued to be paid on a cost-reimbursement basis, approaches are being developed to include capital in the prospective payment system. In addition, as mentioned above, State certificate-of-need laws for the most part apply to hospitals but not to ambulatory sites, such as physicians' offices or ambulatory diagnostic imaging centers. Since both payment and planning policies constrain hospitals much more than ambulatory settings, the predictable result is an increased tendency to install expensive new technologies such as MRI outside of hospitals. It is noteworthy that after a comparable period of diffusion in the United States, 18 percent of X-ray CT scanners v. 39 percent of MRI units were in nonhospital settings (449). Total charges for MRI scans, consisting of a technical (facility) fee and a professional (physician) fee, have ranged from $450 to $1,000 (234). There is virtually no Medicare experience in paying for MRI. HCFA has approved paying for the use of MRI for certain purposes only since November 22, 1985 (20), although a few Medicare carriers apparently chose to accept MRI claims before this date (234). At present, the use of MRI does not increase payment to ASCs or to hospi Ch. 3-0verview of Alternative Physician Payment Methods Under Medicare: A Framework for Evaluation • 93 tals using it to diagnose inpatients.5 HCFA is developing guidelines for carriers regarding paying physicians a technical as well as a professional fee for MRI performed in nonhospital settings (55). Eighty percent of the top 30 commercial insurance companies were paying for MRI services on a routine or case-by-case basis in January 1985 (234), but only 20 percent of the 70 Blue CrossBlue Shield plans were paying for MRI in July 1985 (210). Extracorporeal Shock Wave Lithotripsy (ESWL) Like MRI, ESWL is a new and expensive procedure that has excited considerable interest. Unlike MRI, its costs and applications are relatively simple to define. It has only recently been approved as a reimbursable procedure by Medicare (301), and most carriers do not yet have any experience paying for it. Its cost-saving potential, however, has made most payers-including Medicare-eager to include it as a covered service. ESWL uses shock waves produced outside the body to disintegrate kidney and other upper urinary stones, eliminating the need for traditional open surgery in most cases (18). The current model of the device used for ESWL is large and expensive to purchase and requires its own facility. Nevertheless, if used by enough patients (over 1,000 per year), it results in a per-patient treatment cost considerably lower than that for open surgery, primarily because it requires a very short hospital stay (18). Some centers even offer ESWL to ambulatory patients. Because of anticipated lower costs per treatment, ESWL promises to be a profitable technology for those hospitals that provide it, particularly if these cases are reimbursed at the same rate as open surgery. However, the high fixed costs of the extracorporeallithotripter (about $2 million for purchase and installation of the current model) make it less expensive than the alterna •At present, the use of MRI itself does not increase Medicare payment to a hospital, even though use of MRI for a patient may increase that hospital's costs. A possible alternative form of payment for MRI, which has been recommended in principle by the Prospective Payment Assessment Commission, is a budget-neutral DRG addon for cases in which MRI is used. tives only at high volumes of use. Because the number of kidney stone patients is limited, it is probable that more devices will become available than are justified strictly by the number of patients who would have undergone stone surgery otherwise. If this is the case, the eligibility criteria for ESWL might be expanded to include many patients with less serious stones in addition to those otherwise eligible for surgery, leading to an increase in demand for the service (431). In the future, the technology itself may be applied to patients with lower urinary stones and gallstones, but the present device is not approved by the Food and Drug Administration for these purposes (379). Medicare reimbursement for ESWL is similar in structure to that for surgery. Medicare's share of the capital costs of its purchase and installation are reimbursed at cost through Medicare Part A, though it is possible that these costs will be incorporated in the DRG rate in the future. The hospital's costs of operating the device and of caring for lithotripsy patients are reimbursed (also under Part A) at the rate of the applicable DRG (#323 or #324 if no adjunct surgical procedures are performed). Physicians' charges for performing the procedure, of course, are reimbursed under Part B. ESWL technology is undergoing rapid change. Although only one manufacturer, Donnier Systems, currently has approval from the Food and Drug Administration to market the device, a number of other companies are developing competetive devices. Medicare's per-case hospital payment system, which presently pays for ESWL at a DRG rate that is much lower than the rate for open surgery for kidney stones, makes these alternatives highly promising and has probably helped stimulate their development. Only a few hospitals can provide extracorporeallithotripsy; fewer than 60 devices will be in place in the United States by the end of 1985 (378). A few nonhospital ambu latory centers are providing ESWL, but it is not an approved procedure in ASCs, and Medicare will not pay for its facility-related costs in this setting. Other alternatives to open surgery besides ESWL are also expanding rapidly. Endoscopic procedures that can withdraw kidney stones through a narrow tract, rather than a large inci 94 • Payment for Physician Services: Strategies for Medicare sion, are proliferating simultaneously with ESWL. Like traditional surgery, these procedures require a surgical suite and require specialized endoscopic instruments costing up to $50,000 as well (3). A major issue at present is what and how physicians should be paid for ESWL. More specifically, payers are questioning whether physicians should be paid the same for performing ESWL as for the open surgery it replaces, since ESWL requires additional training on the part of practicing physicians but appears to take less time to perform (18,431). The few carriers thus far with any ESWL reimbursement experience are reimbursing the procedure at rates ranging from approximately $1,200 to $2,000, at or slightly lower than the surgical rate. In most cases, the rates were based on consultations with outside urology experts and negotiations with the respective lithotripsy centers (431). HCFA is developing guidelines to help carriers establish an approved charge for the service (431). For the most part, kidney stone surgery, like most other surgical services, is reimbursed as a package that includes some preoperative and postoperative care by the urologist. Under the present system, an effort to reimburse for lithotripsy at a lower rate might stimulate some "unbundling," or redefinition of the service that results in physicians' billing for the procedure separately from some of the preoperative or postoperative visits now included in a single bill. Conversely, if ESWL is reimbursed at the same rate as major surgery, the physicians who perform it will reap a considerable profit. The existence of ESWL in a few regional centers, if it continues, could result in some form of price level negotiations between carriers and urologists performing ESWL, regardless of the structure of physician payment (431). ,· ",~" ," ::" , Chapter 4 Mo'diticau()'~,S to Cu§torri8ry, P~evaUJng;,2and:Re,Qsontble· .· 1 'Charge Payment '' ~~ -<~ -~·~.· ,~,'::.~ :::::.:. .,, "··~ ~:~:,'<~>_'"· ',: ,;0::~,:-e· ..);)~~;•• past 1~ prologu~.,"(l -William Shakespe~·re, The Tempest li&U:#UJ\.1 Contents Page Introduction .......................... ··:-.., ......... .: ......................·. 97 Implications of Alternative,. Methods ofModifying CJ>R........................ ·' <98 C6htrollirtg' ~~pro~ed CR~rges for All'Services ..: . · . . . . . . . . : ...... : : .:',too Controlling Approved Charges for Selected Services . ; . . . . . . . . . . . . . . . . . . . . 103 Negotiated or Discounted Fees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Technological Change . . . . . . . . . . . . . . . . . . . . . . . . . . . ......... 114 Administrative Feasibility : : . . . . . . . . . . . . . . . . . . ,. . . . . • . ;.>.11.5 't0!1c;IUsion .. :<":,::,' ,·, ., ..... >-~',::~::,:~,:~ "· .' ........,. . . . . . . . . ."; ~ .-.' ~ . .':,,~::,'~,;i17 List of Tables.> Tqf.11e.Np. :•.<•):. . : : ,: ... ·.···. ... . .·· ,j~ ::,;.,•.'(. 4~1~ Methods' fi§£:Modifying:CPR Paymeht Iritended'tH·iCortttol N1ed1care . Expenditures or To Reduce Variati6n's in ReimbU:~sement for Services .... 4-2. Medicare Approved Charges and Assignment Rates for Physicia~s'.. Services, by Type and Place of Servi~e( 1981 ............... 104 4::3i Medicare Approveq Cgatge~, Perc~~t :QJstributiori; qf . . ·~ .•... •··· .: Approved:Ehcifges, •and:Assigrtmel):t<~~tes for PP.ysicians' ServJces, 'by Combirt~tions of Place and Type.of Service, 1981 .................. . 4-4. Medicare Weighted Mean Prevailing Charges for the Five Most Common Services, Specialists/Nonspecialist, Calendar Year 1982 ........ 108 :,:, ;,: , ~ ,"o i ,' t:~~~;;,,,;;~ ' ,,, \,',:;:-\,, Chapter 4 Modification to Customary, Prevailing, and Reasonable Charge Payment INTRODUCTION This chapter examines possible changes to Medicare's customary, prevailing, and reasonable (CPR) charge method1 of paying physicians that would continue the historical pattern under CPR of computing distinct charges for individual physicians. Most of the modifications are intended to constrain the rate of growth of expenditures for physician services.2 Some of the changes identified could also reduce the substantial variation in Medicare payment rates for selected services. As noted in chapter 2, the variations in Medicare payment rates for some services suggest possible inequities in the distribution of benefits and inefficiences in the program. From the start, Medicare's CPR payment system has included several features intended to limit program expenditures for physician services. One such feature is a restriction on the amount that Medicare pays for physician services. Medicare's approved charge3 for a physician's service is the lowest of the physician's billed (or actual) charge, the physician's customary charge, or the prevailing charge in a locality. Indeed, in fiscal year 1984, Medicare-determined approved charges were, on average, 24 percent lower than physicians' billed charges (69). A second feature of Medicare's original payment system intended to limit program expenditures is the requirement that beneficiaries assume responsibility for a portion of physicians' approved charges, namely, by paying a deductible and coinsurance. A third feature of Medicare's original payment system that has cost-containment attributes is assignment. Medicare expend 'The CPR method, the principal method that Medicare uses to pay physicians, is described in app. C along with other facets of Medicare's physician payment process. 'The chapter considers controlling Medicare's expenditures for physician services by controlling Medicare's payment to physicians and does not consider other means, such as revising beneficiary payments. 3Under Medicare, reasonable charges, approved charges, and allowed charges are synonymous terms. Approved charges will be the term used in this chapter. itures are not directly affected by assignment, but by accepting assignment, physicians are in fact accepting a reduction in the payment for any service for which their billed charge exceeds Medicare's approved charge. Medicare has made further attempts to constrain program expenditures by amending CPR in various ways. Past approaches have included temporarily freezing all fees, as mandated by the Deficit Reduction Act of 1984 (Public Law 98369); lowering the percentile at which all prevailing charges are set; and applying the Medicare Economic Index (MEl) to limit annual increases in all prevailing charges. Although Part B expenditures have risen despite these measures, they might have increased more if controls had not been imposed. Medicare has not in the past attempted to moderate the growth in program expenditures or to redress perceived imbalances in relative payments by reducing differentials in payment rates for selected services. As noted in chapter 2, Medicare payment rates tend to be higher for procedural and inpatient services than for nonprocedural and ambulatory services, reflecting the program's benefit package that emphasizes high-cost acute and inpatient care. The rates also tend to be higher for specialist and urban services than for generalist and rural services in order to reflect local differences in physicians' fees. Another untried approach in reducing the rate of growth in program expenditures is for Medicare to give beneficiaries the option of receiving care from preferred provider organizations (PPOs). Medicare could take advantage of the increasingly competitive market and contract, either directly or through carriers or other entities, with only those physicians or groups of physicians who would agree to accept Medicare payments below the level of approved charges as payment in full. 98 • Payment for Physician Services: Strategies for Medicare This chapter explores, in the context of today's conditions, variations of the CPR method previously or currently used by Medicare to restrain program expenditures for physician services. It also analyzes the potential for controlling program expenditures and modifying perceived imbalances in Medicare payment rates for services that vary by type, site, specialty, and geographic location. Negotiated or discounted fees are also considered as a cost-containment approach. IMPLICATIONS OF ALTERNATIVE METHODS OF MODIFYING CPR Under CPR, the rate of growth in Medicare expenditures for physician services could theoretically be restrained by controlling approved charges for all or selected services (see table 4-1). Approved charges could be controlled by changing the manner of updating prevailing or customary charges, for example, by freezing them.4 Approved charges could also be controlled by lowering the percentile for calculating prevailing charges, which is now at the 75th percentile of customary charges.5 •A freeze on physician charges is but one way of changing the manner of updating charges. Other ways include changing the frequency of updating customary or prevailing charges and capping prevailing charges. 5Another method of controlling the rate of growth in Medicare expenditures would be to reform Medicare's coding system, which encourages physicians to bill separately for each activity undertaken in the care of the patient and may stimulate coding for more complex services. Coding problems are found in all fee-for-service meth- Under both methods, assignment (physicians' acceptance of Medicare's approved charges as payment in full) could be voluntary or mandatory, and mandatory assignment could apply to some or all services. 6 The implications of controlling Medicare approved charges for all services and for selected services are evaluated below with respect to dimensions indicative of the performance of the health care system: cost and efficiency, quality of care, access, technological change, and administrative feasibility. ods, and the issue is discussed in ch. 6. In addition, the possibility of imposing an aggregate expenditure cap is discussed in connection with fee schedules in ch. 5. •The analysis that follows assumes the retention of voluntary assignment unless otherwise mentioned. Table 4·1.-Methods for Modifying CPR Payment Intended To Control Medicare Expenditures or To Reduce Variations in Reimbursement for Services Scope of change All servicesa. . . . . . . . . . . . . Selected services . . . . . . . . Change manner of updating prevailing and/or customary charges (e.g., by freezing) 1. Freeze prevailing and/or customary charges for all services 2. Freeze prevailing and/or customary charges for selected services such as: • procedural, ' • inpatient, • specialist, and • urban 3. Freeze prevailing and/or customary charges for selected services such as: • procedural, • inpatient, • specialist, and • urban and increaseb prevailing and/or customary charges for other services, such as: • nonprocedural, • ambulatory, • generalist, and • rural Change percentile for calculating prevailing charges 1. Lower percentile for calculating prevailing charges for all services 2. Lower percentile for calculating prevailing charges for selected services, such as: • procedural, • inpatient, • specialist, and • urban 3. Lower percentile for calculating prevailing charges for selected services, such as: • procedural, • inpatient, • specialist, and • urban and raise the percentile for calculating the prevailing charge for other services, such as: • nonprocedural, • ambulatory, • generalist, and • rural aModlfications that affect all charges for services will not reduce variations in charges between services. bsuch selective increases can be accomplished by an add-on to frozen charge screens. SOURCE: Office of Technology Assessment, 1985. Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 99 The implications of controlling approved charges in approved charges. Thus, if Medicare reduced for selected services are also examined with rethe approved charge for a service from $100 to spect to redressing perceived payment imbalances $80, coinsurance would be reduced from $20 to by type of service, by site of treatment, by phy$16 (i.e., 20 percent of the $20 reduction). If a sician specialty, and by geographic location (see physician refused to take assignment with the table 4-1). Controlling Medicare payments for lower approved charge, he or she might continue selected services by reducing the variation in apto bill the beneficiary $100. Beneficiary unassigned proved charges among them could also pertain liability would then be $20. Despite the $4 decrease in coinsurance, total out-of-pocket costs to constructing a fee schedule (see ch. 5). Indeed, conversion to a fee schedule would afford an opfor the beneficiary would increase to $36. Only portunity to make any corrections in relative apif the physician billed between $81 and $85 would proved charges. the decrease in cost-sharing be more than the increase in unassigned liability. Thus, beneficiaryNone of the modifications to CPR payment dis out-of-pocket expenses might well increase withcussed in this chapter would change the financial lower approved charges. incentives that CPR gives physicians to provide additional services to generate income. In decidThere is no theoretical or empirical evidence to ing whether or not to provide a service, physiindicate that physicians would increase their cians would be likely to respond to changes in charges to non-Medicare patients if Medicare lo payment level or in relative payment rates. Morewered approved charges for Medicare patients. over, any decrease in the growth of Medicare exIndeed, non-Medicare patients might not be will penditures would be of short duration. CPR paying to purchase physician services if fees to them ment per se and the modifications discussed in this were raised (188,357), particulary in an era of in chapter encourage physicians who respond to ficreasing physician supply. But physicians might nancial incentives to raise their billed charges to shift their time and provision of services to non Medicare patients, thus increasing non-Medicarebeneficiaries, since such increases are later re flected in Medicare's approved charges. aggregate expenditures. A confounding factor in examining the effects Lowering approved charges would lower benof controlling approved charges on costs and eficiaries' financial access to care. Reducing the other dimensions is the uncertainty surrounding ratio of approved to billed charges has reduced the relationship between lower payment rates and assignment rates (158,184,315,357,394). The adchanges in the volume of services beneficiaries reditional costs associated with seeing physicians ceive (see ch. 2). How physicians and beneficiwho do not take assignment would diminish acaries would respond to lowered approved charges cess to care. Access could also decline if, as areis uncertain. As suppliers of services, physicians sult of lower Medicare approved charges, physiwould be expected to react to lower payment rates cians chose not to treat Medicare patients for by providing fewer services. But physicians also certain services. exert control over services used and might seek A decrease in the assignment rate could also into maintain their incomes by providing or billdirectly affect quality by curtailing access. Ifing for additional or more highly priced services. access to appropriately used services, e.g., extraLowering approved charges would lower bencoporeal shock wave lithotripsy (ESWL) for cereficiary coinsurance payments, and if out-of-pocktain renal stones, was reduced, quality could be et expenses fell as a result, beneficiaries would be lowered. On the other hand, if access to inapproexpected to seek more care. But if assignment conpriately used services, e.g., routine skull X-rays tinued to be voluntary, increases in beneficiaries' for minor injuries, was reduced, quality could be out-of-pocket expenses for unassigned liability improved. In addition, lower approved charges would most likely exceed reductions in coinsurcould directly affect quality by influencing the ac ance. The decrease in beneficiary coinsurance tions of some physicians who take assignment. would apply to only 20 percent of the reduction Physicians might include financial considerations 100 • Payment for Physician Services: Strategies for Medicare in choosing and providing services where the medfrom non-Medicare to Medicare patients; changesical and ethical decision is unclear (194). For exin physician opportunity costs; and changes in ample, some physicians might spend less time with billing practices, such as billing for a more comMedicare beneficiaries and more time with paplex procedure than actually provided or billingtients for whom their time is more highly paid. separately for items customarily included underone procedure (see ch. 2) (28,158,259). Controlling Approved Charges for All Simulations have examined the effect of the Services MEl on controlling approved charges and proA freeze on fees for physician services could be gram costs (see Paringer in box 4-A). The datadesigned and implemented in a variety of ways. have to be extrapolated with caution, since theVariables include charges to be updated (e.g., the MEl "caps" payment and is only partially analprevailing or the customary and the prevailing), ogous to a freeze. The MEl, a looser form of conthe frequency of updating, and the method of uptrol than a freeze, allows for inflation in the gendating. For example, customary and prevailing eral economy and in physician practice costs. Thecharges could be frozen for 2 years, and the upiP.dex has had a decided effect on lowering the andates could allow increases only in billed charges nual increase in the prevailing charge for somefor the first year. There would also be discretion procedures. Nonetheless, a large percentage of theabout the concept of participating physicians (see increase in Medicare program costs-47 percentch. 2). Although the specifics of a particular freezfrom 1980 to 1983-was due to higher prices foring method would influence its effects, the disindividual services (70).cussion below for the most part is confined to the The effect of the physician fee freeze enacted general implications of a fee freeze. under the Deficit Reduction Act of 1984 on Medi Lowering the percentile for calculating prevailcare costs and other dimensions of beneficiaries' ing charges could also be accomplished in a numhealth care has yet to be measured and reported. ber of ways. One strategy would be to lower the Preliminary evidence shows a reduction in the rate current prevailing percentile and retain the curof growth in expenditures per beneficiary for phy rent MEL Another strategy, to lower the current sician services for fiscal year 1984 (84). These data prevailing percentile and eliminate the current might reflect changes in the health field, such as MEl, would decrease provider and beneficiary changes in Medicare's payment methods for hos confusion and moderate the uneven effects of the pital services or an increasing competitive envi index on approved charges (see ch. 2). ronment. Whether changes in market incentives resulting from an increased physician supply and Costs and Efficiency from alternative organizational and delivery sys tems would favor decreases in approved charges As noted above, short-term savings to the Medi and would be strong enough to overcome tradi care program could theoretically be achieved by ~ional patterns of physician practice is conjectural. freezing charges for all services. But the empiri cal research on the U.S. and Quebec health care Lowering the percentile at which prevailing systems suggests that this approach has been incharges are calculated could produce short-termeffective in constraining the rate of growth in ex-reductions in the growth of Medicare expendi-penditures for physicians' services (158). These tures. Ifthe prevailing percentile is lowered fromresearch findings are not conclusive, since an inthe 75th to, for example, the 50th, Medicare excrease in the number and complexity of services penditures for physician services would be rebilled may have masked the effects of constrainduced to the extent that approved charges are curing payment rates on expenditures. The mecharently higher than the 50th percentile, assumingnism driving these changes in service quantity and that the volume and complexity of services are intensity is a matter of uncertainty and debate. not increased. The magnitude of the decrease inExplanations put forward include physician-in the rate of growth cannot be determined. To the duced demand; patient-initiated demand; a shift extent, if any, that the volume of services in Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 101 102 • Payment for Physician Services: Strategies for Medicare creased, the decrease in the growth of Medicare physicians perceived the lower approved chargeexpenditures would be lessened. as providing insufficient net revenues, they could either refuse assignment or orient use to patients In some cases, the magnitude of Medicare sav ings would be influenced by the MEL For those with private, higher paying insurance (234,431). procedures with indexed prevailing charges (the An exception might be made in providing MRI prevailing charge adjusted by the MEl) now at or for certain elderly patients with specific conditions above the 50th percentile, the MEl would have that are the target of research protocols, but in no effect on the amount of short-term savings. that case, access would be sporadic. Decreasing For procedures with indexed prevailings that are access to ESWL could have cost implications for the Medicare program, because ESWL might be lower than the 50th percentile, maintaining the less costly than surgery for certain renal stone care MEl after reducing the prevailing percentile to the (431). 50th would protect short-term savings. There would be no additional savings for services for There are effects on access and quality unique which the indexed prevailing is now below the to the specifics of the freezing method. Physicians' 50th percentile. If the MEl was eliminated, how reaction to a comprehensive freeze would depend ever, short-term program savings would be less both on the effect of the freeze on their real in because of an increase in payments for procedures comes and on the medical economic environment. that had been capped below the 50th percentile The longer the freeze lasted, the greater the num by the MEL ber of physicians who would be hurt financiallyEven in the short term, establishing the prevailand the greater the number of physicians whoing at the 50th percentile would not decrease the would be likely to refuse assignment.vrevailing charge for those procedures that have A freeze on approved charges could also affect a very small spread of customary charges between access and quality through the method of updat the 50th and 75th percentile. In effect, the 50th ing charges, the relationship of assignment to up and 75th percentiles of customary charges are the dating, and the extent of assignment. If physicians same for such services. Anecdotal evidence indi were required to accept assignment for all serv cates that a small spread is typical of procedures ices during a freeze, fewer physicians would ac that are controlled by a physician specialty, e.g., cept assignment during a freeze period than if cardiac nuclear procedures (347), and of specific assignment could also be accepted on a claim-by localities (521a). claim basis.Long-term savings produced by lowering theprevailing percentile to the 50th percentile are unIf there was a participating physician component comparable to that of the freeze imposed by likely. Over time, an increase in billed charges the Deficit Reduction Act of 1984 (see ch. 2), only would lead to increased prevailing charges, which nonparticipating physicians could refuse assign in time could be as high as the indexed prevail ing would have been. ment during the freeze. However, future access could be decreased to the extent that participat ing physicians refused to renew their participa Access and Quality tion agreements. In a strongly competitive area,A decrease in assignment rates in response to physicians might be more willing to accept assignlowered Medicare payment rates for all physician ment and renew participation agreements. Theservices would decrease beneficiary financial accare provided by participating physicians mightcess to care. Access would also be negatively afnot change, because their charges will be updatedfected if physicians choose to provide a ·service at the end of the freeze. Unless the net revenuesonly to non-Medicare patients. For example, the for discretionary services were generous at the onuse of magnetic resonance imaging (MRI) and set of the freeze, the clinical decisions of nonESWL for Medicare beneficiaries would depend participating physicians for such services takenon the level of Medicare's approved charge. If on assignment might be affected. Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 103 Controlling Approved Charges for Selected Services The variation in approved charges for selected services could be reduced by lowering approved charges for procedural, inpatient, specialist, and urban services (higher priced services) either with or without raising the approved charges of nonprocedural, ambulatory, generalist, and rural services (lower priced services). Both approaches could modify perceived imbalances in approved charges among such services. Reducing the Variation in Approved Charges by Type and Site of Service The concept of reducing the perceived disparity in approved charges between procedural and nonprocedural services is initially attractive in considering Medicare expenditures. Some evidence indicates disproportionate differences in the cost and the price of certain procedural services (46,227). When new technologies, in particular equipment-intensive and surgical procequres, are introduced, they are often priced at a high level (403,424,588). Initially, a high fee may be appropriate because the new procedure may require special skills and much professional time. Although experience and technological improvements over time often lower the level of expertise and amount of time needed to perform the procedure, initial payment levels are not reevaluated. In this regard, it would be informative to trace the evolution in prices over time for MRI and ESWL, which were both approved for Medicare coverage in 1985. The establishment and maintainence of high prices for services whose costs have declined over time is thought to have contributed to the wide differences in approved charges for procedural and nonprocedural services. Medicare has also continued to provide more generous payment for inpatient services than for services in other sites. This policy has not kept pace with recent Medicare initiatives, e.g., increased coverage for home health services, that encourage out-of-hospital care. The comparability of inpatient and ambulatory services, particularly visits, is still undecided. A rationale for paying more for visits in a hospital than in an office is that the visits differ. Patients in hospitals tend to be sicker than ambulatory patients and require more physician attention. On the other hand, physicians do not pay overhead costs for treating patients in hospitals, although their time and transportation costs may be higher than when caring for patients in their offices. Lowering approved charges for procedural services or inpatient services over which Medicare has market power could be an interim step in reducing the growth of Medicare expenditures or could be an independent modification of CPR. Medicare in 1983 had 74 percent or more of the market share for seven high-priced surgical procedures, including cataract surgery, and 40 percent or more of the market share for four highvolume diagnostic procedures (see table 2-12 in ch. 2) (69). Furthermore, the elderly accounted for anywhere between 26 and 37 percent of the performance of nine other surgical procedures and five other diagnostic and therapeutic procedures. Costs and Efficiency.-The fact that approved charges for procedural services and inpatient services constitute a major part of Medicare's expenditures for physician services suggests that reducing such charges has the potential for restraining the overall rate of increase of Medicare expenditures. National data for 1981 indicate that considerably more than half of Medicare's approved charges for physician services nationwide are for procedural services and that almost 64 percent of these charges are for services provided in inpatient settings (see table 4-2).7 If approved charges for inpatient medical care (primarily visits), which represent 20.6 percent of Medicare's approved charges for physician services (see table 4-3), and payment for all procedural services, which rep resent 48.2 percent of these charges (see table 4 2) were constrained, 68.8 percent of Medicare's approved charges would be affected. How reducing approved charges for procedural and inpatient services would affect Medicare costs is not clear, in part because the effect of price on use of services is still a matter of debate.8 If the 'South Carolina 1983 Part B claims data suggest that an even higher percentage (66 percent) of approved charges are for procedural services (247). •However, the relation of use to expenditures is clear from Medicare Part B data from 1975-1983. Figures on the contribution of increa~ed vol.u~e per enrollee to the growth in approved charges for surg1cal, chmcallaboratory, diagnostic, and X-ray services ranged from 39 to 44 percent; the increase in volume of services per enrollee for medical care (primarily office visits) was 22 percent (248). 104 • Payment for Physician Services: Strategies for Medicare Table 4·2.-Medicare Approved Charges and Assignment Rates for Physicians' Services,by Type and Place of Service, 1981 Approved charges Percent of Assignment(in $000s) approved charges rate Type of service: Nonprocedural services: Medical care ........................................ $ 4,517 40.2% 51.4% Consultations8 ..•...•.•.•...•.•..••.•....••...•.•.•. 381 3.4 59.7 Subtotal ......................................... 4,898 43.6 Procedural services: Surgery ............................................ $ 3,635 32.3% 47.6% Diagnostic radiology ................................. 865 7.7 57.5 Diagnostic laboratory ................................ 834 7.4 48.1 Radiation therapy ................................... 151 1.3 62.3 Anesthesia ......................................... 535 4.8 44.7 Assistant-at-surgery ................................. 196 -1.7 48.1 Subtotal ......................................... 6,216 48.2 Other: Other medical servicesb ............................. . $ 127 -1.2% 61.9% Total for all services ............................... $11,241 100.0% 50.5% Place of service: Office ................................· · ... · · · · · · · · $ 3,203 28.5% 37.1% Inpatient hospital ................................... 7,144 63.6 53.7 Outpatient hospital .................................. 532 4.8 68.7 Home .............................................· 71 0.6 57.0 Independent laboratory ............................. . 39 0.4 40.1 Skilled nursing facilityc .............................. 150 1.3 83.2 Other .............................................. 102 0.9 79.3Total for all services ............................... $11,241 100.0% 50.5%aconsullalions involve nonprocedural services primarily.bother medical services Include the rental of durable medical equipment, the purchase of durable medical equipment, the use of ambulance services, and the rentaland sale of internal and external prostheses and supplies.cThls category also Includes physicians' services rendered in nonskilled nursing homes.NOTE: Columns may not add to 100.0 percent due to rounding. SOURCE: I. Burney and G. Schieber, "Medicare Physicians' Services: The Composition of Spending and Assignment Rates," Health Care Financing Review, forthcoming. The original table listed service by type of service without categorization as nonprocedural and procedural. volume of services increased, lowering approved In addition, the practice of medicine is not alcharges for procedural and inpatient services ways precise. There is general agreement aboutwith or without raising approved charges for nonthe need for some services for specific conditions procedural and ambulatory services-could in(e.g., in vitro cultures for suspected urinary tractcrease Medicare costs. If the volume of services infections) and the need for providing services in did not increase, lowering approved charges for certain sites (e.g., treatment for hip fractures inprocedural and inpatient services could decrease the hospital). Changes in approved charges wouldthe rate of growth in total Medicare expenditures: be unlikely to affect the provision of such services. For many presenting conditions, however,If approved charges for nonprocedural and amphysicians:must use their judgment in choosingbulatory services were raised simultaneously, the among po~sible diagnostic and therapeutic servgrowth in Medicare expenditures would increase ices and sites.. The finding that the cystoscopic rateor decrease depending ·on the magnitude of the for urologic conditions in one medical market areachange in approved charges and in the use of each in Maii'ie'is[~tnore than double the rate for the Statetype of service. However, the proportion of pro~ as a whole, !While the cystoscopic rate in another cedural to nonprocedural and inpatient to 'ammedical ina'rket area is only about half the averbulatory services among physician services is age, for ex~ir}ple, indicates the discretionary naunknown and might change with a change in apture ofcys'toscopy (568). A procedure that canproved charges. be perform~disuccessfully either as an ambulatory Table 4·3.-Medicare Approved Charges, Percent Distribution of Approved Charges, and Assignment Rates for Physicians' Services, by Combinations of Place and Type of Service, 1981 Inpatient Outpatient Skilled nursing All places Office hospital Home hospital Independent facility Other places Approved charges (In SOOOs) All types of services ........ $11,239.8 $3,202.8 $7,143.9 $71.0 $532.2 $39.0 $150.1 $100.8 a Medical care ............. 4,516.7 1,780.9 2,319.3 60.2 181.5 -127.4 47.4 Surgery .................. 3,643.4 365.1 3,125.8 2.0 137.9 _a 3.2 0.4 a Consultation ............. 381.0 63.8 304.0 0.6 7.1 -5.3 0.2 Diagnostic radiology ...... 864.9 358.4 385.4 1.0 110.5 1.0 8.1 0.5 Diagnostic laboratory ...... 834.1 536.9 222.6 2.4 27.2 38.0 2.9 4.1 ~ a a _a .I>, Radiation therapy ......... 150.9 47.2 53.9 -47.0 a a a I Anesthesia ............... 535.0 2.1 529.6 0.1 3.1 -- a a a 3:: Assistant-at-surgery ....... 195.8 2.0 192.2 0.1 1.5 ---0 a Other medical services .... 127.0 46.4 11.1 319.3 16.4 -3.2 45.3 9: Percent distribution of approved charges 2 ~ All types of services ........ 100.0% 28.5% 63.6% 0.6% 4.7% 0.3% 1.3% 0.9% a _b ;:, Medical care ............. 40.1 15.8 20.6 0.5 1.6 1.1 0.4 (I) b b b Surgery .................. 32.3 3.2 27.8 -b 1.2 -- b b b b 0 Consultation ............. 3.4 0.6 2.7 -0.1 -- b b b Diagnostic radiology ...... 7.7 3.2 3.4 -1.0 -0.1 b -_b ~ Diagnostic laboratory ...... 7.4 4.8 2.0 -b 0.2 0.3 -0 (I) b b b -b Radiation therapy ......... 1.3 0.4 0.5 -0.4 --3 b b b b b b Ill Anesthesia ............... 4.8 -4.7 ---- b b b b b _b ~ Assistant-at-surgery ....... 1.7 -1.7 --- b b b 1J Other medical services .... 1.1 0.4 0.1 -0.1 --0.4 Cil Assignment rates (percent) ~ All types of services ........ 51% 37% 54% 57% 69% 40% 83% 79% :;· Medical care ............. 51 34 59 36 81 -b 81 85 ~ b b -b -b Surgery .................. 48 42 48 -54 b b b b ~ Consultation ............. 60 41 63 -57 ---Q. b b _b Diagnostic radiology ...... 57 38 71 -67 -96 b b b Diagnostic laboratory ...... 48 40 64 -66 40 - b b b b Radiation therapy ......... 62 46 72 -67 ---I ;:, _b b b b b b Anesthesia ............... 45 45 ---- b b b -b -b -b ~ Assistant-at-surgery ....... 48 -48 - iii b -b -b Other medical services .... 62 44 39 -93 77 Q aless than $0.1 million. Ill bless than 0.05 percent. ca NOTE: Columns and rows may not add to 100 percent due to rounding. Q) SOURCE: I. Burney and G. Schieber, "Medicare Physicians' Services: The Composition of Spending and Assignment Rates," Health Care Financing Review, forthcoming. '3 ~ Q) :a • ~ 106 • Payment for Physician Services: Strategies for Medicare or as an inpatient service is cataract surgery (161). Increasing the approved charge for nonprocedural and ambulatory services might affect the choice of services and sites in cases where the choice is discretionary. The effect on total volume of services and expenditures cannot be estimated, because among other factors, the number of such discretionary services is unknown. If approved charges for procedural and inpatient services were lowered, beneficiary costs would increase whether or not there was an increase in use, because beneficiaries' increase in nonassigned liability would almost always be greater than their decrease in coinsurance. If approved charges for procedural and inpatient services were lowered as approved charges for nonprocedural and ambulatory services were raised, the net effect on beneficiary costs would be uncertain. The change in beneficiary unassigned liability and coinsurance would depend on the extent to which assignment for procedural and inpatient services decreased and assignment for nonprocedural services increased and on the absolute changes in approved charges and the magnitude of any changes in use. Access.-If approved charges for procedural and inpatient services were reduced, the ratio of approved to billed charges would decrease, assignment rates would fall, and access could decrease. On the other hand, competition among providers of many procedural services is likely, given the current and projected supply of most surgical specialties and some internal medicine specialties. Most national studies project a continued growth in the supply of these physicians and an oversupply by 1990 (176). Competition among physicians in the form of taking assignment could be financially rewarding, if beneficiaries considered the differences in their liability between assigned and unnassigned claims when choosing physicians. Furthermore, one study found that surgical assignment rates were not significantly related to payment levels for surgical services (393). 9 Also, since •The relationship of assignment and reimbursement rates for laboratory and X-rays services is unclear. The main finding of the Rice study is that there is a significant positive relationship between changes in reimbursement rates for medical services and changes in assignment rates (393). Although changes in the assignment rates for laboratory and radiological services appear to be significantly correlated with changes in the reimbursement rate for medical serv a reduction of 10 to 20 percent in payment rates for many procedures and inpatient hospital visits would still give physicians high Medicare net revenues, assignment rates might not decline substantially if approved charges for such services were lowered (166). The relationship between Medicare approved charges and the price paid by other insurers also affects access to procedural services. If Medicare's lower approved charge for a service was much below the price allowed by other insurers, some physicians might choose not to provide the service to Medicare beneficiaries. For this situation to occur, however, there would have to be an adequate non-Medicare market for the service, such as there is in the case of MRI (234) and ESWL (431). Access to hospital-based, procedure-oriented physicians-radiologists, pathologists, and anesthesiologists-might not be affected by controlling approved charges for their services. Pathologists and radiologists currently have very high assignment rates (68). Although anesthesiologists accept assignment less frequently than thoracic surgeons, anesthesiologists accept it as often as surgical specialists such as urologists and orthopedic surgeons (68). Competition might be aminor factor in assignment decisions for some radiologists and pathologists. The Graduate Medical Education National Advisory Committee projected that specialists in anesthesiology, pathology, and therapeutic radiology would be in near balance with supply in 1990; diagnostic radiology was projected to be a specialty in oversupply (57). Anesthesiologists were originally projected to be in undersupply, but during the last few years residency programs have grown to such an extent that anesthesiology may be in oversupply in the near future (350). Raising approved charges for nonprocedural or ambulatory services would increase assignment rates and hence access to these services. The use ices, the finding may be an aberration of the claims system. The Medicare program prohibits physicians from assigning only a portion of services that are delivered to a beneficiary at the same place and time. Thus, if laboratory and radiological services were provided at the same time and place as a medical service, which is likely, they would most likely be listed on the same claim, and accepting assignment for these services would be directly connected to accepting assignment for the medical service. Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 107 of pneumococcal vaccination might increase, although its low use seems related more to the lack of physician and beneficiary knowledge of its effectiveness than to a low payment level. Quality.-Effects of reducing the variation in approved charges by type and site on quality of care would depend, in part, on the extent to which assignment rates were affected, and, in part, on the appropriateness of services.10 Reviews of the literature have concluded that there is excessive use of hospitals, some surgical services, and inpatient laboratory services in teaching hospitals and to a lesser extent in nonteaching hospitals (108,109,110,581). However, there is a problem in determining the appropriate use of specific procedural and inpatient services, as illustrated by the great variation in the practice of medicine and the lack of scientific norms of medical care (568). For example, in Iowa, the chances that a male resident 85 years old will have had a prostatectomy range from 15 percent to 60 percent in different medical service markets (568). This large variation suggests that for some patients a prostatectomy may have been inappropriate treatment and may have constituted poor quality of care. Surgery and hospitalization are not without risk: the mortality rate attributable to a prostatectomy, for example, ranges from 1.2 percent to 4 percent (568). If lowering approved charges reduced the inappropriate use of procedures and inpatient care, quality could be improved (184). However, there is the danger that cutting the payment level for all procedural and inpatient services might reduce the provision of necessary as well as unnecessary services (108,109,110,581). Patients with severe illnesses that require much specialized, procedural care might be harmed by such a change (194). Quality related to the use of nonprocedural and ambulatory services could also be influenced by raising approved charges for nonprocedural and ambulatory services and increasing access to such services. Because the need for an increase in use has not been identified, the effect on quality of increasing access to such services is not clear. ' 0ln most of the studies, inappropriate services are defined as services that "provide no significant benefit or . . . could be rendered in a less costly lower level institution or outpatient setting" (163). Reducing the Variation in Approved Charges by Specialty and Location11 In the 1970s, a major concern of Congress was rationalizing the distribution of physicians by specialty and by location (492) by reducing the variation in approved charges for similar services provided by generalists and specialists and provided in different geographic localities, particularly within States. Recently, policy interest has been focused on reducing such variations as a cost-containment mechanism. 12 For the most part, differences in approved charges are relevant for services that are provided by physicians of many disciplines: the greatest overlap in services provided by generalists and specialties lies in the visit category, which nationwide accounts fo:r; 41 percent of Medicare approved charges (69). In 1982, the prevailing charges nationwide for different types of visits, the five most common procedures, averaged 24 to 73 percent higher for specialists than for generalists (see table 4-4) . 13 Almost all the empirical evidence indicates that physicians practicing in urban and suburban areas usually receive higher Medicare approved charges for similar services than physicians practicing in "The focus of this discussion is on reducing the variation in prices within States, since this geographic division best reflects urban/ru ral price disparities, which are a policy issue of interest. 12Equity to providers could be a reason for attempting to modify the wide differentials in payment levels between generalists and specialists and among geographic localities. Opinions on Medicare's responsibility in this regard differ. The opinions are based on both a philosophical stance and practical considerations of access and costs. If equity among providers were one of the Medicare program's concerns, the program's ability to act as a prudent buyer, i.e., to provide its beneficiaries with the most appropriate services avail able at the lowest possible cost to the program, could be constrained by the need to assure equitable revenues to providers. On the other hand, Medicare would be concerned if disparate charges among providers and among areas decreased beneficiaries' access to appro priate health services. 13There are many problems in analyzing national Medicare data based on carrier data, because of the variety of ways in which car riers classify specialists. A specific problem is that the specialty stand ing of family physicians varies among carriers. In calculating pre vailing charges, Medicare carriers usually categorize general practitioners as generalists and internists as specialists, but nation wide information on carrier practices about the categorization of family physicians on the carrier level for reimbursement purposes is not available. Conversations with staff of the Inspector General's Office of the Department of Health and Human Services suggest that carriers could categorize family physicians as a specialty, but not all carriers do so (542). 108 • Payment for Physician Services: Strategies for Medicare Table 4·4.-Medicare Weighted Mean Prevailing Charges for the Five Most Common Services, Specialist/Nonspecialist, Calendar Year 1982 Percentage specialistService Nonspecialist Specialist differential Brief Ftua hospital visit .......... . $16.63 $23.90 43.7% Limited FlU hospital visit ......... . 19.63 25.88 31.8 Limited FlU office visit ........... . 16.99 21.05 23.9 Brief F/U office visit ............. . 13.58 17.67 30.1 Minimal F/U office visit .......... . 16.11 27.92 73.3 aF/U -Followup. SOURCES: U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Program Opera!· ing, unpublished computer tabulations from the Directory of Medicare Prevailing Charges, W. Merasholl, personalcommunication, June 19, 1985. rural and inner city areas (71). As an example, in 1975 Medicare approved charges for specialists averaged 23 percent higher in metropolitan than in nonmetropolitan areas. When adjusted for costof-living differences, the payment level difference was reduced to 8 percent (71). More recent data found that fees for first office visits were 52 percent higher in urban areas than in rural areas and median fees for revisits were 7 percent higher in urban areas than in rural areas (354). A study by Pennsylvania Blue Shield on physician pricing patterns under Medicare in Pennsylvania had less definitive findings. lnterarea price variations by specialty varied with the procedure, and, although prices tended to be highest in the Philadelphia urban area, prices in rural areas were not always the lowest (372). Costs and Efficiency.-lf there was an increase in volume, lowering approved charges for specialist and urban services with or without raising approved charges for generalist and rural services could increase Medicare costs. But, available' data suggest that if there were no increase in volume, lowering charges for specialist and urban services could constrain the rate in growth of Medicare expenditures for physician services. If apprbved charges for generalist and rural services were raised at the same time, the effect on Medicare expenditures would be uncertain. Medicar~ expenditures could increase if the costs saved by the program due to a decrease in approved charges for specialist and urban services were less .than the costs added to the program by the increase in approved charges for generalist and rural ~ervices. On the other hand, Medicare expenditures could decrease if the costs saved by the program I due to a decrease in approved charges for specialist and urban services were more than the costs added to the program by the increase in approved charges for generalist and rural services. One approach to lower approved charges for specialist services and to raise approved charges for generalist services would be to calculate a single prevailing charge for all physicians in a locality. If a single prevailing charge were calculated for generalist and specialist services in a locality, the effect on Medicare expenditures would depend on the proportion of generalist and specialist services in the locality and the distribution of customary charges for generalists and specialists. If the distribution of customary charges for generalist and specialist services was narrow calculating a single 75th percentile for both generalists and specialists would be about the same as calculating a separate 75th percentile for each and averaging them. In this case, calculating a single 75th percentile for specialists and generalists would not affect Medicare expenditures. Or a single prevailing charge could be calculated for all physicians in a State as a way of lowering approved charges for specialist services and raising approved charges for generalist services. This approach would also lower approved charges for urban services and raise approval for rural services. Research on the cost effects of reducing variations by specialty is sparse and has not considered the effect of changes in prevailing charges on volume of services. One study found no significant differences in Medicare costs when prevailing charges were computed separately for each Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 109 specialty as compared with computing them for physicians grouped into three broad categories (331). Program outlays were reduced about 2 percent in a simulation that eliminated specialty differentials by computing a prevailing charge for all physicians in a county (330). If assignment continued on a claim-by-claim basis, approximately half of the beneficiaries would have had an increase (averaging 17 percent) in out-of-pocket expenses (330). Evidence on the effect of reducing variations in prevailing charges by locality is equally scanty. Unfortunately, the results of the few available studies are mixed and inconclusive, leaving unanswered the question of how reducing variations within a State would affect Medicare program costs and beneficiary liability. The major issue of volume response also remains unresolved. When prevailing charges were calculated on a statewide basis rather than by localities within a State, prevailing charges for physicians in the major urban areas decreased and the prevailing charges for physicians in small urban and nonurban areas of the State increased as expected (394). However, total Medicare expenditures were not reduced: physicians billed for a greater number of services and more complex services. A nationwide study performed for the Civilian Health and Medical Program of the Uniformed Services (CHAMPUS) examined the cost effects of reducing variations in prevailing charges by specialty and by location (504). Although the findings cannot be generalized to the Medicare program for reasons such as wide differences in patient population and program administration between the two programs, the findings are of interest. An increase in CHAMPUS program expenditures was estimated if separate prevailing charge screens for specialists, and, where applicable, separate charge localities in a State were eliminated and statewide prevailing charge screens for all physicians were established. The new method of calculating prevailing charges was adopted in 1980. The increase has been, as projected, approximately 5 percent annually over any increase in CHAMPUS expenditures in the absence of the policy change (241). A corresponding increase in beneficiary cost-sharing occurred. Although increased physician participation had been anticipated when charges were increased, there has been no increase in assignment rates (241). Access.-A major policy issue has been improving access to primary care services provided by physicians of appropriate training.14 Reducing variations in approved charges by specialty and within States would affect access to general practitioners, family physicians, and internists differently, because carriers classify general practitioners as generalists and internists as specialists. Carriers can classify family physicians as specialists, but not all carriers do so. If there were no increase in volume, assignment rates and access to primary care provided by general practitioners could increase if approved charges for the primary care services of general practitioners were increased at the same time as approved charges for such services provided by specialists were lowered. In localities where family physicians are classified as generalists, increasing approved charges for the primary care services of generalists could increase the assignment rates of family physicians and access to their primary care services. If specialists' approved charges were lowered, for example, by calculating a single prevailing charge for generalists and specialists, and there was no increase in volume, there is a strong possiblity that assignment and beneficiary access to primary care provided by family physicians (if they were in the specialist prevailing charge screen) and internists could decrease. The effect of reducing specialty differentials on physicians' decisions to train in the primary care specialties is problematicY If approved charges 14Although the term "appropriate" training is difficult to define, Federal policy, mainly through the channel of training grants for primary care residencies, has explicitly accepted that primary care services are to be provided by physicians trained in primary care and has defined primary care physicians as general and family phy sicians, general internists, and general pediatricians. The first three medical disciplines are of import for the elderly Medicare population. 15The number of primary care physicians (general practice, fam ily practice, general internal medicine, and general pediatrics) in creased at about the same rate as total physicians and grew from 56 physicians per 100,000 population in 1970 to 70 physicians per 100,000 population in 1981 (544). The Graduate Medical Education National Advisory Committee has stressed training primary care physicans to improve the balance of physicians across specialties. 110 • Payment for Physician Services: Strategies for Medicare for specialist services were decreased and there were no increase in the volume, there would be a decrease in Medicare expenditures for services provided by internists and family physicians, in those areas where family physicians are classified as specialists. If approved charges for generalist services were increased as well, there would be an increase in Medicare expenditures for services provided by general practitioners.16 If there were an increase in the volume of services, the effect on Medicare revenues for each specialty would be uncertain. Furthermore, the effect of financial incentives on specialty choice in today's economic environment has not been investigated. The early literature indicates that financial incentives played only a minor role in specialty choice (334). However, those studies were performed when debts for medical education and malpractice insurance rates were much lower. Access to primary care services provided by primary care physicians is also affected by the geographic distribution of these medical disciplines. However, the ability of or need for statewide prevailing charges to attract more primary care physicians into poorly served areas is not clear. Numerous factors affect location decisions. Although a series of studies in the 1960s and 1970s suggested that economic factors were a minor factor (334), later analyses of Canadian physicians reported a positive, although small, relationship between income level and location choice (36,184). Berry found that if gross earnings increased 1 percent in medical service areas in Quebec, net immigration increased by 3.4 percent (36). Comparable results were found by Hadley in Canadian provinces; a 1 percent increase in net income could expect to attract 3.3 percent more new physicians (184). More recent research showed that more physicians in the United States have located in nonmetropolitan areas (427). Some researchers have concluded that the economic forces of an increase in the total supply of physicians and the overall growth rate of each specialty determines the geographic dispersion of the specialty (344). 16The-increase in approved charges for general practitioner services is a moot point in the long term, since few if any general practitioners are now being trained. Although between 1970 and 1979, the number of general practitioners and family physicians in the smaller towns declined, in 1979 nearly every town with a population of 5,000 had a general practitioner or family physician, and 85 percent of towns with a population of 2,500 to 5,000 had a general practitioner or family physician present (344). In 1982, there were still131 counties housing 3.5 million people (approximately 2 percent of the population) without an active physician (124). The Newhouse study showed that very few rural residents lived far from a physician and predicted an increase in the diffusion of family physicians into the smallest towns as their numbers grew (344). The Bureau of Health Professions has predicted that the diffusion of primary care physicians is expected to reduce overall shortage area needs in the coming years, although needs will persist in many currently designated shortage areas ( 546). As noted earlier, the evidence suggests that approved charges in rural areas, for the most part, are lower than in urban areas. Higher practice costs in urban areas could explain the differences, however, the data on the costs of operating physicians' practices in different locations are conflicting (50,334,355,512). Indeed, the latest data indicate that practice costs are higher in rural than in urban areas (355). To the extent that urban/rural differences in approved charges exceed differences in urban/rural practice costs, physicians might be discouraged from practicing in rural areas. If one believes that more family physicians are required in sparsely populated areas, reducing the variation in approved charges within States and thereby increasing payment rates in rural areas might be sufficiently effective on the margin to increase the interest of family physicians in settling in such localities. The number of family physicians increased 22 percent from 1977 to 1985 (344), and graduates of family practice residency programs are more likely to settle in smaller and nonmetropolitan areas than are other specialties. Since established physicians are not likely to move, the location choices of young physicians are most apt to be influenced by financial considerations. Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 111 A much lower percentage of communities with 2,500 to 5,000 people had a practicing internist than had a general practitioners or a family physician in 1979, although the percentage of communities with an internist had increased 35 percent from 1970 to 1979. If one considered it necessary to further increase the expected rate of movement of internists into rural areas, narrowing the difference in internists' prevailing charges between metropolitan and nonmetropolitan areas might influence the location decisions of new internists. Medicare is an important source of revenue for internists. Other than thoracic surgeons, internists derive the largest percentage of gross income (29 percent) from Medicare of any specialty (see table 2-8). Based on precedent, internists in urban areas might further decrease their assignment rate if modifying geographic differentials meant a decrease in approved charges in urban areas. Therefore, at the same time that reducing geographic differentials might interest new internists to locate in less populous areas, such a change might decrease internists' assignment rates and beneficiary access in urban areas. It should be noted that reducing the variation in approved charges by specialty might also affect access to primary-care services provided by nonprimary-care specialists (1). Most nonprimary-care physicians have higher assignment rates than primary care physicians. If these data indicate that nonprimary-care physicians are more satisfied with Medicare payment under assignment than primary care physicians, access to primary care services provided by nonprimary-care physicians may not be unduly affected by a decrease in payment. If, on the other hand, these data are more reflective of the nonprimary-care services of nonprimary-care physicians, the assignment rate for, and access to, nonprocedural services provided by nonprimary-care specialists might decrease if approved charges for such services were decreased. Quality.-Anticipating the effects on quality of reducing the variations in prices for services by specialty is confounded by the unresolved issue of whether generalists and specialists deliver "similar services," or whether specialists provide higher quality care than generalists. If specialists provide the same services as generalists, both groups of physicians should be paid at the same level. However, the degree of similarity between services cannot easily be distinguished from the coding system, the basis of payment under any fee-for-service system. Even though the code for a service provided by both general practitioners and specialists is the same, the service provided under the code may be different. For example, almost all visit codes do not differentiate by diagnosis or the content of the service provided. One school of thought is that services provided by generalists and specialists are similar for payment purposes (336,420). Medicare's coverage policy is consistent with this view, since Medicare does not limit coverage for most services according to specific training, but usually permits all physicians to provide all services (414). Another view is held by those Medicare's carriers that use different fee screens for specialists and general practitioners on the grounds that services delivered by specialists are different, more intensive, or of higher quality than those provided by generalists and that the more extensive training of the specialist warrants a higher level of payment. The issue of "similar services" for payment purposes has not been resolved by court action. The Michigan Academy of Family Physicians successfully sued for the elimination of specialty screens by their Medicare carrier. The District Court found that if and when services are found to be similar, prevailing charge payment should be the same regardless of who provides the service, thus implicitly rejecting the argument that differences in charging patterns among specialties are indicative of different services' being performed under the same procedure code (309). Although the U.S. Court of Appeals, Sixth Circuit, partially affirmed the District Court's ruling, the decision is currently being submitted to the Supreme Court for the second time for reconsideration (534)Y 17Another pertinent court action took place in Florida where the Medicare carrier does not employ any specialty reimbursement differentials. The Dade County Society of Internal Medicine sued to force the carrier to use specialty screens for internists, and by implication, for other medical disciplines. The plaintiffs argued that Medicare beneficiaries in Florida receive lower allowances than in other areas of the country when beneficiaries use the services of 112 • Payment for Physician Services: Strategies for Medicare Studies of care given by different specialists provide some but inconclusive support for the position that specialty training is associated with better quality of both ambulatory care and hospital care when specialized physicians are practicing within the domain of their training (369,370,392,398). The evidence that physician performance is improved by specialization per se is weaker (194) and contradictory (416). An analysis of visits found that nearly all specialists have significantly longer visits than general practitioners and family physicians (322). However, there is no evidence to support that differences in time result in differences in quality. Indeed, time,per se may not be an important measure of quality. In addition to the lack of definitive evidence on whether specialists provide better care than generalists, there are other unknown and confounding factors that stand in the way of determining how reducing the variation between generalists' and specialists' approved charges would affect quality. These factors include the extent to which specialists provide care for conditions outside of their own specialty, the inability to determine specialists' care because of the variation in the number of prevailing charge screens among carriers, and the relationship between quality and financial incentives. Quality may depend not only on the kind of physician, but also on the interaction among the kind of physician, the kind of patient, and the kind of service. The quality issue is further complicated by the lack of a standard definition of a specialist by Medicare for payment purposes. A General Accounting Office survey of three carriers found that roughly 50 percent of physicians who classify themselves as specialists are not board certified in their declared specialty. About one-fourth of the physicians that classified themselves in one subspecialty of internal medicine were not board certified in either internal medicine or the board specialty (475). However, there are no data on the relationship of quality to board certification as compared with quality and board eligibility or quality and self-declared specialization. specialists, and, therefore, beneficiaries allegedly suffer both direct economic loss and possible injury when they are discouraged from consulting a specialist. After 5 years of litigation, the Dade County Society of Internal Medicine withdrew the case in the spring of 1985 (534). Negotiated or Discounted Fees Some private sector payers and Medi-Cal (California's Medicaid program) have recently adopted selective contracting with providers, primarily as a cost-containment measure. Under selective contracting, payers contract with selected providers, usually hospitals, who agree to accept either a negotiated fee or a flat discount from their charges as payment in full. The experiences of private sector organizations and Medi-Cal with selective contracting may provide insight into the potential of using this method as an option in the Medicare program, while CPR continues as Medicare's major way of paying physicians. The discussion that follows considers two questions. Does the evidence show that the method is worthy of consideration for Medicare? Furthermore, does Medicare have the authority and capability to implement a discounting system7 Preferred Provider Organizations Negotiated or discounted fees in the private sector have been utilized as a financial component of PPOs and not as a payment method used by traditional indemnity plans. A PPO "may be an organization, a delivery system, or an arrangement between providers and third-party payers" (156). PPOs are characterized more by their diversity than by their similarity (see app. D). The third party pays the PPO plan, which in tum makes arrangements to pay the providers. One of the ways in which PPOs vary is how they pay providers. Payment mechanisms for hospitals range from negotiated per diem reimbursement (244) to capitation (52) and discounted charges (156). Currently, most PPOs reimburse physicians either by negotiating a discount from an established fee schedule or by discounting from usual, customary, and reasonable charges18 (53, 156), with discounts ranging from 5 to 30 percent (156). Two payment methods that are gaining in use are relative value scales constructed specifically for a PPO and capitation (144). Indeed, some experts consider discounting a transitory payment "The usual, customary and reasonable charge method is basically the same as Medicare's CPR method of paying physicians, but uses different nomenclature (see app. C). For clarity of discussion, this chapter substitutes Medicare terms for those used by the private sector Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 113 methodology and expect that many PPOs will evolve into health maintenance organizations (HMOs), as risk is increasingly shared with providers (53). Relation of PPO Payment Methods to the CPR Method PPOs that discount from tees that are determined by usual, customary, and reasonable methods use payment methods that are very similar to those used by Medicare to constrain prices. If there is no available claims history, the PPO may reduce physician charges based on their billed charges. This method is used by a few PPOs that have been established by organizations other than insurers, but its use appears to be declining for a number of reasons, including potential antitrust considerations (177). The method is similar to taking assignment under Medicare-both methods reduce payment from an individual physician's actual charges for a service when the actual charge is greater than the approved charge-and to Medicare's method of paying physicians who choose to become participating physicians under the arrangements mandated by the Deficit Reduction Act of 1984. Like Medicare, PPOs that discount from fees reduce charges for all services provided by participating physicians when their actual charges are greater than the program's approved charges. Furthermore, Medicare's participating physicians have signed an agreement for a year's participation, an acceptance of assignment on all claims, which is similar to physicians' contracting with a PPO. PPOs that have a claims history of physician payment usually evaluate physician charges in an area and declare some reduction on an areawide basis. Again, the methods used are similar to those used or those that could be used by Medicare. Some PPOs lower the percentile at which the prevailing charge is calculated (462a); some offer a percentage of an area's prevailing charge (29); and some establish payment at the mean of the prevailing charge (7). A very few have lowered the percentile at which the prevailing charge is calculated for medical services less than they lower the percentile at which the prevailing charge is calculated for surgical services (244). Both PPOs and Medicare use utilization review to control the volume of inpatient services, and many PPOs go beyond Medicare controls to include prior authorization before hospitalization. In a survey of members of the American Association of Preferred Providers, almost half of the 143 operational members reported that they had a utilization program in place for cost-containment purposes (7). Relatively few PPOs have developed "adequate protocols to review clinical efficiency or appropriateness of care" (51). It is unlikely that PPOs utilization review programs have been extended to cover office-based physicians' services. Although under Medicare utilization review for office-based physicians is required, the implementation of the review varies considerably among carriers and is often ineffective in containing costs (475)_19 The major way that PPOs differ from Medicare in physician payment is not in pricing method, but in the negotiation of contracts with selected providers. Under Medicare, almost any physician who chooses can participate in the program, but in a PPO arrangement, the PPO chooses the providers with whom it wishes to contract. The methods for selecting participating physicians vary. Some, but far from all, PPOs limit physician enrollment on the basis of performance standards (53). Future PPOs may attempt torestrict members to practitioners with cost-effective practice patterns (51). Very few PPOs have the standards or the technology capable of systematically evaluating the cost-effectiveness of physicians. Dimensions of Evaluation Evidence of the effect of discounting by PPOs in the private sector on quality, access, and costs is primarily limited to information supplied by sponsors of PPOs and other interested parties. The number of PPOs and their rate of growth are indicators of private sector interest in PPOs, and '•Sections 7500-7535 of the Medicare Carriers Manual, Part C, Claims Process provides instructions for utilization review by carriers. The General Accounting Office concluded that the Health Care Financing Administration's policies and practices have tended to provide disincentives to carriers for performing effective utilization re view (475). 114 • Payment for Physician Services: Strategies for Medicare since the emphasis in the private sector is on controlling costs through price competition, an oblique expression of the private sector's evaluation of the cost-containment potential of PPOs. Despite wide variations in estimates (see app. D), it is evident that PPOs have increased considerably since the acronym was first established in 1981 (7) and will increase in the future. To the extent that PPOs employ discounting as a payment methodology, physicians' acceptance of PPO discounting is indicated by their enrollment in PPOs. One study estimates that overall 5 percent of the Nation's doctors had a contractual arrangement with a PPO in 1983 (156). A much higher percentage of physicians was involved in California, which is the State where the largest number of PPOs are located.20 The level of physician participation would appear to be linked with the degree of competition in an area. Information on patient acceptance of PPOs is just developing, and available numbers are too small to be statistically significant. In 1983, only 14.6 percent of the physicians in Calif0rnia who had signed contracts with PPOs had seen PPOlinked patients (374). Of all physicians, obstetrician/gynecologists and orthopedists were most likely to have seen a PPO patient. The likelihood of patient participation was also higher among the larger metropolitan areas than in the semi-urban and rural areas, and varied among metropolitan areas (374). Objective analysis of the cost-effectiveness of PPOs is sparse. A study of Stouffer Corporation's PPOs found significant cost savings the first year. In addition to discounting, the PPOs had a rigorous monitoring system of utilization review in participating facilities, which had lower charges than the community norm before the PPOs were functioning (53). Selective contracting by Medi-Cal with hospitals on a negotiated rate basis yielded substantial savings to the State from reduced payments, with 200ne-fourth of physicians contacted in California to join PPOs have signed a contract with one or more organizations. If physicians not yet contacted signed up at the same rate as those that were contacted, 35.5 percent of physicians in California would have signed PPO contracts (374). little evidence of reduction of quality of care (245). The Medi-Cal program does not contract with physicians, although there is legislation authorizing it to do so. Although private payers have signed contracts with physicians, by early 1985 the practice was still not sufficiently widespread to have affected office practice patterns. Assessment of the effects of private sector contracting on access to physicians and quality of care remains to be done. Selective contracting with negotiated or discounted fees would allow Medicare to use its leverage in the market place to bargain with physicians about the price of services and assignment (242). Furthermore, Medicare could technically implement a discounting method based on selective contracting (see following section on administrative feasibility). One problem, however, is that although PPOs in the private sector are often designed to reward patients for using member physicians, Medicare might have problems in implementing a reward system. Reducing deductibles and coinsurance of those beneficiaries who use less costly physicians might not be effective because of the extensive use by Medicare beneficiaries of Medigap insurance that covers their costsharing liability. On the other hand, Medicare beneficiaries might choose not to purchase Medigap insurance if they were able to obtain the coverage they want from Medicare. Reducing premiums for some beneficiaries and not others might be politically troublesome. Technological Change An important condition affecting the development of technologies is the potential market, which is determined in large part by third-party reimbursement ( 487). CPR reimbursement provides physicians with financial incentives to provide technology, particularly equipment-intensive and surgical technology, to the extent that their net revenues are higher with greater use. Although the CPR payment method has been instrumental in the development of cost-increasing technology, it has provided little incentive for physicians to choose cost-saving technology. By reducing Medicare's approved charges, the modifications to CPR described in this chapter Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 115 could change physicians' financial incentives to adopt and use medical technology and, in turn, affect the market for medical technology.21 The effect on the market for medical technology cannot be ascertained, however, because the effect of lowering approved charges on the volume of services provided is uncertain. If the volume of services provided to Medicare beneficiaries in response to lowered approved charges were to increase, the effect on the market would depend on the extent to which increased Medicare revenues to physicians from the increased volume compensated for the decrease in Medicare revenues to physicians from decreased approved charges. If the volume of services did not increase, constraining approved charges would decrease physicians' demand for services that they use in their practice. Generally, to the extent that physicians' Medicare net revenue for a technology decreased, physicians' financial incentives to refer patients for services to other facilities would increase. Beneficiaries receiving highly sophisticated, expensive and new technology, such as MRI, would probably be referred to a regional facility. Beneficiaries receiving more established technologies that physicians might have in their offices but need replacement, such as mammographic, electrocardiographic, and X-ray equipment, could be referred locally, for example, to an office that only performs mammography located in the same building as their physician. Manufacturers could respond by developing cost-saving office-based equipment as well as continuing to develop technology for larger ambulatory facililites. Officebased laboratory testing equipment, for example, is already being developed in response to financial incentives, such as Medicare's prospective payment for inpatient services, to move care to less financially constrained ambulatory sites (332). A number of factors could work against a shrinking market if Medicare's approved charges "The financial constraints on the use of technology imposed by Medicare's prospective payment system for hospitals has had an effect on the market for technology, and has played a role in shifting some technology into ambulatory sites (489). Changes in physicians' Medicare revenues would also be expected to affect the market, since physicians influence the use of technology both in inpatient and ambulatory sites. 56-119 0 -86 -5 QL 3 were lowered. First, the modifications to CPR might restrain approved charges for only a relatively short period, and, hence, might not affect physicians' adoption and use of technology. Second, the competitive nature of today' s health care system and the fear of malpractice claims would undoubtedly affect physician behavior and might soften the decrease in physicians' demand for technology. Lastly, the non-Medicare market might be sufficiently large to override any changes in the diffusion of the technology. Moreover, the effects that lowering Medicare's approved charges would have on technologies for which Medicare has market power and that provide a large part of physicians' incomes might be somewhat different from the effects for other technology. For example, about 80 percent of the cataract surgeries performed in the United States are covered by the Medicare program, and decreasing prices for such surgery would have a dramatic effect on the incomes, as well as the Medicare revenues, of ophthalmologists who perform the surgery (161). For cataract surgery, unlike MRI and ESWL, there does not appear to be another large population who could be provided with the procedure if physicians perceived a lower level of remuneration as unsatisfactory. Although some ophthalmologists might reduce the amount of cataract surgery in response to lower approved charges, others might increase the number of such procedures. In any case, the growing supply of ophthalmologists would propbably increase the aggregate supply of ophthamologic surgery, so that reducing Medicare prices is unlikely by itself to decrease the use of the procedure and the use of implantable lenses (161). Because of the potential for volume increase, the financial incentives for the development of lower priced lens implants are not clear. Administrative Feasibility The administration of the Medicare program's physician payment system is complex, cumbersome, and characterized by extreme variation among carriers along a series of parameters and by confusion among beneficiaries and physicians (see app. C and ch. 2). Although any of the above modifications of the current CPR method are administratively feasible with current computer tech 116 • Payment for Physician Services: Strategies for Medicare nology, changes with the greatest promise of simplifying administration are those that would reduce the number of factors for discriminating among physicians in determining their level of payment. The disapproval rate for physician claims is much higher under Medicare than in the private sector partly because private sector payers make fewer distinctions among physicians, making payment more consistent, uniform, and understandable (488). Competitive pressures among insurers to pay subscribers' bills is also instrumental in the lower disapproval rate. Perhaps of greater import is the fact that Medicare's prevailing charges are calculated at the 75th percentile of customary charges, whereas the corresponding private sector prevailing charges are calculated at the 80th to 90th percentile. Both freezing customary or prevailing charges and lowering the percentile at which prevailing charges are calculated could be easily and inexpensively implemented, because they require no new data or basic changes in Medicare's physician payment system or in claims processing. Nonetheless, these methods retain all the features that make the system so difficult to administer. Furthermore, if such modifications resulted in prevailing charges that were appreciably lower than now, carriers could have increased administrative expenses due to an increase in telephone calls, letters, and appeals from physicians. Reducing the variation in approved charges for services by type and site of service might make the CPR payment method even more complex than at present. Lowering approved charges for some services but not others, or lowering approved charges for some services and raising approved charges for others would require the identification of specific services and would generally increase the number of factors used to determine physicians' charges. The Health Care FinancingAdministration (HCFA), with expert advice, could identify those services whose appproved charges would be lowered and those services , whose approved charges would not be changed or would be raised. The carriers would then have to implement HCFA's decisions. Although not outwardly obvious, lowering approved charges for inpatient services would also complicate the administration of CPR, because in practice carriers construct one prevailing charge for surgical services irrespective of where performed (347). Thus, additional prevailing charges would have to be developed for surgical services performed on an ambulatory basis. Again, physicians who were adversely affected by the modifications might require carriers to spend time and effort in answering their complaints. Reducing the variation in approved charges by specialty and within States could simplify administration. Either change would increase the uniformity of payment among physicians, although changes in payment level might initially cause a negative reaction from physicians whose approved charges were lowered. The concept of selective contracting by negotiating fees or ·discounting from charges is very new to Medicare. Medicare might have some of the technical capability to implement a discounting method for physicians' services based on selective contracting. Claims administration for PPOs in the private sector has proven to be more complex and costly than many insurers had anticipated (246). Although Medicare might have the ability to identify lower cost physicians from historical data, the possibility of establishing a utilization review system for ambulatory services, a system necessary for cost saving, is less certain in the short run. Although HCFA appears to have the jurisdiction to negotiate directly with physicians (174), such direct negotiations run counter to precedent. Since carriers have traditionally been HCFA's contact with physicians, the most likely approach is for carriers to undertake selective contracting with providers or provider groups who would lower their allowed charges (preferred providers). Establishing a category of preferred providers would require supplying physicians with copies of their customary and prevailing profiles. It would also require establishing two pricing systems for claims processing-one for physicians who would be paid by the traditional method and another for physicians who would be paid on a contract basis (347). Ch. 4-Modifications to Customary, Prevailing, and Reasonable Charge Payment • 117 CONCLUSION The effects on Medicare expenditures of lowervided to non-Medicare patients to maintain total ing approved charges for all or selected servicesrevenues. whether by freezing customary and/or prevailing charges or by lowering the percentile at which prevailing charges are calculated-are uncertain. The relation of payment rates to volume of services has not been established in the theoretical or the empirical literature. If the volume of services does not increase in response to a reduction in approved charges, lowering approved charges would decrease the rate of growth in Medicare expenditures. If the volume of services does increase when approved charges are lowered, however, the effect on Medicare expenditures would depend on the magnitude of the costs saved by the program due to an decrease in approved charges compared with the magnitude of program costs incurred due to an increase in the use. Lowering approved charges for all or selected services by freezing charges or lowering prevailing percentiles would have only a temporary effect at best in terms of reducing Medicare expenditures. Under CPR, increases in physicians' billed charges are later reflected in Medicare's approved charges, thereby encouraging physicians to raise their billed charges to beneficiaries. None of the identified modifications would change this feature of CPR. Freezing charges or lowering prevailing percentiles would be likely to increase beneficiary costs regardless of whether the volume of services provided to beneficiaries changed. Since lowering Medicare payment decreases assignment rates, beneficiary unassigned liability would be likely to increase. Although beneficiary coinsurance would decrease with lower approved charges, the increase in beneficiary unassigned liability would most likely exceed the decrease in coinsurance. An increase in beneficiaries' out-of-pocket expenses would decrease their financial access to services. Quality of care would be decreased to the extent that access to an appropriate level of services fell. If the volume of services provided to Medicare beneficiaries did not increase in response to lowered charges, physicians might increase their time spent and volume of services pro- Reducing the variation in approved charges for selected services would address the problem of perceived inequities in payment rates between certain services. Lowering approved charges for services over which Medicare has market power could be undertaken as an interim step or as an independent modification. The effects on cost, access to care, and quality of reducing the variation in payment rates among services by lowering the approved charges for procedural services, inpatient services, specialist, and urban services would be similar to the effects of lowering approved charges for all services. Access to nonprocedural services, ambulatory services, generalist, and rural services, however, might not be affected. If the variation in approved charges among services was reduced by lowering approved charges for procedural services, inpatient services, specialist, and urban services and raising the approved charges of nonprocedural services, ambulatory services, generalist, and rural services, the cost and access effects would be different. The effect on Medicare program expenditures would be unpredictable and would depend on whether the cost saved by the program due to a decrease in approved charges was equal to, greater than, or less than the costs added to the program by the increase in approved charges. Beneficiary liability would increase for services with lower approved charges and would decrease for services with higher approved charges. The effect on net beneficiary liability is uncertain and would depend on whether the increase in beneficiary liability as a result of lowering approved charges for some services was equal to, greater than, or less than the decrease in beneficiary liability as a result of raising approved charges for other services. During the process of reducing the variation in approved charges between procedural and nonprocedural services, Medicare could adjust approved charges for technologies whose initial pay 118 • Payment for Physician Services: Strategies for Medicare ment level has been maintained although the physician time, skills, and resources required to perform the procedure have declined. Medicare could also periodically review and adjust approved charges for such technologies whether or not variations in approved charges between procedural and nonprocedural services were reduced. Reducing the variation in approved charges by specialty and location has aspects that differ from reducing the variation in approved charges by type of service and site of service. One way of reducing the variation by specialty would be to calculate a single prevailing charge for all physicians in a locality. This approach would simultaneously lower the approved charges for specialist services and raise the approved charges for generalist services. The change in total Medicare expenditures would depend on the relative numbers of generalist and specialist services in the locality and the distribution of customary charges for generalists and specialists in the locality. Similarly, the change in Medicare expenditures from reducing locality differentials by calculating a statewide prevailing charge for all physicians would depend on the relative number of services in the different localities and the distribution of customary charges by locality. Reducing the variation by specialty does not appear to be an effective way to stimulate physicians to train as primary care physicians; however, the effect of payment rates on specialty choice in today's economic environment has not been investigated. Reducing the variation within States might marginally influence family physicians and internists to locate in small towns. Determining the quality effects of reducing the variation in approved charges by specialty is confounded by the unresolved issue of whether specialists provide better quality care than generalists. Controlling the approved charges of all services by providing beneficiaries with the option of receiving care from preferred providers appears to have the potential for constraining expenditures. The effects on quality of, and access to care, however, are unassessed as yet. Medicare could adopt this new method as an optional payment method for Medicare beneficiaries. Chapter 5 Payment Based on Fee Schedules INTRODUCTION In one sense, Medicare's customary, prevailing, and reasonable (CPR) charge determination system can be thought of as being neutral with respect to prices in the physicians' services market; Medicare approved charges are simply established by identifying particular prices from the existing distribution of fees charged by the physicians themselves. As a result of this approach, however, even within a single locality and within a single specialty, any two physicians who perform a particular procedure may have different maximum approved charges. ln.fact, it is possible-although highly improbable-that every physician performing a particular procedure would have his or her own unique Medicare approved charge. Because one year of a physician's billed charges are used to set the next year's Medicare approved rates, the CPR system has obviously not been neutral with respect to physicians' billed charges in the succeeding years of its implementation. An alternative to a neutral payment system might be designed to take advantage of Medicare's substantial potential market power with respect to phy- THE CONCEPT OF FEE SCHEDULES A fee schedule can be viewed as an exhaustive list of physician services in which each entry is associated with one specific monetary amount. (Two basic variations on the fee schedule theme involve possible multiple monetary amounts for each service depending on the geographic location or specialty of the involved physicians.) A concept closely related to a fee schedule is that of an RVS. An RVS is an exhaustive list of physician services in which each entry is associated with one specific numerical value that expresses the value of the service in question relative to an arbitrary numeraire. An RVS can be converted to a fee schedule by multiplying the relative value of each service by a monetary conversion factor. An ordering sequence for the list of services is generally provided by a procedural coding and sicians services. Further, such a system might be much simpler to understand for both the physicians and the beneficiaries. In the sections that follow, the notions of fee schedules are reviewed. The chapter begins with an explanation of the concepts of fee schedules, relative value scales (RVSs), and procedural coding and terminology systems. Also discussed are the potential uses of a fee schedule for reimbursement purposes. The initial issues arising prior to the implementation of any fee schedule are enumerated, as are issues revolving around the problems of maintenance of a fee schedule via updating or occasional appropriateness checks for possible recalibration. Two somewhat arbitrary categories for methods of constructing particular fee schedules are then discussed: 1) relative-valuebased methods, and 2) "competitive" methods. The concluding sections of the chapter address the potential impacts of all of the various fee schedule options and review the prospects for fee schedules as a whole. terminology system, a taxonomy of physician services. The most commonly used procedural coding and terminology systems are: 1) the various versions of the California Relative Value Studies; 2) (510) the system primarily used for diagnostic coding but which also includes the procedural coding scheme used in Medicare's prospec tive hospital payment system; and 3) the Current Procedural Terminology, 4th Edition (CPT-4) (85) the coding system developed under the auspices of the American Medical Association and currently incorporated in the HCFA Common Procedure Coding System (HCPCS). 1 'By HCFA policy, by July 1984, all carriers were to have converted to the use of HCPCS for all Medicare Part B data to be submitted to HCFA central office in Baltimore. 122 • Payment for Physician Services: Strategies for Medicare Fee schedules offer a method of fee determination within the context of fee-for-service reimbursement that can address many of the problems currently perceived within CPR. These include such issues as variations in approved charges, unpredictability of payment amounts, confusion on the part of beneficiaries and providers, and limited Government control over rising price levels for physician services. Because under a fee schedule a single fee is paid for a particular service to any physician (within a particular peer group in a particular jurisdiction), variations in approved charges are eliminated within that peer group and jurisdiction. In an extreme form, a national fee schedule that did not recognize specialty distinctions for payment purposes could provide a single payment rate for a specific service for all physicians in all parts of the country. There would be no variations in payment. More likely forms of fee schedules would involve some geographic distinctions for payment purposes, such as fee schedules applicable on a statewide or carrier-wide basis. Under some circumstances, specialty distinctions for payment purposes could be a feature of fee schedules. The establishment of a set of fee schedules could also highlight differences in payment levels for various services, such as those observed between procedural and nonprocedural services. Because the relative approved charges for any two specific services would be identical across physicians given a fee schedule, it would be easier to identify potential discrepancies in fees in the schedule compared to discrepancies under CPR. In implementing or updating a fee schedule, one could resolve such discrepancies. Discrepancies in payment for a particular physician service by site might also be easier to resolve under the administration of a fee schedule. Because the payment amount provided as a Medicare benefit for a particular physician service could be known in advance for both beneficiaries and physicians, there would be much less uncertainty about beneficiary coinsurance liability and physicians' expected receipts from Medicare carriers. As a result, one would expect much less confusion on the part of beneficiaries with respect to their financial obligations. Knowing their unassigned liability in advance would also enable beneficiaries to become better buyers. Under such a system, physicians' billings could proceed on a more expeditious basis under fee schedules because payment amounts could be better known in advance. Given a fee schedule system of payment, a single parameter could be used to revise the level of payments to take account of changes in the costs of producing physician services and perceived changes in the value of those services. This is in sharp contrast to the fee revisions under CPR, which result from the interactions of individual physicians' billing decisions, changes in medical practice and medical practice costs, and departures, if any, from relative values observed in Medicare localities in calendar year 1971. Even under a relative value system with multiple conversion factors for the various types of physician services, there would be potentially greater control of increases in the prices paid by Medicare for physician services. Uses of Fee Schedules for Reimbursement Purposes Three alternative approaches to the use of fee schedules for the purpose of determining reimbursements can be identified: • a schedule of maximum allowances, • a schedule of absolute reimbursements with no permitted additional patient liabilities, and • a schedule of Medicare reimbursements without regard to potential patient liabilities. These alternatives are not mutually exclusive. Furthermore, any or all of these alternatives might also be combined with an expenditure cap, which might be implemented by either disallowing claims above the cap or by discounting claims until there was a reasonable expectation that the cap would not be exceeded. In effect, Medicare's current reasonable charge process operates as a schedule of maximum allowances, with individual maximum allowances available for each procedure provided by any physician (or physician practice). For physicians whose customary charge for a particular procedure exceeds the adjusted prevailing charge, the value of the maximum allowance is equal to that of the adjusted prevailing charge. For a physician whose customary charge is at or below the adjusted prevailing charge, the value of the maximum is equal to that of the customary charge. For all physicians, however, for any claim submitted with a charge below the lesser of the customary or prevailing charge, the approved charge is the submitted charge. In calendar year 1984, 18.3 percent of all Part B claims were submitted at or below the CPR limits (535). Alternative Reimbursement Approaches A fee schedule implemented as a schedule of maximum allowances would set upper bounds on approved charges for specific services. For example, were the fee schedule amount for cataract excisions with intraocular lens insertions to be established at $1,500, the approved charge for a physician who billed for that procedure would be set at the lower of the submitted charge or $1,500. As under the current system of coinsurance, beneficiaries would have an incentive to secure needed physician services from a provider who would bill for an amount lower than the approved charge. This incentive would be diminished for those beneficiaries with Medigap coverage that "filled in" coinsurance amounts. A fee schedule implemented as a schedule of absolute reimbursements with no additional patient liabilities permitted would involve a significant departure from the present Medicare system of physician reimbursement. This option would involve a form of mandatory assignment-in effect, a prohibition of physician billing above the Medicare allowance. Under such a system a physician would receive only that portion of the fee schedule amount above the coinsurance (and any deductible) regardless of the submitted charge. The submitted charge, if any, might be disregarded; only the procedure code for the service would be used in determining the appropriate reimbursement.2 Other things being equal, physician price 'Under a comparable system used for pharmaceutical reimbursement under the Medicaid program in California, providers billed for specific services often without specifying a charge, since that charge was irrelevent with respect to reimbursement. Ch. 5-Payment Based on Fee Schedules • 123 under such a system would have no effect on beneficiaries' decisions with respect to individual physicians since there would be no difference in beneficiary liability for specific services. The third alternative with respect to establishing reimbursement amounts from a fee schedule would involve an even more radical departure from the present Medicare system of determining approved charges for physician payment. A fee schedule implemented as a schedule of Medicare reimbursements without regard to potential patient liabilities would in effect be universal nonassignment. This new arrangement would involve payment of only the fee schedule amount (above the deductible and any coinsurance) regardless of the physician's submitted charges. (Although physicians might still bill carriers directly, there would be no implication that the approved charge in such cases would necessarily be payment in full.) Because the beneficiaries would be responsible for paying for the difference between the physician's bill and the Medicare allowance under this kind of system, beneficiaries would have a substantial incentive to seek physicians with low submitted charges for needed services. Such a system might also be implemented to allow a beneficiary to keep any difference between the fee allowed by the schedule and any lower fee charged by and paid to the physician. Expenditure Cap Any or all of the three methods of using a fee schedule for Medicare reimbursement might be modified to implement an aggregate expenditure cap for physician services. One form of such a system has been employed under the health insurance program in the Canadian province of Quebec (388). Under an expenditure cap system, reimbursements might be made at some fraction of the relevant amount as long as there was a possibility that the expenditure cap might be exceeded. Most likely (and comparable to the compensation schemes used by some individual practice associations (IPAs)) would be a discounting program involving payments at, say, 85 to 95 percent of expected amounts with rebates to physicians (based on billing volume) if the expenditure cap exceeded total interim payments. A somewhat unlikely version of an expenditure cap might in 124 • Payment for Physician Services: Strategies for Medicare valve payments at 100 percent of the expected level until the cap had been reached, after which no claims would be paid. (It is alleged that some Medicaid programs, in effect, employed such a system by deferring until their next fiscal year payment on all current year claims starting from the time that their expected budget limit had been reached.) Another alternative might involve payments at 100 percent during the initial quarter of the year with quarterly downward adjustments, if needed, based on projections of anticipated claims in succeeding quarters. Unfortunately, this might have the effect of producing "gaming" behavior by physicians with patients who presented afflictions during the last quarter of the year. In this regard, in Quebec it is reported that some physicians at or near their billing limits join "billing-pools" to take advantage of unused billing quotas of other colleagues at the end of a billing period (388). One other issue that might arise in the implementation of an expenditure cap implemented through discounting would involve beneficiary coinsurance and nonassigned liability. If beneficiary coinsurance were calculated on the basis of the discounted approved charge, there would be a net decrease in expected beneficiary liability and, possibly, an increase in beneficiary utilization in response to the change in price. Other things being equal, a budget neutral proposal would retain beneficiary coinsurance liability with respect to the undiscounted charge. A more serious problem might be anticipated with respect to nonassigned liability under a discounting system. If physicians collected from the beneficiaries the full difference between their submitted charge and the discounted approved charge, the later rebates, if any, would involve double payments to physicians since the rebate amount would already have been collected from the beneficiaries. Further, even if beneficiaries were "indemnified" in this process by being reimbursed for the entire undiscounted approved charge on unassigned claims, under this system physicians would have an increased incentive to not accept assignment. Having the certain beneficiary payment in lieu of the potential rebate would minimize the "loss" to the physician that might occur if the expenditure cap were exceeded. Initial Implementation Issues In addition to issues with respect to the ability to administer a fee schedule on a continuing basis (to be addressed later in this chapter), there are a variety of issues that relate to problems attendant solely to the initial implementation of a fee schedule. Such issues include the following: • who might participate in the development of a fee schedule (specifically involving antitrust related prohibitions with respect to physician organizations); • whether the method of fee schedule construction needs to be the method of fee schedule maintenance over time; and • how to handle the transition from CPR to a fee schedule. 3 The last issue prompts the question of exactly how close to a fee schedule is the current distribution of approved charges? The Antitrust Issue As a purely mechanical exercise, any Medicare carrier could be instructed to estimate average approved charges for each service that it has reimbursed. A listing of the resulting charges by service could be used as a fee schedule. However, because of technological change in medical practice this fee schedule would soon become inadequate. Continuing input from physicians would be necessary to update the fee schedule, both with respect to new procedures and to changes among the established ones. Physician input in the development of a fee schedule clearly is useful and probably is essential. The method through which that input is obtained, however, may be suspect because of possible violations of one or more of the antitrust 'Basically, there would be few administrative difficulties in converting from CPR to a fee schedule. The major complication would be what policies, if any, would be used in the case of physicians whose approved charges would be reduced following the conversion. Previous physician payment reform proposals have suggested the use of "hold-harmless" measures that, in effect, would freeze individual physician's approved charges rather than reducing them until the time when increases in other charges brought the frozen charges into proper alignment. Another alternative would involve blending the new rates with the established ones as has been used in the conversion of hospital payment policies under the prospective payment system. statutes. It is hard to imagine physicians' establishing a fee schedule as something other than pricefixing. In fact, the Federal Trade Commission (FTC) has sued several medical associations with respect to their actions involving the publication of relative value studies or participation in fee review efforts. FTC has also issued anumber of advisory opinions that have had the effect of circumscribing concerted physician action with respect to the development of fee schedules. The effect of these opinions is not to prohibit physician input into the development of fee reforms. Individual physicians and medical societies may not negotiate fees but may discuss reimbursement issues-including relative values-with thirdparty payers without running afoul of antitrust prohibitions (93). FTC has modified its consent orders with several physician associations to note specifically that a physician association is not prohibited from "providing information or views, on its own behalf or on behalf of its members, to third party payers concerning any issue, including reimbursement" (554). What has been proscribed by FTC orders are agreements between physician associations and third-party payers, "whether extracted by negotiation or coercion, and any conduct in furtherance of such a result" (554). At the outset, it should be noted that the Medicare program (and any State Medicaid program) cannot be held to be in violation of antitrust prohibitions. If the Health Care Financing Administration (HCFA) unilaterally issued a fee schedule without physician input or if it adopted without modification the 1974 California Relative Value Study, there would be no violation. There is a "deemed repeal" of the antitrust acts for organizations established through the direct actions of the U.S. Congress. State actions (such as those that might involve Medicaid) are also exempt (377). Procuring physician aid even in a legal fee schedule development process, however, might be somewhat convoluted. The antitrust laws were instituted to prohibit "unreasonable" restraints on trade and competition (377). The drafters of those acts can be presumed to have believed that vigorous competition among many sellers would be the preferred state in any market because a sys- Ch. 5-Payment Based on Fee Schedules • 125 tern of competition would foster efficiencies unless restricted by private agreements or actions. However, that competition in the (physician) market might not produce good results is, in and of itself, not an acceptable antitrust defense. That the alternative, for example, to a fee schedule "competitively" derived from bilateral monopoly negotiations between a private market insurer and a medical society might not involve perfect competition is also not relevant. Therefore, that physicians might perceive an agreement to cooperate in the development of a relative value scale -much less a fee schedule-to be an antitrust violation might inhibit needed physician cooperation even though many types of physician contributions to such an effort would not be perceived by FTC itself to be potential antitrust violations. Three ingredients are needed to prove an antitrust violation: 1) there must be an agreement between two or more otherwise independent parties (usually in the same line of business); 2) the agreement must restrain trade or competition; and 3) the agreement must be "unreasonable" in terms of its effects on competition (267). An illegal agreement would be one that suppresses or destroys competition, not merely an agreement that regulated the behavior of the parties concerned while promoting competition. FTC has promulgated its judgment that RVSs for physician services may have anticompetitive consequences including the following (554): • establishment of price relationships without regard to quality, efficiency, or demand differences; • fragmentation of billing categories, with separate charges for individual services resulting in higher prices; • concerted or interdependent adherence to relative value scales by physicians; and • establishment of a "starting point" from which collusion may occur. In addition, FTC also noted in its advisory opinion to the American Society of Internal Medicine (ASIM) that an agreement by ASIM's members to adhere to its proposed "relative value guide" would do the following (556): • tamper with market pricing structures; • pose a danger of higher prices with respect to some medical services; 126 • Payment for Physician Services: Strategies for Medicare • stabilize prices artificially; or • restrict output of certain services, viz., procedural services, and possibly restrict the output of nonprocedural services as well. The major objections involve the possible effects on the price structure in the markets for physician services.4 In fact, any relative value scale adopted by Medicare would likely find use in the private market by both physicians and other health care insurers. Physicians, insurers, and health care financing researchers continue to use the California Relative Value Study even though its publication has been enjoined by FTC since 1979. Should HCFA initiate fee negotiations or request or be granted congressional authorization to conduct fee schedule negotiations with one or more medical societies, the implied repeal of antitrust violations would be effective. However, were HCFA to issue a solicitation in the form of a Request for Proposals for an RVS, some medical societies that considered responding would be unlikely to respond because they might consider themselves to be in danger of being sued by FTC or a competing physician association for violating antitrust prohibitions. Implementation v. Maintenence Clearly, any particular method of creating a fee schedule could be replicated any number of times as needed to adjust for changes over time. Because of this, it might be possible to establish a fee schedule system for which the method of updating fees was identical with the method of original implementation. An easy example might be the use of one year's average submitted charges to estimate a next year's fee schedule. Some methods, however, do not lend themselves to easy or at least inexpensive replication, viz., empirical estimates of resource costs associated with specific proce •The fragmentation issue arises in the evolution of procedural coding and terminology systems; it is not a function of RVSs. The output restrictions referred to in the FTC's advisory opinion to ASIM involve procedural services most likely performed by physicians who are not internists. One infers from the FTC opinion that surgeons, for example, would rationally reduce the supply of their services if their payment rates declined. If ASIM members or other physicians, however, were successful in raising the prices of their own services attendant to publication of their relative value guide, buyers might reduce their purchases of those services (555). dures. In such cases, replication as a means of updating might imply a very expensive system-perhaps, therefore, an infeasible system. Replication, however, is not the only means of updating. The Medicare Economic Index (MEl), for example, which is used in the process of updating Medicare prevailing charges, could be used to update a fee schedule regardless of the process used to derive that schedule. Other price or cost indexes might also serve this function. Use of an index might allow for the establishment, for example, of an RVS through a one-time physician consensus development process for each procedure or set of procedures. This process would not have to be repeated every year. Replication of the original process for the reconsideration of relative values (or relative fees) might be necessary only to establish levels for newly introduced procedures or for other practice changes that were believed to warrant such reconsideration. A varied mix of methods might be used to improve the rationality of any particular fee schedule over time. For example, one might initially change to a Medicare fee schedule by having carriers estimate average approved charges for each procedure to establish a baseline RVS. For payment purposes, this RVS might be converted to a fee schedule that might be updated each year using the MEL New procedures might be given interim payment rates following a consensus development process. Final payment rates could be establish~d following estimations of resource costs, perhaps 18 to 36 months after the interim rates had gone into effect. Finally, the members of an independent physician payment review commission might review and recommend changes to correct any interjurisdictional or interspecialty differences brought to their attention. Transition From CPR to a Fee Schedule If a particular fee schedule were identified;,and deemed to be desirable, an initial problem would involve the transition from the current system to that schedule of fees. The expectation under the current system is that for approximately no less than 25 percent of the Medicare volume for any procedure, the approved charge is equal in value to that of the adjusted prevailing charge, with the rest of the distribution of charges at a variety of lower levels. For some time, however, there has been speculation that Medicare payment levels were moving in the direction of de facto fee schedules because of the implementation of the MEL To the extent that this phenomenon has occurred, a transition to a de jure fee schedule might be less of a problem. By the early 1970s, it was clear that the use of one year's submitted charges to establish the next year's customary and prevailing charges provided an incentive to accelerate fee increases. As a result, there was a concern expressed that Medicare fees were fostering inflation in medical care prices, rather than merely following changes in the costs of providing physician services. To attempt to ensure that increases in Medicare approved charges followed rather than led inflation in physician fees, legislation was passed to institute a proce-: dure to cap prevailing charges. The level of the cap would be changed each year through the use of an "economic index," which explicitly estimated both increases in the costs of providing physician services5 and increases in general earnings levels. The MEl was mandated in section 223 of the Social Security Act Amendments of 1972 (Public Law 92-603). Because of the imposition of the Economic Stablization Program in 1972, the provisions of the MEl were not implemented until July 1, 1975. Prevailing charges in effect at the passage of the legislation provided the initial caps on approved charges. Thus, the base year for the MEl was July 1, 1972 through June 30, 1973, fee screen year 1973.6 In any subsequent fee screen year, the "adjusted" prevailing charge for any service would be the lower of the 75th percentile of the distribution of volume weighted customary chargesnow known as the "unadjusted" prevailing-or a value equal to the product of the prevailing charge from fee screen year 1973 multiplied by the current value of the MEL For example, for a procedure that had a fee screen year 1973 pre 'The components of physician practice expenses that are included in the MEl are staff salaries, rental costs, automobile expenses, supplies, professional liability insurance, and "all other" costs. 'Approved charges for that time period had been established through statistical manipulations of physician charges submitted during calendar year 1971. Ch. 5-Payment Based on Fee Schedules • 127 vailing charge of $100 and for which the fee screen year 1982 "unadjusted" prevailing charge was $185, the "adjusted" prevailing charge would have been $179-the value of the MEl times the base year prevailing charge (116). From the MEl base year through June 1983, physician prices as measured by the Consumer Price Index ( CPI) increased 258 percent while the MEl increased 206 percent. Because of this disparity, it has been assumed that the MEl might ultimately transform the CPR system into a fee schedule based on the fee screen year 1973 prevailings. However, because the particular limit (submitted, customary, prevailing, or other charge) used to establish the approved charge for any physician bill to Medicare has not generally been recorded by carriers during the payment process until recently, there has never been a complete national source of statistics on the constraints imposed by the MEL Thus, it has been impossible to distinguish whether an MEl induced fee schedule will be achieved or merely approached asymptotically. The available evidence is equivocal with respect to how close the current system is to a fee schedule. For some years, the Medicare Directory of Prevailing Charges (532) has included an indicator to identify for 110 common physician services those prevailing charges that have been established through the use of the MEL In fee screen year 1984, 55 percent of all prevailing charges listed in the Directory for general practitioners and 62 percent of the procedures for specialists were established by the MEl (532). These numbers, however, have been relatively stable if not declining since at least 1981, a pattern that is not indicative of the imminent coming of fee schedules for all services. Using the MEl indicators and other data collected for the fee screen year 1984 Directory, the Congressional Budget Office (CBO) estimated that 60 percent of approved charges in the Medicare program are priced at levels determined through the MEL They estimate that by 1990, this will increase to 70 percent. Those estimates, however, are probably somewhat upward biased because of peculiarities in the data definitions in the in 128 • Payment for Physician Services: Strategies for Medicare structions to Medicare carriers for collecting these data. 7 An alternative source is an analysis of calendar year 1983 carrier data from the State of South Carolina (247). This analysis of data on physician services excluding anesthesiology showed that 43.2 percent of approved charges were established at the level of the adjusted prevailing. Because the adjusted prevailing is the lower of the MEl cap or the actual 75th percentile of the distribution of volume weighted customary charges, 43.2 percent must be considered an upper bound estimate of the impact of the MEl in that State. In addition to this aggregate estimate, Juba estimated comparable percentages for a variety of types of services. These ranged from 65.2 percent and 64.6 percent for office and hospital visits, respectively, to 38.9 percent and 30.3 percent for radiology (professional component only) services and surgery, respectively. These statistics suggest that the MEl may be closer to producing a fee schedule for physician visits and other nonprocedural services than for surgeries and some of the more technical services. It does not suggest that a fee schedule is at hand as a result of the MEl. If this interpretation is correct, however, transition to a fee schedule may become both easier 'Data for the Directory submitted by the carriers for each of 110 services include: the adjusted prevailing charge, the 50th and 75th percentiles of the distributions of volume weighted customary charges, and the total number of services whose prices were used to establish the prevailing charge. By assuming that the distribution of customary charges is statistically normal or near normal, one can estimate the actual percentile of the prevailing. The total units of service can then be used to aggregate expenditures over the entire set of procedures. This is basically the CBO procedure. Because the 50th and 75th percentile estimates are established by identifying the lowest customary charge that is no less than (i.e., equal to or greater than) the desired percentile, the resulting CBO percentile estimates will be biased upward by varying degrees. Further, to the extent that procedures introduced since 1971 have been less affected by the MEl, the 110 procedures included in the Directory will be less representative of the distribution of all physician services provided to Medicare beneficiaries, again contributing an upward bias to the estimates. Finally, of the 110 procedures included in the Directory, inpatient surgical procedures tend to be underrepresented, because the surgeries included in the Directory are a much smaller proportion of approved charges for all surgeries than the comparable proportion represented by the specific types of physician visits included in the Directory. Because recent evidence (247,294) suggests that visits are relatively more constrained by the MEl than surgeries, the underrepresentativeness of surgeries in the Directory will impart an additional upward bias to the resulting estimates of MEl impact. and somewhat more complicated. The ease in transition would be found in the problem of establishing fees for the office visits and hospital visits, services responsible for significant fractions of Medicare expenditures. To the extent that there is relatively little variation in approved charges with respect to individual visit types, intraspecialty disputes over appropriate prices may be lessened. Standard deviations with respect to average approved charges for the four most common office and hospital visits (in South Carolina) were found to be between $2.35 and $3.40 (247) (see table 5-1). If the distribution of approved charges is roughly normal, approximately two-thirds of the approved charges for any of those visits are within $3.40 or less of the average. In fact, 85 percent of the limited followup office visits exhibited approved charges within 25 percent of the State mean approved charge across all specialties, and 94 percent were within 10 percent of the relevant specialty mean. Thus, establishing a fee schedule amount at the average approved charge would not imply substantial changes in unit payments. On the other hand, standard deviations for some of the surgical procedures, for example, are 10 to 100 times greater than those of the most common visits. This relationship implies that for a particular patient or-for some physiciansall patients, a single fee schedule amount, even if based upon the average, might involve a nontrivial loss of unit revenue. Such a prospect might cause a physician to change his or her clinical decisions about the patient's therapy or his or her entrepreneurial decisions about assignment or participation in the Medicare program. To the extent that this problem exists, it may be advisable to phase-in a change to a fee schedule. In the past, proposed Medicare physician payment changes have been designed to be phasedin through the use of "hold-harmless" provisions. Under this approach, the payment for a particular procedure to a physician whose approved charge would otherwise exceed the fee schedule amount is frozen at the previous approved charge level until such time as approved charge increases for other physicians bring the fee schedule amount to that level. This approach has the effect of temporarily rewarding physicians whose fees are above average. If the expenditures for those Ch. 5-Payment Based on Fee Schedules • 129 Table 5·1.-Mean Approved Charges and Standard Deviations for Selected Medicare Services,• South Carolina, 1983 Percent of total Mean approved charges approved Standard in State charge deviation Office visits: 90080 Comprehensive: established patient ............. . 1.03% $ 42.48 $ 15.47 90020 Comprehensive: initial patient .................. . 0.81 38.11 17.53 90060 Intermediate: established patient ............... . 1.37 18.23 3.68 90050 Limited: established patient .................... . 9.57 12.83 2.35 90040 Brief: established patient ...................... . 0.72 11.54 2.74 Hospital visits: 90220 Comprehensive examination ................... . 4.56 54.63 13.76 90250 Limited: followup ............................. . 11.86 16.26 3.40 90240 Brief: followup ............................... . 0.53 13.99 2.59 Other medical procedures: 93547 Selective angiography ......................... . 0.67 563.27 46.64 90620 Consultation: initial comprehensive ............. . 1.85 63.01 10.91 90630 Consultation: initial complex ................... . 0.53 63.01 9.43 99174 Critical care: extended ........................ . 0.65 39.69 10.53 99173 Critical care: intermediate ...................... . 0.73 36.56 7.97 93000EKG ........................................ . 1.56 23.27 3.20 Surgery: 33513 Quadruple bypass ............................. . 0.65 3,691.17 175.67 33512Triple bypass ................................. . 0.92 3,617.33 344.82 27130 Athroplasty .................................. . 0.66 2,009.57 257.69 66980 Lens prosthesis: cataracts ..................... . 7.15 1,335.70 139.51 27244 Femoral fracture .............................. . 0.89 1,003.70 93.50 44140 Colectomy ................................... . 0.67 984.54 123.80 27236 Femoral fracture: proximal end ................. . 0.52 835.26 55.86 66920 Cataract removal: extraction lens ............... . 0.61 794.57 40.62 52601 Transurethral resection of prostate .............. . 2.44 792.16 35.72 47605 Cholecystectomy with cholangiography .......... . 0.69 702.04 72.98 43239 Upper G.l. endoscopy with biopsy .............. . 0.52 229.48 34.54 43235 Upper G.l. endoscopy ......................... . 0.72 208.59 35.79 Radiology: 74240 Upper G.l. tract and exam ...................... . 0.51 31.12 3.22 77405 Therapeutic: intermediate ...................... . 0.51 24.10 3.05 71020 Two-view chest X-ray .......................... . 1.34 13.76 1.44 71010 Single-view chest X-ray ........................ . 1.01 9.71 .77 Pathology: 82947 Glucose test ................................. . 0.53 5.59 1.04 81000 Urinalysis ....... : ............................ . 0.70 3.79 .48 aProcedures that account for at least 0.5 percent of approved charges In the State. SOURCE: D. Juba, "Analysis of Issues Relating to Implementing a Medicare Physician Fee Schedule," prepared for the U.S. Congress, Office of Technology Assess· ment, Washington, DC, November 1985. "above average" fees are used, in effect, to reduce the increases allowed for other physicians, the hold-harmless approach penalizes those physicians whose fees were below average. An alternative would involve blending fee schedule payments with CPR payments during a transition period. This approach allows for a faster transition to single payment rates than would "hold harmless" provisions, while reducing the magnitude of any windfall losses or gains that might attend an "overnight" implementation of a fee schedule. Updating, Maintenance, and Appropriateness Checks As indicated earlier, the method of fee schedule origination need not be the method of updating. For this reason, relatively costly methods of creating fee schedules or RVSs could be consid 130 • Payment for Physician Services: Strategies for Medicare ered to take advantage of any of their potential design features. (Replication could remain a method of updating either on an annual basis or for less frequent or partial recalibrations.) In the absence of replication, there are two general problems that can be anticipated in updating a fee schedule: 1) identifying appropriate aggregrate changes in the level of fees, and 2) identifying appropriate changes in relative fees within the schedule. (One might note that these are the two primary functions given to the Prospective Payment Assessment Commission (ProP AC) under Public Law 98-21, which established the prospective payment system for Medicare Part A.) If the market for physicians' services were perfectly competitive and if CPR did not contain incentives to raise billed charges in one year to increase approved charges in the next, CPR would have a theoretical advantage with respect to maintenance of payment levels. Other things being equal, if the costs of practice of all physicians rose, billed charges would also rise appropriately to reflect input cost increases, and approved charges would follow. If the costs of producing q particular physician service rose more than other services, one should observe a greater increase in approved charges for that service under CPR. However, it has been noted that CPR's incentives can influence billed charge levels. Further, although competitive, the market for physicians services is not perfectly so. Given a conversion to a fee schedule by Medicare, some other alternative to sole reliance on the prior year's billings would have to be adopted for fee schedule updating. Aggregate Changes Over Time The model of a perfectly competitive market can be used to examine how prices should change over time in an efficient economy. Such an examination can provide guidance in the development of policy for updating a fee schedule. Specifically, in a perfectly competitive market, suppliers would behave as if they were minimizing the costs of producing their services for any level of total output. Increases in input prices would be reflected in changes in suppliers' cost functions, 8 from 'A cost function denotes the mathematical relation between input prices and the minimum cost of production of a particular level of output for a particular production process. which one could infer the price increase that would be anticipated in a competitive market with a fixed level of output. The mathematical results of this exercise are the following: the expected proportional change in cost for a cost minimizer given changes in input prices is equal to the weighted sum of proportional changes in input prices, where the weights are the shares of total cost of the various inputs. Hence, one could develop an index to estimate the most "efficient" increase in fees that would be appropriate given observed increases in physicians' costs of practice. There are two available indices that relate to physicians' costs and prices. They are the Professional Services Index of the Medical Care Component of the CPI and the MEL The former is somewhat better known to the general public and has been computed on a monthly basis longer than the Medicare program has been in existence. It is based on 79 somewhat general physician services, 9 the billed charge for which is requested on a monthly or bimonthly basis from a fixed cohort of roughly 650 physicians located in urban areas across the United States. For historical reasons, the services of ophthalmologists are included in a separate vision care index, and the services of anesthesiologists and pathologists are included in the Hospital Price Index subcomponent of the CPl. For the purpose of updating a fee schedule, the CPI professional service subcomponent does have the advantage of being an index of fees that physicians charge their patients. Because it is based on a fixed basket of services, for a fixed cohort of physicians who are asked prices charged to private-pay patients, it may even be biased downward as an index of physician fees in general. In any case, it does not directly reflect changes in the costs of physicians' practices. The MEl was mandated by the Social Security Act Amendments of 1972 (Public Law 92-603) in response to concerns that increases in Medicare approved charges led rather than followed inflation in physician fees. To break this pattern, the Senate Finance Committee had proposed to limit increases in Medicare prevailing charges by com •The exact number of specific services included is much larger, since each physician practice in the sample provides his or her billed charge for a specific service within one or more of the somewhat general categories. paring the prevailings to an index based on increases in the costs of producing physician services and increases in general earnings levels. The Finance Committee did not specify the exact form of the index, but it did suggest that the weighted sum of the price changes for various practice in puts might be an acceptable approach. The notion is common sensical: if the prices of 40 percent of one's inputs are increasing by 10 percent and the remainder are increasing by 15 percent, then on average input costs are increasing by 13 percent (13,251). Although neither the Senate staff nor the Social Security Administration staff who developed the MEl (118) began with a cost function analysis, the index that was developed is a closer analog than the CPI to a predictor of the price increases expected from efficient physicians who faced increasing input prices. There are a number of refinements that might be introduced in the MEl, particularly with respect to the question of productivity changes, but the existing MEl might be an appropriate index for use in updating the general level of fees in a Medicare fee schedule. In an RVS-based fee schedule, one would simply multiply the change in the MEl by the existing conversion factor to obtain the appropriate increase in the conversion factor. Recalibration The index approach to fee schedule updating is administratively easy, but it embodies the im plicit presumption that relative fees within the schedule are correct and remain correct. At this point, one could reprise the justifications for lo cality and specialty differentials, restate the argu ments for using the payment system to encourage the provision of some services and to discourage others, and review the appropriate way to estab lish and monitor approved charges for new pro cedures that enter the repertoires of a significant number of physicians. Because the circumstances that underlie these issues are dynamic, one would want the fee schedule system itself to have a mech anism for responding to such dynamics. For example, if the Medicare approved charge for a particular service were $25 in Manhattan and $20 ill northern New Jersey, there could be a peri odic review of the need to continue such a dif ferential. Similiarly, specialty differentials for spe- Ch. 5-Payment Based on Fee Schedules • 131 cific services could be reviewed. The approved charges of new procedures not only could be reviewed over time to verify efficiencies that could be expected to evolve, but the approved charges of any procedures that are replaced by new ones could be examined to determine any continued justification for paying different prices for services with equal results. Keeping Fee Schedule Levels and Capitation Levels Commensurate Within the framework of the fee schedule as a method of payment for physician services, aggregate price levels and relative price levels remain the two basic issues. However, even if fee-forservice continues as the predominant method of payment, whether by fee schedules or not, there are a substantial number of Medicare beneficiaries whose physician services will be provided under capitation arrangements, such as competitive medical plans (CMPs) or health maintenance organizations (HMOs). Comparisons of the expenditures for physicians' services under the two systems may provide another means of assessing the appropriateness of fee levels under fee-forservice. If there were HMOs that maintained dis aggregated data on their costs of treating specific ailments on an ambulatory basis, such costs might be used to examine approved charges for the physician services used in those treatments. The comparisons might also be used to exam ine the appropriateness of payments made under prepayment arrangements. For example, in Cali fornia it was recently observed that the State pays more per Medi-Cal (Medicaid) recipient enrolled in HMOs than it does for recipients who receive services in the fee-for-service sector (74).10 None theless, because the level of costs of CMPs may rise to the level of prepayment amounts, one might justifiably use fee schedule payment level changes to assess proposed changes in prepayment levels.11 10This appeared to be a result of State stringency in raising fee levels for fee-for-service providers rather than as a result of HMO inefficiencies. "Under a worst case scenario, average adjusted per capita cost (AAPCC) levels for competitive medical plans (CMPs) would be overestimates because of beneficiary selection favorable to the CMPs. CMP costs, however, could rise even further as they compete for healthy patients by offering additional benefits or amenities. AAPCC levels based on non-CMP enrollees would also rise due to exacer bated adverse selection. As a result, neither CMP costs nor aggregate expenditure levels for the nonenrolled beneficiaries would be an appropriate guide to future CMP prepayment levels. 132 • Payment for Physician Services: Strategies for Medicare APPROACHES TO THE INITIAL CONSTRUCTION OF FEE SCHEDULES For the purpose of discussion, methods to conThus, two services might have the same relativestruct fee schedules will be partitioned into two value, but be assigned different fees. The healthcategories. The first includes all approaches based insurance programs in France use this type of syson the concept of an RVS-whether a chargetem (115).based, resource-cost-based, or consensus RVS.The second labeled "competitively" developed fee RVSs for physician services are a relatively re schedules, includes four methods that for the most cent phenomenon. The Casualty Actuarial Socipart are based on either implicit or explicit use ety developed RVSs for commercial insurers inof market mechanisms to develop a set of fees. the 1940s (430). The best known of the RVSs areThe first two involve unilateral buying policies those that were published by the California Med ical Association. Separate editions were published that might be adopted by the Medicare programin developing a fee schedule. The third involves in 1956, 1957, 1960, 1964, 1969, and 1974. (Assoliciting for competitive bids from physicians or noted above, the California Medical Associationother suppliers of physician services from which was enjoined from publishing any further editions a fee schedule would be constructed. The fourth in 1979.) Other professional societies, such as the alternative involves direct negotiations between American Society of Anesthesiologists and thethe Medicare program and physician groups to American College of Obstetrics and Gynecology,explicitly develop a fee schedule. have also developed RVSs. Relative Value Scales The Value in Relative Value The concept of value embodied in any RVS is An RVS, in and of itself, is not a fee schedule. important. Differences in the concept to be usedGiven a procedural coding and terminology sysmay lead to quite different sets of relative values. tem listing all physician services, an RVS is a One might argue that the values in an RVS shouldcardinal ranking of each of those services with rereflect differences in the costs of producing the spect to some conception of valueY For exam ple, a total hip replacement might have a rankservices. This approach would tend to establish RVS based fees to physicians that did not distorting of 40.0 compared to the ranking of an inguinal their clinical decisionmaking. On average thereherniorraphy of 9.0 (76). Each service's ranking sulting approved charges would be a constantallows an ordering of that service relative to all multiple of estimated costs and there would beothers. The difference between any two services' no expectation that any one set of services wouldrankings in some sense is a measure of a differbe particularly encouraged by the payment sysence in value (192). tem. However, even if this type of RVS were toConversion of an RVS to a fee schedule is relabe based on the costs of the most efficient waystively straightforward. Assigning a monetary conof producing the services, there might be an obversion factor to a relative value unit allows the jection that some services of little or no medicalcomputation of a fee for any service: the fee is benefit to patients should not be valued at cost.simply the product of the service's relative value Alternately, therefore, one might argue thatin units multiplied by the conversion factor. Alvalues in an RVS should reflect differences in theternately, there might be different conversion facstatistically expected value of a change in healthtors associated with different types of service. status (compared to not receiving the service) of 12The units of relative value for any RVS are arbitrary. Although a patient who receives a particular service. Phyone might choose a numeraire service (228), the choice of a numer sicians might be able to acheive some concensus aire service would itself be arbitrary and none of the issues of the California RVS, for example, was based on such a numeraire. The on this issue, although patients' perceptions of thenumber of RVS units for any service has no meaning except in relavalue of physicians services might well be varied,tion to the number of units of some other service. and might differ from those of the physicians, as well as from the values that might be discerned by the Medicare program. In addition, from the latter perspective, the value of a particular service might be placed in the context of all of the other nonphysician services that might be provided in conjunction with the services in question. A service provided in an ambulatory care setting, for example, might be preferred to an apparently equivalent service provided in a hospital because the total cost to the program would be lower. Relative values, therefore, could reflect not only the costs of efficient production, including the costs of physicians' time, but also the preferences and costs of patients, the Medicare program, and probably those of society as a whole. It is unlikely that any set of RVSs would meet each of these requirements. (One might argue that, if nothing else, because historical charges represent the resultant of: 1) physician costs; 2) Medicare, insured, and private-pay patients' preferences; and 3) Medicare rules and regulations, the relative values implicit in charge histories are an appropriate source for an RVS.) As a practical matter, however, there are a variety of ways of constructing an RVS. These methods can be assessed in terms of the derivation of their relative values and possible impacts of their use for establishing a fee schedule. At the outset, it should be noted that attempts to date to compare relative value scales from various sources have found few aggregate differences among alternative RVSs (191,227). Some of those differences, however, may be significant for the choice of RVS or modification of an RVS that might be employed in establishing a fee schedule. Similarity among alternative RVSs strengthens the case for using a relatively inexpensive method of constructing an RVS. Charge-Based Relative Value Scales One first option for fee schedules would involve the use of carriers' patient history data to establish an RVS. Estimation of a central tendency measure (mean, median, specified percentile) for each physician service would establish that service's relative value. Based on the total approved charges for all procedures, a single conversion factor would be established that would make the re- Ch. 5-Payment Based on Fee Schedules • 133 suiting fee schedule budget neutral compared to CPR. Hadley and colleagues found that the choice of the central tendency measure does not appreciably affect RVS scores (191). This would argue for the use of average charge values, which are mathematically easier to compute than percentiles. If the incidence and magnitude of outliers were found not to be similar across procedures, use of the median charge might replace the use of averages. Hadley and colleagues examined submitted charges, prevailing charges, and reimbursements for specific procedures and found that the choice of charge measure also had little effect on the resulting RVS scores that might be computed from history data (191). Data used for that analysis included fee screen year 1982 national data from the Medicare Directory ofPrevailing Charges and 1978 Medicare claims data from the State of California. In light of the recent findings that allowed charges for visits appear to be a smaller fraction of billed charges than for the more technical services (247,294), one might expect that an RVS based on submitted charges would differ from one based on allowed charges, expecially with respect to visits. To the extent that submitted charges reflect current private market values that source would be preferred as a source of relative values. 13 This option is the only one for which there exists empirical data on any of the effects of a change from the current CPR system. Claims data from the State of South Carolina14 from calendaryear "Regardless of the choice, approved charges would be used to determine the conversion factor to preserve a budget neutral change to this type of RVS-based fee schedule. 14Although the State of South Carolina is relatively small and approved charges per claim in that jurisdiction are 14 percent lower than the national average, its implementation of the CPR system for determining approved charges is not believed to be unrepresentative of all carriers. In March of 1983, for example, the net claims assignment rate in South Carolina was 56.7 percent compared to 53.2 for the United States as a whole (530). In the first quarter of fiscal year 1983, the approved charges as a percentage of billed charges in South Carolina were 78.1 percent and 76.2 percent, respectively, on assigned and unassigned claims. The comparable U.S. statistics were 76.1 percent and 76.6 percent, respectively. Where South Carolina's claims processing system is different from the nation's as a whole is in its early introduction of the use of CPT-4 as the procedural coding and terminology system for physician services, a system that is now required of all carriers. For that reason, data analysis of potential chan~es in South Carolina may be representative of national effects that may be forthcoming. 134 • Payment for Physician Services: Strategies for Medicare 1983 to assess the effects of a change to a fee schedule based on average approved charges without specialty differentials (247). By design, the system was budget neutral, so there was no change in estimated Medicare expenditures. The major effect of the simulated change to a fee schedule was to increase payments to general and family practitioners and to reduce program payments to internists. Payments to surgeons and radiologists were largely unaffected, i.e., total payments for the services of those specialists remained within 1 percent of actual payments under CPR (see table 5-2). With no changes in physicians' assignment decisions following the change to a fee schedule, anticipated total Medicare revenues of physicians would change by smaller amounts than the change in Medicare payments. The reason is that beneficiary costs on unassigned claims would increase for some of the patients of physicians whose approved charges had been reduced. Juba's examination of aggregate estimated changes in physicians' Medicare revenues showed that for individual physician practices roughly two thirds of all physicians would have observed either no change in Medicare revenues or a change of less than 5 percent (see table S-3). A total of 6.4 percent of all physicians would observe losses of more than 10 percent percent, including 14.3 percent of all internists. Nearly 10 percent of all physicians would observe increases in excess of 10 percent, including 19 percent of all general practitioners and 11.3 percent of family practitioners, but only 3.8 percent of radiologists and 1.5 percent of all internists (see table S-3). Similar results were found by Sulvetta in simulating a fee schedule based on average approved charges using California data from 1980 (455). Total anticipated Medicare revenues for four out of five specialties15 studied were changed by less than 1 percent; internists' Medicare revenues were reduced by 1.64 percent. Of greater interest is the range of gains and losses within each specialty: 86.2 percent of physicians were found to experience revenues under the fee schedule within 5 percent of their previous experience (with 29 percent of physicians experiencing no change). However, 6 percent of physicians were found to experience gains of more than 5 percent, and 7. 7 percent to experience losses greater than 5 percent. The latter group included 12 percent of the internists and 10 percent of the orthopedic surgeons. Resource-Cost-Based Relative Value Scales It has long been recognized that sound reimbursement principles require that (physician) payment levels not be greater than needed to procure sufficient, high quality physician services, but also not be less than needed to reflect the costs of efficiently producing those services, including a return on physicians' investments in training. Hence there has been interest in the development of a resource-cost-based RVS. On the face of it, the steps involved in estimating resource costs should be straightforward. One begins with the enumeration of the constituent re 15The five specialties were general practice, general surgery, internal medicine, orthopedic surgery, and ophthalmology. Table 5·2.-Simulated Percent Changes in Medicare Program Payments Following Conversion to a Fee Schedule8 From CPR Payment, South Carolina, 1983 All General Family Internal General Orthopedic specialtiesb practice practice medicine surgery surgery Ophthalmology Radiology Office visits ......... 0.0 19.6% 16.6% -16.5% 1.2% -6.0% Hospital visits ....... 0.0 17.4 11.5 -8.8 6.6 Surgery ............. 0.0 8.4 0.1 -0.9 0.0 Radiology ........... 0.0 -0.1 Pathology ........... 0.0 1.3 -1.8 1.8 0.1 -4.5 All types of servicesb c ........ 0.0 16.5 11.9 -7.5 1.0 -0.6 0.1 -0.2 -Procedures in the cell account for less than 5 percent of total approved charges for that specialty. aFee schedule based on statewide average approved charges without regard to physician specialty. blncludes physicians in listed specialties and others. clncludes other medical services; excludes anesthesia. SOURCE: D. Juba, "Analysis of Issues Relating to Implementing a Medicare Physician Fee Schedule," prepared for the U.S. Congress, Office of Technology Assess· men!, Washington, DC, November 1985. Ch. 5-Payment Based on Fee Schedules • 135 Table 5·3.-Simulated Percent Changes in Physicians' Medicare Revenues Following Conversion to a All specialtiesb ........ General practice ....... Family practice • 0 ••••• Internal medicine ...... General surgery ....... Orthopedic surgery .... Ophthalmology ........ Radiology ............ Fee Schedule8 From CPR Payment, South Carolina, 1983 Reductions greater -11% to -6% to -1% to No +1% to +6% to than 25% -25% -10% -5% change +5% +10% 0.5% 5.9% 6.5% 21.9% 23.1% 22.3% 10.1% 0.0 0.4 1.0 1.0 26.0 33.8 18.7 0.0 1.4 0.0 2.8 20.9 39.4 24.1 1.0 13.3 13.0 43.5 16.4 8.3 3.1 0.0 0.9 7.9 29.0 16.4 29.4 6.1 0.0 0.0 1.9 33.7 34.6 15.4 4.8 0.0 1.0 4.9 22.6 41.2 12.8 7.8 0.0 0.0 7.6 50.6 14.2 15.2 7.6 Increases +11% to greater +15% than 25% 7.3% 2.5% 13.7 5.3 9.2 2.1 0.9 0.6 8.4 1.9 7.7 1.9 7.8 2.0 3.8 0.0 aFee schedule based on statewide average approved charges without regard to physician specialty. blncludes physicians in listed specialties and others. SOURCE: D. Juba, "Analysis of Issues Relating to Implementing a Medicare Physician Fee Schedule," prepared for the U.S. Congress, Office of Technology Assess· men!, Washington, DC, November 1985. sources that comprise the costs to be measured. These variables can be readily identified. At a very basic level, two categories have been identified: 1) all overhead costs such as salaries, rents, utilities, supplies, professional liability insurance, and other services; and 2) physicians' own time resources. There is a general consensus on the relative total costs of physician and overhead resources within individual physician practices. Net physician revenues have consistently been found to be approximately 60 percent of gross professional revenues. This result has remained virtually constant since the American Medical Association has published data on physician practice costs. HCFA survey data have also been consistent with this results. Comparable data are reported in the journal Mecial Economics based on survey data. The most recent findings published in that journal indicate that the median practice expense proportion was 38.2 percent (355). There is not much variation in this ratio among specialties, nor is this ratio much affected by legal form of organization. The current empirical literature, however, is sparce specifically with respect to resource costs for particular procedures. There has been little or no attempt to assess resource costs for actual practices. Wagner describes the difficulties involved in identifying and collecting data for assessing resource costs (561). She describes it as a "bottomup" approach since it involves measuring the quantity of each type of input involved in producing each kind of product. Unit prices are also needed for each type of input. The vector product of all of the units of input and the unit prices of those inputs yields an estimate of the costs of the final product. Crucial assumptions involved are: 1) that the observed level of utilization of capacity (of both equipment and personnel) be optimal, 2) that the organization and technology of the observed settings be optimal, 3) that the proficiency of the performers in the observed settings be optimal, and 4) that the observed quality of services provided be optimal. Any deviation from these assumptions at a minimum would introduce statistical noise into estimates derived from several different practice locations. In fact, violations of one or more of the assumptions would involve comparing different products or different inputs. For example, if measured by the gross professional revenue per patient contact minute, the apparent cost of physician time for a fully occupied physician may appear to be much less than a second physician with an identical income but with some free time for seeing additional patients. In fact, the contrary is the correct view. The opportunity costs16 for the first physician can be seen as the greater because slack time generates no additional revenue or output of patient services. Differing quality levels in the estimates obviously involve different products, but even differing technologies used to produce seemingly iden 16"0pportunity cost" is a concept used in economics generally defined as the return available from the best alternative use of a particular resource. One is told that "there is no such thing as a free lunch," because there are remunerative or at least satisfying alternatives to being treated to an otherwise "free" meal. 136 • Payment for Physician Services: Strategies for Medicare tical products may also imply different products. Similarly, practitioners of different competencies may also produce "identical" services that are far from identical. In theory this might invalidate any micro-costing study as a guide to more general application, specifically its use in a relative value study. In fact, however, some theoretical compromises are necessary once it is recognized that there is some statistical variation about any measures of average performance that may become available. Wagner examined micro-costing studies of ra diology procedures in two large teaching hospi tals, clinical laboratory tests in a British hospi tal, and six hospital-based obstetrics/gynecology procedures in a large U.S. teaching hospital (561). She did not find a great deal of correspondence between the relative values produced by these studies and the relative values published in the 1974 California RVS. Unfortunately, there was no way to examine the variations in relative values that might have been found through these studies since they were based on such small samples and each of the studies was focused on a different set of procedures. The most relevant attempt to date to estimate resource costs for physician procedures was the study by Hsiao and Stason for HCFA (227). In this study, data from the Study of Surgical Services in the United States were assembled to estimate average physician patient contact time for a selection of 50 surgical services. The authors estimated total resource cost relative values as the product of estimated average physician time, relative complexity, imputed physician opportunity costs (to correct for differences in the length of specialty training), and relative overhead by specialty:17 resource cost-relative = MD timei x complexityi x opportunity costi X overheadi valuei i where i and j refer to the ith procedure and the jth physician specialty, respectively. 17Average patient contact time in the operating room was derived directly from Study of Surgical Services in the U.S. statistics. Estimated pre-and post-operative patient time was developed by a consensus measure of visit time for any operation. Procedure complexity measures were developed using a Delphi method with a panel of 25 physicians from the Boston area. Training length estimates came from American Medical Association data, and overhead estimates came from Medical Economics survey data. This formulation assumes that the average patient contact time estimate is a fairly reliable estimate for all physicians within a specialty; that all sources of variation in required physician skill levels can be accounted for in a single complexity measure; that a physician's opportunity costs are solely related to length of training; and that overhead by specialty is uniformly related to all services within that specialty. The simplifying assumptions were necessary for any estimates to be produced. Within that context, the exercise was useful in identifying the issues involved in estimating resource costs and in demonstrating that a plausible set of estimates could result (see table 5-4). In general, there has been no controversy with respect to the specific estimates produced by Hsiao and Stason in 1979. Any current objections to the results involve the timeliness of the data used and data refinements that might increase the potential realism of the assumptions. The data for the physician time estimates used for each service were collected in the early 1970s and may no longer reflect current physician practices. The simplifying assumptions from the original study might now be relaxed or refined given better data on individual physician practices and their finances. There are some other perspectives, however, on the general problems of estimation of resource costs for payment purposes that are presented in the following sections. Should It Cost What It Costs?-Another range of issues involves the contrast of concerns between providers and Medicare as a payer. "Resource costs" is fundamentally a supply side concept; reimbursement is more often the subject of demand side considerations. In particular, one must exercise a general caution in applying any resource cost estimation methodology for the purpose of establishing relative payment levels, particularly with the purpose of identifying a "just price" (187). Empirical studies in health care and other indus tries have verified the economists' theoretical prediction that the costs of production will rise (and occasionally fall) to the level of the purchase price. The most common example cited is the cost of producing airline services prior to deregulation. When the Civil Aeronautics Board established price levels for certain airline trips, the commercial carriers' competition for passengers drove up the provision of in-flight amenities, and hence the Ch. 5-Payment Based on Fee Schedules • 137 Table 5·4.-Resource Cost Relative Values and Relative Reimbursements per Hour Implied by Medicare Prevailing Charges, Massachusetts, 1978 Relative Medicare Payment Physician service (and specialty) value prevailing charge per hour Hemorrhoidectomy . . . . . . . . . . . . . . . . . . . . 0.9 $271 $193 Inguinal hernia repair. . . . . . . . . . . . . . . . . . . 1.0a 339 218 Appendectomy . . . . . . . . . . . . . . . . . . . . . . . . 1.0 339 272 Cholecystectomy . . . . . . . . . . . . . . . . . . . . . . 1.6 570 275 Hysterectomy . . . . . . . . . . . . . . . . . . . . . . . . . 1.9 640 279 Lens extraction........................ 1.7 678 679 Suprapubic prostatectomy . . . . . . . . . . . . . . 1.8 720 399 Transurethral prostatic resection . . . . . . . . 1.6 678 475 Initial office visit (general practitioner) . . . 0.19 20 40 Routine brief office visit (general practitioner) . . . . . . . . . . . . . . . . . . . . . . . . . 0.08 10 40 Initial office visit (internist) . . . . . . . . . . . . . 0.21 34 68 Routine brief office visit (internist) . . . . . . . 0.09 15 60 aey design, inguinal hernia repair was selected as the numeraire service thereby establishing its relative value as 1.0. SOURCE: W.C. Hsiao and W.B. Stason, "Toward Developing a Relative Value Scale for Medical and Surgical Services," Health Care Financing Review 1(2):23·38, fall 1979. cost of producing those trips. Similar results have been found with respect to the costs of in-center maintenence dialysis given the HCFA's fixed limit on dialysis payments (56). This result does not even require explicit and direct competition between sellers. To paraphrase one of the earliest neoclassical economic theorists, Alfred Lord Marshall, if the price that the final purchaser is willing to pay is relatively high and relatively flexible, the sellers' purchases of product inputs will also tend to be relatively high and relatively flexible (289). Thus, if Medicare's reasonable charge levels for endoscopic examinations, for example, were initially in excess of resource costs, physicians who purchase endoscopes would be less likely to try to bid down the price of that type of equipment. Other things being equal, over time the "costs" of producing endoscopic exams would rise to the level of the purchase price, the reasonable charge. Other Resource Cost Issues.-There are several issues that have been identified with respect to the problem of resource cost estimation: • demand side adjustments to resource cost estimates that might be introduced in a relative value scale such as -identification of physician services gener ally believed to be ineffective, -identification of services whose provision should be encouraged, -identification of sets of "equally effective" physician services; • task delegation and physician time estimates; • variations in practice input unit costs; • variations in physician incomes with respect to specialty and experience; • variations in physician practice styles; and • variations in estimates of resource costs for specific physician services. Potential Demand Side Adjustments. -As noted above, there might be a need for an adjustment factor in a resource-cost-based RVS to reflect differences in the general effectiveness of specific services. Although some generally effective services may not prove to be effective with respect to a particular patient, there are some services that are generally believed to be ineffective and some, for example, gastric freezing, that have been shown to be ineffective. The costs of ineffective services-however inexpensive-need not warrant equal treatment with the costs of services of proven efficacy. On the other hand, one might also want to examine a multiplier adjustment for such services as effective preventive care such as pneumococcal vaccination, if it were believed that payment levels above costs would lead to the additional provision of such services. Finally, a possibility exists that within the set of generally effective physician services, there will be sets of equally effective services that are substitutes for one another. In this case, a pure demand side ap 138 • Payment for Physician Services: Strategies for Medicare proach would require that payment for the two services be equal and set at the level of the lower priced service. Physician Time Estimates and Task Delegation. -It is expected that the most significant observable resource involved in the production of physician services would be individual physician time. The varieties of styles of medical practice, however, suggest that physicians have a great deal of choice with respect to whether they individually perform certain tasks or delegate those tasks to appropriately trained staff. For example, many medical practices routinely delegate blood pressure testing to nurses or physician assistants. If not properly accounted for, the substitution of staff time for physician time could confound any resource cost estimates, or at least introduce additional variation in the estimates of the averages. A common approach has been to measure only direct physician time and to allocate all costs for support personnel to physician overhead without regard to the specific services in which those personnel may participate. This relatively simple approximating strategy has some merit. The single most common category of employee in physicians' offices is "secretary, receptionist, bookkeeper" (499). These employees are unlikely to be directly involved in the production of medical services. However, there are a significant number of physician assistants employed in medical practices, and appropriate weighting strategies for including those costs would have to be explored (490). Variations in Practice Input Unit Costs.-Surveys of physician practice costs and incomes by HCFA, the American Medical Association and Medical Economics have been successful in eliciting information from physicians on their total annual expenditures for practice inputs. The survey approach has been somewhat less successful in estimating unit costs because of data definition problems and difficulties in determining the appropriate measures of units. Annual expenditure statistics are clearly less burdensome to collect than would be unit costs: there would also be less required detail. As a result, however, the impact of differences in unit costs on practice decisions is unknown. The notion of the estimation of average resource costs of specific physician services from a sample of practicing physicians would rely on the assumption that there is some degree of uniformity in the production of those services. Although there exist certain (recipe driven) production processes in which there are no choices with respect to amounts of inputs, physicians can and do make choices with respect to the organization of their practices. Although for the most part physicians' business decisions rather than their clinical decisions are involved, these choices affect the costs of providing clinical services and may affect the estimates of the costs of those services. A simple example can be seen in decisions with respect to office space. In an area of relatively low rentals, a physician practice may acquire office space with relatively large rooms. The net impact on actual rental overhead or absolute rent expenditures is uncertain. Costs could be higher, lower, or identical with the corresponding expenses of practices in areas with higher rental rates. However, in the absence of good data on the unit costs of each practice, use of average rental rates for estimation will make it appear that this practice carried either a relatively larger overhead for office rental expense or absolutely larger rental expenses for each physician service performed in the office. Variations in Physician Incomes With Respect to Specialty and Experience.-The remaining unit cost of interest is the physician cost. Except for the case of salaried physicians, for the most part this statistic is not directly available. Relevant estimates that are currently available relate to physician net incomes after practice expenses-that often include deferred income. As might be expected, net incomes are computed as residuals. The unit cost measure that results is net professional revenues per year or per hour. Actual net income is only one of several available means of valuing physician time. The most simple method of valuing the physician resources employed in the provision of a specific service would be to multiply the average net income per hour and the time, in hours, used to produce the service. Actual incomes embody the results of so Ch. 5-Payment Based on Fee Schedules • 139 many individual choices, however, that computation based on any small sample of practices might not lead to accurate results. The alternate method of establishing a relative value of phyician time used in the literature is to base those values on relative complexity and opportunity costs (227,271). Variations in Physician Practice Styles. -One of the implicit assumptions involved in the development of any resource cost method for potential use in payment reform would be that there are not too many clinical options with respect to performing specific physician services. The underlying distribution of tasks and time would exhibit only some limited variation. If, however, there is a continuum of clinical options, then the averages generated from the observations in an estimation study might not represent any one style of practice. Payment rates derived from such estimates might be too high for some styles of practice and too low for others. For example, in estimating the total .approved physician charges associated with Medicare beneficiaries hospitalized in particular diagnosis-related groups, Mitchell, et al. found that the costs attributed to any observation were significantly affected by the presence of an assistant at surgery (320). Payments based on the average would be too high for cases without an assistant, too low otherwise. Given variation in practice styles across physicians performing the same service, one might expect a comparable result. Variations in Estimates of Resource Costs for Specific Seroices.-Finally, for all of the reasons discussed above plus any other natural occurrences of variation, one would expect that there would be variation about any average resource cost estimate that can be produced for an individual physician service. The relative size of the variations among a set of services whose resource costs have been estimated is crucial. Even with the most accurate estimation method, resource costs estimates that are not found to be significantly different from one another due to inordinate variation will not prove to be a compelling guide to reimbursement reform. Relative Value Scales Achieved Through Negotiations/Consensus Development Both the charge-based and resource-cost-based RVSs would be derived primarily from empirical analysis of quantitative data. The former would involve a somewhat mechanical determination of central tendency values from distributions of charges; the latter would involve a somewhat more thoughtful examination of physician practice cost data, perhaps supplemented with information on the relative complexity of various procedures. An alternative that would place greater reliance on physicians' professional judgments might involve explicit negotiations or consensus development processes to achieve an RVS. The developers of previous RVSs have employed professional judgments in modifying the results obtained from statistical manipulations. For example, in the preface to the 1960 version of the California Relative Value Studies, it is noted that although basic relative values were established by statistical analysis of data from 6,800 physicians with respect to roughly 600 procedures (75): [I]n a few instances it was apparent from anal ysis of the data and from consideration of sub sidiary data that strict adherence to the survey values would produce unrealistic results. In such -instances, values were set by consultative means. Thus, the use of professional judgments in the establishment or revision of an RVS through negotiations or consensus development procedures is not unprecedented. In fact, considerable physician consensus with respect to relative values has been shown to exist. In the 1950s, Horton demonstrated this fact through analysis of surveys of physicians in Connecticut and Montana (225,226). Hsiao and Stason found such consensus within a set of surgical services although not between surgical services as a class and office visits as a class (227). Recently, Egdahl and Manuel have used a consensus development process to rank surgeries with respect to complexity and severity (119). Types of Decisions.-One should make distinctions between the various types of decisions that 140 • Payment for Physician Services: Strategies for Medicare might be the subject of an RVS negotiation or consensus development process. (For the purpose of discussion they will be described as changes in relative values. The same types of decisions would also have to be made in the initial development of an RVS.) Negotiations with respect to the value of a conversion factor or factors used to transform an RVS into a fee schedule are one example, but will be discussed in a later section. Other examples are as follows: • changes in relative values with respect to a numeraire service whose price is to remain fixed, • changes in relative values for a class of services while the price of all other services remains fixed, and • changes in relative values subject to the constraint that projected total relative values be fixed given anticipated volumes for each service. The relation or lack of relation between the RVS and one or more prices in a fee schedule changes the nature of the negotiations; hence the distinctions made above. The first example might involve specific services perceived to be either undervalued or overvalued with respect to actual payment levels; hence the need for revisions in a relative value scale without a general change in conversion factors. The second example is itself best exemplified in the current discussions with respect to the relative values of nonprocedural services. One might expect discussions on the merits of raising the relative values of those services while holding the conversion factors for all other services constant, the expectation expressed by FTC (554). Alternatively, one might discuss the merits of reducing the relative values of all other services, holding the conversion factor for nonprocedural services constant. The final example would involve "pure" changes in relative values, but this would imply much less certainty about results, since all prices in any fee schedule derived from the RVS in question would be subject to change.18 18 The process might be as follows: Given a revised set of relative values, one could sum the anticipated total of relative value units, assuming that the volume of services from a prior time period would remain unchanged. Given that volume, one would compute the con version factor implied by a specific budget target. Since any set of relative values would imply its own conversion factor, all prices in the resulting fee schedules would be changed. Possible Outcomes.-Berenson has discussed various options that might be employed in establishing a relative value scale through group-decisionmaking processes (33). He suggested that although costly, a commission representing the community at large might be better able to develop an RVS that reflected a broad view of what should constitute the values of a range of medical services than a commission constituted of "experts" representing each of the various specialties. He also noted that there are no models available that could be used to predict the impact on the relative value of any service that might result from such complex decisionmaking processes. Experience to date with negotiating systems used in other countries gives little additional guidance on outcomes. The West German sickness funds have tried without much success to reduce the relative fees paid for laboratory services and to raise fees for basic medical services so as to increase the relative incomes of general practitioners (162). However, most governments that have been involved in such negotiations have been perceived to be more interested in adjusting fees to control expenditures in the aggregate rather than in finetuning with respect to individual services (33). "Competitively" Developed Fee Schedules The CPR system, to a certain extent, is neutral with respect to price competition in the markets for physician services. Physician fees evolve, for whatever reasons, and the CPR process educes a set of approved charges from the middle to upper ranges of the fee distribution. At best, this is a passively competitive posture: Medicare as a price taker without searching for the lowest price. However, more competitive postures for the program are available. These would include pricing policies that would be more directly analogous to perfect competition, bilateral monopoly, or the use of the purchasing power of a monopsonist. 19 19Perfect competition describes an idealized market for a homogeneous good in that there are a substantial number of both (cost minimizing) buyers and (profit maximizing) sellers, no one of which has a direct effect on the price of the good in question. (Hence, the phrase "price taker.") Bilateral monopoly describes a market in which there is a single seller and a single buyer. Monopsony describes a market in which there are many sellers but only a single buyer. Perfect competition and monopoly (or monopsony) are not opposite ends of a spectrum with respect to competition. The op HCFA has used various analogs of such reimbursement policies for the purchases of pharmaceuticals, durable medical equipment, and laboratory tests (292). Four approaches to "competitive" physician payment policies are discussed below. The first two involve unilateral buying policies that might be adopted by the Medicare program in developing a fee schedule. The third involves soliciting for competitive bids from physicians or other suppliers of physician services from which a fee schedule would be constructed. The fourth alternative involves direct negotiations between the Medicare program and physician groups to explicitly develop a fee schedule. Lower the Price In theory, a virtue of the CPR system is that it allows beneficiaries in any locality the expectation that a significant fraction of the charges for a physician service will be covered by Medicare almost without regard to the beneficiary's choice of a physician. Only those beneficiaries who receive services from the most expensive doctors should expect to have substantial additional liabilities, and those liabilities would apply only on nonassigned claims. This flexibility was one of the reasons that prompted the National Association of Blue Shield Plans to adopt the UCR20 concept in the mid-1960s. Perhaps equally important, UCR offered the Blue Shield Plans the opportunity to sell a product that would nearly always give subscribers paid-in-full benefits without the additional cost of setting (national) fee schedule payments at a level high enough to guarantee physician participation in all parts of the country (312). Obviously, the same would be true for Medicare under CPR even if the jurisdictions in question were each of the States as an alternative to the Nation as a whole. posite of competition-whether perfect or monopolistic-involves a lack of (low) price searching by buyers and a lack of purposive behavior by sellers to either maximize profit, surplus, or market share. 20UCR stands for "usual, customary, and reasonable charges," the pricing concept used primarily by Blue Shield plans. It was developed prior to the introduction of Medicare, and was the model adopted for CPR. Blue Shield's "usual" charge became Medicare's "customary" charge, Blue Shield's "customary" charge became Medicare's "prevailing" charge. Ch. 5-Payment Based on Fee Schedules • 141 Although the degree of physician participation was and is an important marketing consideration for Blue Shield Plans, it is not clear that the Medicare analog-physician acceptance of assignment -is as important for the Medicare program given that: 1) carriers do not act as insurers that underwrite the Medicare program, and 2) there is recent evidence from the Medicare participating physician program that a significant number of physicians will agree to accept assignment on 100 percent of claims (521). Assignment is important to both Medicare beneficiaries and to the Medicare program. However, a fee schedule alternative to CPR need not involve a relatively high price uniform in all jurisdictions to guarantee high assignment statistics. A relatively low price may suffice in some localities. Specifically, if an approved charge of $2,000 is necessary to elicit an assignment rate of 50 percent for cataract extraction operations in New York City, that does not imply that $2,000 should be the approved charge for that operation for all of New York State, much less all of the country. Lower prices might elicit equal or higher assignment rates for that operation in jurisdictions outside of New York City. One might establish fee schedule payment levels below current prevailing charge levels. In fact, a relatively low percentile level might be selected, such as the 50th percentile of approved charges or lower. (Under Medicare's "lowest charge limitations" applied to certain laboratory tests and items of durable medical equipment in the late 1970s, payment levels were restricted to the 25th percentile.) With few exceptions, use of the 50th percentile would produce a fee schedule comparable to that that would be produced by using average approved charges to develop a relative value scale. If the distributions of approved charges for individual physician services are skewed to the left, selection of the 50th percentile as the fee schedule standard might have a slight downward effect on total Medicare expenditures for physician services. Such a decline, however, would be moderated if there were volume increases observed with respect to physicians' experiencing reductions in approved charges. Find the Lowest Sufficient Price A potential difficulty with the use of the 50th percentile as the fee schedule amount is that in 142 • Payment for Physician Services: Strategies for Medicare some jurisdictions that amount will be too high Solicit Bids and in others too low to secure sufficient access If there were some uncertainty that physicians for beneficiaries to needed health care services. would supply sufficient services to Medicare benWere the level too high, one might observe an eficiaries under either of the two empirical meth increase in the number of physicians becoming ods of establishing fee schedules from existing dis Medicare "participating practices" and an increase tributions of approved charges, an alternativein the percentage of claims for which assignment approach to competitively procuring such servwas accepted. In contrast, were the level too low, ices would be to solicit bids. This approach might one would expect to observe a decrease in physi take the form of exclusive or semi-exclusive bidcian participation or acceptance of assignment. ding. Under the former, a single physician group Rather than merely reacting to these changes or consortium of groups willing to supply up to in beneficiary access, a policy might be initiated a specific quantity of a particular service for a to explicitly identify the lowest fee schedule fixed unit price would bid for the exclusive right amounts subject to the constraint of achieving to provide those services to the Medicare popucomparable expected levels of beneficiary access lation in a specific geographic area. Obviously this in each jurisdiction. For example, carriers might would imply restricting beneficiary freedom of choice in that area. Partly for this reason, it might be instructed to identify the lowest approved charge for a particular physician service that be particularly applicable to such services as exwould include 25 percent of the physicians who tracorporeal shock wave lithotripsy (ESWL), had provided such services to Medicare beneficiwhere a relatively small number of providers can aries. (Once the fee schedule amount had been be expected in each market. This approach, however, would seem to be inappropriate for a serv established, however, any and all physicians could provide the service at that price.) Alternatively, ice such as the provision of pneumococcal vac the carriers might be instructed to identify the cine that is easily and generally available from lowest approved charge (for each service) that many providers. would encompass a particular fraction of assigned Alternatively, a semi-exclusive approach mightservices in each locality. be tried. Under that approach bids for particular By design, these types of procedures for estabphysician services would be solicited from all lishing fee schedules would draw maximum fees practices in a particular locality. These bids would from the lower end of the distributions of apbe in the form of both a price and an expected proved charges rather than the upper end as in quantity of service to be supplied at that price. the current CPR system. The pricing philosophy All physician bids would be examined to identify in this case is analogous to that used in the Maxithe lowest bid price sufficiently high to provide mum Allowable Cost program that HCFA has imthe expected utilization of the service in question. plemented for purchases of pharmaceuticalsAgain, any and all physicians in that locality primarily in the Medicaid programs (261). might be allowed to provide the specific service at that price. In fact, much as in the U.S. Treas To remain competitive within the Medicare sysury's auctions for its bills, physicians might bidtem, physicians would have to restrain their fees "the auction price" for their expected provisionor possibly subject their Medicare patients to subof the service in question, explicitly accepting the stantial amounts of nonassigned liability. Alter price to be determined by the bidding. This apnatively, the physicians whose fees were at or beproach would work well for pneumococcal vac low fee schedule amounts might find additional cinations. However, it might vitiate any poten Medicare patients requesting their services in lieu tial Medicare market advantage in procuringof continuing to obtain services from the relatively lithotripsy or magnetic resonance imaging servhigher priced physicians in a particular locality. ices if all bidders knew they might participate atIn contrast, under the current CPR system (as supthe winning price.plemented by Medigap insurance), there is little or no advantage -to a physician in having relaA problem with either of these potential bidtively low fees. ding schemes is the multiplicity of both services and localities. Conduct of the bidding even on just a one-time basis could be extremely complicated. Further, under a simpleminded version of the exclusive bidding option, one might expect absurd results, such as one practice in the northern part of a city with exclusive rights to limited office visits, while the rights to limited hospital visits might be won by a competing practice on the other side of town. There is also the potential problem of the creation of a local monopoly for a single winning bidder. If market entry required substantial resources, a single winning bidder without competition might successfully resist subsequent Medicare cost containment initiatives. For these reasons, semi-exclusive bidding might be conducted with respect to relative value unit conversion factors for a complete set of physician services. Alternatively, exclusive bidding for a relatively compact set of related services might be conducted with any additional services not in the bidding set to be priced based on the RVS conversion factor implied by the bid. Finally, exclusive bidding might be used only for relatively homogeneous services with high expected volume or expenditure levels, such as cataract excisions (with intraocular lens insertions) or laboratory tests. All other services might be priced using one of the other alternative approaches to developing a fee schedule. Negotiate The final "competitive" option would involve explicit negotiations between the Medicare program and physicians providing services to Medicare beneficiaries. In theory, this could take the form of service-by-service discussions to arrive at a fee schedule, although more likely would be negotiations with respect to conversion factors to be used with an existing RVS. The latter has been Ch. 5-Payment Based on Fee Schedules • 143 the more commonly observed pattern among government programs in other countries (33). In Canada, for example, the Ontario Health Services Insurance Plan has adopted every version of the RVS promulgated by the Ontario Medical Association, negotiating primarily with respect to the conversion factor (578). Due to FTC interpretations, this is an option that exists only for Medicare or other governmentsponsored programs, such as Medicaid. Blue Shield negotiations with physician groups, for example, would be prohibited although physician input in the form of discussions about fees would not be prohibited. The problems facing the Medicare program in implementing negotiations, however, would start with identifying a group with whom to negotiate. The American Medical Association is the largest single association of physicians in this country, but its membership includes just more than half of all U.S. physicians. A new physician group might have to be constituted to sit on the other side of the negotiating table. Other countries with experience in negotiating fees have tended to recognize existing physician associations. In Canada, the Medical Care Act of 1966 established that a uniform schedule of fees would be negotiated periodically between the medical association of each province and the provincial agency responsible for their payment (28). In West Germany, the Cost Containment Act of 1977 mandated the establishment of a National Health Conference including all major interest groups active in the health care sector, specifically including the associations of sickness funds' physicians. A national relative value scale is periodically negotiated between the association of sickness funds and the associations of sickness funds' physicians (387). IMPLICATIONS OF PAYMENT BASED ON FEE SCHEDULES The dimensions by which to assess the converbe speculative, because of the lack of data availsion of Medicare physician payments from CPR able to examine even the initial changes in fees to a fee schedule are: quality of care, access to that might be wrought by a conversion to fee care, cost, technological change, and administraschedules, much less to project behavioral changestive feasibility. Much of the following analysis will that might be induced thereafter. In addition, al 144 • Payment for Physician Services: Strategies for Medicare though the conversion to a fee schedule for the most part would be a quantitative rather than a qualitative change in payment policy, in the aggregate the expected effects of this type of change would be small. The CPR system is basically a fee schedule system with physician-specific fees for many services of most physicians and localityspecific and/or specialty-specific fees for the remaining services/physicians. In that both CPR and any of the alternative fee schedules involve fee-for-service payments, the impacts of a switch to the latter may rest solely on any difference in the level of payment rather than in the method of fee determination. Given that a physician's clinical choices with respect to specific services are also influenced by considering patient preferences, available practice resources, and medical indications with respect to modalities within the physician's repertoire, other things being equal, the effects on clinical choices of changes in that physician's approved charges for specific services should be small. More likely are changes in physicians' entrepreneurial decisions with respect to agreeing to provide services to Medicare patients in the first place and/or accepting assignment on those services. Paradoxically, conversion of Medicare physician payments from CPR to a fee schedule system would both make more rigid the structure of relative values for physician services and allow more Government flexibility in changing those relative values. The use of a fee schedule for payment purposes would imply that the ratio of the approved charge for any service to the approved charge for any other would be a constant. If a fee schedule were based on a single national relative value scale, such ratios would be fixed for all services in all jurisdictions. Because of this, changes in a single fee in the schedule could dramatically change relative values. In theory, correcting any perceived imbalances in approved charges, such as those involving procedural/nonprocedural differences, would be simpler in a fee schedule world than in a world of CPR. But because the effects of such changes would be more pervasive under fee schedules, there might be more resistance to such change. For this reason, advocates of such changes might prefer that procedural!nonproce dural imbalances be corrected in the initial implementation of the fee schedule. Quality and Access There are no data with respect to the relationship, if any, between quality of care and method of payment. And even any such relationship between quality and level of payment as exists would be unlikely to be discerned if a fee schedule conversion led to relatively small differences in payment levels. In terms of the technical quality of performance of specific services-once the choice has been made to provide those servicesprice can be expected to be of little importance in the short run. 21 In the long run, however, lower Medicare payments might lead physicians to obtain lower quality supplies, facilities or personnel. The quality impact of price-in this case, the level of the Medicare approved charge under a fee schedule-is likely to be indirect through its effects on access to particular physicians and the facilities in which they practice. That impact will depend on the opportunity cost to the individual physician of the use of his or her time to attend to an alternative, non-Medicare patient. To the extent that private insurance pays a physician higher amounts than Medicare and to the extent that patients with such insurance demand the physician's services, that physician's Medicare patients may not receive as much time or attention as otherwise. And to the extent that those private patients recognize quality and demand the services of physicians perceived to provide relatively high quality services, the opportunity costs for those physicians of attending to Medicare patients will be higher. If it were the case that physicians who provided relatively high quality care indeed perceived higher opportunity costs in the private market (regardless of the level of their Medicare approved charges relative to peer physicians) and responded by not participating in the Medicare program, quality of care for Medicare patients might decline. 21The "short run" denotes a period of time during which physicians' capital and other resources cannot be changed. By construction, therefore, the costs of discriminating between patient payer classes with respect to quality could be substantial. How the resulting level of quality would com pare to that under CPR is uncertain. The major theoretical difference between a fee schedule and CPR is that CPR would allow a higher payment to a physician who, because of superior quality, had a higher customary charge. Of course CPR allows higher payments to any physician with higher customary charges, regardless of origin. Hence, it is unclear that only the physicians of highest quality are disadvantaged by Medicare payment levels. Within the category of fee schedules, any options that eliminated specialty differentials or locality differentials might also affect quality in this regard. In general, quality could be enhanced to the extent that physicianswho provide relatively high quality services respond to increases in their approved charges by increasing their participation in the Medicare program. Other physicians, however, might also respond comparably. Further, quality could be reduced in response to an aggregate increase in approved charges to the extent that those increases lead to a increase in the probability of beneficiaries' receipt of services of inappropriately high risk or of little effectiveness. These gross effects can be anticipated with any physician payments regardless of whether there are increases or decreases in average approved charges. Which effects will predominate cannot be predicted, a priori. If the initial estimates from the resource cost based RVS approaches are correct (227), one might expect that the approved charges for office visits would increase relative to surgical services under a fee schedule derived from such a study. If the approved charges were realigned by raising average approved charges for office visits relative to current levels without changing the aver age approved charges for other services, one might expect an increase in the provision of the nonprocedural services. This increase would improve quality of care only to the extent that the expected value of changes in health status attendant to such visits exceeds current approved charges. (There is no evidence, however, of current "underuse" of such services given current levels of approved charges.) If average approved charges were lowered for the surgeries leaving of- Ch. 5-Payment Based on Fee Schedules • 145 fice visit average approved charges unchanged, one would not expect an improvement in quality resulting from an increase in the provision of primary care. However, to the extent that surgical services that are not risk-free are provided in response to current approved charges that exceed either costs to the physicians or benefits to the Medicare patients, a decline in surgeries might lead to an improvement in quality for Medicare beneficiaries. Access and Assignment As indicated in chapter 2, a positive correlation between the level of Medicare approved charges and assignment has been well established. Conversion of Medicare physician payments from CPR to a fee schedule based on average or median (or some other central tendency measure of) approved charges would result in increases in approved charges for some physicians for some services and decreases for others. Therefore, one would expect a decrease in the probability of assignment, being accepted in those instances where approved charges were reduced and an increase where approved charges were raised. (Similarly, one would expect an increase in the probability that a physician would become a "participating physician" if his or her approved charges had been increased.) Unfortunately, use of the available models of assignment to make projections can only provide aggregate expected effects; in particular, a budget neutral fee schedule of any variety would be estimated to have an expected zero net impact on assignment. Within such models individual beneficiaries would be projected to experience increases or decreases in assignment with the accompanying changes in liabilities for physician services. More refined models than those currently available would have to be developed and validated to estimate specific supply responses and allow a more realistic estimation of aggregate changes in response to conversion to a fee schedule embodying a specific level of aggregate fees. If Medicare converted to fee schedules and imposed mandatory assignment, some physicians could be expected to no longer provide services to Medicare beneficiaries. Those physicians who 146 • Payment for Physician Services: Strategies for Medicare dropped out of the Medicare program would be If a fee schedule conversion embodied a seleclikely to be those with relatively high billed tive reduction in average approved charges-e.g., charges compared to their peers in individual loa reduction in approved charges for services for calities. This could reduce beneficiaries' access to which costs were believed to have declined subcertain types of physicians. If, on the other hand, stantially since their introducton-savings might the fee schedule amount was established only as accrue to Medicare. If there were no change in the Medicare allowance and not necessarily imthe volumes of such services, the savings would plying payment in full, beneficiaries' access to be proportional to the reduction in approved physician services would become primarily a charges. If physicians reduced the provision of question of their personal finances. Beneficiary such services, the savings to Medicare would be greater. Further, to the extent that the financial financial barriers to access to the services of physicians with relatively high billed charges could incentives in the current high payments cause inbe increased. With respect to any single physician appropriately high utilization levels for such services where there is also patient risk, a reduction so affected, this would hurt his or her poorer Medicare patients more than the more affluent in use might imply an improvement in quality for ones. Medicare beneficiaries. If physicians responded to reduced approved charges by increasing volumes the cost reductions and potential quality enCosts hancements would be smaller than otherwise. Medicare Program Costs In order to examine and estimate changes that might occur under a budget-neutral conversion With respect to Medicare expenditures for phy to fee schedules that simultaneously reduced some sician services at any point in time, the cost im procedural/nonprocedural imbalances, a simula pacts of a change to fee schedules would depend tion analysis was conducted (247). Assuming there more on the level of payment than the method were no charges in volume, if approved charges of fee determination. Assuming that a budget neu were unchanged on average, but the fee schedule tral conversion to fee schedules were imposed, one introduced was designed to "adjust" approved might expect little initial impact on total Medi charges to increase payments for office visits care Part B expenditures. On the other hand, one chosen to illustrate the effects of raising the relamight speculate that physicians whose approved tive approved charges of nonprocedural services charges were constrained would respond by in to levels commensurate with the estimates from creasing the intensity or quantities of services earlier studies (227), total Medicare costs, by as billed, such as billing for longer, and more expen sumption, would be unchanged, but payments for sive, visits or providing additional ancillary servoffice visits would nearly double (247). Revenuesices; hence there might be some increase in costs. for general practitioners and family practitioners (Those physicians who experienced an increase would increase 50.8 and 37.3 percent, respec in approved charges might not raise their charges tively. Internists' revenues from Medicare wouldso much as otherwise in future years, but they be nearly constant, but radiologists and surgeons would not be expected to bill for fewer or less ex would experience declines. pensive services on average in the year of the con version.) As indicated in chapter 2, the evidence If the fee schedule were initially based on aver with respect to physicians' volume responses to age approved charges, but-for illustration-ap changes in approved charges is equivocal. Unless proved charges for office visits were increased as there was a substantial volume response, under above, holding all other fees in the schedule con a fee schedule conversion conducted to coincide stant, total Medicare physician payments might with the advent of a new fee screen year (that increase by 3.9 percent, including a 36.7 percent would be accompanied by higher aggregate apincrease in payments made for office visits assum proved charges regardless of the conversion), any ing that carriers paid the lower of the billed charge initial cost impacts might be undiscernable. or the fee schedule amount (247). Paying the fee schedule amount in all cases would require additional increases in expenditures. Changes in Medicare costs over time might be influenced by a change to fee schedules for two reasons. The first involves the fee screen updating process. Under CPR this is a mechanical, if not mindless process. Increases in approved charges are somewhat limited by MEl, but there are still a considerable number of services not constrained by the Index. Further, average approved charges can increase by more than the increase in MEl even for those services where the prevailing charge is established by MEJ.22 A potential virtue of a fee schedule is that the entire price structure can be controlled during the fee schedule updating process. In fact, a study of physician payments in Medicaid programs found that expenditure increases were lower in those States that used fee schedules compared to those that used CPR approaches to fee setting (215). A second, and much less likely, reason why Medicare expenditure increases might be reduced under a fee schedule regimen involves the relative approved charges for preventive care. This argument suggests that if, for example, under a resource-cost-based RVS, approved charges for office visits were increased, a greater number of preventive care services would be provided. As a result, there would be a reduced need for acute curative services in later time periods. Although there might be an initial increase in expenditures given the increase in approved charges for the nonprocedural services, the rate of increase in total expenditures-if not the level of expenditures -would decline. The cogency of this argument is reduced, however, by recent evidence that has not verified that those persons who forgo preventive care in one time period experience greater costs in future time periods (343,348). Furthermore, if greater use of preventive services increased life expectancy, the total Medicare expenditures would probably increase as survivors incurred medical expenses in their additional years of life (437,485,576). 22The MEl will constrain the increase in the maximum approved charge for a given service. UntillOO percent of the volume of a particular service is limited by the MEl, the average approved charge for that service can increase faster than the maximum. 56-119 0 -86 -6 QL 3 Ch. 5-Payment Based on Fee Schedules • 147 The additional effects on the Medicare program's costs of any changes in assignment policy that accompanied a conversion to fee schedule should be neglible compared with a fee schedule conversion without assignment changes given that Medicare only pays that portion of the bill equal to the approved charge. To the extent, however, that a mandatory assignment policy reduced the participation of physicians with relatively high charges, declines in expenditures that might otherwise have been made for the services of such physicians might exceed the increase in payment levels for physicians whose prior approved charges had been relatively low. Beneficiaries who formerly received services from physicians with above average approved charges who elected to no longer accept Medicare patients would be expected to either switch to less expensive physicians or forgo the use of services that might otherwise have been provided. Both effects would tend to reduce aggregate Medicare obligations. A net increase in Medicare expenditures would be expected only if the above average charge physicians who remained in the program increased volumes by more than enough to offset the reductions effected by beneficiaries' receiving services at or below the previous average approved charge. Under a fee schedule implemented as the Medicare allowance only, beneficiaries might reduce their utilization of services in the aggregate. This might lower Medicare expenditures, but primarily by shifting Medicare costs back to the beneficiaries. Whe~her an. expenditure cap would, in fact, cap expenditures IS an open question. For example, the evidence from the Canadian province of Quebec has been interpreted to both support and refute the effectiveness of an expenditure cap in the form of individual physician revenue limits-a system with direct rather than indirect physician incentives under a payment system with a single payer rather than many payers as in the United States. It is alleged to have produced gaming behavior on the part of the physicians (388), but other Canadian observers conclude that the limits were set so high that they may not have had any aggregate effect (135). An expenditure cap system with less direct incentives for individual physicians would be unlikely to be more effective in constraining expenditure increases. 150 • Payment for Physician Services: Strategies for Medicare duction of health care services requiring inputs in addition to physician services except to the extent that those inputs are complements to physician services. Nor would greater efficiency be expected in the combination of services used to treat a medical condition or in the weighing of the costs and benefits of services. Technological Change Technological change may be one area in which method of fee determination can have an effect in addition to level of payment. In this regard, as long as payment rates are determined prospectively without regard to costs, it will always be the case that the higher the level of potential payment, the greater the potential return to innovation, particularly cost-saving innovation. Interest in cost-saving innovation, however, may be greater the less is the difference between payments and current costs. Quality-enhancing innovations that involve increases in cost would probably be advanced more under current CPR than any other alternative physician payment system because payments are based on charges that can be increased to reflect increases in costs. Such innovation would thrive more under most fee schedules than under packaging or capitation. A new fee for a new service would probably be introduced within a fee schedule system. But under packaging and capitation, adoption of cost-increasing technologies would add to cost but not to revenue, and payment recipients would have little financial incentive to adopt such technologies except to prevent losing patients. Where the physician payment includes both professional and technical components, payment levels can provide (or fail to provide) incentives for technological change, particularly with respect to cost-saving innovations. However, where total payments for a service are split between physicians' professional components and facility or equipment expenses, the effects of physician payments on technological change are uncertain. Further, the specific effects of Medicare physician payment policies may be negligible with respect to any technologies where Medicare beneficiaries are only a small fraction of the relevant patient population and hence where innovation and diffusion may be driven by the policies of private insurers. Even where Medicare policies may make a difference, as long as Medicare institutional payments to hospitals, outpatient departments, and ambulatory surgical centers are large compared to payments to physicians for particular services dependent on acquisition of resources paid by Medicare intermediaries under either Part A or Part B, innovation in such physician services may depend more on institutional payment policies than physician payment policies. If those facility or equipment expenses are not recognized as covered services or if institutional payment levels are too low, many physicians may be unable to secure access to such equipment for their Medicare patients regardless of how remunerative the Medicare approved charge for the professional service may appear to physicians. (Although hospitals may continue to compete for physicians and patients by attempting to acquire "prestigious" and costly new equipment, there have been neither studies nor anecdotes to suggest that such acquistions have been associated with the payment levels for physician services associated with the use of such equipment.) Only where the physician payment levels were too low might physicians not adopt certain technologies for their Medicare patients even where the institutional payment policies did not inhibit acquisition of the resources required for the technology in question. Within the fee schedule options, treatment of new services will almost always be an incidental matter. Obviously, vendors of new services can be expected to recommend high approved charges rather than low ones. Therefore, advocates of any new potentially cost-saving service, such as extracorporeal shock wave lithotripsy, will want to argue on relative value grounds for establishing an approved charge based on an existing substitute service. Advocates of potentially cost-increasing innovations such as MRI for many conditions would prefer resource-cost-based approaches. Were exclusive competitive bidding to become a generally accepted means of establishing fees, the potential advantages to providers of cost-saving innovati~:ms would increase, spurring additional innovation along those lines. However, to the ex tent that competitive procurements were periodically reopened for bids, the potential physician returns to innovation can be expected to be bid to zero. The diffusion of technologies whose approved charges are lowered is likely to be retarded. Administrative Feasibility Compared to CPR and packaging, fee sched ule administration would be easier. CPR, in par ticular, requires the equivalent of maintenance of individual fee schedules for each physician prac tice. Under all of the fee schedule approaches con sidered, the need to retain physician-specific fee data would be eliminated although aggregate fee data might be retained for updating purposes. In addition, beneficiary and provider inquiries should be reduced because the payment levels for any service can be established and disseminated in advance. With respect to updating, relative ease of administration would depend on whether replication was selected as the means to update the fee schedule over time. Some of the fee schedule options would require fairly elaborate construction efforts for an initial implementation. Replication in those cases would be costly. None of the fee schedule approaches would necessarily require replication for updating. In fact, updating through the use of an index such as the current MEl or a more refined index could be performed as a purely ministerial exercise for any fee schedule option. Periodic examination of fee schedules for recalibration within the schedule or for proper evaluation of new services would be a useful adjunct to any of the options. Here again, replication is an option for such periodic examinations, but not CONCLUSION The primary potential advantages of fee schedules are rationality, predictability, and simplicity, and therefore ease of understanding for both beneficiaries and providers. In addition, a fee schedule system would not involve the maintenance of what amount to individual price sched- Ch. 5-Payment Based on Fee Schedules • 151 a requirement. Combinations of fee schedule development methods for this purpose would not be illogical. As indicated in an earlier illustration, one might initially change to a Medicare fee schedule by having carriers estimate average approved charges for each procedure to establish a baseline RVS. (For payment purposes, this RVS might be converted to a fee schedule that might be updated each year using the MEL) New procedures might be given interim payment rates following a consensus development process. Final payment rates could be established following estimations of resource costs, perhaps 18 to 36 months after the interim rates had gone into effect. Finally, the members of an independent physician payment review commission might review or recommend changes to correct any interjurisdictional or interspecialty differences brought to their attention. There are major differences among the fee schedule options in terms of the efforts required for implementation. Use of historical charge data by the carriers clearly would be the easiest method. Estimation of the lowest prices needed to procure certain levels of (assigned) utilization would be straightforward, but would require greater effort. Development of resource-cost-based relative values for even a significant fraction of the over 7,000 available procedures in HCPCS would be a substantial undertaking. Unfortunately, the efforts required for consensus development, competitive bidding, and fee schedule negotiations cannot be estimated at this time. There is little or no experience in the use of these methods specifically for the purpose of pricing physician services in the United States. They are likely to require more effort than the purely data driven approaches. Whether they would require more or less resources than resource-cost-based estimates is uncertain. ules for each physician as is required under the current CPR system. Further, fee schedule updating could be accomplished using methods that would allow greater control over annual increases in average price levels than are currently possible under CPR even with the use of the MEL Chapter 6 Payment for Packages of Services INTRODUCTION Packaging is an approach to physician payment that involves redefining the payment unit from the individual service to a broader "bundle" of services (313).1 This approach could control both costs and utilization by reducing the number of service units billed and encouraging the judicious use of services within packages. Under packaging, the financial risk would be borne by the individual physician or other recipient of payment (547). This chapter examines variations in packages of services and discusses potential effects of packaging alternatives on quality of care, access to care, costs and efficiency, technological change, and administrative feasibility. Also considered in this chapter are the potential effects of paying for physician services via collapsed procedure codes.2 One objective of paying physicians a specified rate for a group of services would be to give the Medicare program more control over program costs. Unless the rates for packages were set at the same level as or below the mean of current charges, however, payment for packages of services would not necessarily result in a reduction of Medicare expenditures. In most cases, rates for packages would be prospectively determined and would include ancillary services (e.g., clinical laboratory tests, X-rays, injections) so physicians might think carefully about ordering a marginal test or requesting a consultation. The major difficulties of paying a rate for a package of services stem from the potential for underuse of needed expensive services or denial of care to very ill and potentially resource-intensive patients. The use of appropriate case-mix 'This chapter uses the term "packaging" synonomously with the term "bundling." 'Collapsed procedure codes would not produce a "true" package (319), because the unit of payment under collapsed codes would remain the individual service. Although collapsing procedure codes is compatible with Medicare's customary, prevailing, and reasonable (CPR) payment method (ch. 4) or fee schedules (ch. 5), the concept is discussed here as a means of introducing the concept of packaging. measures-measures of the relative frequency with which physicians treat patients with different types of medical conditions-should result in higher payments to physicians who treat more complex patients and should obviate some of the negative effects. Relative to the present customary, prevailing, and reasonable (CPR) Medicare fee screen method, packaging would create situations where physicians might gain or lose income, due to the "averaging effect."3 So that physicians faced a fixed amount of revenue for each package of services, mandatory assignment would be necessary. Otherwise, physicians would be able to shift the financial risk to Medicare beneficiaries by billing them more than the allowed packaged rate. Some physicians, for example, surgeons, already provide much care that is paid on the basis of a global or package rate. However, there has been little empirical research testing the applicability of packaging to broader areas of physician payment. To address the lack of research on packaging physicians' services for inpatients, the Social Security Amendments of 1983 (Public Law 98-21) mandated a study by the Department of Health and Human Services (DHHS) to examine the feasibility of using a diagnosis-related group (DRG)4 type of classification to pay for inpatient services provided by physicians to Medicare beneficiaries. In addition, the Office of Research and Demonstrations in the Health Care 'Relative to the present, paying an average rate for a package of services would reduce payment for some physicians and increase payment for other physicians. •Diagnosis-related groups (DRGs) are groupings of diagnostic cat~ egories drawn from the International Classification of Diseases and modified by the presence or absence of a surgical procedure, patient age, presence or absence of significant comorbidities or complications, and other relevant criteria. DRGs are the case-mix measure mandated for Medicare's prospective hospital payment system by the Social Security Amendments of 1983 (Public Law 98-21) (141,489). A later section of this chapter discusses the applicability of DRGs to physician payment. ~ • (J) ::. - 0' .., ~ ~ c;· Table 6·1.-Variations in Packages of Services iii" ::. Collapsed Ambulatory episode Inpatient episode Total episode g>of care Variable procedure codes Ambulatory visit Special procedure of care of care ~- Unit of payment Procedure Visit Procedure Ambulatory episode of care Inpatient episode of care Total episode of care (J) Case-mix None AVGs,a reasons for Diagnoses, DRGs, PMCs,b AVGs, diagnoses, or DRGs, staging,d APACHE,e AVGs, DRGs, or ICD-9-CM !'! visit, or visit types or severity-of-illness reasons for initial visit PMCs, severity-of-illness codesg May need new ~ indexc index, or MEDISGRPSf classification system iil adjustor Primary physician or Primary physician or Primary physician or Physician, physician group, Primary physician or ~ Recipient of Physician or physician group payment physician physician group physician group physician group medical staff, hospital, ~combined medical group hOspital staff 0' .., Scope of Procedure Visit and ancillaries Physician services and Physician services and Inpatient P,hysician Inpatient and ambulatory ancillaries included for ancillaries entire episode servicesh physician services and ~ services ambulatory patients but of ambulatory care ancillaries ~ excluded for inpatients Ill (iJ Approach to Prospective or Prospective Prospective PrC>Spective Prospective Prospective payment retrospective Time period lmmediate1 Immediate Fixed irttervali Episodic or fixed interval Episodic Episodic or fixed interval aAmbulatory visit groups (AVGs), a classification system similar to DRGs, define similar visit types in relation to the amount of time a physician spends with a patient (140). bpatient management categories (PMCs) are based on patients' clinical charaoteristics and severity of illness (586). crhe severity-of-illness index reflects the patients' overall severity of illness, not just the severity of each Cllagnosis (219,220). dstaging was developed to measure the biological progression of an illness within diagnostic categories (173,489). erhe Acute Physiology and Chronic Health Evaluation, Modified Version (APACHE) consists of 12 commonly used physiologic measures weighted to produce a total score for an intensive care unit patient. frhe Medical Illness Severity Grouping System (MEDISGRPS) groups patients !>y severity on the basis of data acquired after admission (59). gThe International Classification of Disease, Clinical Modification, 9th edition (ICD-9-CM), developed in the late 1970s, is a diagnostic lexicon of 10,241 five-digit codes, that encompass the realm of diseases known at that time. ~Hospital ancillaries are assumed to be incorporated in the hospital DRG. 'Immediate time period refers to the services related to a single patient-provider encounter. iThe services associated with the interval surrounding the procedure. SOURCE: Adapted from J.B. Mitchell, K.A. Galore, J. Cromwell, et al., "Alternative Methods for Describing Physician Services Performed and Billed," prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, November 1983. Ch. 6-Payment for Packages of Services • 159 physician office visit ambulatory care, and spemade mandatory, could be the individual physicialized ambulatory care including chemotherapy cian, a single specialty physician group, a multior ambulatory surgery (270). A study is currently specialty physician group, a facility, or a combeing conducted in California to develop emerbined facility-physician corporate entity.10 Algency department groups for both hospital and though Medicare can determine how it will payphysician costs of emergency room treatment (78). for physicians' services, the Medicare program Investigators have examined the effects of three cannot control how physicians are paid within a different case-mix measures on the creation of group setting. For instance, although a group of ambulatory-visit packages. Mitchell and colleagues9 physicians may bill on a fee-for-service basis, the found that reason for the visit, diagnosis comgroup may pay its members a salary or offer a salary plus a percentage of income earned above bined with the visit type, and ambulatory patient groups (the name was later changed to ambulaa base figure. How an individual physician is paidbears particular importance for packaging, be tory visit groups, were not superior to diagnosisalone in explaining the variation in services assocause the intended positive incentives of packagciated with an office visit (319). The number of ing, such as those for the judicious use of servcategories created by the different methods varices, may be diluted if physicians are far removed ied significantly. Using diagnosis/visit type profrom the direct payment (364). On the other hand, duced hundreds of packages; ambulatory patient removing physicians from the negative incentives groups produced 154; and reason for visit proof a particular payment mechanism, such as for duced 14 (319). Substantial variation in services the underuse of services, may be beneficial. remained even after adjustment for case-mix (319). Paying an individual coordinating physician for The simplest special-procedure packages would some of the more complex packages would in volve substantial financial risk to that physician. be based on collapsed procedure codes and com bined services; therefore, case-mix might not need For instance, if a patient's episode of care for my to be adjusted (319). For instance, a package might ocardial infarction was complicated by another include the surgeons' services, the anesthesiolochronic illness, such as diabetes, the coordinat gists' services, and assistant surgeons' services as ing physician would be financially liable for adwell as X-rays. If case-mix were adjusted, DRGs ditional visits and payment for other physicians'consultative services. Empirical research has or patient management categories might be usedfor inpatients, and other categories could be choshown that many physicians have small inpatient sen for ambulatory patients. An inpatient episode caseloads and may experience larg~ losses because of care might be classified by DRGs or patient of random variation in case-mix severity (313). management categories. New classification sysAlthough a large group of physicians might be tems would need to be developed for a total epibetter able to handle these variations in payment, sode of care. an individual physician may have difficulty do ing so (320). Payment to a larger entity, such as Packaged payment may introduce new adminthe medical staff, produces greater opportunity istrative and competitive arrangements for phyfor risk pooling and averaging (314). Medicalsicians depending on who is paid. The recipient staffs could form an individual practice associaofpayment for a packaged fee, if assignment were tion, an organizational form that has becomemore common in recent years.'These investigators used the 1979 and 1980 National Ambulatory Care Survey as a means of examining the services associated The scope ofservices covered by a package may with different types of office visit packages (319). The National Ameither be narrow or broad (see table 6-1). An exbulatory Care Survey asks a nationally representative sample ofoffice-based physicians to provide information on all services orample of a package with a narrow scope is an dered, even if the physicians do not provide them. There are anumambulatory-visit package, which is limited to theber of J:mitations to the data: volume of service cannot be deterservices associated with one visit to a physician.mined; no physician fee data are collected; there are no data onpatient office followup; no ambulatory department (hospital) information is collected; and all possible ancillary services are not 10lf assignment remained optional, the beneficiary could also re covered. ceive the payment. 162 • Payment for Physician Services: Strategies for Medicare California converted from the 1964 California relative value scale to the expanded 1969 version for Medicare billing purposes, billed or approved charges attributable to terminology changes increased 5 percent for office visits and 7 to 8 percent for hospital visits (442). If procedure codes were collapsed, the 11 existing codes for visits or the 9 for chest X-ray might be reduced to fewer categories for payment (see table 6-2). A group of experts could be convened to determine which codes to collapse on the basis of current codes used. For example, payment could be based on the most frequently billed code for a particular category (569). In other cases, a group of codes being considered for collapsing might have an equal or near equal distribution, and calculations of the payment rates could be based on a weighted average (319). Physicians could either continue to bill with the multitude of codes as they do now (and codes could be collapsed at the carrier level)13 or they could be given new code books. The potential effects of collapsed procedure codes on quality of care, access to care, costs and efficiency, technological change, and administrative feasibility are discussed below. Quality of Care Since payment for physician services under collapsed procedure codes would be similar to the current payment system, the payment level would be more likely to affect quality of care than the collapsing per se. If only 3 visit codes instead of the current 11 were allowed and payment rates 13ln the past, carriers collapsed codes by default. From the inception of the Medicare program until HCPCS was required in 1984, carriers used different coding systems: 1) one of two early versions of the California relative value scale, 2) a national Blue Shield Association coding system, 3) CPT, or 4) carrier adaptations of coding systems, such as the 1964 California relative value scale. Those carriers using systems other than CPT might have been billed by physicians with CPT codes. In order to pay physicians, carriers would have needed to collapse the CPT codes to fit into the California Relative Value Studies scale (569). Some carriers may have collapsed codes on a predetermined basis, and others may have done so on an ad hoc basis (55). For the first year of HCPCS, carriers are also, in effect, collapsing payment. Carriers who in the past used coding systems with fewer visit codes than CPT-4, would have only historical charges for those codes. Therefore, the carriers would have to assign a visit code to one of the categories for payment purposes (58). By the second year, charges would then exist for the 11 CPT-4 visit categories. Table 6·2.-Procedure Codes for Office Medical Services (Visits) and Chest X·Rays Codes for office medical services (visits): New patient: 90000 Brief service 90010 Limited service 90015 Intermediate service 90017 Extended service Established patient: 90030 Minimal service 90040 Brief service 90050 Limited service 90060 Intermediate service 90070 Extended service 90080 Comprehensive service Codes for chest X·rays: 71010 Radiologic examination, chest; single view, frontal 71015 stereo, frontal 71020 Radiologic examination, chest, two views, frontal and lateral; 71021 with apical lordotic procedure 71022 with oblique projections 71023 with fluoroscopy 71030 Radiologic examination, chest, complete, minimum of four views; 71034 with fluoroscopy 71035 Radiologic examination, chest, special views (e.g., lateral decubitus, Bucky studies) SOURCE: S.B. Clauser, C.M. Fanta, A.J. Finkel, et at. (eds.), Physicians' Current Procedural Terminology, 4th Edition, CPT4 (Chicago, IL: American Med· leal Association, 1985). were set at the mean,14 physicians who earned less per visit than they had in the past might either provide unneeded laboratory tests or bill separately for previously included laboratory tests (332). In addition, some physicians might reduce the time spent in face-to-face contact with their patients in order to see more patients per day. These incentives would not apply to those physicians who earned more per visit. Access to Care If payment rates with collapsed procedure codes appeared reasonable to physicians and if specialists were still allowed to bill different rates from generalists, access to care might remain stable (319). Specialty-specific billing in certain ways serves as a partial proxy for case-mix adjustment (319). Without specialty-specific rates, 80 to 90 percent of the specialists in one study would have lost money under collapsed procedure codes relative to the present system (319). If payment rates "if payment rates for all packages were set at the median, some of the skewing that an average produces might be avoided (569). did not appear reasonable to physicians and specialty-specific billing was not maintained, patient access to care might suffer. Analysis of South Carolina Part B data from 1981 showed that overall assignment rates for collapsed office visit codes would fall only slightly if specialty-specific billing were maintained (319). The Medicare market share in an area and in a physician's practice might determine whether access is a problem or not (319). In South Carolina, one-fifth of the physicians, for instance, provided one-half of all visits to Medicare beneficiaries (319). Medicare would be able to exert considerable leverage over these physicians. Costs and Efficiency Although collapsed procedure codes could control the rate of increase of Medicare expenditures, they would not necessarily reduce Medicare expenditures. The effect of collapsed procedure codes on Medicare expenditures would depend on the nature of the collapsed codes, utilization patterns, and the patient's severity of illness. With payment rates set at the mean, the effects would be similar to those described above. Mitchell and colleagues collapsed 12 codes for colonoscopy15 (a type of colon examination) in two different ways (319): • into a single collapsed procedure code (with a weighted average of charges for all12 procedures), and • into two different codes based on the extent to which the fiberoptoscope was inserted during colonoscopy. With a single code for all 12 procedures, the price for a colonoscopy was $247. With two codes based on distance into the colon, the price was $165 for the less complicated procedure and $293 for the more complicated procedure (see table 63). Ifall12 colonoscopy codes were collapsed into one, physicians who lost money relative to the current system would be paid $21 to $102 less than at present; physicians who gained money relative to the present system would earn between $14 and $106 more. Medicare might save the amounts "The newest edition of CPT -4 lists 13 codes for colonoscopy. Ch. 6-Payment for Packages of Services • 163 listed for physicians who lost income relative to the present on colonoscopies that go higher into the large intestine. Conversely, Medicare would have some losses for lower level colonoscopies. Should present assignment rules continue, a beneficiary would be likely to have higher costsharing liability if a simple colonoscopy was perf~rmed_ but a lower liability for a colonoscopy h1gher mto the large intestine. The same would apply to Medicare program costs. If assignment was not mandatory, physicians who stood to lose money on particular cases might refuse assignment in order to be able to bill patients for additional amounts. Mitchell and colleagues also collapsed 11 visit codes in two different ways (319). In the first situation, 11 visit codes were collapsed into 2 types of visits according to the type of patient seen (new or established). In the second situation, the 11 visit c_odes were collapsed into 5 codes (2 for new patients and 3 for established patients). Rates were set using· a weighted average of charges in the various visit categories. When specialty was taken into account and 5 codes were used, the amounts paid to physicians would have been comparable to the present. Results similar to those for colonoscopy occurred when visit codes were all collapsed into one code.. Patients' cost-sharing liability would rise if their visits were classified in a category with a higher average charge, and cost-sharing liability would fall for patients in a category with a lower average charge than in the past. Technological Change The incentives for technological change under collapsed procedure codes would be similar to thos~ _under the present system. If the colfapsing of v1s1t codes was coupled with the ind'b.sion of certain laboratory tests in the visit rate. (as was done in Quebec (28)), physicians would have an incentive. to use fewer and less expensive laboratory tests.16 16in Quebec, Canada, between 1971 and 1976, fhe a.verage l'lum~er of base. services, such as visits to physicians, provi~d to pati_ents rema~ed stable even with fee constraints and rising physiCian expenditures. But the number of associafe& Plantation, FL ...........................Staff-salary and bonus Charges6 Plan shares with parent corporation None Premium and drug copayment ~() ~ Group Health Plan of Southeast Michigan Troy, MI ...............................Staff-salary, referrals-some capitated Negotiated6 Plan bears allb None for staff; capitated specialists Premium; copayment for drugs and mental C/) ~ Metropolitan Health Council of Indianapolis Indianapolis, IN ..........................Staff-salary, referrals-<5% of contracting specialists on capitation Per diem Plan bears allb for professional services only None for staff; capitated specialists risk costs of professional services only, plan splits hospital savings health Premium and drug copayment ~ a;· (I) 0' .... s: (I) 9: 2 Cil with hospital Medical East Community Health Plan , Braintree, MA ...........................Staff-salary, referrals on capitation Per diem (State ratesetting) Plan bears allb None for staff; capitated specialists' Premium and drug copayment professional services only Genesee Valley Group Health Association Rochester, NY ...........................Group-members paid salary and bon usc Per diem (State ratesetting) Plan bears allb None tied to utilization Premium and nominal copayments for visits Genesee Health Plus Flint, Ml ...............................IPA receives capitation for physician services, 12% withheld for excess Charges paid by HMO, IPA shares no risk for Plan shares risk in AMCRA risk pool1 IPA bears risk for professional services Premium and nominal copayments utilization fund; individual physicians inpatient hospital and ancillaries; hospi paid capitation or FFS services tal surplus shared with IPA; individual providers not at risk Family Health Program, Inc. Long Beach, CA .........................Staff-salary and bonus9 Discounted6 Plan bears allb None Premium for high option only Delmarva Health Care Plan Easton, MD .............................IPA-capitated case managers; referrals at FFS charges Unspecified Plan shares in AMCRA risk pool1 Case managers at risk for maximum $1,000 Premiums and copayments services per patient, including hospital charges Table 7·3.-Capitation Payment for Medicare Beneficiaries in Demonstration Projects, as of Mar. 31, 19858 -Continued Unit of payment Unit of payment Government-plan Plan-provider Beneficiary Plan to physicians to hospitals risk-sharing risk-sharing cost-sharing Health Options of South Florida Per diem Plan shares in AMCRA Case managers at risk Premium and nominal Miami, FL ................... . .. HMO pays IPA, which pays case copayment managers capitation, minus withhold risk pool1 for withhold in referral and hospital fund,to fund hospitalization and referrals; pooling risk amongreferred specialists paid FFS case managers Central Massachusetts Health Care. Inc. .. IPA pays members UCR, minus 20% Unspecified Plan shares in AMCRA 20% of ambulatory Premium and nominalWorcester, MA risk pod fees and 25% of copaymentsfrom ambulatory services and 25% hospital fees at risk from hospital services to fund risk pool aBeginning Apr. 1, 1985, these plans converted to TEFRA risk contracts. Conversion was completed by June 30, 1985. Payment to plans covered all ambulatory and inpatient services covered under parts A and B of Medicare, except that payment was limited to Part B services for beneficiaries who were eligible only for Part B. Plans may have covered additional services not covered by Medicare. An additional 7 plans were included in earlier Medicare Capitation Demonstrations. bReinsurance limits catastrophic losses. cNot linked to utilization. dNa discounts cited. eMethod of payment unspecified. fAMCRA (American Medical Care and Review Association) sponsored a risk-sharing pool for its member health plans participating in Medicare demonstrations. gperformance related, possibly to utilization rates. SOURCE: A. Brewster, K. Langwell, P,. McMenamin, et al., "Evaluation of the Medicare Competition Demonstrations: Preliminary Implementation Case Studies of Four South Florida AHPs," Mathematica Policy Research, Inc., Washington, DC, Apr. 30, 1984; J.P. Hadley, Office of Demonstrations and Evaluation, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Dec. 30, 1985; P. Doyle and K. Langwell, "National Evaluation of Medicare Competition: Preliminary Implementation Case Study of Genesee 1-iealth Care," Mathematica Policy Research, Inc., Washington, DC, Jan. 28, 1985; 1. Lubitz and S. Nelson, "Evaluation of the Medicare Competition Demonstrations: Preliminary Implementation Case Study of Genesee Valley Group Health Association," Mathematica Policy Research, Inc., Washington, DC, August 1984; K. Langwell, "National Evaluation of Medicare Competition Demonstrations: Preliminary Implementation Case Study of Delmarva Health Care Plan," Mathematica Policy Research, Inc., Washington, DC, Jan. 28, 1985; K. Langwell, "National Evaluation of Medicare Competition Demonstrations: Preliminary Implementation Case Study of Family Health Program, Inc., Long Beach, CA," Mathematica Policy Research, Inc., Washington, DC, Jan. 28, 1985; S. Nelson, "National Evaluation of Medicare Competi tion Demonstrations: Preliminary Implementation Case Study of Central Massachusetts Health Care, Inc.," Mathematica Policy Research, Inc., Washington, DC, March 1985; S. Nelson, "National ~ Evaluation of Medicare Competition Demonstrations: Preliminary Implementation Case Study of South Florida Group Health, Inc.," Mathematica Policy Research, Inc., Washington, DC, May 1985; L.F. Rossiter and K. Byrd, "National Medicare Competition Evaluation: Implementation Case Study Report for: Group Health Plan of Southeast Michigan," Medical College of Virginia, Richmond, VA, September 1984; L.F. Rossiter and K. Byrd, "National Medicare Competition Evaluation: Implementation Case Study Report for: Metropolitan Health Council of Indianapolis, Inc.," Medical College "'I of Virginia, Richmond, VA, Aug. 31, 1984; L.F. Rossiter and G.C. Wood, "National Medicare Competition Evaluation: Implementation Case Study Report for: Medical East Community Health Plan, Braintree, Massachusetts," Medical College of Virginia, Richmond, VA, February 1985. ~ ~ g. i ;:, CD ;:, - • ..... Oo "' 188 • Payment for Physician Services: Strategies for Medicare services outside the plan and for Part A services. Thus, the experience with these plans does not relate to the incentives of prospective capitation payment. Nevertheless, for five of the seven plans, Medicare expenditures including in-and out-ofplan care were lower for plan enrollees than for control groups (91,566). Unlike the two other plans, these five owned or controlled their own hospitals and operated outside of New York City. In all instances, Medicare expenditures per enrollee were higher for physician and related services but lower for inpatient services and, with one exception, for hospital outpatient services (91). Conflicting hypotheses have been advanced to predict utilization and medical expenditures for people who enroll in HMOs (279). According to one theory, people who feel at higher risk of expensive care are more likely to join HMOs for the comprehensive coverage and financial protection offered. The greater coverage of HMOs may also attract people who favor preventive care and are likely to seek care when illness occurs. Other explanations would predict that HMO enrollees may be predisposed to use less expensive care than average either because they prefer not to use medical care or because enrollees are less likely to have ties to an existing provider, an indication that they were in good health. Studies of non-Medicare enrollees have indicated the importance of an existing physician relationship, but have found conflicting results concerning indicators of prior health status. The second set of Medicare studies examined the prior use and cost of beneficiaries who joined four risk-sharing plans during the mid to late 1970s (120, 121). Enrollees in a plan with a preTEFRA risk-sharing contract had inpatient use and expenditures about 50 percent lower than a comparison group (120). Two out of three plans studied from the Medicare capitation demonstrations also had indications of nonrandom selection, since their enrollees had previously had 20 percent lower Medicare reimbursements than comparison groups (121). A greater degree of biased selection by low-risk people into HMOs was found in four Medicare demonstration projects in which HMOs were permitted to screen the health of applicants before offering high-or low-option coverage (278). The results of these studies and of others indicating biased selection by low-risk non-Medicare enrollees into HMOs may not be generalizable (32,278). First, all cases of lower risk enrollment concerned prepaid group practices. Higher risk people may be attracted to the individual practice association type of HMO since they can maintain previous relationships with physicians. Second, with one exception, only one fee-for-service alternative was offered. A study of Federal employees' enrollment in health plans found lower users gravitated not to HMOs but to low-premium, high-cost-sharing, fee-for-service plans (433). Third, with one exception, the studies reported on cases where an HMO option was being offered for the first time. Furthermore, countervailing information for people under age 65 suggests, for example, that although enrollees of prepaid group practices in the West were younger than those in competing plans, prepaid group practice enrollees in most age-sex and diagnostic categories had greater need for hospital care ( 47). Enrollment in HMOs of beneficiaries likely to have medical expenses much below average could occur because of the preferences of enrollees or the marketing techniques of the plan. Holding an open enrollment period, during which beneficiaries choose among options and plans must accept people on a first-come first-enrolled basis, has been suggested as a partial means of avoiding biased selection (129). In some of the instances cited above, however, apparently nonrandom selection resulted from an open enrollment process. It should be noted that Enthoven also called for community rather than experience rating, limitations on switching plans, and minimum benefit coverage. As part of the Medicare capitation demonstrations, beneficiaries in the Minneapolis-St. Paul area had the choice of enrolling in one of four HMOs or of retaining traditional fee-for-service arrangements. In 1982, a comparison of HMO enrollees and nonenrollees found that beneficiaries who had joined an. HMO during the early part of the demonstration tended to have had less private health insurance than nonenrollees and were more dissatisfied with their usual source of care, especially in the areas of cost and paperwork (155). The beneficiaries who chose HMOs char acterized themselves as healthier than those who remained with fee-for-service arrangements. HMO enrollees were more socially active and less likely than nonenrollees to report having a serious medical condition associated with a greater likelihood of high medical cost. This finding may have been related to the fact that until1984, HCFA permitted the four HMOs to screen beneficiaries' health before enrollment in a high-option plan. The issue of biased enrollment is of great importance to the Medicare program. Whether there Ch. ?-Capitation Payment • 189 is a selection bias will atfect whether Medicare gains or loses from greater enrollment in risksharing plans. Medicare will gain if it pays plans 95 percent of the AAPCC for enrollees who would have incurred greater expenses outside of the plan, but Medicare will lose if it pays plans 95 percent of the AAPCC for beneficiaries whose expenses would have been far below that amount outside of the plan. ESTABLISHING THE CAPITATION RATE It is in the interest of the Medicare program, its beneficiaries, and society as a whole to pay for medical care in such a way that plans have financial incentives to work with providers to deliver care efficiently. If great differences exist in the profit that can be made from enrolling lowrisk vs. high-risk beneficiaries, plans will find it in their interest to stress marketing strategies rather than efficient delivery of care. Competing through marketing techniques to enroll low-risk people and to avoid high risks would be consistent with the expertise of health insurance companies, which may sponsor risk-sharing plans. Because of the substantial variation in medical expenses among Medicare beneficiaries, risksharing plans could face great losses. In 1977, Medicare paid $4,000 or more per patient for only about 5 percent of its beneficiaries, but the expenses of these beneficiaries accounted for 52 percent of Medicare's total expenditures (182). Medicare paid more than $20,000 per patient for only 0.09 percent of all beneficiaries, and these payments totaled 2.8 percent of program expenditures. A risk-sharing plan with a random group of beneficiaries could thus be hurt financially if even a small number of these expensive cases enrolled in excess of the predicted average. The risk would be greater for smaller plans because they would be subject to relatively more random fluctuation. An HMO with 1,000 enrollees could experience a 5 percent or greater cost increase from the enrollment of three beneficiaries instead of the expected one in the category over $20,000. Such a plan would probably bear an even greater risk because HMOs, unlike Medicare, typically provide catastrophic coverage. At the same time, a risk-sharing plan has the potential to reap sizable gains. In 1977, Medicare incurred no expenses for about 36 percent of its beneficiaries (182).6 In addition to people who used no services during the year, these beneficiaries included people who did not reach the deductible and people who did not bother to sub mit claims. However, most of the variance in Medicare expenditures (about 92 percent) was associated with differences among people who incurred costs, especially with whether or not people had hospital costs (about 38 percent of the total variance). A risk-sharing plan uncertain whether or not Medicare's payment will cover the expenses of enrollees may attempt to enroll lower risk beneficiaries. A greater degree of favorable selection will be required to cushion a plan against such uncertainty the smaller the plan's enrollment, the less uncertainty the plan's management is willing to bear, the lower the savings that the plan expects from more efficient delivery of care, and the less the difference between the AAPCC and Medicare's payment (182). On the other hand, by reducing hospitalization rates, HMOs may realize greater net revenues (revenue from AAPCC payments minus expenses of care) from high-risk en 'In 1982, Medicare incurred no expenses for 31 percent of elderly beneficiaries and 45 percent of disabled beneficiaries (525). 190 • Payment for Physician Services: Strategies for Medicare rollees who are especially likely to use hospital care under fee-for-service arrangements. It should also be noted that the regulations implementing TEFRA prohibit plans from using health screen ing or discriminatory marketing practices. Private health insurance has not developed means to adjust rates for different risk groups that are applicable to Medicare (461). Only about 15 percent of all individual applications for medical insurance are rejected because of adverse medical histories, are offered policies that exclude certain medical conditions, or are insured for an excess premium that reflects such high-risk factors as health status and occupational hazards. Because companies have rejected the worst risks or have restricted their coverage, the experience of the private sector provides little basis for calculating rates according to risk category. On the other hand, the insurance industry's experience with coverage for affinity groups may be transferable to the Medicare population. Insurers have provided coverage for employers who wished to furnish health insurance to their elderly workers. More importantly, insurance companies have sold policies to beneficiaries to supplement their Medicare coverage. A related development is coverage of long-term care as an insurance product. A desirable approach to establish Medicare capitation rates would contain incentives for plans to deliver cost-effective care and attract plans to participate, while allowing Medicare to reap some of the savings from more efficient delivery of care. The approach would also discourage manipulation by plans, including selection of low risks, and would require minimal additional data and expense to implement and administer. Alternative Methods of Refining the Average Adjusted Per Capita Cost (AAPCC) In 1983, an actuarial firm advising HCFA concluded that the AAPCC method was the best then available for determining Medicare payment to risk-sharing plans (311). However, the AAPCC explains very little of the variation in Medicare expenditures for its elderly beneficiaries (32). Ap plied to 1979 expenditures per beneficiary, for example, age, sex, and welfare status explained only 0.6 percent of the variation.7 Research is underway to refine the AAPCC to incorporate factors that would identify low-and high-risk beneficiaries, namely prior use of medical care (19,32), functional limitations and disability from chronic illness (181,456), or demographic and socioeconomic characteristics ( 48). One model used as prior use variables the hospital days used in the previous 2 years, whether or not the enrollee was hospitalized in the previous year, and whether or not the Medicare Part B deductible was met in the previous 2 years (32). Although this model was superior to the current AAPCC and to whether or not a person was hospitalized in the previous year, the hospital daysPart B model explained only 4.3 percent of the variance in Medicare expenditures per beneficiary. The prediction for groups of enrollees is the im portant result because Medicare would pay risk sharing plans for groups of beneficiaries. As one would expect, the predictions for groups were more accurate.8 The hospital days-Part B model was again superior to the others. For groups of beneficiaries biased by prior use, age, or welfare status, the ratio of the predicted reimbursement for the biased group to the actual reimbursement for the population from which the biased group was selected differed no more than 5 percent. This model is being used in a demonstration project in Minneapolis, in which an HMO is trying to attract frail elderly beneficiaries (see table 7-4) (278). The main shortcoming of this refinement is that it does not distinguish high users who had selflimiting acute conditions from those with chronic conditions that will continue over time (183,278). If anHMO would enroll fewer (more) chronically ill people among its high users, Medicare would be likely to have greater losses (gains). Another refinement of the AAPCC related to prior use would add information about diagno 'The proxy for welfare status was whether or not States had purchased Medicare Part B coverage for beneficiaries who were also eligible for Medicaid. Institutional status, although used in the AAPCC formula, was excluded because it is available only from special surveys. •The random errors of predictions for individuals tend to cancel out for large groups (278). Ch. 7-Capitation Payment • 191 Table 7·4.-Medicare Demonstrations of Capitation Payment, December 1985 Prior Use Modification of AAPCC. Medicare Payment at 85 Percent of AAPCC. Senior Health Plan, Minneapolis, MN Finlay Health Plan, Miami, FL General Description: For all Medicare enrollees who have been in the program at least 2 years, i.e., those 67 or older, data from the Health Insurance Master File on prior hospital days used and whether they met the Medicare deductible, rather than institutional category, are used along with age, sex, and Medicaid status to produce an individually adjusted Average Adjusted Per Capita Cost (AAPCC). Medi· care's capitation payment to the plan is 95 percent of the adjusted AAPCC. Medicare pays for enrollees without 2 years of utilization experience in the Master File at 95 per· cent of the standard AAPCC, adjusted for age, sex, Medicaid, and institutional status. Scope of Services: Medicare Parts A and B, with no limit on days of use for inpatient and nursing facility services. Plan makes arrangements for some community-based social services, but not to the same extent as a Social/Health Maintenance Organization. Payment to Physicians: Network model. Plan pays a capitation payment for both physician and hospital services to 6 riskpools made up of both physicians and hospitals. Each pool uses different arrangements to share risks and pay providers. Payment to Hospitals: In all cases the plan puts both the hospital and primary care physicians acting as gatekeepers at risk for some portion of their services. The level of risk borne and the payment rate are negotiated between the plan and representatives of the risk pool. Physician referrals are paid according to arrangements made by each risk pool. However, the plan withholds a negotiated portion of the pool's capitation payment to fund referals. Risk-Sharing With HCFA: None. Plan may arrange independently for stop-loss insurance. Beneficiary Cost-Sharing: Premium is $21.75, in addition to standard Part B premium. Beneficiary is not liable for copayments or deductible. General Description: Medicare pays plan a capitation payment of 85 percent of the AAPCC, instead of 95 percent as mandated by regulations implementing the provisions of Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA) (Public Law 97-248). In return, the plan can keep any surplus revenues from the capitation payment, rather than having to report an Adjusted Community Rate and make adjustments required by TEFRA regulations. Scope of Services: Medicare Parts A and B, with no day limits on hospital care and additional coverage for routine medical care, prescription drugs, routine dental care, eyeglasses. Payment to Physicians: Network model. Capitation payments are made to plan-owned clinics with staff physicians and also to contracting affiliated providers (including both group practices and solo practitioners). For Medicare enrollees, the plan makes one capitation payment to cover ambulatory service, and a separate payment to cover inpatient, home health, skilled nursing facility, and hospitalbased outpatient care. For the latter payment, risk is shared between the plan and the providers, with 75 percent of any surpluses or deficits accruing to the plan and the remainder to the contrac'ting unit. Payment to Hospitals: Negotiated per diem contracts. Beneficiary Cost-Sharing: None. No additional premium in addition to Part B premium and no deductible or copayment liability. Risk-Sharing With HCFA: None. Plan may arrange independently for stop-loss insurance. SOURCES: R. Deacon, Office of Demonstrations and Evaluation, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Dec. 4, 1985; J. Laly, Senior Health Plan, Minneapolis, MN, personal communication, Dec. 5, 1985; J. Robinson, Finlay Health Plan, Miami, FL, personal communication, Dec. 4, 1985; and R. Sirmon, Office of Demonstrations and Evaluation, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Dec. 2, 1985. ses associated with hospitalization and indicative of chronic conditions likely to result in substantial future medical costs (183,278). Including diagnosis has raised the explanatory power of the model to 9 percent of the variance (183). The University Health Policy Consortium based at Brandeis University further classified hospitalizations and found that repeated hospitalizations for cardiac conditions, cancer, or musculoskeletal conditions were associated with subsequent expenditures about 3.5 times the average (278). Research is proceeding to predict expenses after enrollment in an HMO and to focus on conditions with large unavoidable expenditures that are not discretionary and hence manipulable by the risksharing plan. It is also important that the approach used does not penalize risk-sharing plans for keeping use and costs of care low. Other Methods of Determining Capitation Payment Besides refining the capitation rate, the AAPCC, to deal with the risk to be borne by plans and possible biased selection by enrollees, Medicare could change its risk-sharing arrangements with plans. Such approaches would incorporate some retrospective adjustments to the prospective payment depending on actual cost of treating beneficiaries (183). 192 • Payment for Physician Services: Strategies for Medicare These approaches would have Medicare share in the risk of high-cost enrollees and would thereby reduce the risk of losses for the HMO from severely ill enrollees. In exchange for Medicare's sharing the risk, the HMO would receive a lower percentage of the AAPCC. In effect, Medicare would be providing reinsurance to the HMO, and Medicare and the HMO would share the profits from greater efficiency. HMOs would be more willing to enroll all beneficiaries. For example, Medicare could pay an HMO well below the AAPCC and in addition pay 80 percent of an individual's care over $5,000 (individual stop-loss reinsurance). Or Medicare could pay a plan 85 percent of the AAPCC and absorb any losses over 3 percent of their total capitation revenue (aggregate stop-loss reinsurance). Both approaches would reduce but not eliminate the financial incentives for the plans to be efficient (183). Alternatively, risk-sharing plans could rely on private reinsurance, which is permitted under the regulations implementing TEFRA, to protect them against losses. Under this approach, Medicare would transfer all the risk to the plan, which would then pay an insurer to help bear the risk. Any savings that Medicare might then share in the form of lower capitation payments would stem from any efficiencies in delivering medical care that were achieved by the plans. Capitation rates could also be determined by competitive bidding. This approach would be more compatible with capitation payment to a fiscal intermediary than to numerous risk-sharing plans. Competitive bidding to select an intermediary would entail disadvantages similar to those enumerated in chapter 5, such as the possibility of unintentionally setting up an organization that would develop monopoly power. IMPLICATIONS OF MEDICARE CAPITATION ALTERNATIVES A substantial body of literature stretching over 25 years pertains to experience with providers paidby prospective capitation payments. Most of the studies concern large, established, nonprofit prepaid group practices, and some of the groups, such as Group Health Cooperative of Puget Sound and many of the Kaiser-Permanente Medical Care Programs (see box 7-B), owned their own hospitals. It is difficult to extrapolate the results of these studies to other plans, both present and future. Although group practices in June 1984 still accounted for 59 percent of all HMOs and 81 percent of total enrollment, their number had stayed almost constant over the previous year, while individual practice associations (IPAs) had grown 27 percent in number of plans and 56 percent in total enrollment (240). More than half of all enrollees were in plans with 100,000 or more members, but by far the greatest growth in enrollment was occurring among plans of 50,000 to 99,999 members (240). Newer plans are also more likely than older ones to be investor owned (for profit) (121) and to have arrangements that place physi dans at financial risk (see box 7-C). Even before the TEFRA regulations, Medicare enrollment in HMOs was increasing more rapidly than general enrollment (212). From June 1983 to June 1984 Medicare enrollment in HMOs increased 36 percent. In addition to HMOs that have expanded their operations to different States, insurers and hospital management companies have entered the HMO market in substantial numbers (see table 7-5). Not only is the relative number of IPAs increasing, but, according to anecdotal information, their sponsorship is as well. Early IPAs were preexisting foundations for medical care or were established by medical societies. By contrast, newer IP As may be outgrowths of fee-for-service group practices or HMOs that wish to grow rap idly. Individual physicians feeling competitive pressure may join an IPA to increase their patient load, and the group practice or HMO may find "direct contract IPAs" a desirable way to expand without great capital investment in facilities. All of these plans will be operating in a market environment much different from the past. As Ch. 7-Capitation Payment • 193 194 • Payment for Physician Services: Strategies for Medicare before, HMOs will be competing for enrollees and physicians with fee-for-service arrangements, but the entire medical care sector is under much greater pressure to reduce the costs of performing services and to constrain total expenditures (see ch. 2). How plan managers, private thirdparty payers, physicians, and lay people will respond is, of course, unpredictable. But the rapid and substantial changes in market context and configuration of plans dictate caution in formulating policy on the basis of past results. This section analyzes the implications of different Medicare capitation arrangements for the dimensions outlined in chapter 3: quality of care, access to care, cost and efficiency, technological change, and administrative feasibility. In the course of the discussion, the implications are noted for five medical technologies: pneumococcal vaccination, clinical laboratory services, magnetic resonance imaging (MRI), extracorporeal shock wave lithotripsy (ESWL), and cataract surgery. The section is divided into two parts according to the recipient of Medicare payment: the first concerns capitation payment to plans, such as HMOs, that in turn would arrange for physicians and perhaps other providers to deliver medical care to enrollees; and the other examines capitation to fiscal intermediaries, such as Medicare carriers, that would arrange for plans and providers to deliver care. In either case the scope of services covered by the capitation payment could vary from physician and ambulatory services to those services plus inpatient care. The discussion throughout highlights potential problems that would warrant attention as Medicare policy is designed and implemented. Capitation Payment to Health Plans Under this approach, Medicare would pay to the plan chosen by a beneficiary a capitation payment for care to be provided during a given time period. Although it is beyond the scope of this project to examine the mechanics of beneficiary choice, this alternative is consistent with Enthoven' s Consumer Choice proposal regarding plans that would provide comprehensive care (129) and with the Reagan Administration's proposals that beneficiaries be given vouchers and select plans (104). Indemnity insurers as well as HMOs and CMPs could receive capitation payments. HCFA might require that all plans meet a minimum benefit requirement and certify their financial viability. It is assumed that one of the beneficiaries' options would be to continue present coverage and present arrangements with physicians. For example, a private insurance company might offer such coverage and accept the capitation payment as the premium. Ch. ?-Capitation Payment • 195 196 • Payment for Physician Services: Strategies for Medicare Quality of Care From the early studies of prepaid group practices in the 1960s through the reviews of IP As and other HMOs of the 1980s, evaluations of practices paid by capitation have found the quality of care provided to their enrollees at least as good and usually better than that of comparison groups (97,107,194,223,279,404,483,579,581). These evaluations have incorporated measures of structure, process, and outcome. The capitation practices have had higher percentages of board-certified physicians, for example; have followed standards for process of care as well as or better than feefor-service practices; and have had comparable or better mortality and morbidity rates. Although people in capitated plans have been as satisfied with the technical aspects of care, they have been less satisfied with interpersonal aspects than those in fee-for-service practices. More recently, a 1984 survey found HMO members more likely than eligible nonmembers to be satisfied with their health care. The greatest differences concerned out-of-pocket expenses, availability of services, and waiting time for an appointment (274). But certain incentives of capitation payment, the results of specific studies, the public sector's experience with prepaid plans, and particular problems of Medicare beneficiaries prevent automatically generalizing the above results to Medicare beneficiaries. No study has examined the quality of care provided to Medicare beneficiaries. Until recently, few Medicare beneficiaries have been enrolled in HMOs, which have marketed their plans mainly to employed populations. The National Medicare Competition Evaluation funded by HCFA includes an evaluation of the quality of care provided by risk-sharing plans (411). Although that part of the evaluation was expanded in June 1985, results are not expected until1987 (541). In addition, an HCFA-sponsored evaluation of Medicaid demonstration projects may also provide relevant information (see table 7-6) (197). Capitation payment provides financial incentives to care for enrollees at low cost. Since per capita revenue is fixed for the time period, the plan's net revenue or profit depends on the costs incurred in providing care. These incentives have many positive implications for quality. There is an incentive to avoid hospitalization and to treat people elsewhere, an approach that reduces exposure to nosocomial infections and to unnecessary procedures. In fact, lower hospitalization rates have been observed in prepaid groups for both medical and surgical care (279). In prepaid groups, admissions were lower for diagnosis and tests and for surgical procedures, including ones that have been associated with unnecessary care (hemorrhoidectomy, surgery for varicose veins, and hysterectomy) (279). However, financial incentives, at least at the plan level, are to reduce cost, and that may be done at the expense of quality. The check on such Ch. 7-Capitation Payment • 197 Table 7·5.-Major Sponsors of Health Maintenance Organizations (HMOs) and Preferred Provider Organizations (PPOs), 1985a HMOsb PPOs Sponsor Total Group Staff IPA Network Total Operational Preoperational Insurers: Aetna ..................................... 3 0 0 3 0 7 7 NA Blue Cross/Blue Shield Affiliated Plans ....... CIGNA Health Plans ........................ 73c 15 14 2 8 8 23 5 28 0 34d 34 NA CIGNA Insurance Co........................ 17 17 NA Equitable .................................. Hancock/Dikewood ......................... John Hancock Insurance .................... 0 4 0 0 0 0 0 3 0 1 4 9 2 5 2 4 Metropolitan ............................... Provident Life Insurance (Health Point Corp.) .. PruCare ................................... 2 0 17 0 0 14 0 0 0 0 0 3 2 0 0 20 7 14 5 6 2 Prudential (PruNet) ......................... Wausau Insurance Co....................... 5 0 0 5 0 30 1 2 1 28 NA Hospital Management Companies: American Medical International •• 0 ••••••••• 0. 2 0 1 1 0 3 3 NA Hospital Corp. of America ................... Humana, Inc . • • • • • • • • • • • • 0 •• 0 •••••••••••••• National Medical Enterprises •••••••••••• 0 ••• 4 10 3 1 0 0 0 0 0 2 2 3 1 8 0 1 21 9 5 0 21 5 1 NA NA Nationwide HMO Networks: HealthAmerica ............................. 36 8 9 13 6 0 0 0 Kaiser Permanente ......................... 11 11 0 0 0 0 0 0 Maxicare .................................. 13 0 0 0 13 0 0 0 U.S. Health Care Systems ................... United Health Care Corp..................... SANUS Corp............................... Whitaker Health Services .................... 3 33 4 10 0 0 0 0 0 0 0 0 3 27 4 10 0 6 0 0 0 1 1 0 0 1 1 0 0 NA NA 0 Totals 248 50 26 106 65 161 118 43 aData obtained from telephone survey, November 1985, except as noted. This list should not be considered exhaustive of sponsoring organizations. bNumber of HMOs owned and operated as lines of business by cited institutions. Does not include HMOs operated under management contracts. CData from Blue Cross and Blue Shield Association, June 5, 1985. and Sept. 1, 1985. dData from Blue Cross and Blue Shield Association, Oct. 1, 1985. estates where Humana Care Plus programs are operational. "Humana Care Plus" plans are operated as HMOs or PPOs in different markets, with varying degrees of selective contracting and free choice of providers. SOURCES: Aetna data: J. Harper, Aetna Insurance Co., Hartford, CT, personal communication, Nov. 12, 1985. American Medical International data: H. Leavit, American Medical International, Beverly Hills, CA, personal communication, Nov. 15, 1985. Blue Cross data: Blue Cross and Blue Shield Association, "Blue Cross and Blue Shield Plan Activities in HMOs, December 31, 1984," June 5, 1985; Blue Cross and Blue Shield Association, "Blue Cross and Blue Shield Plan HMOs Operational After December 31, 1984," Sept. 1, 1985; and Blue Cross and Blue Shield Association, "Blue Cross and Blue Shield Plans Marketing Preferred Provider Products," Oct. I, 1985. CIGNA data: R. Maag, CIGNA Corp., Hartford, CT, personal communication, Nov. 14, 1985, and S. Shulman, Cl· GNA Health Plans, Dallas, TX, personal communication, Nov. 7, 1985. Equitable data: J. Neely, Equitable Life Assurance, New York, NY, personal communi· cation, Nov. 20, 1985; and R. Ullman, Equitable Insurance Society, New York, NY, personal communication, Nov. 19, 1985. John Hancock data: R. Morse, Hancock!Dikewood Health Plans, Boston, MA, personal communication, Nov. 8, 1985; and C. Somers, John Hancock Insurance, Boston, MA, personal com· munication, Nov. 11, 1985. Hospital Corp. of America data: J. Horn, HCA Health Plans, Nashville, TN, personal communication, Nov. 15, 1985. Health Point data: R. Paden, Health Point Corp., Chattanooga, TN, personal communication, Nov. 21, 1985. HealthAmerlca data: G. Nielson, HealthAmerica, Inc., Nashville, TN, personal communication, Nov. 14, 1985. Humana data: M. Hoover, Humana Care Plus, Louisville, KY, personal communication, Nov. 21, 1985. Kaiser data: lnterstudy, Inc., National HMO Census (Excelsior, MN: lnterstudy, Inc., March 1985); and M. Tatge, "HMO Enrollment Up 26.7% to 1.68 Million," Modern Hea/thcare, 15(4):138-141, June 7,1985. Maxicare data: K. Wichser, Maxlcare Health Plans, Hawthorne, CA, personal communication, Nov. 19,1985. Metropolitan data: L. Hyman, Metropolitan Life Insurance Co., New York, NY, personal communication, Nov. 7, 1985; and T. Nimnicht, Metropolitan Life Insurance, St. Louis, MO, personal communication, Nov., 19, 1985. National Medical Enterprise data: S. Tyler, National Medical Enterprises, Beverly Hills, CA, personal communication, Nov. 25, 1985. Prudential data: T. Burke, Prudential Insurance Co., Newark, NJ, personal communication, Nov. 7, 1985. SANUS data: M. Rosen, San us Corp., New York, NY, personal communication, Nov. 15, 1985. United Health Care data: A. Billingstad, United Health Care Corp., Minnetonka, MN, personal communication, Nov. 21, 1985' and S. Conway, United Health Care Corp., Minnetonka, MN, personal communication, Nov. 19, 1985. U.S. Health Care data: D. Richman, "U.S. Health Care Aims at Growth With Push Into New York, Dallas," Modern Healthcare 15(20):50-53, Sept. 27, 1985. Whitaker data: Whitaker Health Services, Marketing Deptartment, Los Angeles, CA, personal communication, Nov. 18, 1985. behavior is that plans would lose enrollment if ter able to judge technical care for conditions or members perceived that quality was below an technologies that they or their friends use freacceptable level. Such enrollee dissatisfaction dequently, but are dependent on physicians' guidpends to some extent on lay knowledge of what ance for infrequently used or new procedures constitutes appropriate care. People would be bet-(366). Table 7·6.-Capitation Payment for Medicaid Beneficiaries in Demonstration Projects, December 19858 ~ • '3 ~ (I) ~ Recipient of 0' government Unit of payment capitation to physician Unit of payment Government-plan Plan-provider "' ~ Plan payment providers to hospitals risk-sharing risk-sharing ~ New Jersey Medicaid Personal o· Physician Plan ...............Physician Capitation fund for case DRGs NA At risk for referrals, ~ managers' services, professional services, and referrals, and ancillaries, up ancillaries up to defined (/) (I) to a maximum. Separate maximum per provider; not fund for hospital services at risk for hospital ~facilities, receives 50% of (1) !'! savings, bonus for outpatient management of ~ inpatient procedures iil Monroe County Medicap ~ Monroe County, NY .............Medicap acts as HMO discretion HMO discretion HMO bids include level of HMO discretion q;· intermediary; seeks bids (county is exempt risk borne; intermediary (I) from 4 competing HMOs from State ratesponsors stop loss pool, 0' for rights to serve Medicaid setting) which HMO can opt out of "' beneficiaries for higher capitation ~ payment ~ Santa Barbara County Health Ill Initiative Cil .......County Health Authority Capitation account to case Per diem Authority at risk b 20% withhold maximum Santa Barbara County, CA (intermediary) manager covers all patient risk to case managers, services with 20% withheld excesses in capitation for risk sharing account shared with county Arizona Health Care Cost Containment System ..........Prepaid plans bid on rates Plan discretion Plan discretion State stop-loss per Plan discretion to serve Medicaid clients beneficiary and medically indigent Missouri Managed Care Jackson, County, MO ...........Plans contract for services Plan discretion Plan discretion Contracting plan bears allb Plan discretion at State-set rates Minnesota Prepaid Medicaid Competition Demonstration ....Plans contract for services Plan discretion Plan discretion For Supplemental Security Plan discretion at State-set rates Income beneficiaries, government pays between 95% and 100% of AAPCCb NA ~Not applicable. aPians may have arrangements for beneficiary cost-sharing. bReinsurance may limit catastrophic losses. SOURCE: R. Deacon, Office of Demonstrations and Evaluation, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Dec. 20, 1985; P.L. Haynes, Evaluating State Medicaid Reforms (Washington, DC: American Enterprise Institute, 1985); J. Meyer, American Enterprise Institute, Washington, DC, personal communication, Oct. 10, 1985; S. Treiger, Office of Demonstration and Evaluation, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communicaton, Nov. 25, 1985; and J. Vertrees, La Jolla Management Corp., Rockville, MD, personal communication, Jan. 3, 1986. As discussed above, the incentives of individual physicians depend on how they are paid by the plan. If physicians continue to be paid fees for their services, financial incentives continue to reward the provision of additional services. Physicians in prepaid groups usually derive part of their income from profit-sharing or "productivity" measures, that is, their use of services and number of patients. The percentage of such income may be quite small; at Kaiser-Permanente in northern California, for example, the incentive compensation payment, which depends on the results of overall operations for the year, has not exceeded 5 percent of the average physicianpartner's income (87). However, in spite of the specific arrangements for compensating physicians, physicians have a personal financial stake in the continued solvency of the organization. Management practices may also be relevant to physician behavior in this regard. Kaiser-Permanente physicians retain the same responsibility for seeing a full load of ambulatory clinic patients regardless of the number of their patients who are hospitalized (223). All other things being equal, a physician in this circumstance would be more likely to prescribe return visits in the office than to hospitalize the patient and increase demands on the physician's time. Hornbrook and Berki have theorized that HMOs would not be expected to skimp on treatment of severe illness, such as colorectal cancer, for which definitive treatment is available (223). HMO physicians are subject to the same community standards of practice as fee-for-service physicians and may face greater malpractice liability exposure since corporations are more likely than individual physicians to be sued. HMOs may also excel in reassuring worried-well patients that their symptoms are self limiting or part of the aging process. But people who are subtly sick, that is, whose conditions cannot readily be identified, may experience delays in the diagnosis of potentially serious disease if HMO physicians face bureaucratic complexities in ordering diagnostic workups or in obtaining tests from outside the HMO. Of course, to the extent that enrollees are dissatisfied and believe that they would receive more prompt care under different arrangements, they can leave that plan and join another. Ch. 7-Capitation Payment • 199 Results of a study in Washington State of patients with colorectal cancer are consistent with the hypothesis concerning delay in diagnosis, but not conclusive (150). After 4 years, no differences in outcome were found between prepaid group and fee-for-service patients, and treatment was comparable once the diagnosis was made. However, a significantly longer period elapsed in the prepaid group between initial contact with a physician and start of treatment, 47 days in the HMO and 14 days in fee-for-service practice. It is possible, under the financial incentives of capitation payment, that delays would occur in resorting to more expensive treatment for a condition for which there were alternative therapies. For treatment of renal stones, which depending on the stone may be treated surgically or medically, there might be a preference for the initial use of a potentially cost-saving technology, such as ESWL, perhaps on an ambulatory basis. Whether or not such a delay would compromise the patient's outcome and quality would depend on the specific situation. Similarly, capitation plans would have an incentive to delay surgery such as cataract removal and to have it performed by physicians who specialize in that procedure. The likely effect on quality is not clear. Delaying. surgery might constitute poorer quality care if the person was unable to function effectively in her or his daily activities. On the other hand, delays in surgery can have health benefits if the surgery, such as appendectomy, is avoided or if the diagnosis is refined. Under capitation payment, plans would have financial incentives to take advantage of economies of scale in locating and using expensive equipment. There would be incentives to send more tests to centralized clinical laboratories, perhaps ones owned by the plan, and to perform far fewer tests in separate physicians' offices. Such a shift has the potential to improve the quality of test results. State standards may be more likely to apply to testing in central laboratories than in physician offices, and appropriately trained technicians may be more likely to perform the test. Although capitation plans have financial incentives to underuse diagnostic tests, like other services, no difference in use has been detected (279). 200 • Payment for Physician Services: Strategies for Medicare Although greater use of preventive services has cause hospitals desire additional admissions and been reported for enrollees of capitation plans, profit from low-cost cases in a given DRG. If capifor the most part these services consisted of antation payment covered only non-inpatient servnual physical examinations, which have questionices, attention to admissions from risk-sharing able efficacy (483). This greater use is consistent plans would warrant the particular attention of with lower financial barriers to initial use because the quality assurance and utilization review body. patient cost-sharing has typically been lower in Although Medicaid programs have increasinglycapitation plans (279). There are no recent studies been adopting capitation payment, historicallyassessing whether HMOs are providing more of few Medicaid eligibles have been enrolled inother preventive services, such as health educaHMOs. Some HMOs have served poor people eftion, nutritional guidance, and counseling. HMOs fectively (194). At least at Kaiser-Permanente inmay cover preventive services and feature that Oregon the program entailed adding substantialcoverage in advertisements as a marketing tool. outreach activities to the regular HMO (178). Vaccinations are often cited as an example During the early 1970s, substantial quality of a cost-effective preventive technology (576). problems occurred in prepaid health plans set up Studies that have examined rates of vaccination, for people eligible for California's Medicaid proincluding influenza vaccination, among HMO engram (483). These problems were addressed by rollees have found no consistent pattern of use: subsequent Federal and State legislation. Amendthe enrollees of some plans had lower rates and ments to the Health Maintenance Organization the enrollees of other plans had higher rates than Act in 1976 (Public Law 94-460) required that all comparison groups (483). Pneumococcal vaccine plans receiving Medicaid funds be federally qualis the only preventive service that Medicare covified HMOs, and California implemented more ers for all beneficiaries. Capitation plans have fistringent regulations for certification and prohibnancial incentives to provide pneumococcal vacited certain marketing and management practices. cine to elderly and high-risk people to the extent More recently, concerns have been about thethat they are likely to remain in the plan long quality of care received by Medicare beneficiariesenough for the plan to reap any savings from disin certain demonstration projects in Florida (477). ease prevention. But pneumococcal vaccine is unThe problems identified in hearings by the Houselikely to save costs, and initial uncertainty about and in a report by the General Accounting Ofits efficacy probably deterred physicians from recfice related primarily to timely enrollment and disommending its use (485). The low use of pneuenrollment (476). The General Accounting Officemococcal vaccine, regardless of payment arrangeis continuing to examine the situation, and Mathement, is consistent with barriers that precede matica's evaluation of Medicare competition dempayment, as discussed in chapter 3. onstrations will also cover the plans involved. The financial incentives for risk-sharing plans The Office of Health Maintenance Organizato reduce the costs of care and perhaps to protions in the Public Health Service determinesvide too few services apply only to the services whether plans meet the conditions of an HMOcovered by the payment. If the capitation payor CMP and are eligible to contract with HCFA ment did not cover inpatient care, the plan afl.d (584). These arrangements have been continued perhaps its physicians, depending on their income under the regulations implementing TEFRA and arrangements, would have increased incentives apply to HMOs and CMPs (both cost and risk compared to the present situation to hospitalize plans) (533). The Public Health Service reviews patients for care. Hospitalization would enable among other things that the plans have in effect physicians to perform tests and therapeutic proquality assurance programs and are financially cedures while incurring the cost only of physician viable organizations. services for the plan. Even overhead expenses would be borne by the hospital. These incentives The regulations implementing TEFRA also give would be compatible with those of hospitals paid utilization and quality control peer review orgaaccording to diagnosis-related groups (DRGs), be-nizations (PROs) the responsibility of reviewing the care provided by HMOs and CMPs (50 FR 1341). Although the specifics are still under discussion, the intention is to tailor review to the different financial incentives and to stress review for underprovision rather than overprovision of services (302). PROs would be able to delegate quality review to committees made up of HMO physicians. The contract that HCFA requires for eligible plans stipulates that as part of its quality assurance program the organization agrees to comply with requirements in the regulations for PRO review of services to Medicare enrollees and to furnish the PRO pertinent data (Article IV General Conditions) (533). The organization also agrees to disclose required financial information and to comply with other reporting requirements designed to monitor continued compliance with the regulations. Prior to TEFRA, HMOs also had to have quality assurance programs and to agree to review by professional standards review organizations (PSROs) (533). In fact, review for HMOs like other practices pertained only to inpatient hospital cases (325). The final set of concerns about quality stem from the characteristics of Medicare beneficiaries. Elderly poor people and elderly people generally have medical and social needs that differ from those of employed populations or even of Medicaid beneficiaries (194). Elderly people are more likely to have chronic illnesses and conditions such as impaired heart or lung function that complicate management of acute illness. Medicare beneficiaries are also more likely than the general population to have motor or sensory impairments that may affect their ability to cope with unfamiliar administrative arrangements, especially those of a large bureaucracy. On the other hand, once a beneficiary becomes familiar with HMO procedures, administrative matters may be more simple than under fee-for-service arrangements because the patient has less paperwork. The evaluation of Medicare competition demonstrations will provide information on whether or not these concerns are well founded. In the meantime, they suggest the need for monitoring the experience of Medicare beneficiaries in capitation plans, especially if Medicare enrollment was Ch. 7-Capitation Payment • 201 expanded into new and rapidly growing plans (194). Factors to monitor would include an increase in preventable deaths, reduction in functional status because of failure to provide services such as physical therapy, and deterioration in quality of life from failure to perform expensive therapies such as coronary artery bypass surgery or artificial hip replacement. If concern lay mainly with plan rather than physician incentives to provide too few services, emphasis could be directed to the availability of resources that hinge on management decisions, such as the number of certain specialists per enrolled population or the availability of certain expensive technologies. Access and Selection Bias If risk-sharing plans continued the low levels of cost-sharing that have typified prepaid groups and erected no additional bureaucratic barriers to access, it is likely that Medicare beneficiaries would have improved financial access to care compared to present Medicare coverage. Enrollees of prepaid groups have been more likely than people in comparison practices to have at least one physician visit during the year (279), a result consistent with the findings of the Rand Health Insurance Study that the likelihood rises with lower cost-sharing (343). The experience has been mixed with enrollees of IPAs (279). Medicaid eligibles in prepaid groups were also found more likely than controls to initiate visits (279). No such pattern has been evident for followup visits (279), a finding consistent with physicians' rather than patients' being more likely to initiate such care. Medicare beneficiaries' geographic access to care might be reduced, especially for specialized services. Plans paid by capitation have an incentive and the ability to match equipment, facilities, and staff to the enrolled population. An example is regionalizing facilities to take advantage of economies of scale in producing technical services. Where regionalization of services was evident in the San Francisco area of Kaiser-Permanente, some larger hospitals were fully equipped, and some smaller ones were equipped for emergency and chronic care (280). Kaiser enrollees may have had longer travel times in some cases. Reduced geographic access could pose problems for 202 • Payment for Physician Services: Strategies for Medicare beneficiaries who have vision or mobility impairments. On the other hand, access could be improved to the extent that more facilities are avail able in one place. A major concern about access stems from the possibility that plans might attempt to enroll lowrisk or low-cost beneficiaries. If such preferred selection took place, beneficiaries with conditions that put them at high risk of using expensive services might have difficulty finding plans to accept them. A standard benefit package and an open enrollment period during which plans were re quired to enroll people in the order they applied could alleviate this problem (129). TEFRA already mandates an annual 30-day open enrollment period during which plans must accept enrollees on a first-come, first-enrolled basis. But biased selection, whether by plans or beneficiaries, may have occurred during previous open enrollment periods (278). Constructing capitation rates that would adequately reward plans for caring for beneficiaries in high-cost categories would reduce plans' preference for low-cost enrollees and might result in plans' preferring to enroll high-cost beneficiaries. Cost and Efficiency Prepaid group practices have achieved savings in total per capita costs (premiums plus out-of pocket expenses) of 10.to 40 percent versus com parison plans (279). Earlier studies could not distinguish the role of differences in benefit cov erage, characteristics of enrollees, payment meth ods, and scope of services (extent of vertical in tegration). However, Rand's National Health Insurance Study assigned people randomly and covered comparable benefits (285). In that study, the expenditure rate for enrollees of Group Health Cooperative, the prepaid group practice, was 25 percent lower than that for nongroup fee-for-serv ice enrollees who received free care. There were no significant differences, however, between ex penditures for the HMO enrollees and for people subject to 95 percent coinsurance for fee-for-serv ice care. People with high cost-sharing had lower visit rates than prepaid group enrollees, but not significantly different hospitalization rates. These results suggest that prepaid group practice and high cost-sharing had similar effects on expenditures and hospital use, but that prepaid group enrollees were not so deterred from seeking care (343). As described earlier, studies have found that, compared to control groups, Medicare payments were lower for beneficiaries in five of seven prepaid groups that had cost contracts (91). Lower costs can be achieved by producing technical services at lower cost or by using a lower cost mix of services to provide the same quality care. In general, HMOs were not found to produce services more efficiently (279), although a recent study reported that within a hospital-based clinic, HMO patients had significantly fewer visits and lower laboratory charges for hysterectomies and appendectomies and lower total charges for appendectomies, but not for cholecystectomy and hernia (23). For the general population and for Medicare beneficiaries, savings have been attributed to lower hospital admission rates for both medical and surgical diagnoses. This phenomemon is consistent with providing a lower cost mix of services. Capitation payment contains an incentive to deliver care in the most efficient setting with the most efficient mix of technologies. Cataract removal, for example, would be likely to be performed almost exclusively in an outpatient set ting, except for patients with complicating comer bid conditions. Expensive technologies with high fixed costs, such as MRI or ESWL, might be regionalized by entrepreneurs or plans in free standing diagnostic or therapeutic centers. As noted above, an implication of having a defined population to serve is that an organization can match facilities and staff to that population. Prepaid groups have historically had lower numbers of surgeons and hospital beds per population (483). This result, of course, may indicate not the efficient use of resources, but the enrollment of a population with lower use of those services. Out-of-plan use has not accounted for the differences in costs between prepaid groups and other practices, and neither ambulatory physician visits nor ancillary use has been markedly lower (279). Nor have HMOs held their rate increases below those of fee-for-service practices; suggest ing that any reductions are one-time savings (279,343) and that capitation plans have been able to maintain a lower level of costs over time. Other organizational arrangements besides prepaid group practice have had low hospitalization rates (436). Two fee-for-service practices that owned their own hospitals, an ambulatory feefor-service group, and physicians in solo practice acting as case managers have achieved low hospitalization rates comparable to those of prepaid groups. These findings suggest the importance of organization as a factor separate from payment method. Lower surgical rates but not savings in expenditures have been found for IP A enrollees compared to people insured with Blue Shield or indemnity plans (279,483). The hospitalization rates of IPAs have been much lower than those of the general population, a mean of 448 days per 1,000 enrollees compared with about 737 per 1,000 U.S. population under age 65 for 1983 to 19849 (240,549). These figures were not age-sex adjusted and, like other comparisons involving HMOs, it is not known whether enrollees were representative of the population. It is also not clear whether consistent definitions of hospital days were used regarding newborns, Medicare beneficiaries, and benefits coordinated with other insurers. Among types of HMOs, network models, in which an HMO contracts with two or more group practices to provide medical services, and group models have had the lowest hospitalization rates, followed by staff models and IPAs (239,240). Since organizational formats besides prepaid group practice appear capable of achieving efficiencies, it is not clear which type of arrangement would predominate in a situation where the incentives of capitation payment pushed plans and providers to operate more efficiently. How plans paid by capitation react would depend greatly on the structure of capitation rates, which, as described in an earlier section, could reward creative marketing strategies to enroll low-cost beneficiaries or could reward efficient delivery of care. Medicare program expenditures over time would depend on which occurred. In any case, 9The information on IPAs covers July 1, 1983, to June 30, 1984. Data for the U.S. population are an average of rates for calendar years 1983 and 1984. Ch. 7-Capitation Payment • 203 program expenditures would be more predictable and controllable than under the current CPR system. Beneficiaries' costs would be likely to fall if risksharing plans, as now, were required to share savings with beneficiaries in the form of increased benefits or reduced premiums. Beneficiaries' costs would not rise in the absence of a Federal policy decision to increase their financial liability under the program. If Medicare's capitation rates were comparable to payments by other payers, capitation payment for beneficiaries would in itself be unlikely to affect the expenditures of other payers. If Medicare's rates were much below the market rate, plans and providers would find non-Medicare enrollees more attractive and would be expected to shift their marketing and provision of services to them. In either case, the ultimate effect on the costs of other payers would depend on the cost-saving activities undertaken by them (see app. D) and by the competitive pressure on providers engendered by such independent changes as increases in physician supply (see ch. 2). Technological Change Capitation payment would expand the changes in market conditions created by Medicare's prospective payment system for inpatients, so that the development of cost-saving technologies would be rewarded beyond the inpatient setting. In recent decades, the prevalence of insurance coverage increasingly provided a secure and growing market for medical technologies (487). In the context of open-ended third-party reimbursement, new technologies, especially those for acute inpatient care, were valued if they provided additional benefits, such as improved diagnosis or treatment. Potential purchasers and users of technology paid little attention to cost because their charges or costs were usually reimbursed. The constraints of capitation payment would make providers more cost conscious about the capital and operating costs of technology. Such a change already appears to be taking place regarding inpatient care as a result of Medicare's payment by DRGs (489). Depending on Medicare's leverage from its market share, Medicare's 204 • Payment for Physician Services: Strategies for Medicare paying for ambulatory care on a prospective baat least the larger prepaid groups have sent their sis would extend these incentives to physicians' patients to facilities outside the plan. Kaiseroffices, freestanding centers, and hospital outPermanente in Northern California used this stratpatient departments. Since DRG payment applies egy for X-ray computed tomography (CT) scan to inpatient operating expenses and some method ning and for open-heart surgery (129,480). If higher rates of use or lower technology cost later of including capital costs seems likely, market incentives would change even if the capitation paymade it cheaper for the plan to provide the servment applied only to ambulatory care and phyice inside the plan, the facilities were added to the sician services. plan. Otherwise, the plan continued to contract outside the plan for those specialized services.Physicians' offices have become a more attractive target for technology development and marUnder more generalized Medicare capitation keting since Medicare DRG payment. Under capipayment, one would thus expect more delay in tation payment manufacturers would attempt to adoption of an expensive technology such as MRI, incorporate cost-saving features in the technolespecially while its demonstrated advantages over ogies for the ambulatory market as well as the alternative modalities are fairly limited (234). As hospital market. For example, physicians who long as use inside the plan was low and the techcontinued to perform clinical laboratory tests in nology remained expensive, a plan would be likely their offices would have greater interest in equipto contract for MRI services outside the plan. ment to perform simple tests that was inexpen Capitation payment could greatly boost the desive to purchase and did not require expensive velopment of managerial technologies (483). Altechnicians to operate. though prepaid group practices have been able to Most technology would continue to be develdeliver medical care at lower cost than comparioped for hospitals, because the most severe, comson practices, the capability of other organizaplex, and expensive cases would be treated there tional arrangements has not been subjected to a market test. As noted above, there are indications and because hospitals would continue to account for a large portion of the medical market. The that other formats, such as multispecialty fee-formedical community would continue to value techservice group practice, can also achieve lower nologies that clearly improved diagnosis or thercosts. And IPAs appear to be evolving in the diapy, even if they increased costs, especially since rection of capitation payment to physicians and physician researchers are typically involved in greater utilization controls. Greater cost contechnology development, evaluation, and initial straints from capitation payment would probably adoption. It is also possible that hospitals would stimulate the development of other arrangements continue to compete for physicians and patients and the spread of those that were successful. by acquiring new technologies. This phenomenon seems to be occurring even under DRG payment in northern Virginia, where the largest hospital Capitation Payment to Geographic wished to purchase an ESWL unit that the hospi tal hoped would serve the Washington, DC, metFiscal Intermediaries ropolitan area (413). Instead of paying individual plans a capitation payment, Medicare could pay fiscal intermediaries Nevertheless, even within that context, the that were willing to assume financial risk for serv more cost-conscious environment would discour age the development and adoption of some techices to beneficiaries in a geographic area (70,242, nologies, especially expensive ones that added to 564). The capitation payment could cover only the cost of care. No evidence has been found that physician services and ambulatory care or could HMOs have been less likely to use expensive techalso encompass inpatient services. The intermediary or carrier in tum would negotiate arrange nologies for their patients (582). But in the early phase of an expensive new technology, when apments with providers in the area and offer bene propriate use was unclear and use rates were low, ficiaries a choice among plans. Ch. 7-Capitation Payment • 205 No such arrangement has existed under MediIt is not clear how several administrative matcare. But in response to an HCFA solicitation, at ters would be handled under carrier capitation,least one organization has submitted a proposal such as enrollment of beneficiaries, establishmentto undertake geographic capitation as a 5-year and updating of capitation payments, policy con demonstration project (290). Under the proposal, cerning case-by-case vs. mandatory assignment,Medicare would pay Blue Cross-Blue Shield of. and sharing of risk between Medicare and theMaryland to insure beneficiaries rather than only carrier.to administer the program in Maryland ( 434). 10 Medicare would pay the plan a monthly amount Quality of Care based on the number of Medicare beneficiaries (434). Half of any profits would go to the Fed A capitated carrier would have financial inceneral Government and the other half would be tives to control the use of providers who condivided between Blue Cross-Blue Shield and bentinued to be paid on a fee-for-service basis. The eficiaries, who would receive a rebate (434). Beneffects on quality would depend on which serveficiaries could choose from several options: ices were constrained or reduced (194). Qualitytraditional Medicare coverage, traditional health could improve if use decreased for services thatinsurance, Blue Cross-backed HMOs, and a Blue provide little or no additional benefit, entail un Cross-sponsored PPO, whose physicians would reasonable risk for the potential benefit, or arecharge lower fees. Other HMOs and CMPs would employed in inappropriate settings. On the other continue to relate directly with the Federal Govhand, quality would be impaired if decreased use ernment to be qualified and to enroll Medicare occurred for services that are now used approbeneficiaries. HCFA is reviewing the proposal, priately or that are underused. An important eleand officials estimate that a decision will require ment would thus be identifying services to targetseveral months (434). for utilization review. The carrier, perhaps withassistance from HCFA, might review the litera Putting the fiscal intermediary at risk formedical expenditures would greatly change present inture and work with panels of expert physicians centives for carriers, who now receive a fixed to select inappropriately used services that wouldamount per claim to administer the program, but be amenable to utilization review. who have no responsibility for the level of proIf the capitation payment did not cover ingram expenditures. To control expenditures, a patient care, providers and carriers would havecapitated carrier could encourage beneficiaries to financial incentives to contain their own expensesopt for lower cost alternatives, such as HMO by admitting patients to hospitals, a move thatmembership or physicians in a PPO; negotiate dis would be welcomed by hospitals paid more forcounts with physicians and other providers in the additional admissions. The entity charged with context of a PPO or HMO; or pursue more strin quality assurance could pay particular attentiongent review of fee-for-service claims (70). HCFA's to institutionalized patients. In addition, the riskguidelines for those who are considering the subsharing arrangements between Medicare and themission of demonstration proposals state that the carrier or between the carrier and providers couldpresent system of payment for physician services share any savings from reduced hospitalization. based on customary, prevailing, and reasonablecharges should be continued as an option for benMedicare would have continuing responsibileficiaries and providers. Although a fiscal interity to monitor the quality of care and to ensurethat appropriate covered services were not being mediary could enlist providers or plans to agree to utilization control or expenditure caps, the indenied to beneficiaries (70). This function wouldhave great importance in a situation that would termediary could not impose such constraints(465). be novel and perhaps initially confusing to beneficiaries. HCFA could draw on the experience oftwo Medicaid programs that operate through car 10Blue Cross/Blue Shield of Maryland now acts as Medicare's intermediary for Part A services and as Medicare's carrier for Part riers, one in Texas and the other in CaliforniaB services. (564). 206 • Payment for Physician Services: Strategies for Medicare Access to Care Beneficiaries could experience problems of geographic access to physicians and other providers if the intermediary contracted with a limited number of providers in an area. Medicare could alleviate this problem by requiring that the intermediary enlist the participation of a minimum percentage of physicians of different specialties and perhaps make arrangements to pay for outof-plan use, or by requiring that present arrangements remain an option for beneficiaries and providers. By contrast beneficiaries' access could be expanded if they were able to choose providers in PPOs and HMOs that were previously unavailable through traditional Medicare arrangements. Depending on the number of physicians and their practice preferences, beneficiaries in rural areas might not have access to an HMO or other practice forms (459). The enrollment process could affect access to certain facilities, providers, and services. Whether Medicare or the intermediary conducted the enrollment process, it would be critical for marketing of options and enrollment of beneficiaries to be conducted fairly, without favoring or slighting any of the intermediary or nonintermediarysponsored plans. Intermediaries would have a financial interest in encouraging enrollment in certain plans, such as the PPO, and in discouraging continuation of traditional Medicare arrangements. One possibility would be for the geographic intermediary to contract with HMOs that it did not sponsor and offer them as options to beneficiaries. That situation could inject some competition into arrangements with the intermediary and reduce the likelihood that low-cost beneficiaries would tend to be enrolled in plans sponsored by intermediary. Regardless of who conducted the process, Medicare could review the marketing material and stipulate certain procedures to be followed. In any case, it would be desirable to limit Medicare requirements to measures needed to protect beneficiaries and not to discourage plans and providers from participating. The enrollment of large numbers of beneficiaries in HMOs in Florida identified administrative problems regarding enrollment and disenrollment (477). Delay in updating beneficiary enrollment records led to initially incorrect payment decisions in some cases. Prior to signing TEFRA risk contracts in April1985, HCFA initiated procedures designed to ensure that HCFA and carrier records are updated in a timely fashion (533). Cost and Efficiency Depending on risk-sharing arrangements and the results of utilization control, the Medicare program could achieve greater predictability over program expenditures and greater control over annual increases. Geographic capitation has been likened to a carrier-wide IPA or CMP that covered both enrolled and unenrolled beneficiaries (70). IPAs have varied tremendously in their success, and specifically in their ability to control use and to reduce total expenditures below that of comparison practices. The achievements of an intermediary-at-risk would, like those of IP As, depend on its ability to negotiate with providers and to control their use of services. Under the Texas Medicaid program, the Texas Purchased Health Program, there has been a substantial increase in total expenditures and an increase in hospital outpatient visits relative to physician office visits (564). The State sets the fees to be paid for services, and the carrier attempts to control volume of services. The capitation payment to the carrier covers hospital and physician services and ancillaries. The Redwood Health Foundation is the carrier for all public assistance beneficiaries in three counties of California (564). Medi-Cal (California Medicaid) costs per enrollee in this area have been below the State average and the average in comparable counties, but slightly above costs per enrollee in prepaid health plans. Medi-Cal authorizes rates of fee-for-service payment. The carrier contracts risk-sharing arrangements with providers and conducts utilization and quality control. Like capitation payment to health plans, geographic capitation should not entail additional costs for beneficiaries and might result in savings in additional benefits or reduced cost-sharing. The cost implications for other payers would depend on the desire and ability of plans and providers to increase their private rates and use. Technological Change Any change in technology would depend on changes in the market for medical services and in the incentives to use different types of care. Such implications are very tenuous, for the reasons outlined above. If geographic capitation resulted in cost constraints and greater cost consciousness among providers, the effects on technology would be similar to those described for capitation payment to health plans. On the other hand, geographic capitation might produce few changes in the delivery of medical care, and present incentives for technological change could continue. Administrative Feasibility Use of the AAPCC to determine capitation rates would be less problematic for Medicare under geo graphic capitation than under capitation to risk sharing plans. A geographic intermediary would receive payments for large numbers of benefici aries across which the risk would be spread. Pay ment to risk-sharing plans would require further refinement in the capitation rate to reduce the pos sibility of biased selection by beneficiaries or plans. Quality assurance activities would differ from those historically undertaken, since the incentive CONCLUSION Capitation payment contains incentives for the recipients to control medical expenditures. Beneficiaries' welfare will be furthered if these incentives are expressed by providing care through a more efficient mix of services, reducing inappropriate care, or treating conditions before they become costly. Past experience of non-Medicare enrollees with HMOs, particularly with prepaid groups, has shown that capitation plans do care for enrollees at lower costs, while maintaining quality at levels equal to or better than comparison practices. However, there is the danger that future plans or intermediaries may constrain expenditures at the expense of quality of care, by reducing or de- Ch. ?-Capitation Payment • 207 of capitation payment is toward underprovision rather than overprovision of services. It is likely that some experience will be gained with these issues as PROs or their designates in the plans undertake the reviews required by the TEFRA regulations. Both rate-setting and quality assurance would most likely require new data, for example, on health status, severity, or outcome measures (459). It would be most reasonable for HCFA to determine from research and demonstrations what kind of data was needed. HCFA might also wish to survey beneficiaries about enrollment to indicate their characteristics and motivations in selecting and changing plans. It would be important for HCFA to be judicious in its requirements for information so that intermediaries were not unduly burdened. Both capitation payment to health plans and to geographic intermediaries would require that procedures be established regarding enrollment. As discussed above, geographic intermediaries may have conflicts of interest with respect to ben eficiary enrollment in nonintermediary-sponsored plans. Under either capitation payment arrange ment, however, it would be vital for HCFA and its intermediaries to coordinate their activities and to have timely, orderly, and accurate procedures for enrollment and disenrollment. laying appropriate services, or at the expense of access to care, by giving preference to low-cost over high-cost enrollees. Similar trade-offs apply to sharing financial risk with providers. Placing greater financial risk on physicians gives them a stronger incentive to contain costs, but also increases the likelihood that appropriate care will be reduced. The major disadvantage of geographic capitation is the substantial market power given to the intermediary, not only with regards to plans competing for beneficiaries to enroll, but also in relation to the Medicare program. A geographic intermediary would be in a strong position in negotiating capitation rates with Medicare because 208 • Payment for Physician Services: Strategies for Medicare of the difficulty that Medicare would face if the intermediary opted out of the arrangement after a few years. From the intermediary's perspective, an important factor would be the reliability of Medicare in continuing this payment approach and in paying rates considered reasonable by the intermediary. The extent to which plans and providers are pushed to be efficient depends on the level of payment as well as the method of payment. At lower levels of capitation payment, there would be a greater likelihood that lower use and cost would be achieved at the expense of quality. Moreover, if relative payment rates diverged from the cost of resources required to care for high-cost and low-cost beneficiaries, plans would be more likely to concentrate on marketing strategies to seek lowcost enrollees and less likely to urge providers to deliver cost-effective care. It is difficult to predict the implications of widespread Medicare capitation payment on the basis of financial incentives and past experience. Medicare enrollment in risk-sharing plans has only recently reached substantial numbers, mostly in demonstration projects that remain to be evaluated. And one cannot assume that new plans, which differ in size, sponsorship, organization, and risk-sharing arrangements from the older, well-studied ones, will achieve similar results in cost, quality, and access. Geographic capitation for Medicare beneficiaries is completely untried as yet, and little experience exists at the State level under Medicaid. Capitation payment, especially to risk-sharing plans, has the potential to moderate the growth in Medicare expenditures while providing beneficiaries with good quality care. The challenge is to develop a method for setting capitation rates that provides incentives for intermediaries and providers to deliver cost-effective care and to provide access to all beneficiaries. The Deficit Reduction Act of 1984, Public Law 98 369 (Sec. 2309), mandated OTA to conduct a study of physician payment under Medicare. In addition, the Senate Special Committee on Aging requested OTA to analyze the effects of physician payment on medi cal technology, with particular attention to payment of physician services under Part B of Medicare. On June 21, 1984, the OTA Technology Assessment Board approved the proposal for this project. During the early part of the project, OTA staff consulted with professional associations for suggestions of candidates for the study's advisory panel. The advisory panels for OTA studies guide OTA staff in selecting material and issues to consider and review the written work of the staff, but the panels are not responsible for the content of the final reports. The advisory panel for this study consisted of members from various interested parties: medical specialties; corporate health benefits; health insurers; carriers; consumer advocacy groups; and scholars in medical ethics, economics, and health policy analysis. Sidney Lee, president of the Milbank Memorial Fund and chair of the standing OTA Health Program Advisory Committee, chaired the advisory panel for this study. The first meeting of the advisory panel was held on September 10, 1984. Before the meeting, the staff began preliminary research into the issues involved in Medicare physician payment and prepared a draft outline for the study. During the meeting, the panel was asked to define and narrow the scope of the task of studying Medicare physician payment. Staff from the Health Care Financing Administration, the Institute of Medicine of the National Academy of Sciences, and the Congressional Budget Office reported on the progress and emphasis of their complementary studies on physician payment to help define the focus of OTA's assessment. After the meeting, the project staff refined the project outline and identified for analysis four alternative approaches to medicare payment for physician services: 1) modifications in Medicare's traditional customary, prevailing, and reasonable method of payment; 2) payment based on fee schedules; 3) payment for packages of services; and 4) capitation payment. The staff also selected five medical technologies for indepth examination of the effects of payment alternatives. Contracts were let for background papers that would each examine one of the five technologies: pneumococcal vaccination, clinical laboratory services, magnetic resonance imaging, extracorporeal shock wave lithotripsy, and cataract surgery. Appendix A Method of the Study On January 29, 1985, a workshop was held to discuss empirical research on the effects of particular payment mechanisms. The workshop, under the chair of Uwe Reinhardt, professor of economics at Princeton University and advisory panel member, included members of the advisory panel and others experienced in the use of databases available from Medicare carriers. In light of the information gained from this workshop, the OTA staff let two contracts for empirical studies on the issues related to fee schedules and capitation payment. The project was discussed further at the February 11, 1985 meeting of the Health Program Advisory Committee, an independent body of experts that advises the OTA Health Program. Discussion centered around the availability of data on physician incentives under various forms of payment and the relative uncertainty about the effects of particular changes. Another set of background papers was commissioned to elicit comparative perspectives on potential methods of paying for physician services. One contractor was chosen to write a background paper on the experience of the Canadian Government in financing a national system of payment for physician services on the basis of fee schedules. Another contractor wrote of the experience of the Kaiser-Permanente Medical Care Program in paying its physicians on a capitation basis. In addition, contractors were chosen to write background papers on the implications of the alternative payment methods for quality of care and ethical issues, matters that are common to all of the alter natives. The second meeting of the advisory panel was held on March 7, 1985, to bring the panel members up to date on the progress of the report. The panel reviewed draft background information intended for the final report. The panel also gave advice on issues for the chapters on the specific payment alternatives. During the spring and summer of 1985, the project staff reviewed the available literature relating to the various payment methods. Draft background papers were also received throughout this time, and the drafts were critiqued by the project staff, by advisory panel members, and by outside reviewers with expertise in the relevant fields. The staff also organized a workshop, held on June 13, 1985, on the administrative issues relating to possible changes in Medicare payment of physician services. The workshop participants, under the chair of Sidney Lee of the advisory panel, included representatives from the Health Care Finane 56-119 0 -86 -8 QL 3 212 • Payment for Physician Services: Strategies for Medicare ing Administration, current Medicare carriers, other insurers, and members of the advisory panel. The staff prepared a draft report, which was discussed at the final meeting of the advisory panel on October 10, 1985, and at the meeting of the Health Program Advisory Committee on October 18, 1985. The draft was also sent to other experts and interested parties for review. During October and November 1985, the project staff revised the report in response to reviewers' comments and sent selected chapters for additional review to members of the advisory panel. After subsequent revision, the staff prepared a final draft, which was submitted in mid-December to the Technology Assessment Board for approval. Other documents in addition to the main report were prepared in connection with this assessment. A case study, Effects of Federal Policies on Extracorporeal Shock Wave Lithotripsy, was prepared by the project staff and will be available through the U.S. Government Printing Office. In addition, the following papers were prepared on contract to OTA to provide background information for the main report and are available through OTA in limited quantities: • "The Frozen North: Controlling Physician Costs Through Controlling Fees," by Morris L. Barer, Robert G. Evans, and Roberta Labelle, University of British Columbia; • "Evaluation of Ethical Implications of Selected Alternatives for Paying Physicians Under the Medicare Program," by Alexander M. Capron, University of Southern California; • "Payments to Physicians in the Permanente Medical Group," by Morris F. Collen, Northern California Kaiser-Permanente Medical Care Program; • "Background Paper on Cataract Surgery and Physician Payment Under the Medicare Program," by Louis P. Garrison, Jr., and Sandra M. Yamashiro, Project HOPE Center for Health Affairs; • "Evaluation of Effects on the Quality of Care of Selected Alternatives for Paying Physicians Under the Medicare Program," by Glenn T. Hammons, Robert H. Brook, and Joseph P. Newhouse, The Rand Corp.; • "Reform of Medicare Physician Payment Policies: Impact on Magnetic Resonance Imaging Technology," by Lisa I. lezzoni, Oren Grad, and Mark A. Moskowitz, Boston University Medical Center; • "Analysis of Issues Relating to Implementing a Medicare Physician Fee Schedule," by David A. Juba, The Urban Institute; • 'The Effects on Clinical Laboratory Services of Selected Alternatives for Paying Physicians Under the Medicare Program," by Lois P. Myers, John M. Eisenberg, and Mark V. Pauly, University of Pennsylvania; • "Implications of Alternative Medicare Payment Methods for Pneumococcal Vaccination," by Michael A. Riddiough, Riddiough & Associates; • "Extracorporeal Shock Wave Lithotripsy: Clinical Applications and Physician Payment," by Jonathan A. Showstack, Eliseo J. Perez-Stable, and Eric Sawitz, University of California, San Francisco; and • "Issues in Capitation: Risks of Financial Ruin for Providers and Ways To Control This Risk," by James Vertrees, Dennis Tolley, and Kenneth Manton, La Jolla Management Corp. Appendix B Acknowledgments and Health Program Advisory Committee This project has benefited from the advice and review of several other people in addition to the advisorypanel. The staff would like to express its appreciation to the following people for their valuable guidance. James Aquavella American Academy of Ophthalmology Rochester, NY ·1 Mary Ann Ba1 y Department of Economics George Washington University Washington, DC Robert Ball Center for the Study of Social Policy Washington, DC James Barnett Bureau of Program Operations Health Care Financing Administration Baltimore, MD Marshall Becker Department of Health Behavior School of Public Health University of Michigan Ann Arbor, MI Ellis Benson Department of Laboratory Medicine and Pathology University of Minnesota Medical School Minneapolis, MN Mark S. Blumberg Kaiser Foundation Health Plan, Inc. Oakland, CA Charles Booth Office of Reimbursement Policy Health Care Financing Administration Baltimore, MD Dan Brock Department of Philosophy Brown University Providence, RI Ira Burney Office of Legislation and Policy Health Care Financing Administration Washington, DC Robert Butler Bureau of Data Management and Strategy Health Care Financing Administration Baltimore, MD Thomas W. Byrne Blue Cross/Blue Shield of Massachusetts Boston, MA James Cantwell U.S. General Accounting Office Washington, DC Sandra Christensen Congressional Budget Office Washington, DC John Clark Bureau of Health Care Delivery and Assistance Health Resources and Services Administration U.S. Department of Health and Human Services Rockville, MD Catherine Grealy Cohen American Society of Clinical Pathologists Washington, DC Eunice Cole American Nurses Association Kansas City, MO Andre-Pierre Contandriopoulos Department de I'Administration de la Sante Universite de Montreal Montreal, Canada Robert Crane Kaiser Foundation Health Plan, Inc. Oakland, CA Patricia M. Danzon Center for Health Policy Studies Duke University Durham, NC Ronald Deacon Office of Demonstrations and Evaluation Health Care Financing Administration Baltimore, MD 214 • Payment for Physician Services: Strategies for Medicare Linda Demlo Office of Demonstrations and Evaluation Health Care Financing Administration Baltimore, MD Richard DiMonda Division of Technology Management and Policy Office of Hospital Management Programs American Hospital Association Chicago, IL Allen Dobson Office of Reimbursement Health Care Financing Administration Baltimore, MD Thomas Dowdal U.S. General Accounting Office Washington, DC Steven P. Dretler Department of Urology Massachusetts General Hospital Boston, MA Charles F. Duvall Private practice Washington, DC Richard Egdahl Health Care Research Unit Boston University Medical Center Boston, MA David Ehrenfried Blue Cross and Blue Shield Association Chicago, IL Alfred Ercolano College of American Pathologists Washington, DC AI Esposito Office of Demonstrations and Evaluation Health Care Financing Administration Baltimore, MD David S. Fedson Department of Internal Medicine School of Medicine, University of Virginia Charlottesville, VA Jon Gabel National Center for Health Services Research and Health Care Technology Assessment U.S Department of Health and Human Services Rockville, MD Paul Ginsburg The Rand Corp. Washington, DC Merwyn Greenlick Health Services Research Center Kaiser-Permanente Medical Care Program Portland, OR Leonard Gruenberg Health Policy Center Heller Graduate School Brandeis University Waltham, MA Paul Gurney Office of Demonstrations and Evaluation Health Care Financing Administration Baltimore, MD James Hadley Office of Demonstrations and Evaluation Health Care Financing Administration Baltimore, MD John D. Haytaian FONAR Corp. Melville, NY Harold Heatherington Office of Reimbursement Policy Health Care Financing Administration Baltimore, MD Alan Hinman Centers for Disease Control Atlanta, GA Mark Hornbrook Health Services Research Center Kaiser-Permanente Medical Care Program Portland, OR Fredrick Hunt Society of Professional Benefit Administrators Washington, DC Steven Jencks Office of Research Health Care Financing Administration Baltimore, MD Lucy Johns Health Care Planning and Policy San Francisco, CA App. a-Acknowledgments and Health Program Advisory Committee • 215 Frank Jolesz Department of Radiology Brigham and Women's Hospital Boston, MA Terry Kay Office of Research Health Care Financing Administration Baltimore, MD Carol Kelly Office of Legislation and Policy Health Care Financing Administration Washington, DC Janet Kline Education and Public Welfare Division Congressional Research Service Library of Congress Washington, DC Mary Nell Lehnard Blue Cross and Blue Shield Association Washington, DC Dieter A. Lehnortt American College of Emergency Physicians Dallas, TX Joanna Lion Health Policy Center Heller Graduate School Brandeis University Waltham, MA A. Russell Localio Risk Management Foundation of the Harvard Medical Institutions Cambridge, MA James Lubitz Office of Research Health Care Financing Administration Baltimore, MD Harold Luft Institute for Health Policy Studies School of Medicine University of California, San Francisco San Francisco, CA Penn Lupovich Group Health Association Washington, DC Leo Marcus MANDEX, Inc. Vienna, VA Cindy Mason Office of Demonstrations and Evaluation Health Care Financing Administration Baltimore, MD John McConnell National Committee on Clinical Laboratory Standards Villanova, P A David Mechanic Health and Health Services Research Coordinating Council Rutgers University New Brunswick, NJ Stephanie Mensch American Academy of Ophthalmology Washington, DC William H. Moncrief, Jr. California Medical Review, Inc. San Francisco, CA Laura Murphy National Electrical Manufacturers Association Washington, DC Robert Mussachio Center for Health Policy Research American Medical Association Chicago, IL Helen Oglesby Blue Shield of California San Francisco, CA Heather Palmer Department of Health Policy and Management Harvard School of Public Health Boston, MA David Plotnick Group Health Association of America Washington, DC Martin Resnick School of Medicine, Division of Urology Case Western Reserve University Cleveland, OH Roger Reynolds American Medical Association Chicago, IL 216 • Payment for Physician Services: Strategies for Medicare Thomas Rice Department of Health Policy and Planning School of Public Health University of North Carolina Chapel Hill, NC Paul Riesel Bureau of 8igibility, Reimbursement, and Coverage Health Care Financing Administration Baltimore, MD Gerald Riley Office of Research Health Care Financing Administration Baltimore, MD Jack Rodgers Congressional Budget Office Washington, DC Louis Rossiter Medical College of Virginia Virginia Commonwealth University Richmond, VA Earl Schwartz Bureau of Data Management and Strategy Health Care Financing Administration Baltimore, MD Anne Scitovsky Health Economics Department Palo Alto Medical Foundation Palo Alto, CA Marc Segal Department of Health Care Resources American Medical Association Chicago, IL Ralph Shaffarzick Blue Shield of California San Francisco, CA Philip A. Shelton Eye Physician Associates Hartford, CT George Silver Health Research Group Washington, DC Malcolm Sneen Bureau of Data Management and Strategy Health Care Financing Administration Baltimore, MD William Sobaski Office of Research Health Care Financing Administration Baltimore, MD George F. Stevenson American Society of Clinical Pathologists Chicago, IL Eugene Stickler Bureau of Data Management and Strategy Health Care Financing Administration Baltimore, MD Rosemary Sweeney The American Academy of Family Physicians Washington, DC Sherry A. Terrell Office of Research Health Care Financing Administration Baltimore, MD Edward R. Thoms Pennsylvania Blue Shield Camp Hill, P A Joan B. Trauner Institute for Health Policy Studies School of Medicine University of California, San Francisco San Francisco, CA Sidney Treiger Office of Demonstrations and Evaluation Health Care Financing Administration Baltimore, MD Leroy Walters Center for Bioethics Kennedy Institute of Ethics Georgetown University Washington, DC Peter Welch Office of Management and Budget Washington, DC Norman Welford Bureau of Medical Devices Food and Drug Administration Silver Spring, MD Howard West MANDEX, Inc. Vienna, VA App. a-Acknowledgments and Health Program Advisory Committee • 217 Richard E. Wild Office of Reimbursement Policy Health Care Financing Administration Baltimore, MD Irwin Wolkstein Health Policy Alternatives Washington, DC David Worthen Veterans Administration Washington, DC Barbara Wynn Bureau of Eligibility, Reimbursement, and Coverage Health Care Financing Administration Baltimore, MD Bernice Young Bureau of Health Maintenance Organizations and Resources Development Health Resources and Services Administration U.S. Department of Health and Human Services Rockville, MD . Donald Young William Pepper Lab Hospital of the University of Pennsylvania Philadelphia, P A Workshop on Administrative Feasibility of Alternative Methods of Paying for Physician Services, June 13, 1985 Charles R. Booth Office of Reimbursement Policy Health Care Financing Administration Baltimore, MD Merwyn Greenlick Health Services Research Center Kaiser-Permanente Medical Care Program Portland, OR Jean Harris Office of Program Operations and Procedures Health Care Financing Administration Baltimore, MD Lisa Iezzoni Health Care Research Unit Boston University Medical Center Boston, MA Stephen Isaacson Policy Division Office of the Civilian Health and Medical Program of the Uniformed Services Aurora, CO Ronald M. Klar Health Services Analysis, Inc. Washington, DC Sidney Lee, Workshop Chair Milbank Memorial Fund New York, NY Arthur Lifson Equitable Life Assurance Society of America New York, NY Chris McEntee American Association of Retired Persons Washington, DC Helen Oglesby Blue Shield of California San Francisco, CA C. Burns Roehrig American Society of Internal Medicine Boston, MA Christie Somers Health Services Alternative Delivery System Development John Hancock Insurance Co. Boston, MA Robert Taylor American Academy of Family Physicians Spartanburg, SC Howard West MANDEX, Inc. Vienna, VA 218 • Payment for Physician Services: Strategies for Medicare Health Program Advisory Committee Sidney S. Lee, Chair President, Milbank Memorial Fund H. David Banta Project Director STG Project on Future Health Technology The Netherlands Rashi Fein Professor Department of Social Medicine and Health Policy Harvard Medical School Harvey Fineberg Dean School of Public Health Harvard University Patricia King Professor Georgetown Law Center Joyce C. Lashof Dean School of Public Health University of California, Berkeley Alexander Leaf Professor of Medicine Harvard Medical School Massachusetts General Hospital Frederick Mosteller Professor and Chair Department of Health Policy and Management School of Public Health Harvard University Norton Nelson Professor Department of Environmental Medicine New York University Medical School Robert Oseasohn Associate Dean School of Public Health University of Texas Nora Piore Senior Fellow and Advisor to the President United Hospital Fund of New York Dorothy Rice Regents Lecturer Department of Social and Behavioral Sciences School of Nursing University of California, San Francisco Richard Riegelman Associate Professor George Washington University School of Medicine Walter Robb Vice President & General Manager Medical Systems Operations General Electric Milwaukee, WI Frederick C. Robbins University Professor Department of Epidemiology and Biostatistics School of Medicine Case Western Reserve University Cleveland, OH Frank E. Samuel, Jr. President Health Industry Manufacturers' Association Rosemary Stevens Professor Department of History and Sociology of Science University of Pennsylvania Appendix C Medicare and Medicaid Payment for Physicians' Services Introduction Third-party payment practices for physicians' services are complex and diverse. Third-party payers in the United States have traditionally paid a fee for each service1 provided by physicians. Nonetheless, there are a variety of approaches in actual payment practices under fee-for-service among third-party payers, including public programs. Diversity is expected because feefor-service is a generic term that includes multiple elements (e.g., payment basis, level determination, and payment updating schedule) that can be combined in numerous ways. Furthermore, public programs have broad policy discretion within Federal legislation, regulation, and guidelines in designing payments for physicians. This appendix describes third-party payment for physicians' services in the public sector, focusing on the Medicare program. A general description of the Medicare program is followed by a summary of the origins of the fee-for-service method adopted by the program and a description of the current payment methods for physician services under Medicare. Although fee-for service by far is the most common method, the Medicare program has adapted it in numerous ways for special circumstances and has sometimes used other payment methods. This appendix also includes a section on physician payment under Medicaid, highlighting similarities to and differences from the Medicare program. Medicare Payment for Physicians' Services The 1965 legislation that established Medicare under Title XVIII (Health Insurance for the Aged and Disabled) of the Social Security Act mandated eligibility for insurance benefits for most Americans 65 years and over.2 On July 1, 1973, the Social Security Amendments of 1972 (Public Law 92-603) extended eligibility to persons under 65 who have been entitled for a period of 24 months to Social Security or Railroad Retirement benefits because they are disabled, and to 'Throughout this appendix, the terms service and technology are used as synonyms. 'Although eligibility for Part A is tied to eligibility for Social Security, at the onset of the program, individuals who were age 65 and not eligible for Social Security were given 3 years to establish eligibility (445). most workers and their dependents with end-stage renal disease (ESRD). Medicare covers hospital insurance benefits (Part A) and supplementary medical insurance benefits (Part B). Table C-1 displays Medicare's current benefits and the financial responsibilities of the program and its beneficiaries under Parts A and B. Part A's primary purpose is to provide insurance against the costs of inpatient hospital care. Other benefits include payment for inpatient psychiatric services, skilled nursing facility services, home health services, hospice services, and comprehensive ambulatory rehabilitation facility services. Payment for most physician services is under Part B, which also includes payment for outpatient hospital services, ambulatory laboratory and X-ray services, ambulatory physical therapy and speech pathology services, and various other limited ambulatory services and supplies, such as prosthetic devices and durable medical equipment (see table C-1). Part B also covers home health services for those Medicare beneficiaries who have Part B coverage only. The law excludes most preventive services and certain other services, such as dental and custodial care. In order to pay for a new technology (service) that is not mandated or prohibited by law, a decision to cover the specific service, or technology, is required. (Coverage is distinguished from payment in that coverage refers to benefits available to eligible beneficiaries, and payment refers to the amount and methods of payment for covered services (585).) Impressive advances in the numbers and types of technologies available to the health care system in recent years has led to an increasing need for coverage decisions. Medicare. decides whether or not to cover a service on the basis of Section 1862 of the Social Security Act, which prohibits payment for items and services that are "not reasonable and necessary for the diagnosis or treatment of illness and injury or to improv~ the functioning of a malformed body member." The criteria Medicare uses to determine if a technology meets the broad statuatory language of "reasonable and necessary" are: 1) general acceptance as safe and necessary, 2) not experimental, 3) medically necessary, and 4) provided according to standards of medical practice in an appropriate setting.3 'The OTA report Medical Technology and Costs of the Medicare Program (486) includes a comprehensive discussion of Medicare's coverage process. 220 • Payment for Physician Services: Strategies for Medicare Table C-1.-Medicare Benefits and Limitations, as of January 1986 Kind of care Medicare pays Part A: Hospitalization 1-60 days 61-90 days 91-150 days (60 day lifetime reserve) After 150 days-no coverage Psychiatric Same as hospitalization Skilled nursing 1-20 days facility 21-100 days After 100 days-no coverage Home health Unlimited visits services Reasonable costs Hospice care Prospective payment rates, per day to maximum of $6,500 average "cap" per beneficiary to each facility Routine home care: $53.17 Inpatient respitea care: $55.33 General inpatient care: $271.00 Total continuous home care: $358.67 Part 8: Home health Unlimited visits services Reasonable costs Physician and other 8@% of approved charges after deductible medical services is met 100% of approved charges for services provided in approved ambulatory surgical center or hospital outpatient department if the physician accepts assignment Immunizations Pneumococcal vaccine Hepatitis B vaccine (for ESRD patients and others at high risk of hepatitis) Chiropractors' Manual spinal manipulation services Most routine foot Nothing care Dentists' services Jaw surgery, setting of facial fractures, treatment of oral infections Dentures Nothing Routine hearing Nothing and eye exams Eyeglasses and Nothing hearing aids Routine physical Nothing examinations Prosthetic devices Those needed to substitute for an internal body organ, or for artificial limbs and eyes, and arm, leg, back, and neck braces Durable medical If rented, approved charges equipment If purchased, monthly payments until Medicare's share is paid or equipment is no longer necessary For long-term use, payment may be made in a lump sum Medical supplies Dressings, splints, and casts Beneficiary pays Initial deductible ($492) Daily copayment ($123) Daily copayment ($246) Same as hospitalization Nothing Daily copayment ($61.50) Nothing 5% of cost to program for: -Drugs and biologicals (not to exceed $5 per prescription) -Inpatient respite a care (per day) (total not to exceed inpatient deductible) SMI basic premium$15.50/mo. Nothing Initial deductible ($75) 20% of approved charges Excess of physician charges above approved charges if physician does not accept assignment of benefits Nothing for covered vaccines, deductible does not apply All costs for all other vaccines All other charges All charges All other charges All costs All costs All costs All costs All costs 20% coinsurance All other costs (e.g., common first aid supplies purchased by patient) Comments Deductible and copayments are adjusted annually Lifetime reserve can be used only once Lifetime limitation of 190 days of coverage Beneficiary must be eligible for Part A Beneficiary may elect hospice care in lieu of other medical care services (with its attendant deductibles and copayments), for two periods of 90 days and one of 30 days, to be taken in that order, upon determination of a terminal illness. Benefit provision expires Sept. 30, 1986. Beneficiary eligible for Part B only May cover other dental services when incident to the provision of covered medical services Examinations may be covered as incident to other diagnostic and therapeutic procedures, e.g., prior to surgery to correct hearing and vision disorders Examinations covered as incident to diagnosis and treatment Equipment furnished by provider is paid by Part A intermediary on a reasonable cost basis Physicians may bill for supplies provided at cost to them App. C-Medicare and Medicaid Payment for Physicians' Services • 221 Table C-1.-Medicare Benefits and Limitations, as of January 1986-Continued Kind of care Medicare pays Blood For all but first 3 pints Outpatient mental 62.5% of reasonable charges up to $500 illness (i.e., $312.50) Outpatient physical In doctor's office, 80% of approved therapy charges after deductible is met From physical therapist, $400/yr. maximum From clinic, home health agency, or other agencies, 80% of approved charges after deductible End-stage renal 80% of prospectively determined, per disease treatment regionally adjusted rates treatments Physicians' services incident to maintenance dialysis, 80% of monthly capitation rates Comprehensive Lesser of 80% of reasonable cost or the outpatient reasonable cost minus 20% of rehabilitation reasonable charges facilities (CORF) Rural health 80% of prospectively determined allservices inclusive per visit rate Beneficiary pays First three pints or replace 37.5% of reasonable charges up to $500, and 100% of charges above $500 $75 deductible and 20% coinsurance All costs above $400/yr. $75 deductible and 20% coinsurance $75 deductible and 20% coi nsu ranee $75 deductible and 20% of customary charges $75 deductible and 20% coinsurance Comments Coverage ends 12 months after the month maintenance dialysis stops or 36 months after month of kidney transplant In order for the beneficiary to receive reimbursement for CORF services, a physician must submit a plan of treatment which must be reviewed every 60 days. Coverage ends when no further progress is being made with respect to the goals specified in the plan aRespite care is defined as short-term (limited to 5 days) inpatient care provided to the individual only when necessary to relieve the family members or other persons caring for the individual during period of hospice election. SOURCE: Commerce Clearing House, Inc., Medicare and Medicaid Guide (Chicago, IL: Commerce Clearing House, Inc., 1985). Part A is an entitlement program and is available without payment of a premium to those eligible. 4 Individuals who are not automatically entitled may voluntarily obtain insurance by paying the full actuarial cost of such coverage ($174 per month in 1985) (471). Individuals eligible for Part A are automatically enrolled in Part B unless they indicate they do not wish to be enrolled. Any citizen or legal alien for 5 years who is age 65 and older, even individuals who are not eligible for Part A, may enroll in Part B, a distinct program under Medicare. Participation in Part Bis voluntary and requires payment of a monthly premium.5 The Part B, premium is deducted automatically from monthly Social Security checks, except in cases where States pay the premium or when work or some other event precludes payment of the monthly benefit check. Participation in Part B is high. In 1982, 99 percent of the eligible elderly and 92 percent of eligible disabled people in Part A were also enrolled in Part B (467). Medicare is administered through private contractors (intermediaries for Part A and carriers for Part B), all of whom maintain a private business as well as the Government contract business. In fiscal year 1984, Medicare had 60 carrier jurisdictions that were serviced by 39 carriers: 28 Blue Cross/Blue Shield Plans and 11 others (514). 'Thirty percent, or 4 million, of State and local employees are the major group of individuals currently not eligible for Part A (27). 'The Part B premium was $15.50/month as of jan. 1, 1985. Fee-for-Service Payment Background.-By the early 1950s, the movement for a national health insurance program for the entire population that began in the 1930s had become a proposal to assist Social Security beneficiaries with the costs of hospitalization. However, despite the limited nature of proposed health insurance legislation, successive attempts at passage failed until 1965, when President Johnson's active interest in a health insurance program during and after his successful bid for reelection and striking changes in the political composition of Congress overcame the resistance of opponents (27). The knowledge that some form of Medicare would pass caused some opponents to facilitate enactment of Government health insurance for the elderly and other opponents to sponsor health insurance bills for the elderly for the first time (287) In early 1965, revised versions of bills sponsored by the Administration (H.R. 1 introduced by Rep. Cecil R. King and S. 1 introduced by Sen. Clinton Anderson) were reintroduced, and Rep. James Burns, a former opponent, sponsored H.R. 4351-a Government health insurance bill. The King-Anderson bill called for compulsory contributions and was closely associated with the Social Security system. It limited benefits to hospital care, nursing home care, and home health care. The Burns bill included physician and other medical services as well as inpatient services. It 222 • Payment for Physician Services: Strategies for Medicare provided for voluntary participation and a Government subsidy, and was separated from the Social Security system (445). The Burns bill was modeled on a high-option Aetna policy available to members of the Federal Employees Health Benefits Program (27,580). In a strategic move, Rep. Wilbur Mills, chairman of the House Ways and Means Committee, proposed incorporating elements of both the King-Anderson bill and the Burns bill into Title XVIII of the Social Security Act.6 Title XVIII retains the basic philosophy of both bills. Although insurance for hospital services, Part A of Title XVIII, is financed by compulsory contributions of employers and employees through the Social Security system, insurance for physician services, Part B of Title XVIII, is voluntary and is financed from premiums paid by the insured and general revenues. The nonregulatory approach of H.R. 4351 was in corporated into the following statutory language (580): ... where payment ... is on a charge basis, such charge will be reasonable and not higher than the charge applicable for a comparable service and under comparable circumstances to the policy holders and subscribers of the carrier . . . In determining the rea sonable charge . . . there shall be taken into consid eration the customary charges for similar services as well as the prevailing charges in the locality for simi lar services. Although longstanding advocates of Medicare legislation had recognized that basing physician payment on physician charges was potentially inflationary, 7 they also knew that it was impractical to contest the method. If Medicare was to be passed, Rep. Mills' support was necessary, and tampering with his package in this major way would jeopardize his approval (27). Furthermore, the logical alternative, i.e., paying physicians on the basis of prospectively determined fees, exercised more control over physicians than a chargebased method and might have adversely affected physicians' cooperation with the program (287). Fee schedules were a traditional payment method for physician services in the United States that had the advantages of uniformity and ease of understanding by both patients and physicians. On the other hand, the charge-based method that Congress adopted was relatively new-it had first been used by a Blue Shield plan in Wisconsin in 1954. Blue Shield had also initiated a "prevailing fee program" for national accounts, which was designed to permit physicians to establish charges for their services without being limited by fee schedules or income levels (312). Furthermore, the 'The "Eldercare" bills, H.R. 3727 and H.R. 3728, became Title XIX of the Social Security Act. 'in the 1965 Senate debate, the leading Senate proponent of Medicare, Sen. Clinton Anderson noted that paying physicians their "usual and customary fees (the Burns suggestion) would significantly and unnecessarily inflate the cost of the program to the taxpayer and the aged" (287). method afforded physicians considerable latitude in establishing payment levels and was considered less intrusive than fee schedules in a physician's financial decisions. Specific definition of the terminology used in the legislation was lacking, but " ... fears of a physicians' boycott and the absence of an obviously attractive alternative, persuaded Senate reformers not to raise further questions about the sensitive issue of what constituted reasonable charges" (287). The congressional intent to make health care for aged citizens available without regard to income level W«S evident in the report of the Committee on Ways and Means to accompany H.R. 6675. The report states, "where payment is on the basis of assignment, the reasonable charge would have to be accepted as the full payment" (478). In the late 1960s and 1970s, the original statutory language was clarified by a series of regulations and administrative guidelines. In attempts to strengthen Government control over physician payment and to restrain the rising costs of the Part B program, tighter controls on the operations of the carriers were developed by increasing the frequency of updating physician charges and by freezing charge limits for various periods of time (27,565). The payment method and its administration described below have perpetuated a loosely administered, decentralized system with differences in payment levels among physicians. The implications of the Medicare payment system for current Part B costs and other effects of the payment system are discussed in chapter 2 of this report. Significant refinements to the current system are considered in chapters 4, 5, 6, and 7. Current Status.-As noted earlier, Title XVIII of the Social Security Act specifies that payment for physician services under Part B of Medicare are to be made on the basis of reasonable charges that are computed from usual, customary, and prevailing (CPR) charges. 8 The Part B program generally pays 80 percent of reasonable charges in excess of the beneficiaries' annual Part B deductible, $75 in 1985. Medicare carriers, private contractors that receive, process, and pay claims for Part B services, have the primary responsibility for determining the reasonable charge for each service provided. Their determinations are to be consistent with the law, regulations, and general principles and guidelines issued by the Health Care Financing Administration (HCFA) (509). Although the basic formula is applied uniformly nationwide, carriers exhibit great variation in executing the method 'Both usual, customary, and reasonable and CPR refer to a general system of computing a payment level based on historical and comparative profiles of physicians' charges. Since CPR is Medicare terminology, it will be used in this report. The terms reasonable, allowed, and approved are used as synonyms in this appendix. App. C-Medicare and Medicaid Payment for Physicians' Services • 223 ology in such areas as locality designation and specialty recognition because of the autonomy offered by the law and implementing instructions. Based on claims information, Medicare carriers maintain records of the services provided and the charges billed by physicians in their charge area. The carriers then develop individual statistical profiles and areawide statistical profiles of physician charges, which are updated annually and are in effect for a "fee screen" year.9 The standards per fee screen year have been based on charges submitted during the calendar year preceding the fee screen year, creating a lag period in updating (509). For example, the charge limits for the fee screen year July 1, 1982 to June 30, 1983, were based on charges received during the preceding calendar year January 1, 1981 to December 31, 1981 (509). The reasonable charge is the lowest of a physician's actual charge, a physician's customary charge (Level 1 fee screen), or the area's prevailing charge (Level 2 fee screen). There are special circumstances when the reasonable charge may not be the lowest of the above three charges. If there are unusual circumstances or medical complications causing essentially different services to be provided, the actual charge for the service may be specified as the reasonable charge even though the actual charge is higher than the customary charge and prevailing charge. The Social Security Act requires that the reasonable charge for a service may not be higher than the charge for a comparable service provided under comparable circumstances to a carrier's non-Medicare subscribers. The actual charge is the charge the physician has billed for the service provided. The customary charge is the physician's median submitted charge during the data collection period preceding the fee screen year. 10 The customary charge is fluid. If a physician revises his or her fees, the carrier will recognize the change when processing claims with the new charges. Until1976, the prevailing charge for a service was the lowest charge for the service that was greater than or equal to a percentile of the distribution of physicians' customary charges weighted by the number of times each physician billed for the service in a locality (designated as a "charge area") the previous calen 'As a result of provisions of the Deficit Reduction Act of 1984, discussed later, the fee screen year as of October 1, 1984, was changed from July 1-June 30th to October 1-September 30th and charge limits for the fee screen year will be based on charges submitted from April1-March 30th of the previous year. "The customary charge for a service provided by physicians beginning a new practice is based on the 50th percentile of the customary charges in a charge area weighted by how often physicians billed for the service (509). In calculating the customary charge, the carrier not only considers charges made by physicians to Medicare beneficiaries, but also considers charges made by the physicians to their patients in general. The amount of non-Medicare data included in the computation varies among carriers according to the size of their non-Medicare business (509). dar year. The prevailing charge limits were originally paid by some individual carriers at the 90th percentile. Medicare later set prevailing change limits at the 83rd percentile in 1969, and they have remained at the 75th percentile since 1971 (496). The Social Security Amendments of 1972 (Public Law 92-603) placed further limits on the yearly increases in prevailing charge levels-because of subsequent congressional action to prevent a rollback in approved charges, these limits were not fully implemented until 1976. The amendments established a Medicare Economic Index (MEl) that relates the rate of increase in physicians' fees to increases in general earning levels and increases in physician practice costs. The index, which is updated annually for a 12-month period beginning July 1, sets an annual cap on prevailings. Prevailing charges are now either the lesser of the prevailing charge ("unadjusted" prevailing) or the product of the 1973 fee screen year prevailing charge multiplied by the value of the current MEl ("adjusted" prevailing) (116). The MEl for 1983 was 2.063. If a prevailing charge for a certain service was $10.00 in 1973, and if the "unadjusted" prevailing was no less than $20.63, the prevailing charge for fiscal year 1983 would be $20.63 (the "adjusted" prevailing). However, if a prevailing charge for a certain service was $10.00 in 1973, and if the "unadjusted" prevailing in 1983 was less than $20.63, the prevailing would be set at the charge that is less than the $20.63 (the "unadjusted" prevailing). In implementing the CPR approach, each carrier is allowed considerable latitude in delineating a charge area. Carriers are expected to delineate localities based on their knowledge of local charging practices, service patterns, and differences in population density, economic levels, and other factors that affect charges for services. Charge areas are usually a subdivision of a State that includes a cross-section of the population (509). Thus, there is no uniform geographic configuration for a charge area. Four types of locality configurations are: 1) statewide localities, 2) regional localities (contiguous counties) without specific regard to urban/rural distinctions, 3) urban and rural localities comprised of noncontiguous areas, and 4) separate localities for major metropolitan areas with nonmetropolitan areas consolidated into one or more localities. Currently there are 240 geographic charge localities (514). Charge areas may also differ according to types and levels of services. For example, a carrier may decide in determining a prevailing fee screen that a State has seven localities for general practitioners, but only one locality (the entire State) for members of a particular specialty. 224 • Payment for Physician Services: Strategies for Medicare Furthermore, carrier practice concerning specialty recognition for the purpose of determining prevailing charges is also extremely variable as it is meant to reflect the existing patterns of charges within a locality. The variation in carrier practice ranges from carriers that use a single prevailing charge screen for services of all physicians11 to those carriers that calculate separate prevailing charge screens for individual specialties. Blue Shield of Pennsylvania, for example, has individual charge screens for more than SO distinct specialties (458). Some carriers calculate a prevailing charge screen for general practitioners and a prevailing charge screen for all other physicians. Other carriers group specialties into other categories, so that there may be one prevailing charge for all surgical specialties and another prevailing charge for medical specialties. Massachusetts constructs prevailing charge screens by type of service and recognizes 25 specialties in constructing prevailing screens for visits and consultative procedures, but only two groups (general and family practitioners, and other physicians) for other procedures (47S)Y The recognition of specialties is a complex issue, confounded by the lack of a clear definition of a specialist within the medical community. 13 Eighty-two specific physician specialties and subspecialties are reported by physicians and included in the Masterfile of the American Medical Association (124). At this time, some carriers define a specialist as one who is board-eligible in a particular specialty, and others limit the designation only to board-certified physicians. Still other carriers define specialists as physicians who classify themselves as such and who limit their practice to a particular specialty (30). "Carriers for the States of Florida, North Dakota, and South Dakota, the State of Kansas excluding Kansas City, western New York, and the combined territories of the Puerto Rico and the Virgin Islands use a prevailing charge screen for the services of all physicians. The American Society of Internal Medicine has brought suit against Florida Blue Cross Blue Shield to force recognition of specialists. The court suit also concerns differentiating between physicians and nonphysicians who use the same codes for a service, and discriminating between specialists and levels of expertise. The suit was withdrawn in 1985. "The variation in carrier practice can be explained by the lack of the specificity of the regulations concerning specialty practice and the wide latitude allowed carriers in implementing instructions. The regulations stipulate that: 1) carriers should be responsive to differentials in levels of charges among different kinds of services in establishing prevailing charge levels; 2) where general practitioners and specialists in a locality have established different levels of fees for their services, the carriers should recognize such differences in establishing prevailing charge screens; and 3) when the physicians have not themselves established fee differentials based on specialty practice, the carrier should not establish artificial ones (42 CFR 405.504). 13An important step on the part of the medical community in defining a medical specialist was taken on Mar. 20, 1984, when the Ad Hoc Committee on Designation of a Specialist of the Council of Medical Specialty Societies released guidelines for the designation of a physician as a specialist. The guidelines do not accept self-designation alone, but list four objective criteria to be used in verifying specialty designation. Medicare carriers frequently rely on relative value studies if there are insufficient charge data about a particular physician's use of a specific service to determine a customary charge screen or if there are insufficient data about the use of a service in a charge area to determine a prevailing charge .14 Physicians also rely on such studies when determining a fee for a new service. Relative value studies express the relationship between services in unit values and not dollar amounts. 15 In determining a physician's customary charge for a service, the carrier multiplies the relative value of the service and a monetary conversion factor that is derived from a physician's known customary charges for similar services in the same category of service (e.g., medicine, surgery, and radiology). In determining the prevailing charge for a service, the carrier multiplies the relative value of the service and a monetary conversion factor derived from the fully adjusted prevailing charges for other services in the same category (509). Thus, an important factor in price determination is the monetary conversion factor which, when used as a multiplier, establishes the price (or payment level) for a service. The conversion factor can be changed to decrease or increase the price of services, and different conversion factors can be used to develop different prices for the same service depending on locality, medical specialty, or other factors. Relative value studies are also procedural terminology documents that health professionals use in describing (coding) services when claiming insurance payment and for other purposes. The number of terms in the various studies has increased dramatically over the years. For example, the number of terms in the Current Procedural Terminology of the American Medical Association increased from 2,084 in 1966 to 6,132 by 1977 and to 7,040 by 1985. The increase in coding terms is intended to provide physicians with more accurate descriptors of the services provided. It also provides physicians with flexibility in describing services. "in order for a carrier to have a sufficient statistical base on which to calculate a physician's customary charge for a specific service, the physician must submit three claims for that service. And, a minimum of four customary charges for a particular service are required for calculating the prevailing charge for the service in a locality (88). 15The antitrust implications of relative value studies have been under examination by the Federal Trade Commission (FTC) and the Justice Department in the past. Continued publication and revision of a number of relative value studies, including the California Relative Value Study, were halted after the settlement of a series of lawsuits in the mid and late 1970s. The antitrust implications of relative value studies depend on the extent to which the members of the groups and output involved in their construction attempt or wish to influence prices and output; thus, the use of relative value studies by the medical professions can be questioned if the intent of the physicians is to fix fees. However, their use by health insurers when determining payment levels for physician services appears to "serve a valid function" (266). Recently, a few medical societies have approached the FTC for advisory opinions concerning the development or updating of new relative value guides and a reexamination ot the previous orders. The current standing of the issue is discussed in ch. 5. App. C-Medicare and Medicaid Payment for Physicians' Services • 225 Recently the Deficit Reduction Act of 1984 (Public Law 98-369) mandated a fee freeze, which started on July 1, 1984, of Medicare customary and prevailing charges for physicians' services. Although the freeze is scheduled to continue only until September 30, 1985, the administration has recommended extending the freeze for another year (552). The conditions of the freeze are dependent on assignment arrangements and are discussed in a section below. Special Provisions.-There are special provisions for hospital-based physicians, teaching physicians, and physician services in intermediate care facilities. Hospital-Based Physicians.-Hospital-based physicians are defined as physicians who provide ancillary medical services in a hospital setting. The three "traditional" hospital-based specialties are radiology, pathology, and anesthesiology, although a number of other types of practices, including emergency medicine, rehabilitation medicine, and cardiology, sometimes meet this definition. Since the beginning of the Medicare program, the Federal Government has made special provisions for paying hospital-based providers, because the program requires the separation of charges for professional and hospital services and because the services hospitalbased physicians provide are so closely allied with hospital services. A physician's professional service-a service that contributes to the diagnosis or treatment of the patient-is paid on a charge basis under Part B. Other services performed by physicians, such as administrative or quality control activities, are considered hospital services and are reimbursed under Part A. In order to simplify reimbursement and claims processing, the 1967 Amendments to the Social Security Act (Public Law 90-248) allowed "combined billing" to be used by hospitals for radiology and pathology services furnished to inpatients, and all physicians' services furnished in hospital outpatient departments. Under combined billing, the hospital uses a single billing form for both the professional and hospital components of inpatient radiology and pathology services. The professional component was identified as a fixed proportion of the total bill for services. Combined billing could be used only if all the physicians in the radiology or pathology departments had a salary or percentage arrangement with the hospital. 16 "Hospital-based physicians are compensated for their services primarily by salary, percentage of departmental revenue, or fee-for service, with feefor-service becoming the predominant important method. Many variations and combinations of methods have been developed to meet specific needs of physicians and hospitals. The 1967 amendments also specified that radiology and pathology professional services rendered to hospital inpatients were to be reimbursed at 100 percent of reasonable charges. Beneficiaries bore no liability for copayment of those services. Because the charges that radiologists and pathologists billed to their carriers continued to be reimbursed at the 80 percent level, this provision was justified as eliminating coinsurance payments by beneficiaries to physicians whose services were not the choice of the beneficiary. It was also intended to reduce hospital-based physicians' incentives for separate billing and thereby reduce processing costs to providers and to the Social Security Administration, even though it made more Federal dollars available for financing hospital-based services (451). The intermediaries paid the combined billing charges, and adjustments were made on an actuarial basis between the two Medicare revenue sources to account for Part B charges being paid by Part A intermediaries. The hospital was paid on the basis of cost, using the charges to compute the cost. The allocation between the Part A and Part B trust fund was based on the physician's allocative agreement. The Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA) (Public Law 97-248) modified the way Medicare pays hospital-based physicians. Section 112 eliminated the provision that Medicare pay 100 percent of the reasonable charges for pathology and radiology services delivered to hospital inpatients. These professional services became subject to the same deductible and coinsurance requirements as other Part B services. Moreover, HCFA determined that the special processing routines required for combined billing were not justified since the option was never widely used, and in implementing regulations eliminated combined billing as of October 1, 1983 (48 FR 39740), More importantly, TEFRA mandated a clearer distinction between Part A hospital services and Part B physician services, regardless of the doctor-hospital relationship. The regulations implementing Section 108 of the legislation restated and clarified the criteria that must be met for a physician's service to be paid on a reasonable charge basis under Part B. In addition to the existing requirements that the service be personally furnished by the physician and contribute to the diagnosis or treatment of an individual patient, a requirement was added that the service ordinarily require performance by a physician (48 FR 8902). If the physician is salaried by the hospital or is on a percentage arrangement, the carrier is required to develop customary charges based on the compensation that the physician receives for the services. TEFRA also mandated that all physician services that do not meet the conditions for charge payment, but benefit a hospital or the patient population as a 226 • Payment for Physician Services: Strategies for Medicare whole, are considered hospital services and are to be reimbursed under Part A. Physicians who receive any compensation from the hospital must have formal agreements that specify time and reimbursement for any Part A services or that provide the basis of allocation of payment between Part A and Part B services. The regulations implementing TEFRA provided that the reasonable cost reimbursement for physicians' services paid under Part A could not exceed "reasonable compensation equivalent" limits that HCFA developed based on physicians' average net income adjusted for specialty, location, and hours worked. The reasonble compensation equivalent limits never became an important factor except for hospital outpatient services, since the implementation of TEFRA and the Social Security Amendments of 1983, which replaced cost-based reimbursement for inpatient services with a prospective payment system based on rates determined by diagnosis; both started October 1, 1983. Part A inpatient physician services are covered by prospective payment just like any other Part A inpatient service. TEFRA also mandated specific provisions governing reimbursement for radiologists, pathologists, and anesthesiologists. The hospital-based radiologist became subject to a limit of 40 percent of the prevailing fee services generally available in radiologists' offices in the community. TEFRA regulations permit payment on a reasonable charge basis to anesthesiologists for up to four concurrent procedures if the anesthesiologist also meets specific guidelines defining appropriate patient care. If an assisting nurse anesthetist is employed by the anesthesiologist, the physician can bill his or her full customary charge as an anesthesiologist. If a nurse anesthetist is employed by the hospital or is self-employed, computation of the anesthesiologist's customary charge is based on one-half time units. (Anesthesiologists bill using a "relative value guide" that combines time units with the relative difficulty and skill involved in procedures.) The greatest changes in TEFRA regarding physicians apply to pathologists. The legislation defined almost all clinical laboratory tests as Part A services, and thus, not reimbursable on a charge basis. Under TEFRA's regulations, only clinical laboratory services meeting very specific criteria can be considered consultative services and reimbursable under Part B; all other clinical laboratory services are reimbursed by Part A. On the other hand, all anatomical pathology services are considered professional services and must be paid on a reasonable charge basis under Part B. Anatomical pathology generally requires examination of body tissue, fluid, or cells by the pathologist. Because anatomical pathology services and some clinical laboratory services, which had previously been combined billed to Part A, are now required to be billed to Part B, carriers have had to quickly establish customary charges using charges for similar services. Teaching Physicians. -Like hospital-based physicians, teaching physicians provide services for the hospital itself (educational and supervisory services) in addition to supplying professional medical services to individual patients. Teaching physicians also tend to be salaried for at least part of their total co.mpensation. Since 1969, with the issuance of Intermediary Letter (IL) #372, Medicare has targeted teaching physicians for special treatment. IL #372 established criteria for identifying the personal, identifiable services that a teaching physician must perform for an individual patient to qualify for fee-for-service payment (6). In 1972, Section 227 of Public Law 92-603 mandated a legislative solution to paying teaching physicians, but was never implemented by regulation. Section 948 of the Omnibus Reconciliation Act of 1980 (Public Law 96-499) repealed Section 227 and essentially codified the requirements in IL #372 that define when a teaching physician may bill for professional services. Section 948 also set forth the manner in which feefor-service payments should be determined for physicians practicing primarily in teaching hospitals. It required that Medicare use the greater of the mean or modal charge collected from non-Medicare patients to determine payment for an individual service. In order to ensure a reasonable minimum for Medicare fees, the Deficit Reduction Act of 1984 amended the 1980 Onmibus Reconciliation Act and set the floor for Medicare fees in a teaching setting at 85 percent of the Medicare prevailing fee in the area. HCFA has not yet published regulations to implement Section 948 of the Omnibus Reconciliation Act of 1980, although they are expected some time in 1985 (252). The Medicare program's only policy that is in effect is its administrative directive, IL #372, which stipulates the conditions for charge payment. There are no promulgated regulations on the level of payment for teaching physicians, and as a result, teaching physicians are today reimbursed by Medicare just like any other physicians. Physician Services in Intermediate Care Facilities.Under fee-for-service, Medicare limits physician payment for visits to beneficiaries in intermediate care facilities (nursing homes) with respect to multiple visits. This restriction was initiated in response to reported abuses early in the program (99). Except when more intensive care can be substantiated, Medicare pays only for one physician visit a month to the same patient in a nursing home, and there is a difference in payment level if more than one patient is visited. If the visit is a routine followup visit, it is paid at the level of a routine followup house call. If the physician App. C-Medicare and Medicaid Payment for Physicians' Services • 227 visits more than one patient ("multiple visits") for routine followup visits, the payment level is lowered to that for a routine followup office visit. If the visit is brief, the Medicare payment may not exceed the payment for brief house calls for single patients and brief office visits for multiple patients (514). Multiple visit rates are 25 percent lower on average than single patient visits (318). Attempts by the Office of the Inspector General in the Department of Health and Human Services to extend the multiple visit limitation to physician visits in skilled nursing facilities17 and hospitals (254) have been unsuccessful to date (99). In addition to limiting payment, Medicare and Medicaid rules also require that patients in skilled nursing facilities be visited at least once every 30 days for the first 90 days following admission, after which the requirement is lowered to once every 60 days (42 CFR 405 1123(b)). Patients in intermediate care facilities must be visited every 60 days (42 CFR 442.346 (b)). Other Payment Methods Although fee-for-service is by far the method used to pay for the great majority of physicians' professional services under Medicare, the program has veered from traditional fee-for-service payment and used alternative payment methods for physicians' professional services to accommodate to special conditions. In certain circumstances, the program has paid an institution, either on a cost or capitation basis, and the institution subsequently has paid the physician a salary or negotiated a fee with the physician for the service or paid the physician per capita. The Medicare program has also paid the physician directly on a nonfee-for-service basis for some services and uses a statewide fee schedule to pay for certain physician provided clinical laboratory services. Physician Services for Kidney Dialysis Patients.Medicare uses alternative methods of fee-for-service in paying for some physician services provided to beneficiaries in the Medicare's ESRD program. Until August 1, 1983, physicians could choose from two methods of payment for maintenance dialysis, the principal service provided in the ESRD program. Under the "initial method," payment for physician services to patients undergoing maintenance dialysis was to the facility, and physicians negotiated a fee with the facility for their supervision or for routine services provided during a dialysis session (nonroutine services were paid "A skilled nursing facility is a specially qualified facility which has the staff and equipment to provided skilled nursing care or rehabilitation services and other related health services. Medicare pays for care in skilled nursing facilities and for physician services provided in such facilities. Medicare, however, does not pay for care provided in nursing homes that are not specially qualified. according to reasonable charge criteria). Or, under the "alternative method," Medicare could pay a comprehensive monthly fee per patient. For patients dialyzed in facilities, the physician's fee was based on a calculation of the customary or prevailing charges for a followup visit, multiplied by 20. For supervision of home patients, the weighting factor was set at 14, to reflect the presumed lower requirements of home patients for physician supervision (405). The payment would be made by the carrier to the physician, if the physician accepted assignment, or to the patient, if the physician did not accept assignment (see discussion of assignment below). In order to provide incentives to the use of home dialysis, the Omnibus Reconciliation Act of 1981 (Public Law 97-35) and subsequent regulations (48 FR 21254) eliminated the "initial method" and require that on and after August 1, 1983, physician services furnished to ambulatory maintenence dialysis patients in a freestanding facility or hospital-based facility or to patients undergoing dialysis at home be paid only under the alternative method. The calculation of the physicians' monthly payments is based on the number of typical dialysis sessions per month, prevailing charges for a medical specialist's brief followup visit for an established patient, and prevailing charges for an intermediate followup visit, weighted by the national averages of patients dialyzed in facilities and at homeY Upper and lower limits on the physicians' monthly capitation payments were established after adjustments for extreme ranges in prevailing charges. The minimum is set at $144 per month and the maximum at $220 per month for both physician services in the home and in facilities. HCFA's intention is not to automatically change the payment levels according to changes in prevailing charges but to review program data and change payment levels if warranted (48 FR 21254). Physician Services for Clinical Laboratory Services. -Prior to July 1984, Medicare payments for clinical laboratory services furnished by a physician or an independent laboratory were made on a reasonable charge basis subject to the Part B deductible and coinsurance.19 The method varied somewhat from Medicare's traditional CPR method of computing reasonable charges. The reasonable charge for ambulatory laboratory services was the lowest of the actual charge, the customary charge, the prevailing charge in the locality, and the lowest charge at which the test is widely and consistently available (which was established for 12 common laboratory tests). "Specifics of the calculation are in 48 FR 21269. "A later section discusses Medicare's deductible and coinsurance under Part B. 228 • Payment for Physician Services: Strategies for Medicare Assignment was permitted on a case-by-case basis, and physicians could bill for laboratory services whether or not they performed or supervised the test.20 When the physician's claim indicated that the test was performed in the office, Medicare would pay the physician as indicated above; when the physician's claim indicated that the test was performed by an outside laboratory, Medicare would pay the physician the laboratory's reasonable charge plus a $3 handling fee. Before July 1984, Medicare payment for laboratory services ordered during hospital outpatient visits was on the basis of reasonable cost. Hospitals providing these services to their outpatients were required to accept assignment; hospitals providing these services to nonhospital patients receiving laboratory services from a hospital serving as an independent laboratory were not required to accept assignment and were paid on a reasonable charge basis. The Deficit Reduction Act of 1984 established a different payment method-a carrier-based fee schedule-for clinical laboratory services conducted in physicians' offices, in independent laboratories, and in hospital laboratories acting as independent laboratories, i.e., furnishing tests to nonhospital patients. The fee schedule was established at 60 percent of the prevailing charge levels for the fee screen year beginning July 1, 1984. After 3 years, a national fee schedule will be formulated with methodology as yet undefined, to serve as the basis of payment. The 1984 law also established a fee schedule for clinical laboratory services conducted by hospital laboratories serving hospital outpatients-the payment level to be set at 62 percent of prevailing charges. After 3 years payment will be on the basis of cost reimbursement unless Congress decides otherwise. Other relevant provisions in the Deficit Reduction Act of 1984 include an annual adjustment of fee schedules to reflect changes in the consumer price index for all urban consumers (U.S. city average) and permission for the Secretary of Health and Human Services to adjust the fee schedules to reflect technological change, emergency services, and other special services. Furthermore, the act also modifies current billing and assignment options. Physicians may bill for services only if the physician personally performs or supervises the test. Anyone who furnishes laboratory services may bill a nominal amount, currently $3 for the collection of the patient specimen; however, only one collection fee per patient encounter will be permitted. Physicians may continue to accept assignment on a bill-by-bill basis, but independent and hospital laboratories must accept assignment. For assigned 20A detailed description of assignment under Medicare is in a subsequent section of this appendix. claims, Medicare will reimburse at 100 percent of the fee schedule and waive coinsurance and the deductible for all assigned tests. Physicians can accept assignment on the laboratory portion of a claim only and not accept assignment for other services. Physician Services and Health Maintenance Organizations (HMOs).-Medicare does not pay physicians directly for their services provided in HMO settings, but contracts with HMOs for physician and other services. The original Medicare legislation authorized payment on the basis of the costs to the organization for providing the specific services to beneficiaries. The Social Security Amendments of 1972 added the option of paying HMOs on a risk-sharing basis for services covered under both Part A and Part B to the existing method of reimbursement on a reasonable cost basis. Under the risk-sharing method, a per capita reimbursement rate that reflects the estimated costs to an HMO for its enrolled Medicare population is compared at the end of the year to the actuarial measure of the costs that would have been incurred by Medicare to serve comparable beneficiaries within the HMO's service area on a fee-for-service basis (the average adjusted per capita cost or AAPCC). If the HMO's costs are less than the AAPCC, the HMO is reimbursed for costs and receives one-half of the excess of the AAPCC over its costs, up to a maximum of 10 percent of the AAPCC. If the HMO's costs are greater than the AAPCC, the HMO has to absorb the entire loss. Both the risk and reasonable cost reimbursement methods required retrospective determination of costs, which is an awkward arrangement for HMOs, which are designed to operate under prospective budgets without extensive reporting requirements to thirdparty payers. Furthermore under the risk-sharing method, although the HMO might have to absorb all losses, it can share in only half of any surpluses. This lack of a strong incentive resulted in only one HMO's entering into a risk-sharing contract with Medicare. As of June 1984, 62 HMOs were reimbursed on a cost basis and an additional 26 were reimbursed on a risk basis under various HCFA demonstration projects. Forty-four other health care prepayment plans had contracts with HCFA on a cost basis for Part B services only (50 FR 1341). In 1982, the Tax Equity and Fiscal Responsibility Act or TEFRA changed the way organizations are reimbursed on a risk basis to permit prepaid capitation payment without retroactive adjustments. The regulations, which became effective February 1, 1985, provide for monthly per capita payments equal to 95 percent of the AAPCC, as adjusted for geographic area and variations within the enrolled Medicare population-age, sex, disability status, welfare status, institutional status, and other relevant factors (50 FR 1369). The orga App. C-Medicare and Medicaid Payment for Physicians' Services • 229 nization is also required to compute an "adjusted community rate"-a rate equal to the premium the organization would charge its non-Medicare enrollees for the Medicare covered services adjusted to reflect the utilization characteristics of the organization's Medicare enrollees. If the organization's adjusted community rate is less than the capitation payment rate, the HMO may keep the entire surplus, but must use it either for providing beneficiaries with additional benefits beyond those required by Parts A and B of Medicare or for reducing premium rates. The.HMO may also put some of the surplus into a benefit stabilization fund or return it to HCFA (see ch. 7). If the adjusted community rate exceeds 95 percent of the AAPCC, the organization may elect to be reimbursed on a reasonable cost basis as in the past. TEFRA also expands the definition of organizations eligible to contract with HCFA for Medicare payment to other medical delivery systems, termed competitive medical plans, that do not meet the restrictive definition of federally qualified HMOs in the Public Health Service Act. Like HMOs, competitive medical plans are required to enroll members on a prepaid capitation basis and assume full financial risk for the full scope of Part A and Part B Medicare benefits. HMOs that have federally qualified must meet other structuralregulations and are required to community rate, rather than experience rate, their premiums for their private lines of business (see ch. 7). Physicians' Acceptance of Medicare Payment Physicians receive Medicare payment for each claim for their services on an assigned or nonassigned basis. When a physician accepts assignment, the physician agrees to accept Medicare's reasonable charge determination as full payment. The physician bills the program directly and, after the deductible is satisfied, is paid an amount equal to Medicare's reasonable charge less the 20-percent coinsurance, which is the patient's share of the bill. If the patient has not had bills sufficient to meet the annual deductible, the patient is also obligated to pay the physician the deductible amount not met.21 When a physician does not accept assignment, the physician bills the beneficiary and the beneficiary requests reimbursement from Medicare. If Medicare's reasonable charge is lower than the physician's actual charge, the beneficiary is responsible for paying the difference between the two charges in addition to the amount of the coinsurance after the beneficiary has satisfied the deductible. The physician's actual charge "The following section on beneficiary financial liability discusses deductibles and coinsurance in detail. is included ih the calculation of the customary charge and the prevailing charge for the next fee screen period irrespective of his or her assignment status. The nationwide net assignment rate of claims declined from a high of 61.5 percent in 1969 to a low of 50.5 percent in 1976 and 1977, and rose to 59 percent of claims and 59.6 percent of charges in 1984 (518). Voluntary assignment is lower than the above rates indicate, since joint Medicare-Medicaid claims are factored into the calculations and assignment is mandatory for Medicaid beneficiaries. The number of the noninstitutionalized dually entitled people, i.e., those eligible for both Medicare and for Medicaid, is estimated at over 3 million or about 15 percent of persons over age 65 (295). However, the exact number of Medicaid beneficiaries who are enrolled in Part B of Medicare is unknown due to insufficient data about the institutionalized elderly and "buy-ins," i.e., beneficiaries for whom States pay the Part B premium. Based on estimates of the dually entitled, voluntary assignment rates appear to average about 11 percent less than indicated by the aggregate statistics (496). Specific estimates of the voluntary assignment rate ranged from 35 (1) to 40 percent in 1982 (486) to 42 to 43 percent in 1982 (174a). The assignment rate varies according to a number of factors, including the following: 1. Physician.-Selected physician reimbursement data from Medicare and Medicaid programs in the 1970s indicate that 28 to 30 percent of physicians never accept assignment, 18 to 19 percent always accept assignment, and 52 to 53 percent make their decisions on a case-by-case basis (71,315). Later data from the American Medical Association show that in 1984 83.9 percent of all non-Federal patient care physicians who treated some Medicare patients sometimes accepted assignment. Slightly over 16 percent did not accept assignment for any patient and 32.1 percent accepted assignment for all of their patients (15). 2. Beneficiary.-Physicians are more likely to accept assignment for disabled Medicare beneficiaries than for elderly beneficiaries and for the older Medicare population than for the younger elderly (494). 3. Geography.-The region and, even more so, the State are factors in accepting assignment. In calendar year 1982 assignment rates ranged from 82.9 percent in Rhode Island to 19.4 percent in Wyoming (494). The geographical variation has been ascribed to historical precedent, physician preference, and administrative practices of individual carriers (496). 230 • Payment for Physician Services: Strategies for Medicare 4. Medical specialty.-The highest assignment rates (about 60 percent) are in the hospital-based specialties of pathology and radiology (297). Among office-based physicians, general surgeons had the highest and otolaryngologists and ophthamologists had the lowest assignment rates (138). 5. Size of bill.-Physicians tend to accept assignment more often on bills of $100 to $200 than on bills lower than $100 or higher than $200 (297,494). 6. Payment Jevel.-Assignment rates increase with an increase in Medicare reimbursement rates, and decrease with a decrease in reimbursement rates (188,315,357,394). A recent, fundamental addition to the assignment process is the establishment of a participating physician program on July 18, 1984, as mandated by the Deficit Reduction Act of 1984. The acceptance of assignment on a bill-by-bill basis still pertains for those physicians who decide not to become participating physicians. However, those physicians who have become participating physicians have voluntarily entered into an agreement to accept assignment for all services provided to Medicare patients for 12-month periods beginning on October 1st of each year (516). There are a number of incentives in the legislation to encourage participation; the major incentive is that participating physicians are exempt from a limitation on future physician charge increases. The statute, as noted earlier, freezes Medicare customary and prevailing charge levels for the services of physicians from July 1, 1984, through September 30, 1985. The law prohibits nonparticipating physicians from raising their actual charges to Medicare patients during the 15-month freeze period and stipulates that Medicare will not recognize any increases in charges during the fee freeze period in calculating customary charges on October 1, 1985, and October 1, 1986. In addition, if nonparticipating physicians increase their actual charges billed to Medicare beneficiaries, they can be excluded from the Medicare program for up to 5 years or be subject to civil monetary penalties. Participating physicians are exempt from some of these limitations. Although no increases in Medicare payment are permitted from July 1, 1984, to Septem ber 30, 1985, participating physicians who increase billed charges during the 15-month freeze were to have the increase recognized in the customary charge up dates on October 1, 1985, and October 1, 1986. 22 Other incentives for participating physicians include "As noted earlier, the Administration's budget proposal for 1986 includes extending the phyician fee freeze for an additional 12 months and delay up dating of participating physicians payments by 1 year. listings in directories made available to beneficiaries, toll-free carrier telephone services, and electronic transmission of claims. January 1985 data show that almost one-third (29.8 percent) of all physicians billing Medicare have chosen to become participating physicians (516). The level of participation varies by specialty and State. As can be seen in table C-2, "other" surgical specialists, anesthesiologists and otolaryngologists have the lowest percentage of participation; and nephrologists, radiologists, and pathologists have the highest percentage of participation. Among the States, physicians and suppliers who practice in South Dakota, Alaska, and North Dakota have the lowest percentage of participation; and physicians who practice in Alabama, Kansas, and the District of Columbia have the highest percentage of participation (516). More recent data indicates that 30.4 percent of all physicians billing Medicare were participating physicians in fiscal year 1985 and 28.4 percent are expected to participate in fiscal year 1986. The assignment rate for all physicians, participating and nonparticipating, in January 1985 had increased to 66.5 percent, a considerable increase over the fiscal year 1984 rate of 56.4 percer\t (352). The 66.5 percent assignment rate includes mandatory assigned claims by clinical laboratories, as indicated earlier, as well as physicians who have accepted assignment on a claim-by-claim basis, and physicians who have accepted assignment for all services to Medicare beneficiaries. The assignment rate for fiscal year 1985 for all physicians has increased to 67.7 percent (521a). Beneficiary Payment Liability Eligible individuals must pay monthly premiums for coverage of physician and other services under Part B and are subject to a deductible and coinsurance for covered services used. If a physician does not accept assignment, the Medicare patient is also liable for the difference between the amount the physician bills and the amount Medicare allows for the service. Services that are not covered for payment by Medicare are the complete financial responsibility of the beneficiary. Beneficiaries' participation in Part B of the Medi care program begins with a fixed monthly premium, which has been rising gradually from $9.60 in fiscal year 1980 to $15.50 on January 1, 1985, a 60-percent increase (see table C-3) (523). The annual out-of-pocket premium payment by the elderly for Part B coverage increased 138 percent from 1977 ($78) to 1985 ($186). At the outset of the program, premiums contributed half of Part B revenues, while general revenues subsi dized the other half. Subsequent amendments limited Part B premium increases to no more than the percent App. C-Medicare and Medicaid Payment for Physicians' Services • 231 Table C·2.-Medicare Participating Physicians and Suppliers Percentage of Percentage of Number of all physicians/ Number of all physicians/ Specialty participants suppliers Specialty participants suppliers Physicians (M.D.s and D.O.s): Limited license practitioners: General practice ................ 13,743 27.3% Chiropractor . . .. . . . . . . . . . .. . . . . 6,217 25.4% General surgery ····· .. 9,491 33.9 Podiatry-surgical chiropody ..... 4,541 38.2 ......... Otology, laryngology, rhinology ... 1,741 24.6 Optometrist .................... 4,541 38.2 Anesthesiology ................. 3,269 21.1 Other limited license practitioners Cardiovascular disease .......... 3,820 35.6 (audiologists, psychologists, Dermatology ................... 2,089 34.0 physical therapists) ........... 2,845 36.8 Family practice ................. 8,820 25.5 Independent laboratory .......... 1,698 28.4 Internal medicine ............... 21,067 32.5 Durable medical equipment Neurology ····· ................ 2,543 34.8 suppliers .................... 5,018 22.7 Obstetrics-gynecology ........... 4,220 27.3 Ambulance service suppliers ..... 2,551 28.6 Ophthalmology ................. 4,220 27.3 Miscellaneous suppliers Orthopedic surgery ............. 4,096 29.0 (orthotists, prosthetists, Pathology ...................... 2,263 39.6 portable X·ray suppliers) ....... 8,555 22.5 Psychiatry ..................... 6,871 30.0 Radiology ...................... 6,658 41.3 Grand total .................... 156,001 29.4% Urology ........................ 2,381 27.8 Nephrology .................... 944 50.8 Total physicians ................ 118,428 29.8 Clinic or other group practice-Total limited license not GPPP .................... 6,795 33.8 practitioners ................. 19,751 34.0 Other medical specialties ........ 6,515 32.4 Total suppliers ................. 17,822 23.8 Other surgical specialties ........ 4,398 18.2 SOURCE: U.S. Department of Health and Human Services, Health Care Financing Administration, HCFA Fact Sheet, January 1985. Table C-3.-Monthly Beneficiary Premium for Medicare Part B Coverage Inflation adjusted Annual increase premium above inflation Period Premium (1980= 100)a (percent change) Fiscal year 1980 ........... . $ 9.60 $ 9.60 NA Fiscal year 1981 ........... . 11.00 9.98 3.96 Fiscal year 1982 ........... . 12.20 10.44 4.61 Fiscal year 1983 ........... . 13.50 11.22 7.47 Calendar year 1984 ......... . 14.60 11.72 4.46 Calendar year 1985 ......... . 15.50 aPremium deflated by Consumer Price Index for Urban Wage Earners and Clerical Workers for the calendar year, as published in U.S. Department of Labor, Bureau of Labor Statistics, Monthly Labor Review, September 1985. SOURCES: U.S. Department of Health and Human Services, Health Care Financing Administration, Medicare Program Statis· tics 1981 (Washington, DC: U.S. Government Printing Office, 1983); and 49 FR 38511. age increase in Social Security cash benefits. As aresult by 1978, the percentage contribution of premiums to Part B costs had dropped below 25 percent (164). The Tax Equity and Fiscal Responsiblity Act of 1982 and the Social Security Amendments of 1983 (Public Law 98-21) temporarily suspended the limitation and increased Part B basic premiums as of calendar year 1984 to a level that results in premium revenues equal to 25 percent of program costs. The Deficit Reduction Act of 1984 extended the requirement that the Part B premium produce income equal to 25 percent of program costs through 1987, with the constraint that the increase in the Part B premium may not exceed the dollar amount of the Social Security cost-of-living adjustment (49 FR 38510). Almost 80 percent of Medicare enrollees 65 years of age and over used physician services in 1982 and 60 percent of Medicare enrollees met the initial deductible (495). In any year, the beneficiary has to incur an initial expense-a deductible-before Medicare will pay for Part B services. In 1982, the deductible was raised from the first $60 to the first $75 of approved charges in a calendar year. Coinsurance is applied each time physician and other Part B services are used and is 20 percent of the remainder of approved charges after the deductible is satisfied. The deductible and coinsurance per enrollee for 1984 was estimated to average at $236, which is an increase of 143 percent from 1977, when they were $97 (see table C-4). Beneficiaries also incur costs when physicians do not accept assignment, since the beneficiary is liable for any difference between the physician's billed charge and Medicare's payment for the service (reasonable charge reduction on unassigned claims). HCFA esti 1\) ~ • i (!) ~ 0' ., ~ ~ c:;· iii' :::. C/) (!) 3. Table C-4.-Medicare Supplementary Insurance: Estimated Total and Per Enrollee Cost-Sharing for the Aged, 1977·848 0 (!) ~ Total in millions Per enrolleeb ~ ii'l Potential Potential liability from liability from ~ Total unassigned Total cost-Total Annual SMI Total cost-unassigned ~-Year Deductible Coinsurance copayments claimsc sharing Deductible Coinsurance copayments premium sharing claimsc 0' ., 1977 ........... $ 969 $1,244 $2,213 $ 804 $3,017 $42 $ 54 $ 97 $ 89.40 $186.40 $32 1978 ........... 1,011 1,454 2,465 912 3,377 43 62 105 95.40 200.40 35 1979 ........... 1,055 1,736 2,791 1,158 3,949 44 72 116 96.65 212.65 43 ~ 1980 ........... 1,103 2,112 3,215 1,538 4,753 45 86 131 101.40 232.40 56 &i' 1981 ........... 1,148 2,576 3,724 1,873 5,597 46 103 148 123.60 271.60 67 iil 1982 ........... 1,525 3,235 4,760 2,281 7,041 60 126 186 139.20 325.20 80 1983 ........... 1,571 3,967 5,538 NA NA 60 152 212 146.40 358.40 NA 1984d .......... 1,616 4,678 6,294 NA NA 61 175 236 175.20 411.20 NA NA = Data not available. aJanuary 1984 current law estimates of copayment amounts based on incurred charges. Data are subject to revision. bAverage annual enrollment is used to calculate these items. Clncludes both aged and disabled beneficiaries. "Potential liability" refers to the fact that physicians who do not accept assignment are free to pursue payment of their billed charges in excess of the Medicare approved charges from the beneficiary, but It is not known how many actually do so. dprojected. - SOURCE: D.R. Waldo, and H.C. Lazenby, "Demographic Characteristics and Health Care Use and Expenditures by the Aged in the United States: 1977-84," Health Care Financing Review 6(1):1-29, Fall 1984. U.S. Department of Health and Human Services, Social Security Administration, Social Security Bulletin, Annual Statistical Supplement, 1982, Table 145 (Washington, DC: 1984). App. C-Medicare and Medicaid Payment for Physicians' Services • 233 mates that the reasonable charge reduction on un assigned claims has increased from $31 per enrollee in 1977 to $85 per enrollee in 1983 (a 214-percent in crease) (563). Of the per capita expenditures on physician serv ices by the aged projected for 1984, the Medicare pro gram is expected to spend 58 percent. The beneficiary out-of-pocket component, when defined as the deduct ible, coinsurance, and reasonable charge reduction, was estimated to be 26.1 percent of the per capita ex penditure in 1984 (563). When the Medicare premium and the payments to the deductible made by Medicare beneficiaries who do not meet the deductible are in cluded, Medicare beneficiaries are estimated as pay ing 60 percent of the cost of physician services under Part B (8). Most of the elderly participants in Part B have some form of supplemental "Medigap" private insurance.23 By 1977, approximately 66 percent of the elderly population had some type of private health insurance to supplement their Medicare benefits (558). Private insurers annually paid $117 in 1984 ($48 in 1977) for physician services for elderly Medicare beneficiaries (563). Most policies are supplementary to Medicare coverage and limited to paying for deductibles and coinsurance ("Medigap"), although there are other forms of "Medigap" insurance that are more comprehensive. To the extent that beneficiary unassigned liability is actually collected, the payment of deductibles and coinsurance by Medigap tends to dilute control of beneficiary use of Part B services through cost-sharing. Medicaid Payment for Physicians' Services The Medicaid program was authorized in 1966 under Title XIX of the Social Security Act as a social welfare program to provide medical assistance to certain categories of low-income people, including the elderly, the blind, the disabled, and members of families with dependent children (the categorically needy). Medicaid is a joint Fe9eral-State program that is administered by individual States under general Federal guidelines that include minimum benefits that must be available to eligible recipients and optional benefits that individual States may elect for their recipients. Both the individual States and the Federal Government provide program funds; the Federal Government contributes "matching funds" for the categorically needy, and, if the State chooses, for the medically in "For a comprehensive discussion of supplementary medical insurance, see app. F of the October 1982 OTA report Medical Technology Under Proposals To Increase Competition in Health Care (483). digent.24 The Federal Government's current contribution to Medicaid payment is estimated at an average of 53 percent with a range from 50 to 78 percent among the States (236,563). State Medicaid programs have considerable discretion in the method to determine payment levels and can use adaptations of either the maximum fee screen method (CPR) or fee schedules. In early 1982, 25 Medicaid programs used various adaptations of the CPR method, 12 of which reimbursed at below Medicare's 75th percentile of prevailing charges (255). The States vary in how often they update prevailing charges, the data sources they use to establish physician profiles, and the percentile at which they set the prevailing charge (214). For example, Medi-Cal (the California Medicaid program) at one time defined the prevailing charge as the 60th percentile in contrast with the 75th percentile then used in the Medicare program (430). State Medicaid programs are required to use the MEl as a screen to limit the rate of increase in prevailing charges (214). In 1982, 24 States reported using a fee schedule (255). Some States derive their fee schedule from Medicare or private insurance payment levels and adjust the schedule over time. Others base the fee schedule on a relative value scale and a conversion factor of their choosing (214). Two States reported using this methodology in 1982 (255). Although in 1979, almost an equal number of States based payment levels on fee schedules as those that used the CPR methodology, 7 of the Nation's 10 largest Medicaid programs used fee schedules, and States with fee schedules accounted for 68 percent of all Medicaid expenditures. States that employed fee schedules increased their fees much less frequently than States that used the CPR methodology to establish payment levels (214). The fee levels in the Medicaid program average only 72 percent of Medicare levels and are as low as 49 percent in some States (204). The State Medicaid programs vary widely in payment levels. When an adjusted weighted average fee for each State adjusted for the cost of living is used, statewide fee indices that aggregated fees across 41 procedures varied from $108.04 for Nevada to $21.68 for Pennsylvania (214). The Medicaid programs also vary according to medical specialty and geographic area (214). For example, payment for an appendectomy performed by a general practitioner ranged from a low of $100 in Pennsylvania to $512.89 in Nevada (255). 24The medically indigent are individuals who meet categorical requirements for Aid to Families with Dependent Children and Aid to the Aged, Blind, and Disabled, but have incomes that a State considers too high to be eligible for cash assistance, but not sufficiently high to pay medical bills. 234 • Payment for Physician Services: Strategies for Medicare Physicians who provide medical care to Medicaid recipients cannot bill recipients, but must accept Medicaid payment as payment in full (236). Medicaid recipients in States with low Medicaid payment levels and low rates of physician participation are more likely to receive care in hospital outpatient departments and clinics, which are usually more expensive than physician's offices (170). Physician fee freezes are utilized by many State Medicaid programs to control program costs. As of July 1984, 17 States had frozen physician fees for 1985 at the allowable rates established for 1984 or an earlier year (214). Some State programs have frozen fees for many years. For example, except for certain procedures, Florida has not increased physician fees since 1983, Michigan since 1977, and Ohio since 1972. Louisiana has frozen fees for 2 years and intends to initiate a flat fee system. Although New York State has frozen physician fees, there is legislation pending to increase physician fees for primary care services (214). In addition to fee-for-service physician payment, State Medicaid programs also have the authority to contract with federally qualified HMOs and with comprehensive medical plans. However, the influence of HMOs has been small; as of September 30, 1984, there were only 65 HMO Medicaid contracts in 21 States (229). Until recently, payment to the HMOs was on a capitation basis only25 and no more than SO percent of the enrollees in a contracting HMO could be Medicaid or Medicare beneficiaries.26 The Omnibus Reconciliation Act of 1981 increased the maximum proportion of Medicare and Medicaid beneficiaries to 75 percent, with provisions for a waiver in special circumstances. Furthermore, regulatory reform efforts have relaxed regulations governing Medicaid reimbursement to HMOs and established procedures for States to contract with HMOs for services provided to Medicaid beneficiaries on a cosF7 as well as on a capitation basis. Until October 1, 1982, when the Arizona Health Care Cost Containment System (AHCCCS) became operational, Arizona was the only State without a Medicaid program. The purpose of AHCCCS is to de velop and test an alternative payment and delivery sys tem that is based on capitation, competition among providers, and a network concept of primary care. AHCCCS contracts with both the public sector (coun ty government) and with the private sector (including individual practice associations (IPAs) and other HMOs) "Although the Social Security Act did not expressly forbid reimbursement arrangements from being on a cost basis, regulations governing Medicaid pay ment to HMOs (42 CFR 431.524) required that all such contracts be on a risk basis. "Social Security Act, Section 1903(m). "48 FR 54013, final rule establishing new Section 42 CFR 434. to provide basic health services to individuals in the Aid to Families with Dependent Children program, individuals on Supplementary Security Income, and single individuals with less than $3,200 annual income (238). The program appears to be encountering problems with respect to costs, the providers' financial performance, and quality of care, and HCFA is imposing tighter controls on its operation (198). The Arizona experiment illustrates some of the changes taking place in the payment and delivery philosophy of the Medicaid program. As described above, the Medicaid program historically has paid physicians on a fee-for-service basis with care delivered by the private delivery system. Beneficiaries have had little or no cost-sharing requirements and were allowed free choice of providers. In the last few years, States, with support from the Federal Government, have turned to systems of case management combined with payment on a capitation basis, restrictions on physician choice, and beneficiary cost-sharing in attempts to contain costs (235). The Omnibus Reconciliation Act of 1981 wrought a major change in State Medicaid programs by modifying Medicaid's long-standing provisions that gave recipients the freedom to obtain services from any provider. As of June 1984, 24 States had restricted recipients' access to all providers by limiting freedom of choice, by requiring recipients to obtain services from a primary care provider or "gatekeeper," or by requiring recipients to receive care only from providers with whom the State had contracts (214). Six of the State programs are in the demonstration or pilot stage. Of the 14 States that will reimburse only those providers with whom the State has a contract, 3 States have limited the provision to selected services, and 1 State has applied the provision only to 6,000 Medicaid recipients enrolled in HMOs (214). Other changes from traditional Medicaid payment procedures include capitation programs and recipient cost-sharing.28 Twelve States now have provisions for paying physicians on a capitation basis rather ,than traditional fee-for-service; four of the States have demonstration capitation programs. And, as of July 1984, 25 States collected a copayment or deductible from Medicaid recipients for selected services that vary from State to State. States have also provided medical care for some recipients by enrolling them in HMOs; 23 "The Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA) altered Medicaid cost-sharing requirements so that States now have the option to require copayment, coinsurance, or deductibles for almost all services to both the categorically needy and the medically needy with certain exceptions, such as categorically needy who are enrolled in an HMO. The regulations also state that no provider participating in the Medicaid program may deny care or services to individuals because of their inability to pay the cost-sharing charges. App. C-Medicare and Medicaid Payment for Physicians' Services • 235 States had exercised this option for some of their recipients as of July 1984 (214). Conclusion Medicare's payment method for physician services can be characterized as a predominantly fee-for-service system, which, in comparison with Medicaid and private health insurance, has remained relatively unaltered in the past 20 years. Until July of 1984, when a temporary treeze on physician fees went into effect, Medicare's major efforts to control physician payment were to reduce the level of prevailings, to delay the updating of physician fees, and to apply the MEl (which became effective in 1976). The prescribed method for computing physician payment based on reasonable charges as determined by the CPR method is extremely variable among Medicare carriers because of the decentralized mechanism established in law and regulations. The variation extends to the delineation of charge areas and the specification of prevailing charges depending on medical specialty. The Medicare program has been flexible in its payment method on a national level. The program has adapted CPR for special conditions, e.g., paying hospital-based physicians. Morever, Medicare has veered from the traditional fee-for-service method, and for some special services (maintenance kidney dialysis and ambulatory laboratory services) has paid physicians on other bases. Medicare has also contracted with HMOs on a cost or capitation basis for physician -and other services. Physician financial involvement in Medicare is influenced by payment practices. The data on physician participation indicate that the number of physicians accepting assignment on a claim-by-claim basis is rising slowly, with extreme variation among medical specialists and among States. Similar variation exists among physicians who have become "participating physicians" in Medicare, i.e., they have agreed to accept assignment for all services provided to Medicare beneficiaries. Over the years beneficiaries' financial liability for physician charges has increased. As a result many elderly Medicare beneficiaries have purchased supplemental medical insurance (Medigap). The other major Federal third-party payment system, Medicaid, differs from Medicare with respect to physician payment historically in having used two ways of setting payment levels, CPR and fee schedules. Almost an equal number of States use each method. Typically, Medicaid pays physicians at a lower rate than Medicare. Another difference is that in the past few years, States, with support from the Federal Government, have used innovative ways of paying physicians and organizing the delivery of medical care. Many States are using systems of case management and retrictions on physician choice as cost-control methods. Appendix D Private Sector Approaches to Physician Payment Introduction Much attention has been drawn to the problems of the Medicare program in maintaining the viability of a medical insurance system for elderly and disabled beneficiaries in the face of rising medical costs. It is important to recognize, however, that rising costs of medical care do not affect Medicare alone. In 1983, private payers for health care services in the United States spent $206.6 billion, or 58.1 percent of the national total of $355.4 billion. Private expenditures grew at an average annual rate of 13.6 percent from 1979 to 1983. Private expenditures for physician services totaled $49.7 billion in 1983, and were projected to increase to $60.4 billion in 1985 and to $88.0 billion in 1990 (21). Many of those private payments are made by thirdparty payers in the form of health insurance benefits provided to employee groups. In 1979, 73.3 percent of the population had some form of private insurance coverage, and 60.6 percent of the population was covered under employer group contracts (83). In 1983, an average of $2,100 to $2,400 was spent by medium and large employers on each employee's health care (448). The sources for health insurance coverage in the United States have been commercial insurance companies, hospital and medical service plans (e.g., Blue Cross and Blue Shield), prepayment group medical plans, such as health maintenance organizations (HMOs), and others, such as employers or labor unions (201). In addition to employee groups, an important part of the market for some insurers is "Medigap" insurance.1 Corporate payers of health care benefits have a stake in controlling the costs of care for the Medicare population, since 58.4 percent of medium and large employers in 1980 maintained health benefit plans for retired employees over age 65 (83). Although Medicare was designed around the model of private health insurance (287), from its inception, Medicare has differed from private third-party payers in the specifics of its benefits, coverage, payment, and other policies. Furthermore, although Medicare policies are determined by laws and regulations, practices of private health insurers have been developed in response to market demand from purchaser groups, by Federal laws such as the Employee Retirement Income '"Medigap" is described in app. C. Security Act of 1974 (Public Law 93-406), and by State laws and regulations (329). In addition, corporate payers of health benefits plans have increasingly turned to self-insurance, or self-funding of benefits, in order to have greater control over the outlays for their employees' benefits. Thus, differences between private insurers' methods of physician payment provide another base of experience for Medicare to examine in considering alternatives for physician payment. The first part of this appendix discusses the range of alternatives in private health insurance. In addition, the recent outcry from corporate payers of health care benefits for controls on health care costs have caused the private sector health insurance industry to respond with many innovations in the provision of benefits and health care services. Although many of these innovations are too recent to have been evaluated for their effectiveness in controlling costs, the private sector may be gaining experience in identifying cost-effective practices and developing methods for rationalizing the provision of health services that may be worthwhile for Medicare to examine. The second part of this appendix discusses these innovations in the private sector. The Private Insurance Market Framework Like Medicare, private health insurance groups the benefits for which it will pay into broad categories. As is illustrated in figure D-1, in addition to the extent of benefit coverage in an insurance plan and the method of determining payment level, the insurance organization's theoretical approach to health insurance is a dimension of payment. Private insurers also make decisions on a claim-by-claim basis for those services, usually new technologies, not explicitly covered or excluded under the terms of the insurance policy. Approaches to lnsurance.-Indemnity insurance and service plans represent the two theoretical traditional approaches to health benefit coverage. Indemnity insurance guarantees the enrollee a fixed amount for a specific service (95,445). The enrollee pays the physician the physician's billed amount for the service, which may be more than the insurer's guaranteed amount (430), and collects payment from the insurer (439). In contrast, service plans assure their members App. D-Private Sector Approaches to Physician Payment • 237 Figure 0·1.-Dimensions of Payment for Physicians' Services in Private Health Insurance (Fee for Service) Fee schedules Variable fee screen (UCR) SOURCE: Office of Technology Assessment, 1985. specific units of service, for example a day of hospital care, for a regular premium in prepayment of those services (95,445). A physician, if a participating provider, contracts with the plan to accept the plan's allowance as full compensation for his or her services (430), and collects payment from the plan (439). However, the theoretical approaches have become entwined; many service plans incorporate some indemnity features, and some indemnity plans have some service plan features (22). Methods of Determining Payment Levels.-Both the indemnity approach and service approach use fee schedules and variable fee screens (usual, customary, and reasonable or UCR) in determining payment levels (95,122). In practice, fee schedules are most common in indemnity policies (325), while the UCR method is most often used to determine payment levels in service plans. Types of Insurance.-In contrast to Medicare, benefits of private health insurance are covered for payment under two principal types of insurance: 1) basic medical expense insurance, and 2) major medical expense insurance. Basic medical expense insurance is further divided into hospital expense insurance, surgical expense insurance, and medical expense insurance. Hospital expense insurance usually includes coverage for hospital room and board and nearly all services provided by a hospital, other than personal convenience items and most physician fees. Surgical expense insurance typically covers the expenses of surgeons and related professional services. Regular medical expense insurance covers physician's nonsurgical services in a hospital, at home, or in the office, and sometimes covers diagnostic services (587). Major medical expense insurance applies broadly to almost all kinds of medical care. Major medical expense insurance is sold either as a distinct policy to supplement existing basic coverage, or it provides coverage for all medical expenses (subject to deductibles and coinsurance) in a single unit. This type ~f insurance, called comprehensive coverage, provides extensive benefits, including coverage for most health care services prescribed by a physician in or out of a hospital (587). Benefits and Coverage.-The specific benefits, the exclusions, and the extent of financial coverage vary from one insurer to another and from policy to policy within each of the organizations. Nonetheless, in general there are some benefits that differ markedly from those offered under Medicare. Dental care, vision and hearing care, outpatient prescription drugs and, more rarely, physical examinations are covered under some private health insurance policies purchased through employer groups (201,203,575). Some major medical policies also offer a relatively high level of protection against catastrophic expenses. Furthermore, although less extensive than other services, coverage for inpatient and ambulatory psychiatric services is considerably greater than that offered under Medicare (414). As can be seen in table D-1, the percent of coverage differed for each category of benefit. A constraint on the benefits offered by private insurance plans is State and Federal laws and regulations which mandate particular types of coverage. Until recently, it had been believed that the Employee Retirement Income Security Act of 1974 and the National Labor Relations Act preempted State laws regulating the content of insurance policies for employees. On June 3, 1985, however, the U.S. Supreme Court upheld a Massachusetts statute mandating certain types of mental health benefits in employee health benefit plans, thus opening the way for States to regulate the content of employee health plans. This ruling is causing concern to nationwide corporations, who believe that the costs of their plans will rise if they are required to meet differing benefits requirements in all of the jurisdictions in which they operate (35,329). Physician Payment in Private Insurance Plans Private health insurance varies from Medicare in relying on both benefit schedules (fee schedules) and variable fee screens in determining payment levels for 238 • Payment for Physician Services: Strategies for Medicare Table D-1.-Group Coverage of Selected Categories of Health Care Percent with benefit Category of Wilensky, Hedger and health care et al.a Schmittb HIAAc Ambulatory prescription drugs ............. . 88.0% 97.5% 75.0% Dental care .......... . 28.3 74.6 46.0 Vision or hearing care .. 11.7 Vision care .......... . 28.0 16.0 Ambulatory psychiatric care ............... . 77.0 93.0 Mental health care .... . 98.7 Physical exam ....... . 6.3 5.2 apercentage of civilian noninstitutionalized individuals with group coverage b(N~13,916); Data from National Medical Care Expenditure Survey (1977). Percentage of 116 health msurance plans that covered 5 million employees in 1979. Survey performed in 1981. CPercentage of 36 employee health insurance plans with 21.8 million employees. Survey performed in 1981 by the Health Insurance Association of America(HIAA). SOURCE: G.R. Wilensky, P.J. Farley, and A.K. Taylor, "Variations in Health Insurance Coverage: Benefits vs. Premiums," Milbank Mem. Fund Q. 62(1)_:53·81, Winter 1984; D. Hedger and D. Schmitt, "Trends in Major Med1cal Coverage During a Period of Rising Costs," Monthly Labor Review 106(7):11·16, July 1983; and Health Insurance Association of America, A Profile of Group Medical Insurance in the United States (Washington, DC: HIAA, no date). their fee-for-service business. Furthermore, as is described below, the variable fee screen method is similar, but not identical, to the method used by Medicare. Fee Screens.-For major medical policies most private insurance companies and health benefit plans use variable fee screens in determining physician payment levels. The amount allowed by a plan or company for a given service is known as the reasonable charge and depends on a fee screen determination of the physician's usual charge and the local customary charge (UCR). In private insurance terminology, the Levell fee screen, the physician's usual charge, is comparable to Medicare's customary charge (583). The socalled customary charge, comparable to Medicare's prevailing charge, is the Level 2 fee screen. The Level2 (customary) fee screen in private health insurance is generally set at the 80th to 90th percentile, that is, the charge level at or below which 80 to 90 percent of the billed charges occur. Medicare, by contrast, is legally limited to the 75th percentile for its Level2 (prevailing) fee screen (Section 1842(b)(3), Soc. Sec. Act). The reasonable charge is generally the lowest of the usual charge (Levell screen) or the customary charge (Level2 screen), both ofwhich are generally calculated from the physician's charge pattern over the previous year (430). In special circumstances, such as an unusually complex treatment, the reasonable fee may be the physician's actual charge, even though it exceeds the fee screens (324). Fee Schedules.-Most private health insurance companies and health benefit plans use benefit schedules (i.e., fee schedules) for determining physician payment for their basic medical expense policies (430). The insurer generally pays the lesser of the listed amount for a service or the actual charge by the physician, irrespective of medical specialty or geographic location (430). If the physician's actual charge is higher than the listed amount, the patient is responsible for the additional payment for the service. Maximum benefit schedules can be set by negotiation2 or according to actuarial calculations. Blue Cross and Blue Shield Plans.-lndividual Blue Cross and Blue Shield plans are members of the Blue Cross and Blue Shield Association, which was formed in 1982 when the Blue Shield Association and the Blue Cross Association merged. All plans are not-for-profit organizations organized under State insurance laws or under special enabling legislation. Blue Cross plans originally covered primarily hospital expenses, but enlarged their scope of coverage to include ambulatory care, other institutional services, and home health care. The Blue Shield plans were founded to cover physician services, but many have expanded their coverage to include other benefits, such as dental services, vision services, and ambulatory services. As of September 1985, there were 67 Blue Shield Plans and 68 Blue Cross Plans, of which 40 are joint plans, making a total of 89 corporations (122). Each of the Blue Cross and Blue Shield Plans is an autonomous organization with its own staff, organizational hierarchies, and decisionmaking processes (22). Although there are specific standards to which a plan must adhere to be designated as a Blue Shield Plan and a member of the national Blue Cross and Blue Shield organization,3 the plans vary considerably. Participation of Physicians.-The concept of par ticipating physicians is a cornerstone of the Blue Cross/Blue Shield plans' philosophy, but there are var iations among the plans in applying the concept. Gen erally, the plan and the physician decide on the pay ment, without direct patient involvement. In return for agreeing to accept the plan's allowance as payment .in full, participating physicians are then paid by the plan rather than by the patient (430). In addition to direct payment by the plan, the non-price incentives for participating in the plans include simplified billing, prompt payment, avoidance of bad debts on covered services, a predictable cash flow, and services of a field staff to handle problem claims (324). 2ln the 1930s and 1940s, insurers sometimes negotiated fee sched ules with medical societies. The current interpretation :of antitrust law requires that schedules be set unilaterally by the carrier (324). 3For example, "A member must be endorsed by the State or county medical societies of the area in which it operates, be nonprofit, main tain free choice of doctor, return at least 75 percent of earned sub scription income to members in benefits, maintain professional re lations and utilization review programs and meet certain financial and reporting requirements" (372). App. D-Private Sector Approaches to Physician Payment • 239 Physicians agree to participate in a Blue Shield serv ice benefit program either independently or because of a decision made by the physicians' medical society4 (although individual physicians may then opt out of their society's participation agreement). About three times as many individual agreements are entered into as are medical society endorsements (439). The find ings of a 1978 study indicate that interest in being a participating physician in Blue Shield plans is quite high. When individual physicians were given the op tion of participation, only 28 percent of office-based physicians declined to do so (440). However, the rate of participation is extremely variable among plans. For example, in 1984 only 16 percent of solo physicians participated in Blue Cross/Blue Shield of Florida (45), although the overall participation rate is 70 percent (122). Although under Medicare a physician may accept or refuse assignment of benefits on a claim-by-claim basis, Blue Shield plans usually require participation on an "ali-or-nothing" basis, i.e., the physician's participation agreement requires that she or he accept the plan's determination of reasonable charges as full payment in all cases. As under Medicare, most Blue Shield plans will also pay for services performed by a nonparticipating physician. However, Blue Shield of Massachusetts cannot, by law, pay nonparticipating physicians, or reimburse subscribers who use them (587). The billing and payment relationship between Blue Shield plans, the subscribers, and the nonparticipating physicians varies. Under some plan arrangements, the nonparticipating physician bills the patient directly, with the plan paying the patient. Under other plans, the participating and nonparticipating physicians are paid on the same basis. Other plans will pay nonparticipating physicians directly if the physician obtains an assignment of benefits from the subscriber (587), but the physician then cannot "balance-bill," i.e., bill the patient for any excess above the plan's allowed charge (122). Beneficiary Cost-Sharing.-Blue Shield plans traditionally have been associated with the service approach to insurance, that is, the plans guarantee to provide services in full, with subscriber cost-sharing limited to levels based on fee schedule allowances or reasonable charges (122). Recently, however, market demands for cost containment have led to an increased emphasis on plans with copayments and deductibles. Florida Blue Cross/Blue Shield, for example, has found that comprehensive insurance combining the medical service coverage of both basic and major medical service plans with subscriber cost-sharing in the form of •This agreement, in order to meet antitrust regulations, must not have anything to do with price setting (122). deductibles and copayments is more marketable to employers than separate basic and major medical plans. Some plans also offer pure indemnity type plans (22,458). Methods ofDetermining Levels ofPayment. -Plans vary in the ways they construct a fee schedule. Blue Cross/Blue Shield of Florida, for example, constructs separate fee schedules for each of its charge areas. Relative values are determined from the 90th percentile of Level2 charges and then multiplied by a separate conversion factor for each charge area to establish a fee (45). The methodology for determining UCR charge levels also varies. Some plans, e.g., Blue Shield of Pennsylania and Blue Cross/Blue Shield of Florida merge their Medicare claims and private claims in creating Level 1 and Level2 fee screens. The construction of specialtyspecific fee screens is another area of variation. Some Blue Shield plans, e.g., Blue Cross/Blue Shield of Florida, calculate a Level 2 fee screen (customary charges) for all physicians regardless of specialty. Other plans calculate separate Level 2 fee screens for paricular specialties; Pennsylvania Blue Shield, for example, calculates discrete level 2 fee screens for 56 specialties. The plans also vary from Medicare in how often they update fee screens. Typically the Levell (usual) and the Level 2 (customary) fee screens are revised every 6 or 12 months. Indeed, the UCR method of payment is analogous to a floating fee schedule; the maximum amount the insurer pays is updated at specified intervals, and the shorter the interval between updates, the higher the reasonable fee (22). Medicare's Medical Economic Index has a parallel in the Blue Cross and Blue Shield Association's membership standard that suggests controlling the Level 2 (customary) fee screen. Some plans employ the rate of increase in the Consumer Price Index as a cap on the Level 2 fee screen, a few plans use other indices (324), and other plans do not use any control on customary charges (45). Commercial Insurance Companies.-At the end of 1981, more than 1,000 private insurance companies, mostly for-profit proprietary companies or subscriberowned mutual companies, were estimated to be offering individual or group health insurance covering over 108 million persons (201). Unlike the majority of Blue Cross and Blue Shield plans, which provide only health benefits coverage and other health-related services, most commercial firms sell other types of insurance as well. In fact, life insurance is the main line of business of the major companies in the commercial health insurance field (587). Moreover, commercial insurers do not have a system of participating physicians. The large number of 240 • Payment for Physician Services: Strategies for Medicare commercial insurers and the lack of standardization in claim forms and benefit programs among insurers make it difficult for physicians to deal with individual companies. Rather, the physicians bill the patients, who must then obtain reimbursement from the insurance company (587). Like Blue Shield, commercial insurers use both fee schedules and the UCR methodology in determining physician payment levels. However, most commercial insurers do not differentiate payment levels by specialty. The physician payment methodology is often part of the specifications of group policies. When selling group policies, the insurer either bids on a series of benefits and specifications designed by the prospective buyer or plans a group's health and welfare program based on the needs and resources of the buyer. Large insurance companies, such as the Metropolitan Life Insurance Co. and the Prudential Insurance Co. generate their own data base on which to base levels of physician payment. Smaller companies usually do not have sufficient claims on which to base a credible5 payment level and often depend on other sources for guidance. One such source is provided by the Health Insurance Association of America Prevailing Healthcare Charges System, which collects, compiles, and publishes data on charges for surgical procedures by physicians (200). Some of the larger insurance companies, e.g., Metropolitan Life Insurance Co., use the surgical charge program as a "back-up" source in responding to physician questions about payment level, or if the number of claims on which to establish a credible prevailing charge is insufficient (303). Both Blue Cross/Blue Shield and commercial insurers are providing new types of insurance coverage through the development of preferred provider organizations (PPOs) (see table 7-5 inch. 7). Payment of physician's services under PPOs through negotiated discounts from the physician's charges or through negotiated fee schedules has added a new source of variation in the methods of payment available under private health insurance plans. In addition, these plans usually provide incentives to the enrollee to obtain care from preferred providers by reducing levels of costsharing for care from those providers. However, since many groups other than insurers are also currently involved in the development of PPOs, they will be discussed in greater detail later in this appendix. 5The credibility of a payment level has legal implications. When the insurer claims reimbursement will cover reasonable fees, or when the insurer claims reimbursement will not create excess physician billing of the patient for charges in excess of the insurer's approved charges, an insurer has to be able to defend the payment methodology as well as the amount of payment if a provider questions an insurers' payment. Self Insurance.-An increasing number of employee benefit plans and other organizations are self-insuring, i.e., they underwrite their own benefits coverage with a budget funded by the organization. Plans can either self-administer or hire an outside firm to process claims and to perform other administrative services. In either case, the plan prospectively determines its medical expenditures for the year. If costs are lower than projected, the plan retains the savings, and if costs are higher than estimated, the plan absorbs the loss. In order to protect against high or unexpected costs, most self-funded plans re-insure their plan, i.e., purchase "stop-loss" insurance that takes effect when a claim for a specific individual exceeds a predetermined amount and when the overall costs of the plan are higher than a prespecified amount. The growth of self-insurance in the past few years has been dramatic, although estimates of market share vary. The Society of Professional Benefit Administrators, the national association of independent thirdparty contract benefit administration firms, estimates that for mid-1984 commercial insurance company fully insured policies have 20 percent of the market for health benefits coverage, Blue Cross and Blue Shield Plans have 35 percent of the market, and self-funded plans have 45 percent of the market. The self-funded statistic includes plans administered by third-party administrators, self-administered plans, and administrative services only arrangements administered by insurance companies (230). Both the commercial companies and Blue Cross and Blue Shield Plans are responding to this potential loss of market share by contracting to administer selfinsured plans for self-funded organizations (149). The Health Insurance Association of America, measuring a subset of self-funded plans, in the form of administrative services only arrangements and minimum premium plans provided by commercial group insurers, has estimated that prior to 1979 only 5 percent of total insurance company group coverage was self-funded insurance. By 1980, these types of arrangements represented approximately 25 percent of total insurance company group coverage; by 1981, they represented 30 percent of such coverage (201). Innovative Private Sector Approaches to the Provision of Health Services Historically, the function of insurance has been to protect individuals from the risk of financial ruin from actuarially predictable untoward events by providing either cash or service benefits to enrolled beneficiaries. Private insurers sought business from large purchasers of group insurance, such as employers, by designing App. D-Private Sector Approaches to Physician Payment • 241 insurance packages, including payment methods, to fit the specific requirements of the purchaser in pro tecting the beneficiaries from such risks. In fields of insurance other than health, competitive pressures to provide the greatest amount of protection at the lowest price force insurers to initiate actions that will reduce the number and cost of catastrophic events (e.g., risk management techniques used by liability insurers) (171). Yet, until recently, most purchasers of health care have not been interested in demanding alterna tives to traditional forms of insurance which would reduce their costs for care (417,454). In the absence of demand, private insurers rarely made initiatives to control costs on their own. One reason suggested for the previous lack of interest in cost containment among insurers and purchasers of insurance is that the potential variability among services and in patient need make it difficult for third-party health insurers to identify discrete episodes with predictably finite costs whose risks and costs they can then work to reduce (157). To ensure that the care delivered during a hospital stay, for example, is as efficient as possible or to ensure that the care delivered during that stay is rendered in the most efficient site (which may not be in a hospital at all) requires that the third-party payer move beyond the financial function of insurance to develop a system for monitoring and controlling the provision of care itself. Rapid growth in the cost of insurance premiums to employers has focused their attention on finding alternative, less costly ways of providing care to their beneficiaries. The increase in the number of self-funded plans, for example, is one example of the response of corporate purchasers of health care to rapidly increasing premiums. The private sector is also adopting innovations in the provision of health benefits for employees that may be viewed as stages in the evolution of various managerial technologies for the provision of health care. The remainder of this appendix will describe each of these approaches in turn: • approaches directed at the beneficiary's choice of medical provider and site of treatment, e.g., changing benefit packages to increase beneficiary cost-sharing or to cover particular services in preference to others; • approaches directed at medical care providers, such as systems of utilization review to monitor the cost, choice, and use of hospital and ambulatory services; • development of alternative provider arrangements, such as HMOs and PPOs; and • development of coalitions of health care purchasers to coordinate activities on a local or regional level. In addition, private sector payers have established other approaches to containing health care costs, such as health promotion and awareness programs, e.g., corporate "wellness" programs that encourage health enhancing behavioral changes in employees in hopes of lowering the groups' utilization of services. Al though these efforts may have an effect on health care costs by lowering the demand for medical services (with as yet unknowable effects on the unit price of such services), they are not germane to our discussion of alternatives for the payment of services when the demand for them actually occurs. To date, research to establish the effectiveness of these changes in private sector policy in controlling aggregate health care costs has been scant, and the evi dence available is mostly anecdotal. Companies have noted individual savings in their insurance premiums (or, in the case of self-funded companies, their bene fit payouts). In a 1984 survey of corporate benefits officers and senior executives of corporations with more than 500 employees, executives of companies who reported changes in their benefit plans to control costs estimated that the changes in health care plans their organizations had instituted had saved between 16 and 18 percent over the last 3 years over what their costs would have been (273). Independent confirma tion of these estimates is not available, however. Nevertheless, the changes in corporate health benefit plans are worth examining as a type of natural experiment in alternative payment for health care and physician services. The very success or failure of those attempts may affect the overall market for physician services in which Medicare must participate. Theresponse of the system to these innovations may be instructive in designing changes in Medicare payment. Changes in Medicare reimbursement may compleinent these changes in the private sector, having a synergistic effect on controlling health costs. It should be noted, however, that these lessons from the private sector in changing benefits apply to a working population. Whether those results may then be applied to an aged Medicare population is a question for further research to answer. Managing the Provision of Benefits To Change Beneficiary Incentives The redesign of health insurance benefit packages to modify beneficiary incentives to seek less costly forms of medical care has two facets. One is to increase employee awareness of the costs of treatment choices by increasing the level of costs the employee must pay out-of-pocket for that care. The other is to encourage or mandate particular providers of care or modes of treatment that are believed to reduce costs. 242 • Payment for Physician Services: Strategies for Medicare Increasing Beneficiary Awareness of Costs.-Increasing beneficiary cost awareness is not per se an alternative method for reimbursing physicians. Yet, the level of coverage a beneficiary receives under his or her insurance plan is an influential factor in the decision to seek the medical care. Health services researchers have noted that the existence of insurance, while perhaps sparing the individual the risk of financial ruin in the event of illness, also serves to insulate the individual and the physician from considering the direct financial consequences of their joint or separate treatment decisions (342,365). Increasing Cost-Sharing. -Unlike Medicare, patient cost-sharing in the form of a deductible amount and coinsurance generally has been characteristic of only part of the insurance coverage offered by private insurance, i.e., major medical insurance. The deductible amount usually takes the form of an absolute amount, or very rarely a fixed percentage of income, and may be applied on an illness, a person, or a family basis (587). Recently, there has been a trend for large businesses to incorporate some form of costsharing in all the insurance that they offer their employees (149). There are a number of reasons why corporate benefits plans may wish to increase the portion of the cost of health care borne by beneficiaries. Increasing costsharing is believed to be relatively easy to implement and administer (209). All other things being equal, sharing more of the costs of care with the employee immediately lowers the costs of care being borne by the company and therefore its health insurance premiums. A second aspect of increased cost-sharing is that it will cause the beneficiary to consider more carefully whether or not to initiate' an episode of care. Thus, increased cost-sharing reduces not only the expendi tures of the third party, but also aggregate health care expenditures. Insurers who were asked to estimate the savings resulting from various cost-sharing require ments said that increasing the deductible from $0 to $500 would save an estimated average of 20.8 percent in claims, and a coinsurance rate of 30 percent would save an estimated 27.5 percent in claims compared to a zero percent coinsurance rate (209). Empirical data from the Rand Health Insurance study seem to con firm this; adult beneficiaries with first-dollar coverage of health care were found to use significantly more health care services than those who paid some coinsur ance or deductibles (343). Because of these cost advantages, corporate benefit plans have increased the level of cost-sharing borne by individual employees. One recent survey found that 50 percent of a sample of corporations employing 500 or more employees increased deductibles in the past 3 years, and that 22 percent of those corporations instituted copayment on medical bills (273). Another survey found 71 percent of 150 companies surveyed raised the amount of deductibles, 53 percent increased the level of coinsurance, and 44 percent increased the employee's share of the premiums paid (147). A 1983 survey of companies found that 48 percent were introducing or increasing cost-sharing for their employees (25). Public sector employers were relativeiy less strict in increasing employee cost-sharing than private sector employers. According to survey data, 30 percent of public sector employers have increased their deductibles, although they are still relatively low. Plans with deductibles of $150 or less covered 65 percent of salaried public employees and 55 percent of hourly public workers, compared with 43 percent of all workers in the public sector (577). Employers were found to be four times more likely to use deductibles over $100 in 1984 than in 1982, and that plans which covered all costs after meeting the deductible declined from 67 percent in 1982 to 42 percent in 1984 (307). The Medicare Part B program already imposes 20percent coinsurance on the beneficiary, while many private benefit plans are only beginning to approach that level. In 1982, 25 percent of employer group health insurance plans had a 20-percent beneficiary coinsurance rate for inpatient care, while in 1984, the proportion increased to 43 percent (307). Although the effect of Medicare coinsurance may be mitigated by Medigap coverage, lower income elderly persons may not carry Medigap coverage if they feel they cannot afford the additional premium. Further, there is substantial evidence that the imposition of cost-sharing has disproportionatly negative effects on utilization of health services by persons of lower income (483). In addition, since the elderly are greater users of medical care, it is believed that substantial increases in beneficiary cost-sharing would thus likely come at the cost of reduced access for Medicare beneficiaries (486). Positive Incentives To Reduce Utilization.-An al ternative to directly increasing the costs of care borne by the beneficiary is to create positive financial incen tives for the beneficiary to constrain utilization. An example of a direct positive incentive would be a bo nus that is paid to a beneficiary if he or she does not submit any claims. Such incentives may be perceived by employees as less harsh because the person decid ing to initiate care may be in dire need of that care and perhaps should not be forced to expost; his or her own resources to risk. Alternatively, indirect incen tives, such as flexible benefits plans, in which the em ployee can choose his or her level of health coverage and take some or all of the employer's contribution App. D-Private Sector Approaches to Physician Payment • 243 as cash compensation, are also being tried in hopes of lowering the employer's expenditure on health care coverage. The use of positive financial incentives for employee beneficiaries to reduce their use of health services is not nearly so common as increasing the level of financial risk borne by the employee. Neither is it known how effective such positive incentives would be in reducing utilization. Positive incentives usually come in the form of a cash bonus given to employees for not using any services during the course of the year. Alternatively, the employer establishes a fund in the employee's name to which the employer makes contributions, which can be used to cover medical cost-sharing or which can be carried over from year to year and used when needed, while any balance can be withdrawn on retirement (153). However, the Internal Revenue Service has recently challenged the tax-free status of such benefit accounts, declaring that the funds in the account must be spent within a calendar or fiscal year (448). A.S. Hansen, Inc., found that only 10 percent of its surveyed companies had instituted such incentive plans (25). A related system of positive incentives to reduce utilization is found in so-called "flexible benefits plans," also known as cafeteria benefits plans, in which employees choose between types of insurance coverage available. One type of flexible benefits plan establishes a "flexible spending account" to which employees contribute pre-tax income as a type of voucher with which they can purchase benefits from a range of options offered by a company. Many employees participating in these accounts have exhibited a preference for greater disposable income rather than for health insurance plans with reduced cost-sharing. The Employee Benefit Research Institute found that health insurance deductibles in these plans averaged $207, versus a nationwide average for all firms of $100 (127). Directing Beneficiaries' Choice of Treatment and Providers.-Health insurance benefits plans are becoming more innovative in the management of their employees' benefits outside the scope of traditional insurance coverage. As was discussed earlier, to direct and manage the provision of health care is a step beyond the insurers' function of providing protection against financial ruin. Several approaches used include the following: • encouraging or requiring second opinions for nonemergency surgery; • encouraging or requiring the provision of certain types of surgery and routine laboratory tests on an ambulatory basis; and • educating employees about and channeling them to efficient providers-i.e., case management. Second Opinions for Surgery.-There are three types of second surgical opinion programs, varying in the level of coercion involved: the passive reimbursement of second opinions obtained at the initiative of the beneficiary, the active promotion by the company of second surgical opinions, and the requirement that the beneficiary obtain a second opinion for all elective surgery (149). Second surgical opinion programs of any type are among the most common of the private sector cost containment activities directed at beneficiaries. A survey of Fortune 500 companies revealed that 71 percent had some sort of second opinion program, 64 percent have a penalty associated with failure to obtain a second opinion, and SO percent mandate that employees use the program before surgery (160). Another survey found that 54 percent of a sample of employers began a mandatory second opinion program in the past 3 years (273). Public sector employers were found to be less likely to use strict second opinion programs with 27 percent having mandatory programs, and 32 percent having only voluntary programs (577). There is still some controversy over the cost-effectiveness of second surgical opinion programs. It is believed that obtaining the concurring opinion of a second, disinterested physician in the necessity for a surgical operation can help to screen out cases for which indications are weak and that may be amenable to less drastic alternative treatments (62). One study found that the mandatory second surgical opinion programs studied exhibited a cost-benefit ratio of 2.63:1 (i.e., $2.63 was saved for every $1 spent to administer the program), while voluntary programs were less effective in reducing costs (146). Some recent studies have determined that 14 to 16 percent of proposed surgeries submitted to second opinions were not confirmed (62). However, one cannot conclude that those nonconfirmations are indicative of unnecessary surgery. Such nonconcurring opinions may often advise delaying surgery, or pursuing medical rather than surgical treatment, and may eventually be followed by the surgery originally proposed (373). At the same time, it is not known whether delay may result in more complicated surgery later (62). The Congressional Budget Office estimated the benefits of a mandatory second surgical opinion program to Medicare at about $80 million (418). The American College of Surgeons said that "it seems only prudent to consider the alleged advantages of the second-opinion concept as unproved and to postpone widespread implementation of programs" (10). Ambulatory Surgery and Testing.-The provision of medical services on an ambulatory rather than on an inpatient basis is also being encouraged by employee health benefits plans. Of the companies re 56-119 0 -86 -9 QL 3 244 • Payment for Physician Services: Strategies for Medicare sponding to the Louis Harris survey, 47 percent had initiated financial incentives of the provision of surgery and testing on an ambulatory basis (273). Gardner, et al., found that 82 percent of Fortune 500 companies. were encouraging ambulatory surgery where possible, and 79 percent were encouraging ambulatory testing (160). It is difficult to distinguish on the basis of these surveys what is meant when it is said that an employer will encourage ambulatory care. Encouragement can be either active or passive; there has been no systematic collection of data on the extent to which private third-party payers simply reimburse for or actively encourage ambulatory surgery testing. Although some insurance carriers maintain lists of procedures that will be reimbursed only on an ambulatory basis, others simply provide information about the availability of the coverage without taking an active role in encouraging it. Some carriers have tried to increase the incentive for physicians to perform ambulatory surgery by increasing the level of reimbursement to physicians for performing surgery on an ambulatory rather than on an inpatient basis (149). Case Management.-ln case management programs, an agent is assigned to the beneficiary to direct and coordinate the provision of medical care for that beneficiary. Although a case manager may be a physician or other provider of care whose services are engaged by the beneficiary directly, in this context case management refers to an agent employed by the corporate benefit plan who arranges and directs the provision of care for the beneficiaries of that plan. The expertise of a case manager is intended to help employees make choices among less costly providers and services and to reduce the cost of care. At the same time, the use of a case manager involves a consider able amount of overhead for the sponsoring plan. Al though a few corporations maintain case management teams in-house, most of those using case management programs contract with outside consultants for serv ices. Costs of case management programs are said to run about 1.2 percent of the level of claims. It is not a commonly used method for managing beneficiary incentives. One survey found that only 1.3 percent of surveyed companies used case management techniques in 1983, although the case manager approach was seen to be growing rapidly (207). Management of Provider Behavior: Utilization Review By monitoring the process of care-giving according to some defined standard, utilization review attempts to manage provider behavior in the provision of care to assure the appropriate use of the plan's resources for the protection of the plan's beneficiaries and of the financial well-being of the plan. Utilization Review: Types.-The focus of most review programs conducted by private insurers, thirdparty administrators, and self-funded employee health benefit plans is on services rendered in a hospital, since those services are the most expensive and the payoff to monitoring services in that site is greater. However, utilization review could be performed in an ambulatory care setting, although it would be likely to be more costly because the site base is so diffuse. Utilization review, as currently used, can be divided into particular types based on when they apply to the patient: preadmission review (requiring approval before an elective admission to a hospital), concurrent review (during the hospital stay), and retrospective review (after discharge from the hospital). Utilization Review: Sources.-Numerous types of private sector organizations provide utilization review for health benefits plans, either under contract to a number of different plans or as a part of the business of providing health insurance benefits (149). Founda tions for medical care and peer review organizations are organizations providing utilization review services that are usually sponsored by physicians and are geographically restricted. They have the advantage of having closer relationships with local providers and may thus have a greater ability to elicit cooperation with the goals of utilization review. Corporations also contract with independent commercial utilization review organizations and with third-party administrators for utilization review services (437). Insurance companies are also developing utilization review programs to meet the competitive challenge of the other organizations (172). The latter three need not be restricted to a particular locality, but may provide services nationwide. Lastly, some employers will organize utilization review programs in-house rather than contracting with outside organizations. Cost-Effectiveness and Prevalence of Utilization Re view.-Empirical assessments of the cost-effectiveness of review programs conflict. Studies performed have usually had methodological flaws that have made it impossible to draw conclusions about the effectiveness of particular programs (142,149). Nor has evidence as to effectiveness of utilization review with regard to Medicare beneficiaries been made available (111). However, anecdotal data available from some com panies' benefit plans report savings in expenditures of 7 to 22 percent resulting from utilization review pro grams, at a cost of about $1 to $2 per employee per month (142,361). Others have found that the total costs of care have increased in spite of utilization re view mechanisms that constrain utilization. One large company found that its utilization review program re App. D-Private Sector Approaches to Physician Payment • 245 Table 0·2.-Surveys of Use of Utilization Review Mechanisms Among Employee Benefit Plans Gardner, et al.a Mechanism (1984) Preadmission certification ........ . 32% Concurrent review ............... . 41 Retrospective review ............. . NR General utilization review programs; type unspecified ..... . 68 Percent of employers using method A.S. Hansen, Louis Harris Hewitt lnc.b & Associatesc Associatesd (1983) (1984) (1982) (1984) 10% 28% 2% 26% 14 NRf 8 34 17 NR 18 40 48 27 MercerMeidinge,.S (1985) 26% 30 auniverse of Fortune 500 companies, sample and respondents unspecified. bsample and respondents unspecified. CPrograms begun in last 3 years; sample of 1,250 companies with 500 or more employees. dsample of 1,185 companies. esample of 256 public employers surveyed In May 1985. fNA = Not reported. SOURCES: S.F. Gardner, J.B. Kyzr·Sheeley, and F. Sabatine, "Big Business Embraces Alternate Delivery," Hospitals 59(4):81·84, Mar. 16, 1985; A.S. Hansen, Inc., 1983 Benefits Survey (Lake Bluff, IL: A.S. Hansen, Inc., 1984); Louis Harris & Associates, Corporate Initiatives and Employee Attitudes on Cost Containment (New York: The Equitable Life Assurance Society of the United States, February/March 1985); Hewitt Associates, "Company Practices in Health Care Cost Man· agement-1984," quoted In J. Goldsmith, "Death of a Paradigm: The Challenge of Competition," Health Affairs 3(3):5-19, Fail 1984; and William M. Mercer· Meidinger, Inc., Healthcare Cost Containment in the Public Sector (New York: William M. Mercer-Meidinger, Inc., 1985). duced corporate wide inpatient utilization by 46 percent over a 5-year period; yet its hospitalization costs per person covered increased 60 percent in that same period (52). (One cannot be certain that costs would not have risen even further in the absence of the company's utilization review program, however.) In spite of equivocal evidence, the use of utilization review in the private sector has been expanding among larger employee benefit plans. Table D-2 summarizes some recent survey data on the prevalence of various utilization review methods. One reason cited by companies for the use of utilization review programs is that, rather than being considered a cost-saving practice in itself, it is seen as a method for collecting provider-specific utilization data that can later be used as a bargaining ·tool for negotiation of preferred provider arrangements and other alternative health care systems (149) (see discussion below). Still, it is believed that utilization review programs possess a great deal of potential for reducing the costs of care, simply because the known degree of nationwide variation in use of services suggests that reductions in utilization are possible without a loss in quality of care (208). A utilization review program can call attention to patterns of care that fall out of line with established norms, and may educate providers in how their practice patterns diverge from those of others. Another reason for the use of these programs may be a belief in the so-called "sentinel effect," which holds that the process of review need not necessarily call particular episodes of care into question. Rather, the fact that the review process exists at all will cause providers to behave more cautiously in prescribing care. There is a further difficulty in instituting and coordinating utilization review programs, particularly for corporations doing business nationwide. Medical service data are not collected in any systematic fashion throughout the country, making it difficult to calculate and compare plan use .with nationwide norms for care. Providers have been unwilling to cooperate with utilization review programs in the past, although that reluctance is lessening as providers come to believe that it is in their own best interests to cooperate (149). Development of Alternative Provider Arrangements Alternative provider arrangements place the choice of treatment in the context of a system for the provision of care. Although individual treatment choices may still be left to the discretion of the patient and the provider, the presence of a superceding organizational structure may force the provider to account for the economic trade-offs between different treatment chQices, if the success of the organization is predicated on the ability to deliver health care in a more costconscious manner. Types of Alternative Provider Arrangements.Although alternative provider arrangements derive from numerous sources, including hospital/physician joint ventures, insurance companies, and consumer groups, corporate benefits plans have recently become leading figures in the establishment of alternative provider arrangements. The major types of alternative provider arrangements considered in this appendix are PPOs and HMOs. Although health maintenance organizations have been in existence for nearly SO years and have been extensively studied (see ch. 7), PPOs are a newer form of alternative provider arrangement that has not been extensively studied. Nevertheless, PPOs have attracted attention because of their paten 246 • Payment for Physician Services: Strategies for Medicare tial for creating financial incentives for beneficiaries to choose cost-effective providers. Preferred Provider Organizations (PPOs). -PPOs include a diverse array of arrangements between a third-party payer and providers of health care, including physicians, hospitals, or both. Estimates of the number of PPOs differ. In January 1985, the American Association of Preferred Provider Organizations identified 143 operational PPOs in 28 States and the District of Columbia (7). According to the Institute for International Health Initiatives, as of June 1985, 229 PPOs were operational, 67 were defined as preoperational, and 38 were of undefined status, in a total of 35 States (237). A conservative estimate of the number of persons enrolled in PPOs in June 1985 was about 5.8 million, a fourfold increase from a December 1984 enrollment estimate of 1.3 million (382). PPOs vary in a sponsorship, membership, and payment methodology. As of June 1985, most PPOs had been sponsored by providers, with 52.3 percent having been sponsored by physicians, hospitals, or physician/hospital joint ventures. Insurance companies and Blue Cross/Blue Shield plans supported 16 percent of the total (237). Physicians providing services under the auspices of a PPO generally agree to fee-for-service reimbursement at some discount from their customary, prevailing, and reasonable charges, although arrangements have included reimbursement according to a fee schedule or on a capitation basis. Of those PPOs responding to the survey, 29 used a relative value scale, 23 used a fee schedule, 18 used individual provider discounts, 18 used "modified fee for service," 3 used gatekeeper reimbursement, and 3 used capitation (7). Among the characteristics that may be involved in a PPO, those features that distinguish it from other types of payment plans are: 1) that the providers agree to accept payment for medical services at some discounted rate, and 2) that providers are willing to accept the scrutiny of utilization review programs in the provision of care. PPOs are believed to be making increasing use of utilization review programs, although most of the efforts at utilization review so far have concentrated on inpatient rather than ambulatory care (382). A 1984 survey showed that of the operating PPOs responding to the poll, 83.1 percent used precertification of admissions, 63.4 percent used discharge planning, 57.7 percent used concurrent review, 54.9 percent used retrospective review, and 35.2 percent used second surgical opinions (425). The Institute for International Health Initiatives found that 73 percent of its respondents had preadmission certification, 74 percent had concurrent review, 66 percent had retrospective review, and 43 percent had mandatory second surgical opinion programs (237). These results suggest that PPOs use utilization review programs more frequently than traditional employee benefit plans (see table D-2). One example of a PPO developed by a private insurer is one sponsored by Blue Cross/Blue Shield of Florida (45). Its strategy involves agreements with selected hospitals that have a reasonably low payment level based on the hospital prospective payment system of diagnosis-related groups (DRGs). If a hospital agrees to participate, Blue Cross/Blue Shield of Florida will then negotiate a DRG contract with it and request the hospital's assistance in signing its medical staff for the PPO. For the PPO program, the plan has divided the State into four regions. A fee schedule based on 90 percent of the average billed charges for 1983 is being developed for each of three regions. If a physician agrees to become a preferred provider, he or she will accept the lower of his or her billed charges or the scheduled fee. The physician also agrees to a system of preadmission certification, certain locally determined surgical procedures requiring to be performed on an ambulatory basis and a medical necessity retrospective review of claims. In return, Blue Shield promises to review, but not necessarily update, the fee schedule annually. A different method had been designed for determining the level of physician payment for physicians in the southern region of Florida, which differs markedly from the other regions in numbers and types of physicians and beneficiaries. For the southern region, a UCR fee schedule is being developed that is based on the 75th percentile of the physicians' 1983 billed charges. Since the 90th percentile is used for determining the reasonable charge for the plan's traditional business, basing the payment on the 75th percentile assures a discount. The payment is tailored to each physician's charges as contrasted with the regionwide fee schedule described above, since each physician will be paid the lowest of the usual, billed, customary, or reasonable charge. Health Maintenance Organizations (HMOs). HMOs are organizational entities that accept payment for the provision of medical services on a per-enrollee capitation basis. The HMO makes arrangements with a panel of physician and hospital providers to provide services to those enrollees, and bears risk for the costs of services in excess of the capitation payment. Except in an emergency or with prior authorization, the enrollee is required to obtain health care services only from those providers with whom the HMO has contracted to obtain care. HMOs increased from 39 to 337 between 1972 and 1984, and the number of subscribers App. D-Private Sector Approaches to Physician Payment • 247 increased from 3.5 million to 16.7 million (240). In 1983, roughly 40 percent of HMO members were in one of the Kaiser plans (239). Commercial insurance companies are actively in volved in a growing segment of the HMO industry, "national HMO firms," which are firms that own or manage separate HMO firms in two or more States (239). Commercial insurance companies sponsored about 10 percent of HMO plans in 1983, which en rolled 11 percent of HMO clients (239). Potential for Cost Savings.-PPOs can be seen as a competitive response by insurers and health care providers jointly to market their services as a unique "product" that may be superior to others' services be cause it is less expensive on a per unit basis, or, more importantly, less expensive in aggregate because of the efficiency of the providers. Beneficiaries enrolled in the PPO are encouraged to use PPO services by benefit ting from reductions in cost-sharing. However, their choice of caregiver is not restricted to these preferred providers; they may choose any other provider out side of the arrangement as long as they pay the appli cable deductible and coinsurance. Preferred providers who simply offer to discount the price of their services while recouping their losses through expanding their volume of services do not offer the same cost savings as preferred providers who both discount their services and maintain strict efficiencies in the provision of their services. Few existing PPOs maintain the necessary sophistication of data collection, however, and the record of PPOs in constraining costs has not been evaluated (52): Nevertheless, many believe that the development of more sophisticated information systems and utilization review mechanisms will allow PPOs to distinguish truly efficient providers and offer their services to the marketplace as a distinct medical product. Development of Health Care Coalitions: Cooperative Ventures in Health Care Cost Containment Coalitions have evolved on a regional basis to address some of the unique variations in health care utilization. Although the name "coalition" connotes the ideal of a merger of a broad range of interests in the health care field, up to this time coalitions have been organized largely by employers (308). Estimates of the number of coalitions differ. The U.S. Chamber of Commerce reported that in 1984 there were 135 coalitions with 6,500 members, an increase of 14 percent from the year before (380). The American Hospital Association found 151 coalitions operating in 1984, with an additional 14 in the development stage (12, 381). This was an increase of 13 percent from the year before. Coalitions, though perceived by some as underrep resenting some of the responsible parties in health care cost containment (308), do provide a mechanism for participating corporations to cooperate in the pursuit of specific goals defined by the membership of that par ticular coalition. To the extent that the coalitions have unique goals specific to the conditions of the health care market in their regions, the members establish their own criteria for success and predicate the con tinued existence of the coalition on those criteria (149). Among these goals is the development of provider specific utilization data systems, which are crucial to the identification of efficient providers. Having iden tified these providers, one can construct an alterna tive, cost-conscious provider system. According to U.S. Chamber of Commerce figures, 80 percent of the coalitions are involved in such activities (380). The American Hospital Association found that 71 percent of the respondents to its survey of coalitions were involved in the development of data systems (12). Coalitions have served as foci for political action in attempts to change local and State regulations in order to foster a more competitive market for health services, and have been instrumental in establishing State all-payer rate-setting regulations in Massachusetts and Connecticut (73,208). Other coalitions have served as a mechanism for employers to establish alternative provider arrangements. According to the U.S. Chamber of Commerce, 70-percent of existing coalitions are active in developing such arrangements (380). According to the American Hospital Association, 44 percent of coalitions responding to its survey were active in developing alternative provider arrangements (12). Conclusion The pace of change in the private health insurance market has become very rapid, as the industry has responded to the demands of its customers to provide new approaches to financing and providing health care. Many of these changes in the financing and provision of care are too recent to evaluate for their effectiveness in reducing costs while retaining quality of care. Corporate benefits managers are taking on the role of being the informed buyer for their employees/beneficiaries amid the plethora of new alternatives in insurance coverage and alternative provider arrangements. Nevertheless, individual insurance beneficiaries will require greater access to information in order to make rational choices about the purchase of health care (483). 248 • Payment for Physician Services: Strategies for Medicare The lessons learned from the private insurance industry in the provision of health care relate to changes in payment to physicians under the Medicare program, but results from that experience should be applied with caution. The effectiveness of many private sector initiatives in controlling costs and preserving quality is unknown. In addition, there are distinct differences between the populations covered by Medicare and those covered by private insurers. Further research will be required on such issues as: • Does increased cost-sharing cause beneficiaries to forgo necessary health care and result in greater expenditures later? • How effective are various forms of utilization controls in assuring cost-effectiveness and quality of care? • How effective are PPOs in restraining costs? • To what extent can the experience of private sector insurers in providing care for their beneficiaries be duplicated for public sector beneficiaries? (368). References 1. Aiken, L.H., and Bays, K.D., "The Medicare Debate-Round One," N. Engl. f. Med. 311(18): 1196-1200, Nov. 1, 1984. 2. Aiken, L.H., Lewis, C.E., Craig, J., et al., "The Contribution of Specialists to the Delivery of Primary Care: A New Perspective," N. Engl. f. Med. 300(24):1363-1370, June 14, 1979. 3. Alder, H. C., Lithotripters: Noninvasive Devices for the Treatment of Kidney Stones, AHA012828 (Chicago, IL: American Hospital Association, 1985). 4. Almy, T.P., 'The Role of the Primary Physician in the Health-Care 'Industry,' "N. Engl. f. Med. 304(4):225-228, Jan. 22, 1981. 5. "Alternative Site Strategies,'' Technology Reimbursement Reports 1(44):4-5, Nov. 29, 1985. 6. American Association Medical Colleges, "Payment for Physician Services in a Teaching Situation," Washington, DC, 1983. 7. American Association of Preferred Provider Organizations, Directory of Preferred Provider Organizations (Washington, DC: AAPPO, January 1985). 8. American Association of Retired Persons, Statement on Medicare Part B, presented to the House Committee on Energy and Commerce, Subcommittee on the Health and Environment, Washington, DC, Apr. 26, 1984. 9. American College of Radiology, "Radiology Practice Survey 1983," Park Ridge, IL, 1983. 10. American College of Surgeons, Socio-Economic Factbook for Surgery 1985 (Chicago, IL: American College of Surgeons, 1985). 11. American Hospital Association, "Report on Status of Capital Expenditure Regulation," Chicago, IL, July 10, 1985. 12. American Hospital Association/Dunlop Group of Six, "Summary Statistics for Health Care Coalitions Responding to the 1984 Survey of Health Care Coalitions," Chicago, IL, November 1985. 13. American Medical Association, Physician Fee and Cost Indicators (Monroe, WI: AMA, 1979). 14. American Medical Association, Profile ofMedical Practice 1981 (Chicago, IL: AMA, 1982). 15. American Medical Association, Department of Health Systems Analysis, Center for Health Policy Research, 'Trends in Medicare Assignment and Data on Participation Rates," Chicago, IL, February 1985. 16. American Society of Internal Medicine, Reimbursement for Physicians' Cognitive and Procedural Services: A White Paper (Washington, DC: ASIM, January 1981). 17. American Society of Internal Medicine, Toward an Incentive-Based Approach for Reforming the Health Care System (Washington, DC: ASIM, April 1982). 18. American Urological Association, Report of American Urological Association Ad Hoc Committee To Study the Safety and Clinical Efficacy of Current Technology of Percutaneous Lithotripsy and Non-Invasive Lithotripsy (Baltimore, MD: AUA, May 16, 1985). 19. Anderson, G., and Knickman, J., "Adverse Selection Under a Voucher System: Grouping Medicare Recipients by Level of Expenditure," Inquiry 21(2):135-143, Summer 1984. 20. Armstrong, G., Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Nov. 18, 1985. 21. Arnett, R.H., Cowell, C.S., Davidoff, L.M., et al., "Health Spending Trends in the 1980's: Adjusting to Financial Incentives," Health Care Financing Review 6(3):1-25, Spring 1985. 22. Arnould, R.J., 'The Effects of Medical Society Control of Blue Shield on Fees in the Physician Service Market: Some Preliminary Evidence," Quart.erly Review of Economics and Business 22(1):32-44, Spring 1982. 23. Arnould, R.J., Debrock, L.W., and Pollard, J.W., "Do HMOs Produce Specific Services More Efficiently?" Inquiry 21(3):243-253, Fall 1984. 24. Arthur Andersen & Co., Study of Reimbursement and Practice Arrangements of ProviderBased Physicians, submitted to the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, December 1977. 25. A.S. Hansen, Inc., 1983 Benefits Survey (Lake Bluff, IL: A.S. Hansen, Inc., 1984). 25a. "Average Net Income and Expenses of Physicians-1981," SMS Reports 1:5, June 1982. 26. Ball, J.R., American College of Physicians, Statement on Physician Payment Before the House Energy and Commerce Subcommittee on Health and the Environment, Apr. 26, 1985. 27. Ball, R., Consultant, Washington, DC, personal communication, Apr. 25, 1985. 28. Barer, M.L., Evans, R.G., and Labelle, R., 'The Frozen North: Controlling Physicians Costs Through Controlling Fees," prepared for the Of 252 • Payment for Physician Services: Strategies for Medicare fice of Technology Assessment, U.S. Congress, Washington, DC, November 1985. 29. Barger, S.B., President, Morgan Bigae Institute, Cincinnati, OH, personal communication, June 12, 1985. 30. Barnett, J., Bureau of Program Operations, Health Care Financing Administration, U.S. De partment of Health and Human Services, Wash ington, DC, personal communication, Decem ber 1984. 31. Becker, G., Human Capital (New York: Nation al Bureau of Economic Research, 1964). 32. Beebe, J., Lubitz, J., and Eggers, P., "Using Prior Utilization To Determine Payments for Medi care Enrollees in Health Maintenance Organi zations," Health Care Financing Review 6(3):2738, Spring 1985. 33. Berenson, R., "Group Decision-Making Meth ods," in J. Hadley, D. Juba, K. Swartz, et al., "Alternative Methods of Developing a Relative Value Scale of Physicians' Services: Year 1 Re port," The Urban Institute, Washington, DC, February 1983. 34. Berkun, M., Medical Director, Empire Blue Cross/Blue Shield, New York, personal commu nication, July 12, 1985. 35. Bermel, J., "More Power for States, Less for Em ployers," Hastings Center Report 15(5):16, Oc tober 1985. 36. Berry, C., Brewster, J.A., Held, P., et al., "Re sponses of Canadian Physicians to the Introduc tion of Universal Medical Care Insurance: The First Five Years in Q!1ebec: Volume I, Final Re port and Executive Summary," Mathematica Policy Research, Inc., Princeton, NJ, February 1978. 37. Berry, R., "A Review of the Medicare Economic Index: An Analysis of Variations in Practice Costs Among Specialties," Issues in Physician Reimbursement, N.T. Greenspan (ed.), HCFA Pub. No. 03121 (Baltimore, MD: Health Care Financing Administration, March 1981). 38. Billingstad, A., United Health Care Corp., Min netonka, MN, personal communication, Nov. 21, 1985. 39. Blue Cross and Blue Shield Association, "Q and A," Washington, DC, July 24, 1984. 40. Blue Cross and Blue Shield Association, Ex tracorporeal Shock Wave Lithotripsy: Clinical Assessment, Utilization and Cost Projections (Chicago, IL: Blue Cross and Blue Shield, May 1985). 41. Blue Cross and Blue Shield Association, "Blue Cross and Blue Shield Plan Activities in HMO's, December 31, 1984," Chicago, IL, June 5, 1985. 42. Blue Cross and Blue Shield Association, "Blue Cross and Blue Shield Plan Activities in HMO's, After December 31, 1984," Chicago, IL, Sept. 1, 1985. 43. Blue Cross and Blue Shield Association, "Blue Cross and Blue Shield Plans Marketing Preferred Provider Products," Chicago, IL, Oct. 1, 1985. 44. Blue Cross and Blue Shield Association, Professional and Provider Relations, Professional and Provider Affairs Division, "Physician Payment Methodologies and Their Relationships to Preferred Provider Products," Chicago, IL, March 1984. . 45. Blue Cross and Blue Shield of Florida, Staff, Jacksonville, FL, personal communication, Dec. 10, 1984. 46. Blumberg, M.S., "Provider Price Charges for Improved Health Care Use," Health Handbook, G.K. Chacko (ed.) (Amsterdam: North-Holland Publishing Co., 1979). 47. Blumberg, M.S., "At Risk for Hospitalization: Differences by Health Insurance Coverage and Income," Advances in Health Economics and Health Services Research, vol. 5, R.M. Scheffler and L.F. Rossiter (eds.) (Greenwich, CT: JAI Press Inc., 1984). 48. Blumberg, M.S., "Measures of Risk for ShortTerm Hospital Days and Part B Covered Charges in the 1977 CMS Aged Sample," Kaiser Foundation Health Plan, Oakland, CA, Apr. 10, 1985. 49. Blumberg, M.S., Kaiser-Permanente Medical Care Program, Oakland, CA, personal communication, July 18, 1985. 50. Bogue, T., Why Not the Most? A Physician's Guide to Locating in Cities With the Most Excessive Medicare Fees in the Country and an HEW Guide to Stopping This Waste of a Billion Dollars (Washington, DC: Health Research Group, Feb. 17, 1977). 51. Boland, P., "How To Negotiate a Cost-Effective PPO," Business and Health ,2(7):18-20, June 1985. 52. Boland, P., 'The Illusion of Discounts in the Health Care Market," Health Affairs 4(2):93-96, Summer 1985. 53. Boland, P., "Questioning Assumptions About Preferred Provider Arrangements," Inquiry 22(4):132-41, Summer 1985. 54. Bombardier, C., Fuchs, V.R., Lillard, L.A., et al., "Socioeconomic Factors Affecting the Utilization of Surgical Operations," N. Engl. ]. Med. 297(12):699-705, Sept. 29, 1977. 55. Booth, C., Director of Reimbursement Policy, Health Care Financing Administration, U.S. De partment of Health and Human Services, Baltimore, MD, personal communication, Nov. 16, 1985, and Dec. 5, 1985. 56. Bovjberg, R., Held, P., and Pauly, M., "Procompetitive Health Insurance Proposals and Their Implications for Medicare's End-Stage Renal Disease Program," Seminars in Nephrology 2:134-172, June 1982. 57. Bowman, M.A., Katsoff, }.M., Garrison, L.P., et al., "Estimates of Physician Requirements for 1990 for the Specialties of Neurology, Anesthesiology, Nuclear Medicine, Pathology, Physical Medicine and Rehabilitation, and Radiology," ].A.M.A. 250(101):2623-2627, Nov. 18, 1983. 58. Bradt, A., Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Dec. 4, 1985. 59. Brewster, A., Langwell, K., McMenamin, P., et al., "Evaluation of the Medicare Competition Demonstrations: Preliminary Implementation Case Studies of Four South Florida AHPs," Mathematica Policy Research, Inc., Washington, DC, Apr. 30, 1984. 60. Brewster, A. C., Jacobs, C.M., Bradbury, R.C., "Classifying Severity of Illness by Using Clinical Findings," Health Care Financing Review, annual supplement, pp. 207-208, November 1984. 61. Brook, R.H., and Avery, A.D., Quality Assessment: Issues of Definition and Measurement (Santa Monica, CA: Rand Corp., 1976). 62. Brook, R.H., and Lohr, K.N., "Second Surgical Opinion Programs: Beyond Cost-Benefit Analyses," Medical Care 20(1):1-2, January 1982. 63. Brook, R.H., and Williams, K.N., "Quality of Health Care for the Disadvantaged," f. Community Health 1:132-256, Winter 1975, as cited in G.T. Hammons, R.H. Brook, and J.P. Newhouse, "Evaluation of Effects on the Quality of Selected Alternatives for Paying Physicians Under the Medicare Program," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, August 1975. 64. Brook, R.H., Williams, K.N., and Rolph, }.E., "Controlling the Use and Costs of Medical Services: The New Mexico Experimental Medical Care Review Organization-A Four Year Case Study," Medical Care 16(1):1-76, January 1978. 65. Brook, R.H., Ware, }.E., Rogers, W.H., et al., "Does Free Care Improve Adults' Health?" N. Engl.]. Med. 309(23):1426-1434, Dec. 8, 1983. 66. Brown, L.D., "Politics as Usual and Customary? Physician Payment in Transition," in Institute of Medicine, National Academy of Sciences, Re- References • 253 forming Physician Payment: Report of a Conference (Washington, DC: National Academy Press, 1984). 67. Burke, T., Prudential Insurance Co., Newark, NL personal communication, Nov. 7, 1985. 68. Burney, I., and Schieber, G., "Medicare Physicians' Services: The Composition of Spending and Assignment Rates," Health Care Financing Review, forthcoming. 69. Burney, I., and Schieber, G., "Medicare Physicians' Services: The Composition of Spending and Assignment Rates," draft, Washington, DC, Apr. 8, 1985. 70. Burney, I., Hickman, P., Paradise, }., et al., "Medicare Physician Payment, Participation, and Reform," Health Affairs 3(4):5-24, Winter 1984. 71. Burney, I., Schieber, G.}., Blaxall, M.O., et al., "Medicare and Medicaid Physician Payment Incentives," Health Care Financing Review 1(1):62-76, Summer 1979. 72. Burstein, P.L., and Cromwell,]., "Relative Incomes and Rates of Return for U.S. Physicians," ]. Health Econ. 4:63-78, 1985. 73. "Business Gets Tough on Costs," Washington Report on Medicine and Health/Perspectives 38(33), Aug. 20, 1984. 74. California, General Assembly, Budget Analysis Office, Report of the Legislative Analyst to the Joint Legislative Budget Committee, Sacramento, CA, Feb. 27, 1985. 75. California Medical Association, 1964 Relative Value Studies (San Francisco, CA: CMA, 1964). 76. California Medical Association, 1969 Relative Value Studies (San Francisco, CA: CMA, 1969). 77. "California Surveys PPO Contracts," Medical World News 25(14):68, July 23, 1984. 78. Cameron, }., "Case-Mix and Resource Use in Hospital-Based Emergency Departments," prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, 1985. 79. Capron, A.M., "Evaluation of Ethical Implications of Selected Alternatives for Paying Physicians Under the Medicare Program," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, October 1985. 80. Cassack, D., "New Horizons in Physician Supply: The Boom in Physician Office Testing," Health Industry Today 47:21-35, 1984. 81. Chacko, G.K. (ed.), Health Handbook (Amsterdam: North-Holland Publishing Co., 1979). 82. "Changing Medical Practice Arrangements?" SMS Reports 2(7), November 1983. 83. Chollet, D.}., Employer Provided Health Bene 254 • Payment for Physician Services: Strategies for Medicare fits (Washington, DC: Employee Benefit Research Institute, 1984). 84. Christensen, S., Congressional Budget Office, U.S. Congress, Washington, DC, personal communication, Dec. 20, 1985. 85. Clauser, S.B., Fanta, C.M., Finkel, A.}., et al. (eds.), Physicians' Current Procedural Terminol ogy, Fourth Edition (CPT-4) (Chicago, IL: American Medical Association, 1985). 86. Coalition for Reducing Medicare Doctors' Charges, Does Your Doctor Take Medicare As signment?: A Directory of the Medicare Assign ment Practices of Doctors and Other Providers . . . (Washington, DC: Gray Panthers of Metropolitan Washington, April 1982). 87. Collen, M.F., "Payments to Physicians in the Permanente Medical Group," prepared for the Office of Technology Assessment, U.S. Con gress, Washington, DC, August 1985. 88. Commerce Clearing House, Inc., Medicare and Medicaid Guide, sec. 3330, pp. 1277-1278 (Chi cago, IL: Commerce Clearing House, Inc., 1984). 89. Commerce Clearing House, Inc., Medicare and Medicaid Guide (Chicago, IL: Commerce Clear ing House, Inc., 1985). 90. Conway, S., United Health Care Corp., Min netonka, MN, personal communication, Nov. 19, 1985. 91. Corbin, M., and Krute, A., "Some Aspects of Medicare Experience With Group-Practice Pre payment Plans," Social Security Bulletin 38(3):311, March 1975. 92. Costilo, LB., "Antitrust Enforcement in Health Care," N. Engl.]. Med. 313(14):901-904, Oct. 3, 1985. 93. Costilo, LB., U.S. Federal Trade Commission, Washington, DC, personal communication, No vember 1985. 94. Cotter, P.S., and Willer, }., "An Analysis of Physician Response to Medicare Participation," Chicago, IL, Sept. 25, 1985. 95. Crandall, O.K., "Payment for Physician Serv ices: Indemnity vs. UCR," Michigan Medicine 82(52): 684-5, November 1983. 96. Crowley, A.E., Etzel, S.I., and Petersen, E.S., "Undergraduate Medical Education," ].A.M.A. 252(12):1525-1532, Sept. 28, 1984. 97. Cunningham, F.C., and Williamson, J.W., "How Does Quality of Health Care in HMOs Compare to That in Other Settings? An Ana lytic Literature Review: 1958 to 1979," Group Health]. 1:4-25, Winter 1980. 98. Davis, C., as cited in LD. Brown, "Politics as Usual and Customary? Physician Payment in Transition," in Institute of Medicine, National 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. Academy of Sciences, Reforming Physician Pay ment: Report ofa Conference (Washington, DC: National Academy Press, 1984). Davis, C.K., Administrator, Health Care Financing Administration, U.S. Department of Health and Human Services, Memorandum to the Inspector General, Office of the Secretary, U.S. Department of Health and Human Serv ices, regarding OIG Draft Audit Report, 'The Medicare Reimbursement Method for Physician Visits to Skilled Nursing Facilities (SNFs) and Hospitals Should Be Revised (ACN 03-42005)," Washington, DC, May 23, 1984 . Davis, K., and Reynolds, R., 'The Impact of Medicare and Medicaid on Access to Medical Care," The Role of Insurance in the Health Serv ices Sector, R. Rosett (ed.) (New York: NBER Universities Conference, National Bureau of Economic Research, 1976). Davis, K., and Russell, LB., 'The Substitution of Hospital Outpatient Care for Inpatient Care," Review of Economics and Statistics 54(2):109120, May 1972. Deacon, R., Office of Demonstrations and Eval uation, Health Care Financing Administration, U.S. Department of Health and Human Serv ices, Baltimore, MD, personal communication, Dec. 4, and Dec. 20, 1985. Denker, L, 'The Final Report of the Interna tional Committee on Cognition in Medicine," Surgical Rounds, pp. 99-104, March 1984. Desmarais, H., Acting Deputy Administrator, Health Care Financing Administration, U.S. De partment of Health and Human Services, State ment before the Subcommittee on Health, Fi nance Committee, Senate, U.S. Congress, Dec. 6, 1985. Dever, R.C., Blue Cross/Blue Shield of Florida, Jacksonville, FL, personal communication, Aug. 23, 1985. "Differences in Practice Characteristics Between Male and Female Physicians," SMS Reports 3(2), March 1984. Donabedian, A., "An Evaluation of Prepaid Group Practice," Inquiry 6(3):3-27, September 1969. Donabedian, A., Explorations in Quality As sessment and Monitoring, Volume I: The Defi nition of Quality and Approaches to Its Assess ment (Ann Arbor, MI: Health Administration Press, 1980). Donabedian, A., Explorations in Quality As sessment and Monitoring, Volume II: The Cri teria and Standards of Quality (Ann Arbor, MI: Health Administration Press, 1982). Donabedian, A., Explorations in Quality As sessment and Monitoring, Volume III: The Methods and Findings of Quality Assessment and Monitoring: An Illustrated Analysis (Ann Arbor, MI: Health Administration Press, 1985). 111. Dowdal, T., General Accounting Office, U.S. Congress, Washington, DC, personal communication, Oct. 28, 1985. 112. Doyle, P., and Langwell, K., "National Evaluation of Medicare Competition: Preliminary Implementation Case Study of Genesee Health Care," Mathematica Policy Research, Inc., Washington, DC, Jan. 28, 1985. 113. Dresch, S.P., "Marginal Wage Rates, Hours of Work and Returns to Physician Specialization," Issues in Physician Reimbursement, N. Greenspan (ed.) (Washington, DC: U.S. Department of Health and Human Services, Health Care Financing Administration, 1981). 114. "DuPont Analyst Blood Analyzer for the Physician Group Practices Market," M-D-D-I Reports 11(41):11-12, Oct. 14, 1985. 115. Duriez, M., Glarmet, C., and Sandier, S., "Physician Compensation in France," The Compensation ofPhysicians: Approaches Used in Other Countries, U.E. Reinhardt (ed.) (Princeton, NJ: Princeton University, 1983). 116. Dutton, B.L., and McMenamin, P., 'The Medicare Economic Index: Its Background and Beginnings," Health Care Financing Review 3(1):137-140, September 1981. 117. Dutton, B.L., and McMenamin, P., 'The Medicare Economic Index: Its Background and Beginnings," Health Care Financing Review 3(1):137-40, September 1981. 118. Dyckman, Z., U.S. Department of Health, Edu cation, and Welfare, Washington, DC, personal communication, April 1980. 119. Egdahl, R.H., and Manuel, B., "A Consensus Process To Determine the Relative Complexity/Severity of Frequently Performed Surgical Services," Surgery, Gynecology & Obstetrics 160(5):403-406, May 1985. 120. Eggers, P., "Risk Differential Between Medicare Beneficiaries Enrolled and Not Enrolled in an HMO," Health Care Financing Review 1(3):9199, Winter 1980. 121. Eggers, P.W., and Prihoda, R., "Pre-Enrollment Reimbursement Patterns of Medicare Beneficiaries Enrolled in 'At-Risk' HMOs," Health Care Financing Review 4(10):55-73, September 1982. 122. Ehrenfried, D., Blue Cross and Blue Shield Association, Chicago, IL, personal communication, April 1985, and Oct. 30, 1985. 123. Ehrenhaft, P., "Payment Methods for Hospital- References • 255 Based and Teaching Physicians," prepared for the Office of Technology Assessment, U.S. Con gress, Washington, DC, April 1985. 124. Eiler, M.A. (ed.), Physician Characteristics and Distribution in the U.S., 1983 Edition (Chicago, IL: American Medical Association, 1984). 125. Eisenberg, J.M., "Physician Utilization: The State of Research About Physicians' Practice Patterns," Medical Care 23(5):461-483, May 1985. 126. Ellwood, P.M., and Paul, B.A., 'Testing the Waters With Competition vs. Regulation," Business and Health 1(5):5-8, April 1984. 127. Employee Benefit Research Institute, "Flexible Benefit Plans Provide Significant Incentives for Health Care Cost Containment," EBRI Issue Brief #43, June 1985. 128. "Employers Rethinking Benefits of HMO Participation," Washington Health Costs Letter 11(17):3-4, Aug. 30, 1985. 129. Enthoven, A.C., Health Plan: The Only Practical Solution to the Soaring Cost of Medical Care (Reading, MA: Addison-Wesley Publishing Co., 1980). 130. Ermann, D., and Gabel, L "The Changing Face of American Health Care: Multihospital Systems, Emergency Centers and Surgery Centers," Medical Care, in press. 131. Etheredge, L., "The Volume of Medicare Physicians Services," paper presented at the Project Hope forum on Health Care Financing Administration Funded Research, Washington, DC, Jan. 16-17, 1985. 132. Etheredge, L., Testimony before the Subcommittee on Health and the Environment, Committee on Energy and Commerce, House of Rep resentatives, U.S. Congress, Washington, DC, Apr. 26, 1985. 133. Etheredge, L., and Juba, D., "Medicare Payments for Physicians' Services," Health Affairs 3(4):132-137, Winter 1984. 134. Evans, R.G., "Health Care in Canada: Patterns of Funding and Regulation,"]. Health Politics, Policy and Law 8(1):1-43, Spring 1983. 135. Evans, R.G., and Barer, M.L., University of British Columbia, Vancouver, British Columbia, personal communication, Nov. 14, 1985. 136. Felch, W.C., "The Internist, Cognoscenti, and Cognition," The Internist, pp. 12-13, JulyAugust 1981. 137. Feldstein, M.S., "A New Approach to National Health Insurance," The Public Interest (23):93105, Spring 1971. 138. Ferry, T.P., Gornick, M., Newton, M., et al., 256 • Payment for Physician Services: Strategies for Medicare "Physicians' Charges Under Medicare: Assign ment Rates and Beneficiary Liability," Health Care Financing Review 1(3):49-73, Winter 1980. 139. Fetter, R.B., "Development of an Ambulatory Classification System," prepared for the Health Care Financing Administration, U.S. Depart 150. ment of Health and Human Services, Baltimore, MD, June 1985. 140. Fetter, R.B., Averill, R.F., Lichtenstein, J.L., et al., "Ambulatory Visit Groups: A Framework for Measuring Productivity on Ambulatory Care," Health Serv. Res. 19(4):415-437, October 1984. 141. Fetter, R.B., Shin, Y., Freeman, J.L., et al., "Case Mix Definition by Diagnosis-Related Groups," Medical Care 18(supp.):1-53, Septem ber 1980. 151. 142. Fielding, }.E., "A Utilization Review Program in the Making," Business and Health 2(7):25-28, June 1985. 143. Figueroa, R., Joint Legislative Budget Commit 152. tee, California Legislature, Sacramento, CA, personal communication, Apr. 19, 1985. 144. Fine, M., American Association of Preferred Provider Organizations, Washington, DC, per sonal communication, Apr. 23, 1985. 153. 145. Fineberg, H.V., "Clinical Chemistries: The High Cost of Low-Cost Diagnostic Tests," in National Center for Health Services Research and Bureau 154. of Health Planning, Public Health Service, U.S. Department of Health, Education, and Welfare, Medical Technology: The Culprit Behind Health Care Costs? S.H. Altman and R. Blendon (eds.), 155. DHEW Pub. No. (PHS) 79-3216 (Washington, DC: U.S. Department of Health, Education, and Welfare, 1979). 146. Finkel, M.L., Ruchlin, H.S., and Parsons, S.K., Eight Years' Experience With a Second Opinion 156. Elective Surgery Program: Utilization and Eco nomic Analyses, HCFA Pub. No. 03095 (Balti more, MD: U.S. Department of Health and Human Services, Health Care Financing Admin 157. istration, Office of Research, Demonstrations, and Statistics, March 1981). 147. Forbes, D., "Companies Cut Back On Benefits: What It's Costing Employees," Dun's Business 158. Month 126(4):64-68, October 1985. 148. Fox, P.O., "Physician Reimbursement Under Medicare: An Overview and a Proposal for Area-Wide Physician Incentives," in Subcom 159. mittee on Health, Committee on Ways and Means, U.S. House of Representatives, Proceed ings of the Conference on the Future of Medi 160. care, WMCP: 98-23 (Washington, DC: U.S. Government Printing Office, Feb. 1, 1984). 149. Fox, P.O., Goldbeck, W.B., and Spies, ].]., 161. "Synthesis of Private Sector Health Care Initiatives," prepared for the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services, Washington, DC, February 1984. Francis, A.M., Polissar, L., and Lorenz, A.B., "Care of Patients With Colorectal Cancer: A Comparison of a Health Maintenance Organization and Fee-for-Service Practices," Medical Care 22:418-429, 1984, as cited in G.T. Hammons, R.H. Brook, and}.P. Newhouse, "Evaluation of Effects on the Quality of Care of Selected Alternatives for Paying Physicians Under the Medicare Program," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, August 1985. Frech, H.E., and Ginsburg, P.B., "Imposed Health Insurance in Monopolistic Markets: A Theoretical Analysis," Economic Inquiry 13:5570, March 1975. Freeland, M., Bureau of Data Management and Strategy, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Apr. 4, 1985. Freiman, M.P., "Cost Sharing Lessons From the Private Sector," Health Affairs 3(4):85-93, Win ter 1984. Freiman, M.P., "The Rate of Adoption of New Procedures Among Physicians: The Impact of Specialty and Practice Characteristics," Medi cal Care 23(8):939-45, August 1985. Friedlob, A.S., and Hadley, J.P., "Marketing Medicare in a Competitive Environment," Health Care Financing Administration, U.S. De partment of Health and Human Services, Balti more, MD, April 1985. Gabel, J.R., and Erman, D., "Preferred Provider Organizations: Performance, Problems and Promise," Health Affairs 4(1):24-40, Spring 1985. Gabel, J.R., and Monheit, A.C., "Will Compe tition Plans Change Insurer-Provider Relation ships?" Milbank Mem. Fund Q. 61(4):614-640, Fall 1983. Gabel, J.R., and Rice, T.H., "Reducing Public Expenditures for Physician Services: The Price of Paying Less," f. Health Politics, Policy and Law 9(4):595-609, Winter 1985. Gallivan, M., "Physician Offices Invade Clini cal Laboratory Market," Hospitals 59(20):84-94, Oct. 16, 1985. Gardner, S.F., Kyzr-Sheeley, B.]., and Saba tino, F., "Big Business Embraces Alternate De livery," Hospitals 59(4):81-84, Mar. 16, 1985. Garrison, L.P., and Yamashiro, S.M., "Back ground Paper on Cataract Surgery and Physician Payment Under the Medicare Program," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, October 1985. 162. Geissler, U., "Physician Compensation in the Federal Republic of Germany," The Compensation ofPhysicians: Approaches Used in Other Countries, U.E. Reinhardt (ed.) (Princeton, NJ: Princeton University, 1983). 163. Gertman, P.M., and Restuccia, J.D., 'The Appropriateness Evaluation Protocol: A Technique for Assessing Unnecessary Days of Hospital Care," Medical Care 19(8):855-871, August 1981. 164. .::;ibson, R.M., "National Health Expenditures, 1978," Health Care Financing Review 1(1):1-36, Summer 1979. 165. Gibson, R.M., Levit, K.R., Lazenby, H., et al., "National Health Expenditures, 1983," Health Care Financing Review 6(2):1-30, Winter 1984. 166. Ginsburg, P.B., Statement before the Subcommittee on Health and the Environment, Committee on Energy and Commerce, House of Representatives, U.S. Congress, Washington, DC, Apr. 26, 1985. 167. Ginsburg, P.B., Rand Corp., Washington, DC, personal communication, Nov. 21, 1985. 168. Ginzberg, E., "A New Supply Policy Is Needed," ].A.M.A. 250(19):2621-2622, Nov. 18, 1983. 169. Glandon, G.L., and Werner, J.L., "Physicians' Practice Experience During the Decade of the 1970s," ].A.M.A. 244(22):2514-2518, Dec. 5, 1980. 170. Gold, M., 'The Demand for Hospital Outpatient Services," presented at the Annual Meeting of the American Public Health Association, Los Angeles, CA, Nov. 2, 1981. 171. Goldberg, L.G., and Greenberg, W., 'The Effect of Physician-Controlled Health Insurance: U.S. v. Oregon State Medical Society,"]. Health Politics, Policy, and Law 2(1):48-78, Spring 1977. 172. Goldsmith, J., "Death of a Paradigm: The Challenge of Competition," Health Nfairs 3(3):5-19, Fall 1984. 173. Gonnella, J.S., Hornbrook, M.C., Louis, D.Z., "Staging of Disease: A Case-Mix Measurement," ].A.M.A. 251(5):637-44, Feb. 3, 1984. 174. Goodman, N., Antitrust Division, U.S. Department of Justice, Washington, DC, personal communication, June 4, 1985. 174a. Gornick, M., Office of Research, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Dec. 7, 1985. References • 257 175. Gornick, M., Hackerman, C., and Newton, M., "Factors Affecting Differences in Medicare Reimbursements for Physicians' Services," Health Care Financing Review 1(4):15-38, Spring 1980. 176. Graduate Medical Education National Advisory Committee, Report of the Graduate Medical Education National Advisory Committee to the Secretary, Department of Health and Human Services (Washington, DC: U.S. Department of Health and Human Services, September 1980). 177. Graugnard, S., American Association of Preferred Provider Organizations, Washington, DC, personal communication, June 11, 1985. 178. Greenlick, M.R., Freeborn, O.K., Colombo, T.J., et al., "Comparing the Use of Medical Care Services by a Medically Indigent and a General Membership Population in a Comprehensive Prepaid Group Practice Program," Medical Care 10:197-200, May-June 1972. · 179. Greenlick, M.R., Hurtado, A.V., Pope, C.R., et al., "Determinants of Medical Care Utilization," Health Serv. Res. 3(4):296-315, Winter 1968. 180. Greenspan, N. (ed.), Issues in Physician Reimbursement (Washington, DC: U.S. Department of Health and Human Services, Health Care Financing Administration, 1981). 181. Gruenberg, L., and Stuart, N., "A Health Status-Based AAPCC: The Disability-Level Approach," Discussion Paper 55, Health Policy Center, Brandeis University, Waltham, MA, December 1982. 182. Gruenberg, L., and Tompkins, C., "An Analysis of Risk-Sharing and Reinsurance in Medicare HMOs," Discussion Paper 59, Health Policy Center, Brandeis University, Waltham, MA, June 1984. 183. Gruenberg, L., Tompkins, C., Wallack, S.S., et al., "A Discussion of Options for Modifying the Current AAPCC Approach for the Financing of Medicare Enrollees in HMOs," draft, University Health Policy Consortium, Brandeis University, Waltham, MA, January 1985. 184. Hadley, J., "An Econometric Analysis of Physician Participation in the Medicaid Program," The Urban Institute Working Paper No. 998-9, Washington, DC, April 1978. 185. Hadley, J. (ed.), Medical Education Financing (New York: PRODIST, 1980). 186. Hadley, J., "Reimbursement, Physicians' Incomes and Physicians' Specialty and Location Decisions," Medical Education Financing, J. Hadley (ed.) (New York: PRODIST, 1980). 187. Hadley, J., and Berenson, R., "Seeking the 'Just' Price: Relative Value Scales and Fee Schedules 258 • Payment for Physician Services: Strategies for Medicare for Physicians' Services," in J. Hadley, D. Juba, R. Berenson, et al., "Final Report on Alternative Methods of Developing a Relative Value Scale of Physicians' Services," The Urban Institute, Washington, DC, October 1984. 188. Hadley, J., and Lee, R., "Physicians' Price and Output Decisions: Theory and Evidence," Urban Institute Working Paper No. 998-8, The Urban Institute, Washington, DC, April 1978. 189. Hadley, J., and Lee, R., 'Toward a Physician Payment Policy: Evidence From the Economic Stabilization Program," Policy Sciences 10:105120, 1978-1979. 190. Hadley, J., Holahan, J., and Scanlon, W., "Can Fee-for-Service Reimbursement Coexist With Demand Creation?" Inquiry 16(3):247-258, Fall 1979. 191. Hadley, J., Juba, D., Berenson, R., et al., "Final Report on Alternative Methods of Developing a Relative Value Scale of Physicians' Services," The Urban Institute, Washington, DC, October 1984. 192. Hadley, J., Juba, D., Swartz, K., et al., "Alternative Methods of Developing a Relative Value Scale of Physicians' Services: Year 1 Report," The Urban Institute, Washington, DC, February 1983. 193. Hadley, J.P., Office of Demonstrations and Evaluation, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Aug. 21, 1985, Nov. 27, 1985, and Dec. 30, 1985. 194. Hammons, G.T., Brook, R.H., and Newhouse, J.P., "Evaluation of Effects of the Quality of Care of Selected Alternatives for Paying Physicians Under the Medicare Program," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, September 1985. 195. Hanlon, C.R., "Is There Cognition Before Operation?" Bulletin American College of Surgeons 68(7):1-2, July 1983. 196. Harper, J., Aetna Insurance Co., Hartford, CT, personal communication, Nov. 12, 1985. 197. Haynes, P.L., Evaluating State Medicaid Reforms (Washington, DC: American Enterprise Institute for Public Policy Research, 1985). 198. "HCFA Approves Arizona Health Care Cost Containment System Third-Year Funding, But Threatens Termination of Demonstration Project If Data Requirements Unmet," The Blue Sheet 27(30):7, July 25, 1984. 199. "HCFA To Allow Delegation of HMO Review," Washington Report on Medicine and Health 39(3):2, Jan. 21, 1985. 200. Health Insurance Association of America, "Fact Sheet," Washington, DC, 1984. 201. Health Insurance Association of America, Source Book of Health Insurance Data: 198283 (Washington, DC: HIAA, no date). 202. Heatherington, H., Office of Reimbursement Policy, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Dec. 13, 1985. 203. Hedger, D., and Schmitt, D., 'Trends in Major Medical Coverage During a Period of Rising Costs," Monthly Labor R. 106(7):11-16, July 1983. 204. Hekman, E.L., State Efforts at Health Care Cost Containment, Health Care Cost Containment Project: Paper #3 (Denver, CO: National Conference of State Legislatures, September 1984). 205. Held, P.J., and Reinhardt, U.E., "Prepaid Med~ ical Practice: A Summary of Findings From a Recent Survey of Group Practices in the United States," The Group Health ]ournal1(2):4-15, Summer 1980. 206. Hellinger, F.J., "Substitutability Among Different Types of Care Under Medicare," Health Serv. Res. 12(1):11-18, Spring 1977. 207. Hembree, W.E., "Getting Involved: Employers as Case Managers," Business and Health 2(8):1114, July/August 1985. 208. Herzlinger, R.E., "How Companies Tackle Health Care Costs: Part II," Harvard Business R. 63(5):108-120, September/October 1985. 209. Herzlinger, R.E., and Schwartz, J., "How Companies Tackle Health Care Costs: Part I," Harvard Business R. 63(4):69-81, July/ August 1985. 210. Hess, T.P., "Blues and Medicare Bend Slowly To Demand for MRI Reimbursement," Diagnostic Imaging 7(7):55-57, July 1985. 211. "HMO Savings Questioned," Medical World News 26(18):12-3, Sept. 23, 1985. 212. "HMOs Grow and Change," Washington Report on Medicine and Health 39(26), July 1, 1985. 213. Hodgson, T.A., and Kopstein, A.N., "Health Care Expenditures for Major Diseases in 1980," Health Care Financing Review 5(4):1-12, Summer 1984. 214. Holahan, J., "Paying for Physician Services in State Medicaid Programs," Health Care Financing Review 5(3):99-110, Spring 1984. 215. Holahan, J., and Scanlon, W., Physician Pric ing in California: Price Controls, Physicians' Fees, and Physicians' Incomes From Medicare and Medicaid (Washington, DC: U.S. Department of Health and Human Services, Health Care Financing Administration, 1979). 216. Holahan, J., Scanlon, W., Hadley, J., et al., 'The Effect of Medicare/Medicaid Reimbursementon Physician Behavior," in Office of Research, Demonstrations, and Statistics, Health Care Financing Administration, U.S. Department of Health and Human Servires, Health Care Financing Conference Proceedings: Physicians and Financial Incentives, J.R. Gabel, J. Taylor, N.T. Greenspan, et al. (eds.), HCFA Pub. No. 03067 (Washington, DC: HCFA, December 1980). 217. Hoover, M., Humana Care Plus, Louisville, KY, personal communication, Nov. 21, 1985. 218. Horn, J., HCA Health Plans, Nashville, TN, personal communication, Nov. 15, 1985. 219. Horn, S.D., Bulkley, G., Sharkey, P.O., et al., "Interhospital Differences in Severity of Illness: Problems for Prospective Payment Based on Diagnosis Related Groups (DRGs)," N. Engl. f. Med. 33(1):20-24, July 1985. 220., Horn, S.D., Horn, R.A., Sharkey, P.D., 'The Severity of Illness Index as an Adjustment to ORCs," Health Care Financing Review, annual supplement, pp. 33-45, November 1984. 221. Hornbrook, M.C., 'Techniques for Assessing Hospital Case-Mix," Annual Review of Public Health 6:295-324, 1985. 222. Hornbrook, M.C., Kaiser Permanente Center for Health Research, Portland, OR, personal communication, July 1985. 223. Hornbrook, M.C., and Berki, S.E., "Practice Mode and Payment Method," Medical Care 23(5):484-510, May 1985. 224. Hornbrook, M.C., Hurtado, A.V., and Johnson, R.E., "Health Care Episodes: Definition, Measurement, and Use," Medical Care Review 42(2):163-218, Fall 1985. 225. Horton, W.H., "Studies To Establish a Professional Services Index," The Connecticut State Medical Journal, April 1953. 226. Horton, W.H., "Professional Services Index Studies: A Comparison of Results in Connecticut and Montana," The Connecticut State Medical Journal, September 1955. 227. Hsaio, W.C., and Stason, W.B., "Toward Developing a Relative Value Scale for Medical and Surgical Services," Health Care Financing Review 1(2):23-38, Fall 1979. 228. Hughes, E.F.X., Fuchs, V.R., Jacoby, J.E., et al., References • 259 "Surgical Work Loads in a Community Practice," Surgery 71:315-327, March 1972. 229. Hull, D., Office of Health Planning, Health Resources and Services Administration, Public Health Service, U.S. Department of Health and Human Services, Washington, DC, personal communication, Jan. 30, 1985. 230. Hunt, F.D., "Revolution in Health Insurance," Society of Professional Benefit Administrators, Washington, DC, no date. 231. Hutton, W.R., "Medicare Reimbursement of Physicians," Statement before the Subcommittee on Health and the Environment, Committee on Energy and Commerce, U.S. House of Representatives, Washington, DC, Apr. 26, 1985. 232. Hyman, L., Metropolitan Life Insurance Co., New York, personal communication, Nov. 7, 1985. 233. Iezzoni, L., and Moskowitz, M., 'The Clinical Impact of DRG-Based Physician Reimbursement," Health Policy Research Consortium, Waltham, MA, Dec. 31, 1984. 234. Iezzoni, L.I., Grad, 0., and Moskowitz, M.A., "Reform of Medicare Physician Payment Policies: Impact on Magnetic Resonance Imaging Technology," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, Sept. 9, 1985. 235. Iglehart, J.K., "Medicaid in Transition," N. Engl. f. Med. 309(14):868-72, Oct. 6, 1983. 236. Ingram, J., and Fischenich, J., "Physician Reimbursement," Topics in Health Care Financing 9:32-53, Fall 1982. 237. Institute for International Health Initiatives, Directory of Preferred Provider Organizations and the Industry Report on PPO Development (Bethesda, MD: American Medical Care andReview Association, June 1985). 238. Intergovernmental Health Policy Project and State Medicaid Information Center, Recent and Proposed Changes in State Medicaid Programs: A Fifty State Survey (Washington, DC: Intergovernmental Health Policy Project, George Washington University, November 1982). 239. Interstudy, Inc., HMO Status Report 1982-1983 (Excelsior, MN: Interstudy, Inc., August 1984). 240. Interstudy, Inc., National HMO Census 1984 (Excelsior, MN: Interstudy, Inc., 1985). 241. Isaacson, S., Office of Civilian Health and Medical Programs of the Uniformed Services, U.S. Department of Defense, Denver, CO, personal communication, Aug. 20, 1985. 242. Jencks, S.F., and Dobson, A., "Strategies for 260 • Payment for Physician Services: Strategies for Medicare Reforming Medicare's Physician Payments: Physician Diagnosis-Related Groups and Other Approaches," N. Engl. ]. Med. 312(23):1492-255. 1499, June 6, 1985. 243. Jenkins, A., University of Virginia Medical Center, Charlottesville, VA, personal communica256. tion, Nov. 26, 1985. 244. Johns, L., Consultant, San Francisco, CA, per257. sonal communication, November 1985. 245. Johns, L., Derzon, R.A., and Anderson, M.D., Selective Contracting for Health Services in California: First Report (Washington, DC: National Governors' Association, 1983). 246. Johns, L., Derzon, R.A., and Anderson, M.D., 258. "Selective Contracting in California: Early Effects and Policy Implications," Inquiry 22(1):2432, Spring 1985. 247. Juba, D., "Analysis of Issues Relating to Implementing a Medicare Physician Fee Schedule," 259. prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, November 1985. 248. Juba, D.A., and Sulvetta, M., "Decomposing 260. Medicare Part B Payments for Physicians' Services," Urban Institute Working Paper 3402-0101, The Urban Institute, Washington, DC, Sep261. tember 1985. 249. Keeler, E.B., Rolph, J.E., Duan, N., et al., The Demand for Episodes ofMedical Treatment: Interim Results From the Health Insurance Exper262. iment (Santa Monica, CA: Rand Corp., December 1982). 250. Kilpatrick, S.J., 'The Distribution of Episodes 263. of Illness-A Research Tool in General Practice?" journal of the Royal College of General Practitioners 25:686-690, 1975. 264. 251. Kirchner, M., "How Much-And How Often -Should You Raise Your Fees?" Medical Economics 58:126-141, May 25, 1981. 265. 252. Knapp, R., Association of American Medical Colleges, Washington, DC, personal communication, Nov. 19, 1984. 253. Kralweski, J.E., Feldman, R., Dowd, B.E., et al., "Employer Perspectives on the Preferred Pro266. vider Organization Concept," Hospital and Health Services Administration, pp. 123-139, July/August 1984. / 254. Kusserow, R.P., Inspector General, U:S. De267. partment of Health and Human Services, memorandum to Carolyn K. Davis, Administrator, Health Care Financing Administration, U.S. Department of Health and Human Services, regarding OIG Final Report-'The 'Multiple Visit' 268. Concept Should Be Applied to Physician Visits to Medicare Beneficiaries in Skilled Nursing Fa cilities and Hospitals (ACN 03-42005)," Washington, DC, Sept. 17, 1984. La Jolla Management Co., "Analysis of State Medicaid Program Characteristics," Rockville, MD, 1982. Laly, J., Senior Health Plan, Minneapolis, MN, personal communication, Dec. 5, 1985. Langwell, K., "National Evaluation of Medicare Competition Demonstrations: Preliminary Implementation Case Study of Family Health Program, Inc., Long Beach, California," Mathematica Policy Research, Inc., Washington, DC, Jan. 28, 1985. Langwell, K., "National Evaluation of Medicare Competition Demonstrations: Preliminary Implementation of Delmarva Health Care Plan," Mathematica Policy Research, Inc., Washington, DC, Jan. 28, 1985. Langwell, K.M., and Nelson, L.M., "Physician Payment Systems: A Review of History, Alternatives and Evidence," Mathematica Policy Research, Washington, DC, Jan. 4, 1985. Leavit, H., American Medical International, Beverly Hills, CA, personal communication, Nov. 15, 1985. Lee, A.J., Hefner, D., Dobson, A., et al., "Evaluation of the Maximum Allowable Cost Program," Health Care Financing Review 4(3):7182, March 1983. · Lee, R.L., and Hadley, J., "Physicians' Fees and Public Medical Care Programs," Health Serv. Res. 16(2):185-204, Summer 1981. Lee, S.S., Quebec's Health System: A Decade of Change, 1967-77 (Canada: The Institute of Public Administration in Canada, 1979). Lee, S.S., and Butler, L.M., "The Three-Layered Cake: A Plan for Physician Compensation," N. Engl. ]. Med. 291(5):253-256, Aug. 1, 1974. Leibowitz, A., Manning, W.G., Keeler, E.B., et al., "The Effect of Cost Sharing on the Use of Medical Services by Children: Interim Results From a Randomized Controlled Trial," Pediatrics 75(5):942-51, May 1985. Lerner, A.N., "Federal Trade Commission AntiTrust Activities in the Health Care Services Field,"]. American and Foreign Antitrust and Trade Regulation 22(2):205-24, Summer 1984. Lerner, A.N., Assistant Director for Health Care, Bureau of Competition, Federal Trade Commision, presentation to the Congressional Research Service, U.S. Congress, Washington, DC, July 1985. Linton, O.W., American College of Radiology, Bethesda, MD, personal communication, Nov. 19, 1984. 269. Lion, J., Health Policy Consortium, Brandeis University, Waltham, MA, personal communication, July '1985. 270. Lion, J., "Prospective Payment for Ambulatory Care Based on Case-Mix," The DRG Dilemma, J.M. Cameron (ed.), forthcoming. 271. Lion, J., Malbon, A., Henderson, M.G., et al., "A Comparison of Hospital Outpatient Departments and Private Practices," Health Care Financing Review 6(4):69-81, Summer 1985. 272. Lohr, K.N., Lohr, W.R., and Brook, R.H., Geographic Variations in the Use of Medical Services and Surgical Procedures (Washington, DC: National Health Policy Forum, October 1985). 273. Louis Harris & Associates, Inc., Corporate Initiatives and Employee Attitudes on Cost Containment (New York: The Equitable Life Assurance Society of the United States, February I March 1985). 274. Louis Harris & Associates, Inc., A Report Card on HMOs: 1980-1984, prepared for the Henry J. Kaiser Family Foundation, New York, no date. 275. Lowenstein, S.R., lezzoni, L.l., Moskowitz, M.A., "Prospective Payment for Physician Services," ].A.M.A. 254(18):2632-2637, Nov. 8, 1985. 276. Lubitz, 1., and Nelson, S., "Evaluation of the Medicare Competition Demonstrations: Preliminary Implementation Case Study of Genesee Valley Group Health Association," Mathematica Policy Research, Inc., Washington, DC, August 1984. 277. Lubitz, J., and Prihoda, R., 'The Use and Costs of Medicare Services in the Last 2 Years of Life," Health Care Financing Review 5(3):117-132, Spring 1984. 278. Lubitz, J., Beebe, J., and Riley, G., "Improving the Medicare HMO Payment Formula To Deal' With Biased Selection," Advances in Health Economics and Health Services Research, vol. 6, R.M. Scheffler and L.F. Rossiter (eds.) (Greenwich, CT: JAI Press, forthcoming). 279. Luft, H.S., Health Maintenance Organizations: Dimensions of Performance (New York: John Wiley & Sons, 1981). 280. Luft, H.S., and Crane, S., "Regionalization of Services Within a Multihospital Health Maintenance Organization," Health Serv. Res. 15(3):231-247, Fall 1980. 281. Luft, H.S., Bunker, J.P., and Enthoven, A.C., "Should Operations Be Regionalized?" N. Engl. f. Med. 301(25):1364-1369, Dec. 20, 1979. 282. Luft, H.S., Garnick, D.W., and Hunt, S.S., References • 261 "Evaluation of Utilization Patterns of Physicians' Practice Arrangements," prepared for the Office of the Assistant Secretary for Planning Evaluation, U.S. Department of Health and Human Services, Washington, DC, January 1985. 283. Maag, R., CIGNA Corp., Hartford, CT, personal communication, Nov. 14, 1985. 284. "Malpractice Claim Risk Has Jumped Since 1980," American Medical News 28(16):10, Apr. 19, 1985. 285. Manning, W.G., Leibowitz, A., Goldberg, G.A., et al., "A Controlled Trial of the Effect of a Prepaid Group Practice on Use of Services," N. Engl. f. Med. 310(23):1505-1510, June 7, 1984. 286. Markel, G.A., Per-Case Reimbursement for Medical Care, Research Report 77-5 (Camp Hill, PA: Pennsylvania Blue Shield, March 1977). 287. Marmor, T.R., The Politics of Medicare (Chicago, IL: Aldine Publishing Co., 1970). 288. Marquis, M.S., "Cost-Sharing and Provider Choice," ]. Health Econ. 4(2):137-158, June 1985. 289. Marshall, A., Principles of Economics (London: The Macmillan Co., 1922). 290. "Maryland Blues Bid To Take Over State Medicare Plan," Washington Health Costs Letter 11(7):4, Aug. 30, 1985. 291. McClure, W., "On the Research Status of RiskAdjusted Capitation Rates," Inquiry 21(3):205213, Fall 1984. 292. McMenamin, P., "On the Welfare Economics of Government Subsidized Demand for Health Care Services," Internal Working Paper #ORI, Health Care Financing Administration, U.S. Department of Health and Human Services, Washington, DC, September 1980. 293. McMenamin, P., "Future Research and Policy Directions in Physician Reimbursement," Health Care Financing Review 2(4):61-76, Spring 1981. McMenamin, P., unpublished tabulations of CY1983 Medicare allowed charges from Minnesota Blue Shield and South Carolina Blue Cross Blue Shield, MANDEX, Inc., Vienna, VA, 1985. 295. McMillan, A., and Gornick, M., "The Dually Entitled Elderly Medicare and Medicaid Population Living in the Community," Health Care Financing Review 6(2):73-85, Winter 1984. 296. McMillan, A., Lubitz, J., and Newton, M., 'Trends in Medicare Physician Assignment Rates, 1968-1984," Health Care Financing Review, forthcoming. 297. McMillan, A., Pine, P.L., and Newton, M., 262 • Payment for Physician Services: Strategies for Medicare Medicare: Use ofPhysicians' Services Under the 314. Supplementary Medical Insurance Program, 1975-1978 (Baltimore, MD: U.S. Department of Health and Human Services, Health Care Fi nancing Administration, March 1983). 298. MD-IPA Health Plan, 1984 Annual Report (Rockville, MD: MD-IPA, 1985). 315. 299. "Medical Practice Indicator Survey Results, Sec ond Half 1983," SMS Reports 3(1), February 1984. 300. "Medicare Assignment: Recent Trends and Par ticipation Rates," SMS Reports 4(1), February 1985. 316. 301. "Medicare Lithotriptor Coverage Includes Ex tracorporeal and Percutaneous Methods," Tech nology Reimbursement Reports 1(14):2, May 3, 317. 1985. 302. "Medicare's New HMO Rules Under Fire," American Medical News, p. 9, Jan. 25, 1985. 303. Metropolitan Life Insurance Co. Staff, New 318. York, personal communication, Jan. 2, 1985. 304. Meyer, J ., American Enterprise Institute, Wash ington, DC, personal communication, Oct. 10, 1985. 305. Meyer, J.A., Johnson, W.R., and Sullivan, S., Passing the Health Care Buck: Who Pays the 319. Hidden Cost? (Washington, DC: American En terprise Institute, July 1983). 306. Meyerhoff, A.S., "Health Care Coalitions: The Evolution of a Movement," Health Affairs 3(1):120-127, Spring 1984. 307. Meyerhoff, A.S., "The Employer's Role in Con 320. trolling Health Care Costs," Compensation and Benefits Management 1(3):235-240, Spring 1985. 308. Meyerhoff, A.S., and Crozier, D.A., "Health Care Coalitions: The Evolution of Movement," Health Affairs 3(1):120-127, Spring 1984. 309. Michigan Academy ofFamily Physicians Versus Blue Cross and Blue Shield of Michigan, 728F 321. 2d 32; 6th Circuit Court. / 310. "Michigan Blues Plan Laboratory ,Services PPO," Business and Health 1(9):47-48, Septem ber 1984. 311. Milliman & Robertson, Inc., "A Review of AAPCC Methodology for Implementing Pro 322. spective Contracts With HMO's," prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Bal timore, MD, Nov. 3, 1983. : 312. Mills, E.S., "History of Private Business Paid in-Full Programs," National Association of Blue 323. Shield Plans, Chicago, IL, Mar. 27, l973. 313. Mitchell, J.B., "Physician DRGs," N. Engl. f. Med. 313(11):670-75, Sept. 12, 1985. Mitchell, ].B., "DRG-Based Case Payment for Inpatient Physician Services," Testimony before the U.S. Senate Committee on Finance, Subcommittee on Health, Hearing on Medicare Physician Payment, Washington, DC, Dec. 6, 1985. Mitchell, J.B., and Cromwell, J., "Physician Behavior Under the Medicare Assignment Option," final report prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, 1981. Mitchell, J.B., and Cromwell, J., "Physician Behavior Under the Medicare Assignment Option," f. Health Econ. 1:245-64, 1982. Mitchell, J.B., and Cromwell,]., "Impact of an Ali-or-Nothing Assignment Requirement Under Medicare," Health Care Financing Review 4(4): 59-78, Summer 1983. Mitchell, ].B., and Hewes, H., "Medicare Access to Physician Services in Nursing Homes," final report prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, Nov. 22, 1982. Mitchell, J.B., Calore, K.A., Cromwell, J. et al., "Alternative Methods for Defining Physician Services Performed and Billed," final report prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, November 1983. Mitchell, J.B., Calore, K.A., Cromwell, J., et al., "Creating DRG-Based Physician Reimbursement Schemes: A Conceptual and Empirical Analysis," Year I report prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, October 1984. Mitchell, J.B., Calore, K.A., Cromwell, J., et al., "Physician DRGs: What Do They Look Like and How Do They Work?" prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, February 1985. Mitchell, J.B., Schurman, R., Cromwell, J., et al., "Analysis of Changes in the Content of Physicians' Office Visits," report to the Health Resources and Services Administration, U.S. Department of Health and Human Services, Hyattsville, MD, June 30, 1983. Mobile Technology Survey, as cited in L.l. Iezzoni, 0. Grad, and M.A. Moskowitz, "Reform of Medicare Physician Payment Policies: Impact on Magnetic Resonance Imaging Technology," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, July 12, 1985. 324. Morris, L.C., "A Third-Party Carrier Perspective on Physician Payment," prepared for the Conference on Strategies for Reform of Physician Payment, Institute of Medicine, National Academy of Sciences, Washington, DC, Oct. 27-28, 1983. 325. Morris, S., Bureau of Quality Assurance, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Aug. 27, 1985. 326. Morrisey, M.A., and Brooks, D.C., "The Myth of the Closed Medical Staff," Hospitals 59(13): 78-81, July 1, 1985. 327. Morse, R., Hancock/Dikewood Health Plans, Boston, MA, personal communication, Nov. 11, 1985. 328. Moshman Associates, Inc., "Medical Procedural Terminology Systems: Development and Characteristics of Three Major Systems for Third Party Payment," Health Care Financing Research and Development Series, Report No.4, Office of Policy Planning and Research, Health Care Financing Administration, U.S. Department of Health, Education, and Welfare, Washington, DC, undated. 329. Moskowitz, D.B., "Health Benefits: Now States Call the Shots," Business Week #2899:39-40, June 17, 1985. 330. Muller, C.F., and Otelsberg, J., '~Alternative Approaches to Physician Reimbursement Under Medicare: A Simulation," final report to the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, Aug. 15, 1979. 331. Muller, C.F., and Otelsberg, J., Study of Physician Reimbursement Under Medicare and Medicaid, HEW Pub. No. (HCFA)03008 9-79 (Washington, DC: U.S. Department of Health, Education, and Welfare, Health Care Financing Administration, September 1979). 332. Myers, L.P., Eisenberg, J.M., and Pauly, M.V., "The Effects on Clinical Laboratory Services of Selected Alternatives for Paying Physicians Under the Medicare Program," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, September 1985. 333. Nalli, G.A., President, MD-IPA Health Plan, personal communication, May 15, 1985, and Nov. 26, 1985. 334. National Academy of Sciences, Institute of Medicine, "Medicare-Medicaid Reimbursement References • 263 Policies," in U.S. Congress, House Committee on Ways and Means, Subcommittee on Health, Social Security Studies Final Report, WMCP: 94-125, Mar. 1, 1976. 335. National Academy of Sciences, Institute of Medicine, Primary Care in Medicine: A Definition, interim report (Washington, DC: National Academy of Sciences, June 1977). 336. National Academy of Sciences, Institute of Medicine, A Manpower Policy for Primary Health Care (Washington, DC: National Academy of Sciences, May 1978). 337. National Academy of Sciences, Institute of Medicine, Reforming PhysiCian Payment: Report of a Conference (Washington, DC: National Academy Press, 1984). 338. National Academy of Sciences, Institute of Medicine, New Vaccine Development: Establishing Priorities, Volume I: Disease of Importance in the United States (Washington, DC: National Academy Press, 1985). 339. Neely, J., Equitable Life Assurance Society of the United States, New York, personal communication, Nov. 20, 1985. 340. Nelson, S., "National Evaluation of Medicare Competition Demonstrations: Preliminary Implementation Case Study of Central Massachusetts Health Care, Inc.," Mathematica Policy Research, Inc., Washington, DC, March 1985. 341. Nelson, S., "National Evaluation of Medicare Competition Demonstrations: Preliminary Implementation Case Study of South Florida Group Health Care, Inc.," Mathematica Policy Research, Inc., Washington, DC, May 1985. 342. Newhouse, J.P., "The Structure of Health Insurance and the Erosion of Competition in the Medical Marketplace," Competition in the Health Care Sector: Past, Present, and Future, W. Greenberg (ed.) (Rockville, MD: Aspen Systems Corp., 1978). 343. Newhouse, J.P., Manning, W.G., Morris, C., et al., "Some Interim Results From a Controlled Trial of Cost Sharing in Health Insurance," N. Engl.]. Med. 305(25):1501-7, Dec. 17, 1981. 344. Newhouse, J.P., Williams, A.P., Bennett, B.W., et al., "Where Have All the Doctors Gone?" ].A.M.A. 247(17):2392-2396, May 7, 1982. 345. Nielson, G., Health America, Inc., Nashville, TN, personal communication, Nov. 14, 1985. 346. Nimnicht, T., Metropolitan Life Insurance Co., St. Louis, MO, personal communication, Nov. 19, 1985. 347. Ogelsby, H., Assistant Vice President, Director of Operations Support, Blue Cross and Blue 264 • Payment for Physician Services: Strategies for Medicare Shield of California, San Fransisco, CA, personal communication, June 13, 1985. 348. O'Grady, K.F., Manning, W.G., Newhouse; J.P., et al., 'The Impact of Cost Sharing on Emergency Department Use," N. Engl.]. Med. 313(8):484-490, Aug. 22, 1985. 349. Orkand Corp., The FEC Factor, A Rapid Growth Health Care Alternative: The First Comprehensive National Study of Freestanding Emergency Centers (Silver Spri~g, MD: Orkand Corp., 1983). 350. Orkin, F.K., Committee on Manpower, American Society of Anesthesiologists, Washington, DC, personal communication, Nov. 15, 1984. 351. "Outpatient Labs Medicare Fee Schedules Increase 4.1%, In Line With Deficit Reduction Act," Technology Reimbursement Reports 1(24):2, July 12, 1985. 352. Owen, C., Bureau of Quality Control, Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Feb. 14, 1985. 353. Owens, A., "How Much of Your Money Comes From Third Parties?" Medical Economics 60: 254-263, Apr. 4, 1983. 354. Owens, A., "Doctors' Earnings: The Year of the Big Surprise," Medical Economics 62:195-il5, Sept. 9, 1985. 355. Owens, A., "Practice Costs: Can You Regain Control?" Medical Economics 62:222-242, Nov. 11, 1985. 356. Paden, R., Health Point Corp., Chattanooga, TN, personal communication, Nov. 21, 1985. 357. Paringer, L., "Medicare Assignment Rates of Physicians: Their Responses to Changes in Reimbursement Policy," Health Care Financing Review 1(3):75-89, Winter 1980. 358. Paringer, L., 'The Medicare Economic Index: Impact on Program Costs and Beneficiary Lia~ bility," Urban Institute Working Paper 1306-01~ 03, The Urban Institute, Washington, DC, June 1981. 359. Paringer, L., "The Distributional Effects of the Medicare Economic Index on Physicians and Beneficiaries," Urban Institute Working Paper 1306-01-03, The Urban Institute, Washington, DC, July 1981. 360. Paringer, L., "The Effect of the Medicare Economic Index on Reasonable Fees: Evidence From California," Urban Institute Working Paper 1306-01-04, The Urban Institute, Washington, DC, July 1981. 361. Paris, E., "Hold That Scalpel?" Forbes 135(10): 35-6, May 6, 1985. 362. Patashnik, B., Office of Reimbursem.ent Policy, Health Care Financing Administration, U.S. Department of Health and Human Services, Balti more, MD, personal communication, Oct. 30, 1985. 363. Patriarca, P.A., Schlech, W.F., Hinman, A.R., et al., "Pneumococcal Vaccination Practices Among Private Physicians," Pub. Health Rep. 97(5):406-408, September-October 1982. 364. Pauly, M.V., "Who Shall Be Paid?" presented at the Project Hope forum on Health Care Financing Administration Funded Research, Washington, DC, Jan. 16-17, 1985. 365. Pauly, M.V., 'The Economics of Moral Hazard: Comment," American Economic R. 58:531-537, June 1968. 366. Pauly, M.V., "Is Medical Care Different?" Competition in the Health Care Sector: Past, Present and Future, W. Greenberg (ed.) (Washington, DC: U.S. Federal Trade Commission, Bureau of Economics, May 1978). 367. Pauly, M.V., The Role of the Private Sector in National Health Insurance (Washington, DC: Health Insurance Institute, 1979). 368. Pauly, M.V., and Langwell, K.M., "Research on Competition in the Financing and Delivery of Health Services: Future Research Needs," in U.S. Department of Health and Human Services, Public Health Service, National Center for Health Services Research, Research on Competitian in the Financing and Delivery of Health Services: Future Research Needs, L.F. Rossiter (ed.), DHHS Pub. No. (PHS) 83-3328-2 (Washington, DC: U.S. Government Printing Office, October 1982). 369. Payne, B.C., Lyons, T.F., and Neuhaus, E., "Relationship of Physician Characteristics to Performance Quality and Improvement," Health Serv. Res. 19(3):307-32, August 1984. 370. Payne, B.C., Lyons, T.F., et al., The Quality of Medical Care: Evaluation and Improvement (Chicago, IL: Hospital Research and Educational Trust, 1976). 371. Pennsylvania Blue Shield, "Medicare B Assignment for Pennsylvania Blue Shield," HCFA-820676 interim report, Camp Hill, P A, July 22, 1982. 372. Pennsylvania Blue Shield, A Study of the Physicians' Services Market in Pennsylvania (Camp Hill, P A: Pennsylania Blue Shield, January 1983). 373. Pfuetze, K.D., " 'Unnecessary' Surgery: Scientifically Erroneous Testimony in Moss Subcommittee Hearings," New York State Journal of Medicine 76(13):2198-2205, December 1976. 374. "Physician Contracting Activities, 1983," Sociaeconomic Report 24(3):1-8, April 1984. 375. "Physicians, Involvement With Medicare," SMS Reports 3(3), May 1984. 376. Pine, P.L., Gornick, M., Lubitz J, et al., "An Analysis of Services Received Under Medicare by Specialty of Physician," Health Care Financ ing Review 3(1):89-116, September 1981. 377. Pollard, M.R., "Antitrust and Physician Payment," North Carolina Medical journal 45(4): 227-231, April 1984. 378. Polzer, E., Dornier Medical Systems, Marietta, GA, personal communication, October 1985. 379. Power, E.J., Effects of Federal Policies on Extracorporeal Shock Wave Lithotripsy, prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, forthcoming. 380. Powills, S., "U.S. Health Care Coalitions Up 14 Percent: Clearinghouse," Hospitals 59(10):40, June 16, 1985. 381. Powills, S., "Labor: A Growing Force in Controlling Health Care Costs," Hospitals 59(21):8285, Nov. 16, 1985. 382. "PPO Enrollment Jumps Dramatically in 1985: Survey," Hospital Week 21(40):1, October 1985. 383. "PPO's Face Antitrust Problems/' Washington Report on Medicine and Health 38( 44): (Perspectives), Nov. 5, 1984. 384. "Profit in Medicine: The Impact on Hospital Staffing Patterns and Physician Practice Behavior," SMS Reports 3(4), June 1984. 385. Prospective Payment Assessment Commission, Technical Appendixes to the Report and Recommendations to the Secretary, U.S. Department of Health and Human Services, vol. 2 (Washington, DC: U.S. Government Printing Office, Apr. 1, 1985). 386. "Recommendation of the Immunization Practices Advisory Committee (ACPI): Update: Pneumococcal Polysaccharide Vaccine Use in the United States," Morbidity and Mortality Weekly Reports 33(20):273-281, May 25, 1984. 387. Reinhardt, U.E., "Health Insurance and Health Policy in the Federal Republic of Germany," Health Care Financing Review 3(2):1-14, December 1981. 388. Reinhardt, U.E. (ed.), The Compensation of Physicians: Approaches Used in Other Countries (Princeton, NJ: Princeton University, 1983). 389. Reinhardt, U.E., "The Theory of PhysicianInduced Demand: Reflections After a Decade," ]. Health Econ. 4(2):187-194, June 1985. 390. Reynolds, R.A., and Abram, J.B. (eds.), Socioeconomic Characteristics of Medical Practice 1983 (Chicago, IL: American Medical Association, 1983). 391. Reynolds, R.A., and Ohsfeldt, R.L. (eds.), Socioeconomic Characteristics of Medical Practice 1984 (Chicago, IL: American Medical Association, 1984). 392. Rhee, S., Luke, R.D., Lyons, T.F., et al., "Do- References • 265 main of Practice and the Quality of Physician Performance," Medical Care 19(1):14-23, January 1981. 393. Rice, T., "Determinants of Physician Assignment Rates by Type of Service," Health Care Financing Review 5(4):33-42, Summer 1984. 394. Rice, T., and McCall, N., "Changes in Medicare Reimbursement in Colorado: Impact on Physicians'· Economic Behavior," Health Care Financing Review 3(4):67-86, June 1982. 395. Rice, T., and McCall, N., "Factors Influencing Physician Assignment Decisions Under Medicare," Inquiry 20(1):45-56, Spring 1983. 396. Richman, D., "U.S. Health Care Aims at Growth With Push Into New York, Dallas," Modern Health Care 15(20):50-53, Sept. 27, 1985. 397. Riddiough, M.A., "Implications of Alternative Medicare Payment Methods for Pneumococcal Vaccination," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, October 1985. 398. Riedel, R. L., and Riedel, D. C., Practice and Performance: An Assessment of Ambulatory Care (Ann Arbor, Ml: Health Administration Press, 1979). 399. Ro, K.K., and Auster, R., "An Output Approach to Incentive Reimbursement for Hospitals," Health Serv. Res. 4(3):177-187, Fall1969. 400. Robinson, J., Finlay Health Plan, Miami, FL, personal communication, Dec. 4, 1985. 401. Rodgers, J., Congressional Budget Office, U.S. Congress, Washington, DC, personal communication, July 26, 1985, and Sept.· 20, 1985. 402. Rodgers, J.F., and Musacchio, R.A., "Physician Acceptance of Medicare Patients on Assignment,"]. Health Econ. 2:55-73, March 1983. 403. Roe, B.B., "The UCR Boondoggle: A Death Knell for Private Practice?" N. Engl. ]. Med. 305(1):41-45, July 2, 1981. 404. Roemer, M.l., and Shonick, W., "HMO Performance: The Recent Evidence," Milbank Mem. Fund Q. 51(3):271-317, Summer 1973. 405. Romeo, A.A., The Hemodialysis Equipment and Disposables Industry, Health Technology Case Study #32, prepared for the Office of Technology Assessment, U.S. Congress, OTA-HCS32 (Washington, DC: U.S. Government Printing Office, December 1984). 406. Rosen, M., Sanus Corp., New York, personal communication, Nov. 15, 1985. 406a. Rosenbach, M.L., Hurdle, S., and Cromwell, J., An Analysis ofMedicare's Physician Participating Agreement Program (Chestnut Hill, MA: Health Economics Research Center, Oct. 29, 1985). 407. Rossiter, L.F., and Byrd, K., "National Medi 266 • Payment for Physician Services: Strategies for Medicare care Competition Evaluation: Implementation Case Study Report for: Group Health Plan of Southeast Michigan," Medical College of Virginia, Richmond, VA, September 1984. 422. 408. Rossiter, L.F., and Byrd, K., "National Medicare Competition Evaluation: Implementation Case Study Report for: Metropolitan Health Council of Indianapolis, Inc.," Medical College 423. of Virginia, Richmond, VA, Aug. 31, 1984. 409. Rossiter, L.F., and Wood, G.C., "National Medicare Competition Evaluation: Implementation 424. Case Study Report for: Choice Care, Cincinnati, Ohio," Medical College of Virginia, Richmond, VA, October 1984. 410. Rossiter, L.F., and Wood, G.C., 'National Medicare Competition Evaluation: Implementation 425. Case Study Report for: Medical East Community Health Plan, Braintree, Massachusetts," Medical College of Virginia, Richmond, VA, 426. February 1985. 411. Rossiter, L.F., Friedlob, A., and Langwell, K., "Exploring Benefits for Risk-Based Contracting Under Medicare," f. Health Care Financial Man427. agement Association, pp. 1-11, May 1985. 412. Roth, M., "Hospitals Seek Kidney Stone Machine," The Fairfax foumal47(120):A1,A6, June 20, 1985. 428. 413. Roth, M., "State Vetos Kidney Stone Machine," The Fairfax Journal 47(171):A1,A4, Sept. 3, 1985. 429. 414. Ruby, G., Banta, H.D., and Burns, A.K., "Medicare Coverage, Medicare Costs, and Medical Technology," f. Health Politics, Policy, and Law 10(1):141-155, Spring 1985. 415. "Rural Medicare Patients Pay More Than Urbanites," American Medical News 24(4):14, Sept. 4, 1981. 430. 416. Sanzaro, P.J., and Worth, R.M., "Measuring Clinical Performance of Individual Internists in Office and Hospital Practice," Medical Care 23(9):1097-1111, September 1985. 431. 417. Sapolsky, H.M., Altman, D., Greene, R., et al., "Corporate Attitudes Toward Health Care Costs," Milbank Mem. Fund Q. 59(1):561-85, Winter 1981. 418. "Savings Estimated for Medicare Second Surgical Opinion," Washington Health Costs Letter 432. 11(15):6, July 19, 1985. 419. Scheffler, R.M., Ruby, G., Paringer, L., et al., 433. 'The Specialty and Spatial Distribution of Physicians: A Review of the Evidence," Health Policy and Education 1:271-289, 1980. 420. Scheffler, R.M., Weisfeld, N., Ruby, G., et al., "A Manpower Policy for Primary Health Care," N. Engl. f. Med. 298(19):1058-62, May 11, 1978. 421. Schieber, G., Burney, I.L., Gorden, J.B., et al., 434. "Physician Fee Patterns Under Medicare: A Descriptive Analysis," N. Engl. f. Med. 294: 1089-1093, May 13, 1976. Schoonhoven, C.B., Scott, W.R., Flood, A.B., et al., "Measuring the Complexity and Uncertainty of Surgery and Postsurgical Care," Medical Care 18(9):893-915, September 1980. Schroeder, S.A., School of Medicine, University of California, San Francisco, CA, personal communication, Oct. 16, 1985. Schroeder, S.A., and Showstack, J.A., "Financial Incentives To Perform Medical Procedures and Laboratory Tests: Illustrative Models of Office Practice," Medical Care 16(4):289-298, April 1978. Schroer, K.A., and Taylor, E., "PPAs: Fewer Start-Ups But Better Operations," Hospitals 59(17):68-73, Sept. 1, 1985. Schumacher, B., Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Nov. 19, 1985. Schwartz, W.B., Newhouse, J.P., Bennett, B.W., et al., 'The Changing Geographic Distribution of Board-Certified Physicians," N. Engl. f. Med. 303(18):1032-8, Oct. 30, 1980. Scott, M., American Society of Anesthesiolo gists, Washington, DC, personal communica tion, Jan. 22, 1985. Shaffert, T., Mitchell, S., and Scalenker, R., "Study of Reimbursement and Practice Arrange ments of Provider-Based Physicians," Physicians and Financial Incentives, HCFA Pub. No. 03607 (Baltimore, MD: U.S. Department of Health and Human Services, Health Care Financing Admin istration, December 1980). Showstack, J.A., Blumberg, B.D., Schwartz, J., et al., "Fee-for-Service Physician Payment: Analysis of Current Methods and Their Devel opment," Inquiry 16(3):230-246, Fall1979. Showstack, J.A., Perez-Stable, E.J., and Sawitz, E., "Extracorporeal Shock Wave Lithotripsy: Clinical Application and Medicare Physician Payment," prepared for the Office of Technol ogy Assessment, U.S. Congress, Washington, DC, Aug. 1, 1985. Shulman, S., CIGNA Health Plans, Dallas, TX, personal communication, Nov. 7, 1985. Shuttinga, J.A., Falik, M., and Steinwald, B., "Health Plan Selection in the Federal Employ ees Health Benefits Program," Office of Health Policy, Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Hu man Services, Washington, DC, September 1984. Sinclair, M., "Blue Cross Makes Bid for Md. Medicare," Washington Post, p. A1, Aug. 3, 1985. 435. Sirmon, R., Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, personal communication, Dec. 2, 1985. 436. Sisk, J.E., "Effects of Increased Competition in Health Care on the Use and Innovation of Medical Technology," Health Care Management Review 9(3):21-34, Summer 1984. 437. Sisk, J.E., and Riegelman, R.K., "Cost Effectiveness of Vaccination Against Pneumococcal Pneumonia," Annals of Internal Medicine, January 1986. 438. Sloan, F.A., "Patient Care Reimbursement: Implications for Medical Education and Physician Distribution," Medical Education Financing, J. Hadley (ed.) (New York: PRODIST, 1980). 439. Sloan, F.A., "Physicians and Blue Shield: A Study of Effects of Physician Control on Blue Shield Reimbursements," Office of Research, Demonstrations, and Statistics, Health Care Financil1g Administration, U.S. Department of Health and Human Services, Health Care Financing Proceedings: Issues in Physician Reimbursement, N.T. Greenspan (ed.), HCFA Pub. No. 03121 (Balt,imore, MD: HCFA, August 1981). 440. Sloan, F.A., and Steinwald, B., "Physician Participation in Health Insurance Plans: Evidence on Blue Shield," Journal on Human Resources 13:237-263, Spring 1978. 441. Smith, S., National Institutes of Health, Public Health Service, U.S. Department of-Health and Human Services, Bethesda, MD, personal communication, October 1985. 442. Sobaski, W.J., Effects of the 1969 California Relative Value Studies on Costs of Physician Services Under SMI, Office of Research and Statistics, U.S. Department of Health, Education, and Welfare, DHEW Pub. No.(SSA) 75-11702, H1-69 (Washington, DC: U.S. Department of Health, Education, and Welfare, June 20, 1975). 443. Somers, C., John Hancock Insurance, Boston, MA, personal communication, Nov. 11, 1985. 444. Somers, H.M., and Somers, A.R., Doctors, Patients, and Health Insurance (Washington, DC: Brookings Institution, 1961). 445. Somers, H.M., and Somers, A.R., Medicare and the Hospitals: Issues and Prospects (Washington, DC: Brookings Institution, 1967). 446. Stano, M., Velky, ]., and Saad, A., "Medicare Assignment Rates in Michigan," Blue Shield of Michigan, Detroit, MI, July 1982. 447. Stason, W.B., and Localio, A.R., "Magnetic Resonance Imaging: Clinical Efficacy, Costs, References • 267 and Policy Consideration," Blue Cross and Blue Shield Association, Chicago, IL, February 1985. 448. Stein, J., "Industry's New Bottom Line on Health Care Costs: Is Less Better?" Hastings Center Re port 15(5):14-18, October 1985. 449. Steinberg, E.P., Sisk, J.E., and Locke, K.E., "X Ray CT and Magnetic Resonance Imaging: Dif fusion Patterns and Policy Issues," N. Engl. ]. Med. 313(14):859-864, Oct. 3, 1985. 450. Steinwald, B., "Hospital-Based Physicians: Cur rent Issues and Descriptive Evidence," Health ·care Financing Review 1(3):63-75, Summer 1980. 451. Steinwald, B., "Compensation of Hospital Based Physicians," Health Serv. Res. 18(1):17 43, Spring 1983. 452. Sulvetta, M., and Scanlon, B., "Tabulated Dis tribution of Group Practices and Solo Physicians by Medicare Assignment Rates," letter to the Of fice of Research and Demonstrations, Health Care Financing Administration, U.S. Depart ment of Health and Human Services, Washing ton, DC, June 11, 1982. 453. Tatge, M., "HMO Enrollment Up 26.7% to 1.68 Million," Modern Health Care 15(4):138-141, June 7, 1985. 454. Tell, E.]., Falik, M., and Fox, P.O., "Private Sector Health Care Initiatives: A Comparative Perspective From Four Communities," Milbank Mem. Fund Q. 62:357-379, Fall 1984. 455. The Urban Institute, "Aspects of Physician Be havior, Service Delivery, and Payment Meth ods in Medicare and Medicaid: Final Report," Washington, DC, undated. 456. Thomas, J.W., Lichtenstein, R., Wyszewianski, L., et al., "Increasing Medicare Enrollment in HMOs: The Need for Capitation Rates Adjusted for Health Status," Inquiry 20(3):227-239, Fall 1983. 457. Thompson, B., 'The Merits of Tracking Pro vider Performance," Business and Health 2(7): 30-31, June 1985. 458. Thoms, E.R., Vice President, Medicare Claims, Pennsylvania Blue Shield, Camp Hill, P A, per sonal communication, Dec. 6, 1984. 459. Tompkins, C., and Gruenberg, L., "A Summary of Major Points Made at the Consultants Meet ing on the AAPCC," Health Policy Center, Brandeis University, Waltham, MA, June 1985. 460. Trainer, ].B., Institute for Health Policy Studies, School of Medicine, University of California, San Francisco, CA, personal communication, July 1985. 461. Trapnell, G.R., McKusick, D.R., Genuardi, J.S., "An Evaluation of the Adjusted Average Per Capita Cost (AAPCC) Used in Reimburs 268 • Payment for Physician Services: Strategies for Medicare ing Risk-Basis HMO's Under Medicare," pre474. pared for the Health Care Financing Adminis tration, U.S. Department of Health and Human Services, Baltimore, MD, April 1982. 462. Trauner, J.B., Preferred Provider Organizations: The California Experiment (Sap Francisco, 475. CA: Institute for Health Policy Studies, University of California, San Francisco, August 1983). 462a. Trauner, J.B., Institute for Health Policy Studies, University of California, San Francisco, CA, personal communication, July 1985. 463. 'Trends in Patient Visits With Physicians: 19751983," SMS Reports 3(7), October 1984. 476. 464. Tresnowski, B.R., "HMO's Are Rolling-And We Can All Benefit," Inquiry 21(3):203-204, Fall 1984. 465. Trieger, S., Office of Demonstrations and Evaluation, Health Care Financing Administration, U.S. Department of Health and Human Serv477. ices, Baltimore, MD, personal communication, Nov. 25, 1985. 466. Tyler, S., National Medical Enterprises, Beverly Hills, CA, personal communication, Nov. 25, 1985. 467. U.S. Congress, Congressional Budget Office, 478. Changing the Structure ofMedicare Benefits: Issues and Options (Washington, DC: U.S. Government Printing Office, March 1983). 468. U.S. Congress, Congressional Budget Office, unpublished estimates based on Health Care Financing Administration, U.S. Department of 479. Health and Human Services, Medicare Directory of Prevailing Charges 1984, Washington, DC, 1985. 469. U.S. Congress, Congressional Budget Office, "Physician Reimbursement Under Medicare: 480. Options for Change," Washington, DC, forthcoming. 470. U.S. Congress, Congressional Budget Office, The Economic and Budget Outlook: An Update (Washington, DC: U.S. Government Printing 481. Office, August 1985). 471. U.S. Congress, Congressional Research Service, "How Medicare Pays Doctors," prepared by J. O'Sullivan, Report No. 84-41 EPW, Washington, DC, Mar. 8, 1984. 482. 472. U.S. Congress, Congressional Research Service, "Medicare," prepared by J. O'Sullivan, Report No. IB84052, Washington, DC, Oct. 15, 1984. 473. U.S. Congress, Congressional Research Service, 483. Medicare: Physician Payments, prepared by J. O'Sullivan, CRS Order Code IB85007 (Washington, DC: U.S. Government Printing Office, Dec. 17, 1984). U.S. Congress, General Accounting Office, Improving Medicare and Medicaid Systems to Control Payments for Unnecessary Physicians' Services GAO/HRD-83-16 (Washington, DC: U.S. Government Printing Office, Feb. 8, 1983). U.S. Congress, General Accounting Office, Reimbursing Physicians Under Medicare on the Basis of Their Specialty, report to the Health Care Financing Administration, U.S. Department of Health and Human Services, GAO/HRD84-94 (Washington, DC: U.S. Government Printing Office, Sept. 27, 1984). U.S. Congress, General Accounting Office, Problems in Administering Medicare's Health Maintenance Organization Demonstration Projects in Florida, GAO/HRD 85-48 (Washington, DC: U.S. Government Printing Office, Mar. 8, 1985). U.S. Congress, House of Representatives, Health Maintenance Organizations and the Elderly: Promises, Problems, and Prospects, Hearing before the Select Committee on Aging, Comm. Pub. No. 98-469 (Washington, DC: U.S. Government Printing Office, 1984). U.S. Congress, House of Representatives, Committee on Ways and Means, "Social Security Amendments of 1965," Report No. 213, 89th Congress, 1st Session on H.R. 6675 (Washington, DC: U.S. Government Printing Office, 1965). U.S. Congress, House of Representatives, Select Committee on Aging, Subcommittee on Health and Long-Term Care, Cataract Surgery: Fraud, Waste, and Abuse (Washington, DC: U.S. Government Printing Office, July 19, 1985). U.S. Congress, Office of Technology Assessment, Policy Implications of the Computed Tomography (CT) Scanner, OTA-H-56 (Washington, DC: U.S. Government Printing Office, August 1978). U.S. Congress, Office of Technology Assessment, Assessing the Efficacy and Safety ofMedical Technologies, OTA-H-75 (Washington, DC: U.S. Government Printing Office, September 1978). U.S. Congress, Office of Technology Assessment, Strategies for Medical Technology Assessment, OTA-H-181 (Washington, DC: U.S. Government Printing Office, September 1982). U.S. Congress, Office of Technology Assessment, Medical Technology Under Proposals To Increase Competition in Health Care, OTA-H190 (Washington, DC: U.S. Government Printing Office, October 1982). 484. U.S. Congress, Office of Technology Assessment, Commercial Biotechnology: An International Analysis, OTA-BA-218 (Washington, DC: U.S. Government Printing Office, January 1984). 485. U.S. Congress, Office of Technology Assessment, Update ofFederal Activities Regarding the Use of Pneumococcal Vaccine, OTA-TM-H-23 (Washington, DC: U.S. Government Printing Office, May 1984). 486. U.S. Congress, Office of Technology Assessment, Medical Technology and Costs of the Medicare Program, OTA-H-227 (Washington, DC: U.S. Government Printing Office, July 1984). 487. U.S. Congress, Office of Technology Assessment, Federal Policies and the Medical Devices Industry, OTA-H-229 (Washington, DC: U.S. Government Printing Office, October 1984). 487a. U.S. Congress, Office of Technology Assessment, Technology and Aging in America, OTABA-264 (Washington, DC: U.S. Government Printing Office, June 1985). 488. U.S. Congress, Office of Technology Assessment, Workshop on Administrative Feasibility of Proposed Changes in Medicare Physician Payment, project on "Payment for Physician Services: Strategies for Medicare," Washington, DC, June 13, 1985. 489. U.S. Congress, Office of Technology Assessment, Medicare's Prospective Payment System: Strategies for Evaluating Cost, Quality, and Medical Technology, OTA-H-262 (Washington, DC: U.S. Government Printing Office, October 1985). 490. U.S. Congress, Office of Technology Assessment, Nurse Practitioners, Physician Assistants, and Certified Nurse Midwives: Quality, Access, Cost, and Payment Issues, forthcoming. 491. U.S. Congress, Senate, 89th Congress, 1st Session, reprinted in 1965 U.S. Code Congression~J and Administrative News, 1943-2227 (St. Paul; MN: West Publishing Co., 1966). 492. U.S. Congress, Senate, Committee on Finance, Social Security Amendments of 1972, Senate Report 92239 to accompany H.R.l (Washington, DC: U.S. Government Printing Office, Sept. 26, 1972). 493. U.S. Congress, Senate, Committee on Finance, The Social Security Act and Related Laws (Washington, DC: U.S. Government Printing Office, April1982). 494. U.S. Congress, Senate, Committee on Finance, and House of Representatives, Committee on Ways and Means, Committee on Energy and References • 269 Commerce, Background Data on Physician Reimbursement Under Medicare, S. Prt. 98-106 (Washington, DC: U.S. Government Printing Office, October 1983). 495. U.S. Congress, Senate, Special Committee on Aging, Medicare: Physician Payment Options, Hearings, Mar. 16, 1984 (Washington, DC: U.S. Government Printing Office, 1984). 496. U.S. Congress, Senate, Special Committee on Aging, Medicare: Paying the Physician-History, Issues, and Options, prepared by L. Etheredge, S. Prt. 98-153 (Washington, DC: U.S. Government Printing Office, March 1984). 497. U.S. Congress, Senate, Special Committee on Aging, Medicare and the Health Costs of Older Americans: The Extent and Effects of Cost Sharing, prepared by A. Langley, S. Prt. 98-166 (Washington, DC: U.S. Government Printing Office, April 1984). 498. U.S. Congress, Senate, Special Committee on Aging, and American Association of Retired Persons, Aging America: Trends and Projections (Washington, DC: U.S. Government Printing Office, 1984). 499. U.S. Department of Commerce, Bureau of the Census, 1977 Census of Service Industries: Health Services (Washington, DC: U.S. Government Printing Office, 1981). 500. U.S. Department of Commerce, Bureau of the Census, Statistical Abstract of the United States, 1984, 104th ed. (Washington, DC: U.S. Government Printing Office, 1983). 501. U.S. Department of Commerce, Bureau of the Census, Decennial Censuses of Population, 1900-1980 and Projection of the Population of the United States: 1982 to 2050 (Advance Report), Current Population Reports, Series P-25, No. 922, October 1982; as cited in Aging America: Trends and Projections, prepared by the U.S. Senate Special Committee on Aging, in conjunction with the American Association of Retired Persons (Washington, DC: American Association of Retired Persons, 1984). 502. U.S. Department of Commerce, Bureau of the Census, Estimates of the Population of the United States, by Age, Sex, and Race: 1980 to 1983, Current Population Reports, Series P-25, No. 949 (Washington, DC: U.S. Government Printing Office, 1984). 503. U.S. Department of Commerce, Bureau of the Census, Demographic and Socioeconomic Aspects of Aging in the United States, Current Population Reports, Series P-23, No. 138 (Washington, DC: U.S. Government Printing Office, August 1984). 270 • Payment for Physician Services: Strategies for Medicare 504. U.S. Department of Defense, Office of the Civilian Health and Medical Program of the Uniformed Services, Program Planning and Policy Division, Individual Provider Reasonable Charges Determined Under CHAMPUS: Alternatives to the Present Methodology (Washington, DC, Feb. 8, 1980). 505. U.S. Department of Health, Education, and Welfare, Public Health Service, Bureau of Health Manpower, A Report to the President and Congress on the Status of Health Personnel, DHEW Pub. No. (HRA) 79-93 (Washington, DC: DHEW, August 1978 and March 1979). 506. U.S. Department of Health, Education, and Welfare, Social Security Administration, Office of Research and Statistics, Assignment Rates for Supplementary Medical Insurance Claims, Calendar Years 1970-72, DHEW Pub. No. (SSA) 78-11702, Hl-46, June 30, 1973 (Washington, DC: U.S. Government Printing Office, 1973). 507. U.S. Department of Health and Human Services, HHS News, Washington, DC, July 31, 1985. 508. U.S. Department of Health and Human Services, Health Care Financing Administration, HCFA Administrators Report, No. 17, Washington, DC, Aug. 27, 1979. 509. U.S. Department of Health and Human Services, Health Care Financing Administration, Determination of Reasonable Charges Under Part B of Medicare: A Basic Text, HCFA Pub. No. HCFA-20036 (2-80) (Washington, DC: U.S. Government Printing Office, 1980). 510. U.S. Department of Health and Human Services, Health Care Financing Administration, ICD-9-CM (Washington, DC: U.S. Government Printing Office, September 1980). 511. U.S. Department of Health and Human Services, Health Care Financing Administration, "Medicare Part B Charges, Overview and Trends, Fee Screen Years, 1976-1980," Feb. 3, 1982. 512. U.S. Department of Health and Human Services, Public Health Service, Health Resources and Services Administration, Diffusion and the Changing Geographic Distribution of Primary Care Physicians, DHHS Pub. No. HRS-P-OD84-1 (Washington, DC: HRSA, November 1983). 513. U.S. Department of Health and Human Services, Health Care Financing Administration, Medicare Directory ofPrevailing Charges, 1983, HCFA-10015 (Baltimore, MD: HCFA, November 1983). 514. U.S. Department of Health and Human Services, Health Care Financing Administration, Medicare Carrier Manual, 1984. 515. U.S. Department of Health and Human Services, Health Care Financing Administration, Prevailing Charge Directory (Baltimore, MD: HCFA, 1984). 516. U.S. Department of Health and Human Services, Health Care Financing Administration, "Medicare Participating Physician and Supplier Program: Fact Sheet," Baltimore, MD, 1984. 517. U.S. Department of Health and Human Services, Health Care Financing Administration, "Geographic Differentials in Medicare Reimbursement for Physician Services," Washington, DC, 1984. 518. U.S. Department of Health and Human Serv-· ices, Health Care Financing Administration, "Medicare Participating Physician and Supplier Program: Fact Sheet," Baltimore, MD, 1984. 519. U.S. Department of Health and Human Services, Health Care Financing Administration as cited in A. Langley, Medicare and the Health Costs of Older Americans: The Extent and Effects of Cost Sharing, prepared for the Special Committee on Aging, U.S. Senate (Washington, DC: U.S. Government Printing Office, April 1984). 520. U.S. Department of Health and Human Services, Health Care Financing Administration, "Medicare Participating Physician and Supplier Program: Fiscal Year 1986," Baltimore, MD, 1985. 521. U.S. Department of Health and Human Services, Health Care Financing Administration, "FY 86 Medicare Participating Physician and Supplier Program: Fact Sheet," December 1985. 521a. U.S. Department of Health and Human Services, Health Care Financing Administration, Staff, Baltimore, MD, personal communication, August 1985, and Dec. 12, 1985. 522. U.S. Department of Health and Human Services, Health Care Financing Administration, Medicare Carrier Manual, section 13512, no date. 523. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Data Management and Strategy, Medicare Program Statistics, 1981 (Washington, DC: U.S. Government Printing Office, 1983). 524. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Data Management and Strategy, Medi care Statistical Files Manual (Baltimore, MD: HCFA, September 1983). 525. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Data Management and Strategy, Annual Medicare Program Statistics 1982 (Baltimore, MD: HCFA, December 1984). 526. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Data Management and Strategy, Office of Financial and Actuarial Analysis, Division of National Cost Estimates, unpublished data, 1985. 527. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Data Management and Strategy, Division of Reports and Analysis, Compiled Carrier Workload Reports, 1985. 528. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Eligibility Reimbursement and Coverage, Office of Reimbursement Policy, Staff, Baltimore, MD, personal communication, August 1985. 529. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Program Operations, Carrier Reasonable Charge and Denial Activity Report, julySeptember 1982 (Washington, DC: U.S. Government Printing Office, January 1983). 530. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Program Operations, Part B Carrier Workload Report March 1983 (Washington, DC: U.S. Government Printing Office, May 1983). 531. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Program Operations, Part B Carrier Reasonable Charge and Denial Activity Report: Fiscal Year 1982 (Washington, DC: U.S. Government Printing Office, August 1983). 532. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Program Operations, unpublished computer tabulations from the Medicare Directory of Prevailing Charges, fee screen years 1981, 1982, 1983, and 1984, W. Merashoff, personal communication, June 19, 1985. 533. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Program Operations, Division of Group Health Plan Operations, Baltimore, MD, personal communication, Aug. 27, Aug. 30, Nov. 8, and Dec. 13, 1985. 534. U.S. Department of Health and Human Serv- References • 271 ices, Health Care Financing Administration, Bureau of Quality Control, Division of Reports and Analysis, Part B Reasonable Charge and Denial Activity Report: january-March 1982 (Baltimore, MD: HCFA, 1982). 534a. U.S. Department of Health and Human Services, Health Cdre Financing Administration, Bureau of Quality Control, Division of Reports and Analysis, Baltimore, MD, personal communication, Dec. 27, 1985. 535. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Quality Control, Carrier Reasonable Charge and Denial Activity Report, JanuaryMarch 1985 (Washington, DC: U.S. Government Printing Office, 1985). 536. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Quality Control, Carrier Reasonable Charge and Denial Activity Report, OctoberDecember 1984 (Washington, DC: U.S. Government Printing Office, 1985). 537. U.S. Department of Health and Human Services, Health Care Financing Administration, Bureau of Quality Control, Report on Medicare Participating Physician/Supplier Claim Workloads: January-March 1985 (Baltimore, MD: HCFA, May 1985). 538. U.S. Department of Health and Human Services, Health Care Financing Administration, Office of Beneficiary Services for Medicare and Medicaid Beneficiaries, "Medicare Medicaid Notes," January 1984. 539. U.S. Department of Health and Human'Services, Health Care Financing Administration, Office of Research and Demonstrations, Health Care Financing Status Report: Research and Demonstrations in Health Care Financing, HCFA Pub. No. 03202 (Baltimore, MD: HCFA, April 1985). 540. U.S. Department of Health and Human Services, Health Care Financing Administration, Office of Research and Demonstrations, Research, Demonstration, and Evaluation Spending Plan (Baltimore, MD: HCFA, Apr.18, 1985). 541. U.S. Department of Health and Human Services, Health Care Financing Administration, Office of Research and Demonstrations, "National Medical Competition Evaluation: An Evaluation of the Quality of the Process of Care," Contract 500-83-0047, Mod. No.6, Exhibit A-1, June 1, 1985. 542. U.S. Department of Health and Human Services, Office of the Inspector General, Baltimore, MD, Aug. 13, 1985. 543. U.S. Department of Health and Human Serv 272 • Payment for Physician Services: Strategies for Medicare ices, Office of the Inspector General, Staff, Bal timore, MD, personal communication, Aug. 20, 1985. 544. U.S. Department of Health and Human Services, Public Health Service, Bureau of Health Professions, Report to the President and Congress on the Status of Health Personnel in the United States, Volumes I and II, 1984, DHHS Pub. No. HRS-P-OD 84-4 (5/85) (Washington, DC: U.S. Government Printing Office, May 1984). 545. U.S. Department of Health and Human Services, Public Health Service, Food and Drug Administration, Office of Biologics, "Summary for Basis of Approval," Bethesda, MD, June 30, 1983, and July 15, 1983. 546. U.S. Department of Health and Human Services, Public Health Service, Health Resources and Services Administration, Diffusion and the Changing Geographic Distribution of Primary Care Physicians, HRS-P-OD-84-1, HRP0904702 (Washington, DC: U.S. Government Printing Office, June 1983; revised November 1983). 547. U.S. Department of Health and Human Services, Public Health Service, Health Resources and Services Administration, Report to the President and Congress on the Status of Health Personnel in the United States, DHHS Pub. No. HRS-P-OD 84-4 (Washington, DC: HRSA, May 1984). 548. U.S. Department of Health and Human Services, Public Health Service, National Center for Health Statistics, Characteristics of Visits to Female and Male Physicians, DHHS Pub. No. (PHS) 80-1710 (Hyattsville, MD: U.S. Government Printing Office, June 1980). 549. U.S. Department of Health and Human Services, Public Health Service, National Center for Health Statistics, unpublished data from the National Hospital Survey supplied by E. Graves, Hyattsville, MD, Nov. 21, 1985. 550. U.S. Department of Health and Human Services, Public Health Service, National Center for Health Statistics, Health, United States, 1983 DHHS Pub. No. (PHS) 84-1232 (Washington, DC: U.S Government Printing Office, December 1983). 550a. U.S. Department of Health and Human Services, Social Security Administration, Social Security Bulletin, Annual Statistical Supplement, 1982, table 145 (Washington, DC: 1984). 551. U.S. Department of Health and Human Services, Social Security Administration, Office of the Actuary, September 1982, as cited in U.S. Senate Special Committee on Aging, and American Association of Retired Persons Aging America: Trends and Projections (Washington, DC: American Association of Retired Persons, 1984). 552. U.S. Executive Office of the President, Office of Management and Budget, Budget of the United States Government: Fiscal Year 1986 (Washington, DC: U.S. Government Printing Office, 1985). 553. U.S. Federal Supplementary Medical Insurance Trust Fund, Board of Trustees, "1985 Annual Report of the Board of Trustees of the Federal Supplementary Medical Insurance Trust Fund," Washington, DC, Mar. 28, 1985. 554. U.S. Federal Trade Commission, Letter to Mr. Rickard F. Pfizenmayer, Esq., Apr. 19, 1985. 555. U.S. Federal Trade Commission, Letter to Mr. William G. Kopit, Esq., Apr. 19, 1985. 556. U.S. Federal Trade Commission Staff, Washington, DC, personal communication, Dec. 11, 1984. 557. Ullman, R., Equitable Insurance Society, New York, personal communication, Nov. 19, 1985. 558. Van Ellet, T., Medigap: State Responses to Problems With Health Insurance for the Elderly (Washington, DC: Intergovernmental Health Policy Project, George Washington University, Oct. 30, 1979). 558a. Vertrees, J., La Jolla Management Corp., Rockville, MD, personal communication, Jan. 3, 1986. 559. Vertrees, J., Tolley, J., Manton, K., "Issues in Capitation: Risk of Financial Ruin for Providers and Ways to Control This Risk," prepared for the Office of Technology Assessment, U.S. Congress, Washington, DC, Oct. 31, 1985. 560. Wagner, D.P., Draper, E., "APACHE II and Medicare Reimbursement," Health Care Financing Review, annual supplement, pp. 91-105, November 1984. 561. Wagner,]., "The Micro-Costing Approach," in J. Hadley, D. Juba, K. Swartz, et al., "Alternative Methods of Developing a Relative Value Scale of Physicians' Services: Year 1 Report," The Urban Institute, Washington, DC, February 1983. 562. Walden, D.C., "Paying Several Physicians for Treating an Episode of Illness," presented at the Annual Meeting of the American Public Health Association, Anaheim, CA, Nov. 13, 1984. 563. Waldo, D.R., and Lazenby, H.C., "Demographic Characteristics and Health Care Use and Expenditures by the Aged in the United States: 1977-84," Health Care Financing Review 6(1):129, Fall 1984. 564. Wallack, S.S., presented at the Project Hope Fo rum on Health Care Financing Administration Funded Research, Washington, DC, Jan. 16-17, 1985. 565. Warner, J.S., "Trends in the Federal Regulation of Physicians' Fees," Inquiry 13(4):364-381, December 1976. 566. Weil, P.A., "Comparative Costs to the Medi care Program of Seven Prepaid Group Practices and Controls," Milbank Mem. Fund Q. 54(3): 339-365, Summer 1976. 567. Welch, W.P., "Medicare Capitation Payments to HMOs in Light of Regression Towards the Mean in Health Care Costs," Advances in Health Economics and Health Services Research, vol. 6, R.M. Scheffler and L.F. Rossiter (eds.) (Greenwich, CT: JAI Press, forthcoming). 568. Wennberg, J., "Dealing With Medical Practice Variations: A Proposal for Action," Health Affairs 3(2):38-45, Summer 1984. 569. West, H., MANDEX, Inc., Vienna, VA, personal communication, Sept. 13, 1985, and Dec. 4, 1985. 570. West, H., and Marcus, L., Development of a Physician Oriented Data Base (Vienna, VA: MANDEX, Inc., Oct. 1, 1984. 571. West, H., Marcus, L., McMenamin, P., et al., Physician and Hospital Reimbursement Study, prepared for the U.S. Department of Health and Human Services, Baltimore, MD, Jan. 31, 1985. 572. Whitaker Health Services, Marketing Depart ment, Los Angeles, CA, personal communication, Nov. 18, 1985. 573. Wichser, K., Maxicare Health Plans, Hawthorn, CA, personal communication, Nov. 19, 1985. 574. Wilensky, G.R., and Rossiter, L.F., "Alternative Units of Payment: An Overview," presented at the Project Hope Forum on Health Care Financing Administration Funded Research, Washington, DC, Jan. 16-17, 1985. 575. Wilensky, G.R., Farley, P.J., and Taylor, A.K., "Variations in Health Insurance Coverage: Benefits vs. Premiums," Milbank Mem. Fund Q. 62(1):53-81, Winter 1984. 576. Willems, J.S., and Sanders, C.R., "Cost-Effectiveness and Cost-Benefit Analyses of Vaccines," ]. Infec. Dis. 144:486-493, November 1981. 577. William M. Mercer-Meidinger, Inc., Health Care Cost Containment in the Public Sector (New York: William M. Mercer-Meidinger, Inc., 1985). References • 273 578. Wolfson, A.D., "Compensation of Physicians in Ontario," in U.E. Reinhardt, The Compensation ofPhysicians: Approaches Used in Other Countries, final report for HCFA Granf No. 95P-97309, Health Care Financing Admimstration, U.S. Department of Health and Human Services, Baltimore, MD, 1983. 579. Wolinsky, F.D., 'The Performance of Health Maintenance Organizations: An Analytic Review," Milbank Mem. Fund Q. 58(4):537, Fall 1'980. 580. Wolkstein, I., Bureau of Health Insurance, Social Security Administration, Washington, DC, personal communication, Apr. 26, 1985. 581. Wyszewianski, L., Wheeler, J.R.C., and Donabedian, A., "Market-Oriented Cost-Containment Strategies and Quality of Care," Milbank Mem. Fund Q. 60(4):518-550, Fall 1982. 582. Yelin, E.H., Henke, C.]., Kramer, J.S., et al., "A Comparison of the Treatment of Rheumatoid Arthritis in Health Maintenance Organization and Fee-for-Service Practices," N. Engl.]. Med. 312(15):962-967, Apr. 11, 1985. 583. Yett, D.E., Der, W., Ernst, R.L., et al., "Physician Pricing and Health Insurance Reimbursement," Health Care Financing Review 5(2):6980, Winter 1983. 584. Young, B., Office of Health Maintenance Orga nizations, Public Health Service, U.S. Department of Health and Human Services, Rockville, MD, personal communication, Aug. 13, 1985. 585. Young, D.A., "Government Payers for Health Care," in Institute of Medicine, National Academy of Sciences, Assessing Medical Technologies (Washington, DC: National Academy Press, 1985). 586. Young, W., "Measuring the Cost of Care Using Patient Management Categories," final report prepared for the Health Care Financing Administration, U.S. Department of Health and Human Services, Baltimore, MD, May 1985. 587. Zelton, R.A., "Health Insurance," in The Physician's Practice, J.M. Eisenberg, S.V. Williams, and E.S. Smith (eds.) (New York: John Wiley & Sons, 1980). 588. Zollinger, R.M., "Holding Fees in Line," Amer ican College of Surgeons Bulletin 1982(4):2-3, April1982. 0 1 !i----------------- Office of Technology Assessment The Office of Technology Assessment (OTA) was created in 1972 as an analytical arm of Congress. OTA's basic function is to help legislative policymakers anticipate and plan for the consequences of technological changes and to examine the many ways, expected and unexpected, in which technology affects people's lives. The assessment of technology calls for exploration of the physical, biological, economic, social, and political impacts that can result from applications of scientific knowledge. OTA provides Congress with independent and timely information about the potential effects-both beneficial and harmful-of technological applications. Requests for studies are made by chairmen of standing committees of the House of Representatives or Senate; by the Technology Assessment Board, the governing body of OTA; or by the Director of OTA in consultation with the Board. The Technology Assessment Board is composed of six members of the House, six members of the Senate, and the OTA Director, who is a nonvoting member. OTA has studies under way in nine program areas: energy and materials; industry, technology, and employment; international security and commerce; biological applications, food and renewable resources; health; communication and information technvlogies; oceans and environment; and science, education, and transportation. llllliH~IIlmilillllll t 3 9072 01782435 4