J—lECS Preliminary Reports d . The National Health Survey Significance, Scope and Method THE NATIONAL HEALTH SURVEY 1955-1936 ,, ‘j’l’i‘vd‘x' . §ignificance, Scope and Method of A Nation—Wide Family Canvass of Sickness in Relation to its Social and Economic Setting ***** Division of Public Health Methods \ Rational Institute of Health The Uhited States Public Health Service hashington, 1938 g. I! f)! < [j A, flfii ; THE NATIONAL HEALTH SURVEY 1935—19361/ . Ccfj 2 ' Significance, Scope and Method of a Nation-wide Family Canvass of Sickness in ‘91 (F51; Relation to its Social and Economic Setting During the fall and winter of 193b-1936 the United States Public Health Service, with the aid of grants from the horks Progress Administration, inquired into the state of the nation's health and underlying social and economic factors by means of a house—to-house canvass of 740,000 families in urban communities in 19 states and 56,000 families in selected rural areas in 3 states. This bulletin, the first in a series of preliminary reports sets forth the purpose and scope of the survey and outlines in some detail the sampling, canvassing, diagnosis verification, coding and card punching proce» dures which made possible both statistical and medical accuracy throughout this very large and detailed study of sickness in its social and economic setting. Subsequent releases in the series will give sickness, medical care, population and related social-economic findings for the nation-wide canvass. The health data cover prevalence and incidence of acute and chronic disabling illness and the receipt of medical care in re- lation to factors such as income, employment status, occupation; age, sex and color; and the amount of time lost from usual activities. The many types of demographic information included in the study will be of great utility for se: age, sex, racial and occupational dis— tribution of the population; the siZc, composition and income and relief status of families; and Various indices of the standard of living of the surveyed households. Significance and lurpose of the Survey Background of the Study. - The Health Survey procedures were based upon techniques developed during twenty years of experi- ence with the family canvass as a method of studying sickness and related economic factors. Earlier important surveys include a series of canvasses in South Carolina cottonémill villages by the United States lublic Health Service, 1916-1918; its Hagerstown studies, 1921-1924; surveys of 9,000 families in 130 communities by the Com- mittee on the Costs of Medical Care, 1925—1931; and the Health and Depression Studies of the United States Public Health Service, 1935, among 11,500 Wage earners' families in 8 large cities and two groups of coal mining and cotton-mill villages. As is seen the previous 1/From the Division of Public Health Methods, United States Public Health Service. The survey was carried out under the direction of Dr. L. R. Thompson, Director of the lational Institute of Health, G. St. J. Pcrrott, Project Director,_and Clark Tibbitts, Field. Director. Others responsible for the technical aspects of the study were Sclwyn D. Collins, Principal Statistician, and Rollo H. Britten, Senior Statistician. The central coding and tabulating i, office was in Detroit, Michigan. ,\ M817074 surveys were limited to about 10,000 families whereas the present one included three—quarters of a million families. If analysis is to be made of health effects 0. the mani— fold differences in climate, racial composition, urbaniZation, in— dustrial activities, and economic well-being which characterize a country of the sizc_of the United States, it is essential to inter— view a great many families distributed in diverse types of communi- tius throughout the land. In studying the occurrence and causes of the less frequent diseases it is also necessary to canvass a very large population in order to obtain reports on enough cases in vari— ous regions and population groups to permit statistical analysis. Thus, the Health Suery by giving opportunity for a large-scale field project met a definite need in public health and disease research. Need for the Survey. - Despite advances in disease control during the past fifty years, a great deal of preventable illness persists and current data on predisposing environmental and social- cconomic factors are in demand by epidemiologists. Although the growing proportion of older persons in the population makes the problem of chronic diseases increasingly important, very little is known as to the number of persons suffering from such conditions, or as to their distribution according to their age and sex. There is also a paucity of information as to the cause and frequency of accidents, especially home accidents, and the disability resulting from them. The adequate provision of modern medical services and facilities for all classes of the population can be efficiently planned only when the unmet need for each kind of service is known. The data necessary for comprehensive analysis of these national health problems are not available from.rcgularly compiled records. Local, state and federal health departments collect facts principally on births, deaths and a limited and incompletely report— ed list of communicable diseases. On the frequency of accidents and disabilities resulting therefrom, only approximate estimates based on insurance company, workmcn's compensation, industrial and safety organizations records have been available. As to tho provision of medical care, records of doctors, hospitals and health agencies lack uniformity and centralization. The true picture of care received in re ation to needs can be citainod only through family reporting. Specific Purposes of the Survey} - The major specific ob- jectives of the Health Suery were to determine (1) the incidence and nature of serious disabling illnesses — those lasting 7 consec— utive days or longer - during a 12-month period among a Iqtionally representative population, (2) the duration of such illnesses, (5) medical care received, (4) the number and type of serious accidents, and impairments resulting from accidents, (5) the prevalence and type of chronic conditions, orthopedic defects, blindness and deafness, (6) the prevalence and kind of disabling illness on the day of the visit, (7) the utilization of certain medical and public health fa- cilities such as health clinics, tuberculosis sanatorie, and public, health nursing services, (8) the relation between disease and social, economic ind other conditions such as income, employment status, occupation, housing conditions, education, age, sex and color, -5- (9) mortality in relation to income and other social and economic circumstances. Utility of the data. — The tabulations were planned to per- mit first an~ahalysis of illness frequency and duration from all causes and to Live rates for broad classifications of disease and impairments. These data, presented against their social-economic background and related to proportionate receipt of medical care by various income classes, throw light on many pressing current problems. The tabulations include, for example, types of information needed by governmental and welfare agencies on illness in relation to unemployment and relief status, and on the nature and extent of invalidity. Other data of interest in connection with employability and possible dependency are the facts on physical impairments and their causes. Health Survey tabulations also provide a basis for estimating the medical needs of the public assistance populations - the relief group, the aged, the blind, crippled children and depend- ent children. The statistics on illness and relative medical needs among self-supporting families in the low income groups are of value for hospitals, doctors and others who furnish medical services. The data on accidents permit classification according to place of occurrence ~ home, occupational, street or other public accident; and according to the type of mishap such as automobile accidents of various kinds, falls, burns and poisonings. _Therefore, the findings may be useful to industries, governmental agencies and safety organizations in studying hazards and planning safety pro- grzms. Also of interest to employers as Well as to workers are the analysis of illness in relation to occupation and time lost. Studies on Specific Diseases and Public Health Problems. - Many intensive studies can be based upon the rich store of Health Survey facts on specific diseases; on diagnoses as verified and am- plified by reports from attending physicians and hospitals and by death certificate facts; on communicable diseases and preventive care such as smallpox vaccinations and diphtheria immunizations; on housing conditions and overcrowding in relation to illness. The Schedule Used. — A copy of the schedule used in col- lecting the facts necessary for the many types of studies indicated above is included in this report as Appendix A, together with an explanatory key, Appendix B. The Population Surveyed and the Selection of the Sample The first problem in making the Health Survey nationally representative, and at the some time financially and administratively feasible, was determination of the size and distribution of the popu- lation to be studied. The reports are based on schedules taken in 84 cities for 740,000 families including approximately 2,660,000 indi-' viduals, or a number equivalent to 8.7 percent of the urban popu- lation of the United States as reported in thc 1930 census. An -4- additional 36,000 families including 140,000 persons were canvassed in 23 primarily rural counties in a Southern state, a North Central state and a South Central state in order that at least indicatory data might also be obtained on the rural health problem. States, Cities and Rural Counties. - The 84 cities in the urban sample were distributed in 19 states regarded as typical of four main geographic regions - East, Central, South and West.’ The cities to be studied and the number of schedules to be taken in each one were determined according to a plan designed to result (1) in a population distributed according to geographic area in approximately the sane proportion as the general urban popu- lation was in 1930 9nd (2) in inclusion of cities distributed wnong four different size-groups: 500,000 population or more; 100,000 to 500,000; 25,000 to 100,000; and less than 25,000 population. The sample was selected as follows! Preliminary to choos- ing the study communities all cities of 2,500 or more population in the United States were classified on the basis of the four regions and according to the size groups within each region. This procedure yielded 10 sub—groups which were reduced to 15 when for Baltimore it was decided to employ a special sampling procedure that would com- plotely cover two health districts but which would not give a cross section of the city. The 84 cities were then selected with the view to getting size and geographic representation similar to that of the 15 suon_reups. In order to avoid over-representation of large city popu— lations, 31 places of more than 100,000 inhabitants and Montgomery, Alabama, were sampled while 53 smaller cities were completely can- vassed. The sampling procedure was recomnendcd also because it per~ mitted a great saving of time and expense and because it was believed that complete coverage of large places was superfluous. The samples taken in the larger cities varied from 5,000 to 45,000 schedules, not according to a fixed ratio but on the basis of (l) the number believed adequate to represent the individual community, and (2) the number required to complete the sample on the basis of regional and size distribution. The proportion of surveyed households to total families ranged from.one in two in the smaller towns to about one in thirty—five in the largest cities. The geographic distribution of the survey sample corre- sponded very closely to that of the 1930 urban population as shown in the left half of the table below. Financial, administrative and time considerations made it impossible to include the very large number of cities necessary to give a sample population distributed g/Tho Health Survey states included in the four regions-are: East — Massachusetts, New Jersey, New York,.?enn3ylvania; Central - Illinois, Michigan, Minnesota, Missouri, Ohio; South - Alabama, Georgia, Louisiana, Maryland, Texas, Virginia; West - California, Oregon, Utah, Washington. ‘ ‘ ' V. 1 -5- by city size-group in the same proportion as the 1930 urban popu- lation. Each size—group was nevertheless well represented. Distribution of the Health Survuy Urban fopuletion and the 1950 Urban Population According to Geographic Area and City SiZc—Group. Regional distribution City—size distribution Region Health 1950 Size Health I 1950 Survey urban Survey urban All .100% 100% All 100% 100% East 37 39 500,000 or more 43 29 Central 35 55 100,000-499,999 51 '23 South 18 18 25,000- 99,999 14 19 West 12 10 Under 25,000 12 29 Appendix C shows the survey cities according to region and size—group. The 23 rural counties which were studied are located in three states and, of courSe, are not representative of the whole rural population. Sixteen are in Georgia and were included at the request of the Department of Health and the medical society of that state; four are in Michigan and three in Missouri. Sampling within the Cities. — After the number of family schedules to be taken in each city had been determined through the process described above, it was necessary to obtain a random dis— tribution of the desired number of families in each of the 51 com- munities which were sampled rather than completely canvassed. The proportion of a city's households to be visited was obtained by di- viding the predetermined survey quota of families by the estimated total number of families in the community.§/ This gave a sampling ratio which was applied to groups of city blocks known as the enumer— ation districts and set up by the Bureau of the Census in 1960. These groups represented units of population in each city and pro~ vided conVcnient assignments for squads of enumerators. Their sam- pling utility was limited, however, because frequently it was im- possible to include enough of them to be sure that each population group within a city would be adequately represented. This difficulty was anticipated, howover, and it was possible partially to correct it by splitting the larger districts. Enumeration districts having s- ~ é/The total number of families was ottainod by divising.thc city population in 1930 by 4, the approximnte average number of per— ec:s per family in the United States in 1950. ' -6... populations of less than 1,000 in 1950 were included as found on the Census Bureau maps; those with populations of more than 1,000 but less than 2,000 were divided geographically into two equal sections; those having populations of from 2,000 to 3,000 were divided into three sections, and so on. When the districts within any city had been subdivided in this fashion, a number was assigned to each and the appropriate sample was then selected from the numerical list. Thus, if it had been determined that the survey was to include one~ eleventh of the population, every eleventh enumeration district was selected from the new list. Collection of the Data The accuracy of the information gathered in a fannly can. vass depends very largely upon the quality of the enumeration ac- tivities. Therefore, the field organization, training methods and canvassing procedures of the Health Survey are of especial interest in view of the difficult technical nature of the inquiry. The 19 states in which the survey was conducted were di- vided into five administrative regions. Each region was assigned to a supervisor who had had successful survey experience and who was given training on the schedule and instructions for nearly two weeks before being sent into the field. Nineteen administrative persons Were selected as state supervisors and were given responsibility for procurement of space and office equipnwnt, preparation of payrolls and accounting. By assignment of administrative matters to state supervisors the time of regional and local supervisors was usually left free for training and supervising field workers. Each of the local units was placed under the direction of a supervisor who had no assistant or one or two assistants, depende ing on the size of his staff. When there were two assistants one was placed in charge of training the enumerators and reviewing their work, and the other became directly responsible for supervising the enumerators in the field. The enumeration was conducted by groups of from five to eight canvassers working under the direction of squad leaders. A control clerk had charge of the schedules while they were in the local office. Selection and Training of Enumerators. - The Health Survey, being financed as a work relief project, drew its canvassing staff from the relief rolls. Through the cooperation of the United States Employment Service and the works Progress Administration it was pos- sible, for the most part, to select mature persons with previous white-collar work experience. Preference was given to those who had been bookkeepers, teachers, nurses, salesmen, and social workers. Many had had experience on other surveys. The individuals tentativelyfselectcd for assignment‘to the survLy were introduced to it through a discussion of its purposes by”- the local supervisor. They were then sent home With an abbreviated instruction manual, tw0 family narratives, a schedule properly filled‘ from one of the narratives, and a blank schedule to be filled from -7- the other narrative. On the following day the home work was reviewed and the prospective enumera ors were asked to fill a second schedule from a third narrative. This was regarded as an examination and those who showed no aptitude for the work were returned to the.employment service. Successful candidates were given additional training, were paired and required to fill schedules covering each ethcr's families, and were required to attend a prepared dialogue showing how an inter- view should be conducted. Following this week of intensive training the canvassers were.sent into the field to make trial enumerations. Further training was given to those who required it. Editing the Schedule. - A reviewing staff in the ratio of one editor to three enumerators was established in each local office to examine schedules for completeness and consistency. Comments were entered on slips attached to the schedules. Unsatisfactory schedules wore returned to the field through the appropriate squad leader, who had daily conferences in the survey office and who met each enumera- tor daily to collect filled schedules and to discuss questions and rejected schedules. Each schedule was edited and then re—edited before being accepted for shipment to the central office. The local review was augmented by further checking in the central office in Detroit. The earliest schedules from each city were examined critically to determine whether incorrect forms of entry or inconsistencies were present which might indicate lack of understanding of field instructions. Errors were reported to the local office. In addition all schedules received throughout the pe- riod of the survey were examined routinely for major errors in the sickness record, such as diagnoses recorded separately which were really part of one illness or inconsistencies in the duration of dis- ability in relation to date of onset and termination. Questionable entries and inquiries referred to the central office by the field staff were discussed in a series of technical bulletins and in corro- spondenec with the individual supervisors. In some instances sched- ules themselves Were returned to local offices in order to illustrate unacceptable entries. Completeness of Enumeration. - As a result of the careful training of enumerators, a vigorous publicity campaign, and the co— operation of public officials and community organizations 98.5 per- cent of the families asked to give information complied with the request. The interest and assistance of state and local health of- ficers and of representatives of medical societies were obtained through announcements of the project sent them by the Surgeon General of the United States Public Health Service. Prepared publicity placed in the hands of the field staff was used freely by newspapers. Radio stations gave discussion time to health officers and members of the survey staff. Announcements were also distributed through schools and churches. ' A thorough check on the completeness of the enumerator's work was made through the use of a control card and a daily record book. Prior to the beginning of the canvass all workers Were Sen' into the ficld‘to enter on a small card the address of every buil in that appeared to be usable for human habitation. These cards were -8... a: i .a. filed in the local office and later, as each schedule was received from the i old, the address was checked against the file. Unoccupied building or dwelling units Were identified in the cnunerator's re— cord book and upon the reported completion of each enumeration dis- trict the control clerk conpared the unmatched cards with the record book entries. An investigation was made whenever a control card was not accounted for by either a schedule or record book entry. Verification cf Diagnoses Confirmation of Diagnoses by Physicians. - In order to check the E23§73fi7§§2$“fh€3£3 ERIEFEEIEE‘EITR reference to illness diagnoses enumerators were instructed to reduest the name of the hospital or the physician who cared for an attended case. Almost without exception, permission to obtain further medical information from these sources was granted, notation of permission or refusal being redo on each schedule. When the schedules with diagnoses to be confirmed reached the central office the sickness entries were grouped according to a 13~fold classification. A separate question— naire had been prepared for each group and the appropriate form was sent to the doctor or institution, if permission for confirmation had been given. Some 555,000 of these inquiries were mailed and 400,000 were returned, with the result that professional reports were ob- tained on 350,000 of the attended illnesses. Copies of the death certificates for fatal illness report- ed in the canvass were secured in order that the stated cause of death might be compared with that on the official record. According- ly, where local records permitted it, copies of death certificates were obtained through cooperation with the registrar. Where local records were not available copies Were obtained from files of state health departments. In the central office nearly 14,000 death certificate copies were attached to the proper schedules and the official data used in coding. Coding and Card Punching Immediately upon their receipt in the central office sched— ules were sent to a special division for complete and detailed re- view. Unacceptable schedules were returned to the field and the rest were prepared for the coding, or transcription of the data into numerical symbols, so that the information might be recorded on punch- ed cards and tabulated by machine. Some conception of the volume of original data which is incorporated in the final tables may be gained from administrative records. Coding of the schedules required 15,000 men-months of work on the part of a staff that reached 1,000 persons at the peak. 'A punched card was prepared for each of the 2,660,000 persons enumer- ated and for each of the approximately 1,100,000 cases of illness reported — a grand total of 5,760,000 cards each bearing 75 to 100‘ items of information. The coding instructions, bound in 20 volumes, totalled 900 pages. I -9- Extreme skill and care were eseeutial in the preparation of instructions because of lack of previous training on the part of the relief workers who were to use them.$/ . An analyzing unit prepared the diagnostic portion of the case data so that it could be coded routinely by the medical coders or referred to a staff of physicians who completed the diagnosis coding for more complicated cases. The analyzers’ work varied from a simple sorting on the basis of type of diagnosis to coding of a highly conplex nature, such as that concerned with accidents, im— pairments, and supplemental information from death certificates and medical reports from attending physicians and hospitals. Personnel for the unit were selected from the best of the coders. Within the unit itself the workers were divided, on the basis of their ability, into groups handling schedules of varying difficulty. Through this procedure it was possible to secure detailed, precise coding of diagnoses. In order to secure the greatest posSible accuracy in pre- paring the punched cards from the coded data, a series of mechanical verification procedures was adopted for the punching operations. Also, as a final precaution before the ards Were used in tabula- tions, a chock was made of about one—fifth of the items punched for each schedule so that possibility of internal discrepancy might be obiiatc . hhen a chance inconsistency was found, such as disagree- ment between the code representing the number of persons per fanily and the number of individual cards punched for that family, reference was made to tho SChOfiulO and the error corrected. Equally careful verification and revorification methods were employed throughout the - machine tabulation proccSSos. Terms and Definitions The meaning of terms used in bulletins and articles report- ing the survey findings will, in general, be apparent from the sched— ule and key sheet appended hereto and from the tables and discussion contained in the various publications. For ready reference purposes, however, a few explanations and definitions are given below. The Study Period. - Enumeration was started early in October 1933?‘333‘3ii‘i££"3191d canvassing was completed by March 30,1936. Each family reported illnesses experienced during the 12 months preceding the day of interview, also family income during the same period and any relief assistance received with the year. Hence the sickness and income information obtained in the survey as a whole pertains to an interval of 18 months extending from October 1, yThc coding instructions were prepared by S. D. Collins, Principal Statistician, and Rollo H. Britten, Senior Statistician, with the assistance of Ruth Phillips, Gerald E. Rein, L.D., H. R. Ogburn, H. M._C. Luykx, David Hailman, Jennie Goddard, and others., The _ coding was done under the supervision of H. R. Ogburn, J. L. McPherson, Ruth Phillips and Dr. V. M. Hoge. -10- 1954 to March 20, 1936. The worst depths of the depression had been passed but widespread unemployment and want still prevailed. In relation to the redical care reported by relief families, it should be noted that the Federal Emergency Relief Administration medical care plan was in operation in numerous cities throughout the nation during 1954 and the first half of 1935. Relief Status. — Families and the members thereof were classified as 55ihg_of-thc relief group if any assistance had been received from an official agency during the twelve months preceding the enumeration date. Income. — Infonnants were asked to estimate roughly the net fandlyrihsome for the preceding twelve months, within the limits of income classes: $5,000 or more, $5,000 to $5,000, $2,000 to $3,000, etc. For coding purposes income was calculated in the central office not only for the related family but also for the total "economic unit" or all persons sharing the household living costs and support. The income of the related family is the information used in the current analyses of survey data. Workers. ~ Included in this term are (1) persons employed in private—industry and in permanent government departments, (2) un- employed persons engaged on work relief, (3) totally unemployed per— sons seeking work. The category also includes persons who were look— ing for their first jobs, a group which is excluded by the Census Bu- reau in defining gainful workers. wage Earners. — These are employed persons with nonpwork .——_ relief 3055; Family Size and Type. ~ These facts were detennined on a basis of persons related to the family by blood, marriage or adop— tion. SickneSs and Impairment Data.§/- The following definitions are of importance in interpreting the significance of the sickness and impairment data. An illness — a continuous period of sickness; such ill- nesses may be due to one or more causes. A disabling illness - an illness that keeps the person from his work, school and other usual activities. No terminated disabling illnesses of less than 7 days duration were included in the survey, with the excep- tion of confinements, hospital cases and fatal cases, which were included regardless of their duration. For §/An analysis of disease and impairment in one Health Survey city is given in: "Chronic Disease and GyOSS Impairments in a Northern Industrial Community," by George St.J. Perrott and Dorothy F. Holland, United States Public Health Service, in the Journal of 3h; American Medical Association, May 29, 1937, pp.18‘76"-"1"8'8‘6._~ w 11 - tho day of visit a rocord was takcn of all disabling illnesses. Disabling illnesses are classified in the Health Survey roports by cause uccording to (l) tho nature of tho sjnclfic diagnosis, and (2) tho duration of symptoms of tno diSc so“ For the purpose of u broad grouping, ill- nossos with synptoms of loss than 5 months' duration ”er classified as "scute" and thosc with symptoms of v . 5 months or more wore clLssificd as ”chronic". Imp irmonts - pormonont handicaps rcsulting from discase;~achidont, or congcnitnl defect, including impLircd or lost manhcrs, or serious dofccts of vision or hcsrirg. both disabling and non—disabling impairmonts aro includod. Physician's Caro - attention received from a doctor of modicinc or other similar practitioner. Such care is in addition to that givon by physicians in hospitals. Private duty nursing care — core exclusive of floor nursing in hospitals but including care given by special hospital nurses. A P P El!) 1 X A (Serial No.) (Do Not Write Here) 1. FAMILY NAME _____________________________________ 2. ADDRESS ______________________________ APT. No. __________ 3. CITY ____________________________ 4. STATE _ (M bBOSFIIiER h id) Egg USUAL EuPLomzm 5:; g: Fox. P11130113 UNDn 25 UNITED STATES 611! ta 0 case a RELA- an m w - Eu- ,_. E sumox Dmnmnu PUBLIC HEALTH g3“ '0 MAR" 33?; Coumv or COL- EDUCA- PLOY- ~38 “3 AD or A0: SIX ’l‘AL o a q B 3 5 I, n .. SERVICE p“. Housm- Sums E o 0 mm on. non num- E 3 5 E a Last Im- son’s Nu“ BOLD chili: Sums Occupation Industry £3“: g8 Case X33551; Case :33;- No. 0 > Age Age Age Age HEALTH 5 e 7 s 9 1o 11' 12 13 14 15 1e 17 18' 19° 20 21 22 23 SURVEY (Given) (Last) 1 l ______ _ ________ ALL INFORMATION IS STRICTLY 2 - _ ____________ _ . CONFIDENTIAL 3 . ............ TO BE USED FOR 4 . _________________ . STATISTICAL PURPOSES ONLY 5 _____ _ 6 ___________ . ______ District __ 7 .1 ____________ _ ____________________ .. 8 _ L“ _ ______ V__,. ____________ DISABLING ILLNESSES Tm: UNABLE m WORK .. NURSE Puwous Arucxs on Punsu: USUAL N ° Doc-ton 2 ' or Sun: DISEASE ACTIVITY—LAST l2 "‘ ES" : Emu-.3: (a; Illness today. Moms 01"“ H fig Number of a (b Illness-f7 days or longer— t 12 months—include all MONTH AND 33 3 El 'I‘Enunu'nox Calls Past "‘ i2”! 32 E a: hoe ital cases and all co emenm. YIAB 0" I ‘5 3‘“: (For Deaths, 12 Months Nun AND Amman-3 or Docmn. HOSPITAL, ETC. 5 5’53 3'?» How 1.5% (a) Hospital, em. now—add to mater. ONSET 0' T315 a o 3", 5 Enter Month '5' g > 8 Long 3 n3 55):: (d) Deaths—Past 12 months—add to roster. DIIBABHNG In Bed 32 §°E and Your) _ § 752 ‘52 An u f‘ ° “‘32 No. (SEPARATI line for EACH illness for man person) mus 111.1{05- Ng‘tlgn Not in a fig To n ha; 551 y. we? :55 a mini ‘ 05- Bed 3 30 Ac Ofllce g Ea » ment "‘53 pital O a Home or q :1? g 3 First 3 ° > Given Name Nature of Ilium or Accident 0 Clinic 2” zih Noticed Q: 36 24 25 28 27 28 39 80 31' 32' 33 34 35 37‘ 38 39 40‘ 41 42‘ d ......................... ___, ______ ____.. _____ . .m. ._ __-,_____,_‘_A__,____ ,...__l_,“__ ______ ,_._.V._ _,.,._ 18 a .................................................................................................... 19 n. ________________________________ . 33 a ........................................... (Name and addm of sick benefit mochnion) (W hat war?) ') (Place 0! death) ‘Means entry (1! malted) should be Yes, No, or Unk. PL A T E | I 5 2—17005 NOTES: 9 M - male F — female 10 S — single '-_-l3 W - white "—- M - married N - negro WDS — wid,div or sep 0th—cther Appendix B m...— ....._...._..._._.. _ __ . United States Public Health Service HEALTH SURVEY Scnedule Key * Means answer should be YES or N0 N0 check (\j) is to be used except where CHECK is stated on schedule HS Form 14C Oct. 7.1935 UNK is to be used only when the answer is un- known to INFORMANT 1 LAST name of head of household 7 write in when spate allCWS as : Head,Son.etc. Abbreviate when necessary as: Dan-daughter ADD: L—in-law Mo-mother S~step Sis-sister Gncrand,great Unrelated: A—adUVted Room—roomer None—unrelated person, not paying for room or meals Ser -servant (living in) Part-partner 8 Age on LAST birthday. Enter in years Under one year enter as: 3/12 — 3 months 2/52 - 2 weeks 3/365- 3 days ._ __.- m- 14 0 — never gone to school K — kindergarten 1,2,etc.—highest grade reached H S-high school~or business school Col—college or university I." J—J. map-cmployed now (not work relief) w R—work relief s w-unsmploycd now. seeking work Rat-retired or not seeking work because of age » Dis-unemployed now. not seeking work because of disability Oth—unemployed now, not seeking work for reasons other than disability. Specify reason in NOTES H w—housewife Sch—school H -home 18 If YES. enter in 18s name and address of sick benefit association 19 If YES, fill in 19a, 19b and 19c Enter army, Navy, M C (Marine Corps), or C G (CoaSt Guard): AND Com (Ccnmissioned Officer) or 0th (any other rank) 20 to Persons under 25 only 23 Enter age at time of occurrence UNK — unknown whether occurred If actual age at occurrence is unknown, get informant's estimate as 6-10 0 — no occurrence 26 If accident. enter cause and type. (Use Code 47b) Persons found by (c) and (d), ADD to Roster and fill in Columns 6-23 for them 28 t: Enter in days. weeks or months (as given 30 by informant) O - none Maximum time is 12 months. A figure is NOT satisfactory unless d for days, w for weeks or m for months is given (small letters) 53 Rec - recovered SUW — still sick, unable to work D — died (After D enter month and year of death). Enter address and city or death in 33a .__.__. 41 Enter in years. Under one year enter as: 470 Appliances. For each person using any appliances because of conditions indicated, use following code: Cr — crutch Cane— cane Br — brace. leg or back Tr — truss Audio—audiophone (or any other mechanical appliance used for deafness) A leg — artificial leg A arm ~ artificial ann None ~ none If any other appliance, specify 6/12 — 6 months, etc. ......,.... .. ......_-.. .. 49 Enter in years or fraction of year 47a PARTS AFFECTED: enter actual part of body (right side paraIYZed.'etc.) If PARTIAL for DEAF is "yes", add: i. if can not hear at church. etc. 51 Enter in days total time unable to work or pursue usual activity 2. if can not hear 2 or 3 ft., etc. 5. if can not hear over telephone l I l i The — tuberculosis ADOD- apoplexy. cerebral hemorrhage Cong— congenital r‘ Write in disease or condition if not in .._.. -—_. -..._ ._-. __., *._——_..._...___.._ 47b CAUSE and TYPE. Specify. Use following cede for place of accident: H - home S - street or other public 0 - occupational If motor vehicle. add A to above If cause is disease. use following: I P — infantile paralysis Rick— rickets 59 For use in checking income class when cruy weekly or monthly figure is known Zearly InconEL Monthly Weekly $5000 or more $417 f $96 f $3000 but under $5000 250-416 58‘95 $2000 but under $5000 167-249 39-57 $1500 but under $2000 125-166 29—38 $1000 but under $1500 84—124 20-28 Under $1000 0-83 0-19 This table will be of use ONLY if person is regularly employed and no change in income during year code ENCIRCLE in Column 5 each person whose .total income is included in the estimnte cf the family income-AND each person whose expenses are paid out of this income Appendix C The 84 Health Survey Cities Arranged According to Region, City Size Group and State; and the Number of Schedules Taken EAST CENTRAL SOUTH 12561 Cities of 500,000 or moreél City Scheds City Scheds gity Soheds City Seheds Bostou,hass. 29,808 Chicago,111. 38,501 Baltimore,Ed. 36,226 Los Angeles,Cal. 26,297 Buffalo,31.Y. 24,055 ‘ Detroit,1iich. 21,706 New York,N.Y. 48,281 St.Louis,Ko. 24,116 Philadelphia,Pa. 32,360 Cleveland,0. 31,993 Pittsburgh,?a. 20,391 Totals 154,895 116,316 56,226 26,297 Cities of 100,000 to 500,0003/ . Flint,Mich. 4,826 Birmingham,Ala. 11,173 0ak1and,Cal. 8,361 Fall River,Mass. 10,481 Grand Rapids,fiich. 5,112 Atlanta,Ga. 10,737 Portland,0re. 10,329 Newark,N.J. 13,999 Minneapolis,fiinn. 12,295 New Orleans,La. 13,192, Salt Lake Cityfi}. 7,775 Trenton,E.J. 7,580 St.Pau1,Hinn. 12,976 Dallas,Tex. 10,898 Seattle,Wash. 9,724 Syrabuso,N.Y. 12,827 Cincinnati,0. 2,549 Houston,Tex. 11,738 Spokane,Wash. 8,127 Columbus,0. 11,058 Richmond,Va. 12,542 Totals 44,887 58,816 70,280 44,316 Cities of 25,000 to 100,0009/ Pittsfiold,Mass. 11,951 Port Huron,flich. 8,295 Montgomery,A1a.§/ 9,742 Salem,0re. 8,142 Lebanon,Pa. 6,412 Springfield,fio. 16,650 ' Monroe,La. 6,985 Lima,0. 11,285 Amarillo,Tex. 11,091 Wichita Falls,T0x.10,793 Totals 18,565 36,230 58,611 8,142 Appendix C (Cont'd) ELST CElTRAL SOUTH IEST Cities of loss.than 23,0002/ Grecnfisld,Mnss. 4,045 Borton,Tll. 2,107 Eufala,A1a. 1,412 Chico,Cal. 2,539 Ipswich,hass. 1,448 Normal,lll. 1,749 Gadsden,A1a. 5,525 Grass Vulley,Ca1. 1,507 Briogoton,N.J. 4,419 Houghtdn,mich. 991 Greenville,Ala. 1,048 Jackson,Cal. 683 Lambertville,F.J. 1,100 Chisholm,fiinn. 1,778 Brunswick,Ga. 3,094 Napa,Ca1. 1,612 Somervillo,H.J. 2,106 Hilmar,fiinn. 1,799 Abbeville,La. 1,373 Vallejo,0a1. 4,626 Hudson,N.V. 3,185 .Einonn,Minn. 6,426 Bossier,La. 917 L1G:ande,0re. 2,264 Newar;,N.Y. 1,661 Chillicothe,Mo. 2,133 Hinden,La. 1,401 St.Helens,0re. 1,279 Penn Yan,fi.f. 1,614 Clinton,Mo. 1,646 Weatherforfl,TeX. 1,497 Bingham Canyon,U. 905 Diryea,Pa. 1,729 Franklin,0. 1,088 Covington,Va. 1,E01 Eureka,Utah 651 Indiana,Fa. 2,490 hilmington,0. 1,499 Farmville,Va. 857 Tooele,Utah 1,143 Ellenburg,Wash. 1,466 Olympia,fiash. 5,721 Totals 23,897 21,218 18,625 22,596 GRAND TOTALS 242,042 232,580 163,742 101,351 é/Sampled. The appreximnte proportions of the population covered are as follows: Atlanta, one—seventh; Baltimore, speCIal sampling procedure; Birmingham, one—sixth; Boston, one—seventh; Buffalo, one—sixth; Chicago, one-twenty’fourth; Cincinnati, one-tenth; Cleveland, one—seventh; Columbus, one~eighth; Dallas, one—seventh; Detroit, one-nineteenth; Fall Hiver, one—third; Flint, one-eighth; Grand Rapids, one-ninth; Houston, one-eighth; Los Angeles, one-eighteenth; Minneapolis, one-eleventh; Montgomery, one-half; Newark,N.J., one-eighth; New York City, one-thirty—eighth; New Orleans, one-ninth; Oakland, one-eleventh; Philadelphia, one—fifteenth; Pittsburgh, one-eighth; Portland, one—ninth; Richmond, one-fourth; St.Louis, one—ninth; St.Paul, one—seventh; Salt Lake City, one—fifth; Seattle, one—tenth; Spokane, one—fourth; Syracuse, one-fifth; Trenton, one-fourth. E/Completely canvassed. ,, , .o» A o e 1 ' Y‘ ..._r- ‘OHISHOLH. —'I_—. . IPSWICH I . GREENFIELD MINNEDOTA HOUGHTON .S'YRACUSE ouum BOSTON ' NEWARK . . ELLéns. fiWlLLMAR - Pl 775er L0 mm MiNNEAPOLI§ ,p ENN YAN HUDSON . ST.PAUL , MAssAcNusU‘rs MucmsAN “IRON“ Nison —m——1_ FALL mvsn - FLINT cpAND RAPIDS - PORY HURON . DETROIT - 'BUFFALO cm CAGO. NEW YORK cLEvaLAND NEWARK . NORMAL DURYEA- souanvu: ' . cmco , LINA O p ENNSYLVAN“ , LAMBERYVILLE GRASS VALLEY ———fi “UN" "'° TRENTON 'N‘" “SALT LAKE any N J 'V‘LLE‘m 5 'TOOELE Lommcoms "mm” 5" 5' szv {numb n ~EINGHAM 01 won 3 “a u 'coLuuaus " PITTSBURGH g ' LEBANON . C AuromuA r——‘—“ ”mum" _FNANNLIN meADELmIA. ' ST. LOUIS ' BENTCN "WWW" x 'CLINTON ' = _ cINcmNAn BALTpaonE 3 MISSOURI 1—,— VINDINIA A MARYLAND . ISFRINGFIE'LD cévacToN RICHMOND :- FANNVILLE ' Los ANGELES ——-1 r cAstzfi ‘ . BOSSIERMONROE ammNcNAu - ATLANTA _AMAR|LL0 _WICH|TA FALLS ‘MlNDEN ’ LoumANA ALAIAIA GEORGIA T MONTGQMERV u ‘3 EUFALA . wcnusnronn . New ORLEANS . . anunswl-K‘ TN: 51250: THE surzms ' ”new“: GREEW'LLE ’ DETERMINED BY THE 1930 POPULA'IION DALLAs . HOUSTON . U.C. BERKELEY LIBRARIES CUHSH'HIWLB