PROGRAMS AND COLLECTION PROCEDURES

Plan and Operation of theHealth and NutritionExamination SurveyUnited States-1971-1973

A description of a national health and nutrition examination survey of aprobability sample of the U.S. population 1-74 years of age:

Part A-Development, plan, and operation.

Part B-Data collection forms of the survey.

-DHEW Publ icat ion No. (PHS) 79-1310

U . S . D E P A R T M E N T O F H E A L T H , E D U C A T I O N , A N D W E L F A R EPublic Health Service

Office of Health Research, Statistics, and TechnologyNational Center for Health StatisticsHyattsville, Md. February 1973

Series 1Number 1Oa

c .I

,- ‘\., .

\

P t

. , . .”

1 I

NATIONAL CENTER FOR HEALTH STATISTICS

DOROTHY P. RICE, Director

ROBERT A. ISRAEL, Deputy DirectorJACOB J. FELDMAN, Ph.D., Associate Director for Analysis

GAIL F. FISHER, Ph.D., Associate Director for the Cooperative Health Statistics SystemsELIJAH L. WHITE, Associute Director for Duta Systems

JAMES T. BAIRD, JR., Ph.D., Associate Director for International StatisticsROBERT C, HUBER, Associate Director for Management

MONROE G. SIRKEN, Ph.D., Associate Director for Mathematical StatisticsPETER L. HURLEY, Associate Director for Operations

JAMES M. ROBEY, Ph.D., Associate Director for Program DevefopnentPAUL E. LEAVERTON, Ph.D., Associate Director for Research

ALICE HAYWOOD, Information Officer

COOPERATION OF THE BUREAU OF THE CENSUS

In accordance with specifications established by the National HealthSurvey, the Bureau of the Census, under a contractual agreement, partici-pated in the design and selection of the sample, and carried out the first stageof the field interviewing and certain parts of the statistical processing.

Library of Congress Catalog Curd Number 72-600207

This report presents a detailed description of the Health andNutrition Examination Survey (HANES). It is intended primarilyto serve as a necessary foundation for understanding and use inconjunction with the substantive findings to be published later inpreliminary reports and in VitaI and HeaZth Statistics, Series 11, ofthe National Center for Health Statistics (NCHS). It will alsoprovide background information for succeeding surveys of thisnature and serve as a guide or aid to others in the planning ofsimilar health or nutrition surveys.

In the planning and operation of HANES, valuable assistancewas received from many individuals and groups. Space does notpermit the recognition of all who participated in the planning,development, and conduct of the many and varied aspects of thesurvey. Their assistance is gratefully acknowledged, however, andan apology is offered for those omitted. Mention should be madeof the important role played by the U.S. Bureau of the Census.Under a contractual arrangement, the Bureau of the Censusparticipated in certain aspects of the sample selection, in theconduct of initial household interviews, and in most of theprocessing of the data. The special role of a task force designatedto formulate a general plan for HANES and chaired by Mr. EarlBryant, Office of Statistical Methods, NCHS, is described in thetext. Also requiring special mention is the role of the Center forDisease Control (CDC), Health Services and Mental HealthAdministration. In addjtion to the advice and assistance providedin the planning operation, particularly by past and presentmembers of the Nutrition Program and by the Division ofLaboratories, Dr. David Sencer, Director, CDC, established aNutrition Laboratory where, under a reimbursable arrangement,essentially all of the laboratory work for the HANES program iscarried out in a highly standardized manner under the direction ofDr. Hipolito Nino.

The overall responsibility for planning the program was that ofMr. Arthur J. McDowell, Director, Division of Health ExaminationStatistics (DHES). The primary responsibility for recommendingthe content of and developing the procedures for the various partsof the detailed component for the examination was that of Dr.Arnold Engel, Medical Advisor for the adult programs of DHES.His counterpart with respect to responsibility for the nutritioncomponent of the examination was Dr. Frank W. Lowenstein,Medical Nutrition Advisor,, DHES. Dr. James E. Kelly, DentalAdvisor, NCHS, and Dr. Lawrence E. Van Kirk, Jr., DentalAdvisor, DHES, had similar responsibilities with respect to thedental component. Other members of the DHES staff who hadresponsibility in specific areas were Dr. Harold Dupuy, Psycho-logical Advisor, Drs. John V. Federico and Lanie E. Eagleton,formerly Medical Advisors, Miss Jean Roberts, Chief, MedicalStatistics Branch, and Mr. Sidney Abraham, Chief, NutritionStatistics Branch.

CONTENTSPart A

Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

RoleoftheNCHSTaskForce . . . . . . . . . . . . . . . . . . . . . . . . .Charge to the Task Force . . . . . . . . . . . . . . . . . . . . . . . . . .Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Background of HES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Basic Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . .Past IIES Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Planning for the Health and Nutrition Examination Survey . . . . . . . . . .General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Development of HANES General Objectives . . . . . . . . . . . . . . . .Development of Some Specific Areas of the Detailed Component . . . . .Development of Specific Areas of the Nutrition Component . . . . . . . .Pilot Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Sample Design . . . . . . . . . . . . . . . . . . . . . . . . . . . e . . . . .General Plan . . . . . . . . . . . . . . . . o . . . . . . . . . . . . . . .Design Specifications . . . . . . . . . . . . . . . . . . . . . . . o . . . .Stratification and Selection of Primary Sampling Units . . . . . 0 . . . . .Within PSU Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Selection of Sample Persons . . . . . . . . . . . . . . . e . . . . . . . .

Other Sampling Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Sampling Features of the Examination . . . . . . . . . . . . . . . . . . . 21Stand Sequencing and Scheduling . . . . . . . . . . . . . . . . . . . . . 21

Advance Arrangements . . . . . . . . . . . . . . . . . . . . . . . . . e . .Professional Relations . . . . . . . . . . . . . . . . . . . . . . . . . . .Public RelationsLogistical Arrangements

. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .

-Household Interviewing Procedures . . . . . . . . . . . . . . . . . . . . . . 24Census Interviewing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24HANES Interviewing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

112

334

4456

101214

151515161818

22222324

iv

CONTENTS-Con.

Page

Appointment and Transportation Procedures . . . . . . . . . . . . . . . . . 25

Examination Center and Field Staff . . . . . . . . . . . . . . . . . . . . . . 26

Examination Center Procedures . . . . . . . . . . . . . . . . . . . . . . . . 26General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26Flow of Examinees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28Physician and Nurse Examination . . . . . . . . . . . . . . . . . . . . . 29Ophthalmology Examination . . . . . . . . . . . . . . . . . . . . . . . . 30Dermatology Examination . . . . . . . . . . . . . . . . . . . . . . . . . 30Dental Examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Dietary Interview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32Laboratory Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . 32Health Technician Procedures . . . . . . . . . . . . . . . . . . . . . . . 32

Quality Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Plans for Analysis and Publication of Data . . . . . . . . . . . . . . . . . . 38

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Appendix I. Technical Notes on the Sample Design . . . . . . . . . . . . . 43Definition of Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43Location of the 65 Health and Nutrition Examination Survey

Stands by Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

~. -i-

, ’

SYMBOLS

Data not available -----11---11------1--------------------- - - -

Category not applicable----L-II-----l-l--------------- . . .

Quantity zero -1-1--1--------1--1---------------------------

Quantity more than 0 but less than 0.05.---- 0.0

Figure does not meet standards ofreliability or precision----1-----11-1----11---------- *I

Vi

PLAN AND OPERATION OF THE HEALTH A

NUTRITION EXAMINATION SURVEY

Henry W. Miller, Division of Health Examination Statis tics

INTRODUCTION

The National Health Survey Act of 1956provides for the establishment and continuationof a National Health Survey to obtain informa-tion about the health status of the population inthe United States, including the services receivedfor or because of health conditions. The respon-sibility for the development and conduct of thatprogram is placed with the National Center forHealth Statistics (NCHS), a research-orientedstatistical organization within the Health Serv-ices and Menta l Hea l th Adminis t ra t ion(HSMHA) of the Department of Health, Educa-tion, and Welfare. Three separate and distinctprograms are employed by NCHS in meeting theobjectives of the Act-a household health inter-view survey, a family of surveys of healthresources, and a health examination survey?

Between the passage of the 1956 Act and1969, numerous studies related to nutritionwere conducted by various sources which re-vealed that within certain areas of the UnitedStates and within certain age and income groupsmalnutrition and undernutrition were stark real-ities. The most recently completed of thesestudies was the National Nutrition Surveyauthorized by Congress with the stated objectiveto

. . . determine the prevalence and location of serious hungerand malnutrition and resulting health problems in lowincome populations in (ten) selected States representing

different geographic regions of the U.S. and to makerecommendations for dealing with such conditions.”

The preliminary results of that survey, thevarious programs being carried out by a numberof Federal agencies to combat domestic hungerand nutrition problems, and the need for dataon the magnitude and distributions of theseproblems in the total U.S. population promptedthe Department of Health, Education, andWelfare in 1969 to establish a continuing na-tional nutrition surveillance system under theauthority of the 1956 Act for the purposes ofmeasuring the nutritional status of the U.S.population and monitoring the changes overtime. The task of developing a plan to carry outthe new program was assigned to NCHS.

ROLE OF THE NCHS TASK FORCE

Charge to the Task Force

A task force was selected by NCHS of staffmembers representing a variety of disciplines toformulate a scientifically sound plan to collect,analyze; and disseminate the data required bythe Department. The charge to the task forcewas to develop a plan specifically to attain thefollowing goals :

“1. The development and implementation of a survey designwhich will permit the use of health data as an objectivetest of programs to improve nutritional status.”

“2. A continuing monitoring of national nutritional statusand related health problems so that the evaluation oftrends and progress over time will be possible and so thatwe will have a beprogram resources.”

tter basis for allocation of scarce

The task force sought the advice of adminis-trators of programs related to nutrition and ofpracticing nutritionists on specific topics vital tothe efficient design of a plan to attain thesegoals. A work paper developed by members ofthe task force which posed specific questionsand also proposed alternative logisticalapproaches in implementing the survey wassubmitted to experts in the field of nutrition.Circulation of the work paper to the RegionalMedical Program Services Advisory Committeeon Nutrit ion and Health and to the Inter-government-Agency Advisory Panel for the Na-tional Nutrition Surveillance System elicited avariety of responses; interested individuals whoattended The White House Conference on Food,Nutrition, and Health also submitted theiropinions.

The experts concluded that nutritional statuscould not be m.easured an.d interpreted by a fewsimple ind.exes; *a person’s nutritional status is acomplex interrelzkionship of clinical observa-tions, biochemical assessments, anthropometricmeasurements,. sociological and psychologicalevaluations, and dietary intake or patterns.Hence the modus operandi employed to measurenutritional status would require a staff of highlytrained professional teams and operating condi-tions conducive to accurate scientific measure-ment procedures. Opinions on the specific testsrequired and procedures to be followed variedamong the consultants, but the sentiment pre-vailed that standard measurements and proce-dures should be developed and maintained andthat continual efforts should be made to im-prove techniques and the overall quality of thedata.

Some differences of opinion arose concerningwhether nutritional status should be estimatedfor the entire population or, alternatively, foronly those segments that are considered to havehigh risk of poor nutrition. Two major reasonsfor measuring the whole population’s nutritionalstatus are (1) that a lack of knowledge presentlyexists concerning the nature and magnitude ofnutritional problems in the various segments of

the U.S: population; and (2) that a base must beestablished with which to compare specificsegments of the population to determine if agroup is relatively disadvantaged and/or in needof special programs. This concept of relativeadvantage or disadvantage is important withinand between time periods.

The consultants stressed that although plansshould be made to measure the nutritional statusof the whole population special attention shouldbe focused on high-risk groups such as preschoolchildren, women of childbearing ages, pregnantand lactatingincome group

women, the aged, and the low-in general.

Recommendations

The principal points and recommendations ofthe plan developed by the task force which wereapproved by the Department and later imple-mented in whole or in part in the survey were asfollows:

4b 1.

66 2.

The National Nutrition Surveillance Survey (NNSS) willbe a continuing national probability sample survey toprovide baseline distributional and trend data on thenutritional status of the one year of age and over,noninstitutional population of the United States, empha-sizing those segments of the population classified as at orbelow the poverty level, and, to the extent possible,consistent with resources and sample design limitations,women of childbearing ages, pre-school and youngschool children, and the aged. Special emphasis impliesthat these groups will be sampled at rates substantiallyhigher than their proportionate representation in thegeneral population. This survey will produce nationaestimates on the nature and magnitude of nutritionalproblems in the specified population; for some groups,estimates would be available for broad geographicregions. The national data will provide planners at thenational level with a rational basis for allocating scarceresources among the programs designed to combatnutritional deficiencies and related health problems.Trend information available from the survey will indicatethe degree to which national goals are met. A survey ofthis nature will also produce data that relate generalhealth and nutritional variables making it possible tostudy relationships between certain health conditionssuch as obesity, atherosclerosis, and dental caries, on theone hand, and measurements of nutritional status, on theother.”

The Department had proposed that the NNSS beestablished under the authority of the National HealthSurvey Act of 1956. For this and for other reasons, thetask force recommended that it “be made an integralpart of the National Health Survey Program through

2

4b 3. The health survey formed by the amalgamation of theHES and the NNSS will incorporate the general objec-tives of the HES, with the objective of the NNSS toestimate the nutritional status of the noninstitutional,civilian, one year of age and over, U.S. population.During the first cycle of the combined surveys, theobjective of the HES program, Cycle IV, is to identifyhealth care needs of the adult population of the U.S. asperceived by the HES sample subjects and as scien-tifically determined by specified examinations ofselected body systems, and to relate both of these to thehealth care actually received. The NNSS measurement ofnutritional status will consist of a clinical assessment bya physician, a number of biochemical tests on blood and(perhaps) urine, anthropometric measurements, andother measures of nutritional status. . . .”

41 4 The HES-NNSS will be a continuing national probabilitysample survey with a two year cycle consisting of twoannual rounds. That is, each annual sample would berepresentative of the population and additive to otherannual samples. The sample selection procedure will bedesigned to ensure that the low income population issampled at a rate several times its proportionate repre-sentation in the general U.S. population. . . .”

66 5. The combined survey will employ four mobile examina-tion centers staffed by specially trained teams ofexaminers including physicians, nurses, dentists, nutri-tionists, and technicians. HES headquarters staff, appro-priately supplemented for the new program, will provideexpertise on nutritional and medical evaluation andplanning. . . . ”

In addition to the preceding items, the taskforce also recommended the collection of cer-

amalgamation with the existing Health ExaminationSurvey (HES). The interrelationship of nutrition andhealth strongly suggests that the objectives of the twosurveys are similar and compatible. The operationaladvantages associated with the amalgamation make thischoice of instrument logical . . .”

tain data through questionnaires: a householdquestionnaire to obtain general demographicdata, a general medical history questionnaire,and a food programs questionnaire to identifyfamilies and specific family members who par-ticipate in food assistance programs.

BACKGROUND OF HES

Basic Characteristics

As stated previously, the Health ExaminationSurvey (HES) is one of the three differentprograms employed by NCHS to accomplish theobjectives of the National Health Survey. Itcollects data by drawing samples of the civilian,

noninstitutionalized population of the UnitedStates and, by means of medical and dentalexaminations and various tests and measure-ments, undertakes to characterize the popula-tion under study. This is the most accurate wayto obtain definite diagnostic data on the prev-alence of certain medically defined illnesses. It isthe only way to obtain information on unrecog-nized and undiagnosed conditions-in somecases, even nonsymptomatic conditions, It isalso the only way to obtain distributions of thepopulation by a variety of physical, physio-logical, and psychological measurements.

In addition to the data collected by theexamining, measuring, and testing procedures, awide range of other data are collected concern-ing each of the sample persons examined.Therefore, it is not only possible to study themany potential relationships of the examinationfindings to one another but also to investigatethe relationships of the examination findings todemographic or socioeconomic factors.

Any information obtained from the surveythat permits the identification of an examinee isheld in strict confidence and, with the exceptionof clinical and other examination findings thatthe examinee authorizes to be sent to hisphysician, dentist, or other source of medicalcare, is used only by persons engaged in thesurvey for the purposes of the survey.

The overall plan of the Health ExaminationSurvey has been to conduct successive, separateprograms in specific age segments of the civilian,noninstitutionalized U.S. population by meansof medical and dental examinations, tests, andmeasurements. These successive programs, re-ferred to as “cycles,” have had a specific agesegment for the target population and have beenconcerned with certain specified health aspectsof that subpopulation.

All HES cycles have made use of a nationwideprobability sample of the population. Thismethod makes it possible to obtain the desiredinformation efficiently and in such a mannerthat the statistical reliability of results is deter-minable. These factors, together with the factthat the examination and measurement proc-esses are highly standardized and closely con-trolled, enable the results of the surveys todescribe the entire population of the UnitedStates on the basis of relatively small samples.

The approach to each cycle has been neces-sarily multidisciplinary in nature. Each hasdrawn on and combined the talents of statisti-cians, physicians of various specialties, dentists,1psychologists, nurses, educators, sociologists,management specialists, and others. In addition,each cycle has involved interagency collabora-t ion : The U.S . Bureau o f the Census hasparticipated in several phases of the surveys; andother Federal agencies (such as the NationalInstitutes of Health, the Office of Education,and the Children’s Bureau) as well as non-Government agencies (such as schools of publichealth, medical research centers, and surveyresearch agencies) have also advised and assistedthe surveys.

The data collected are from national samnlesof the civilian, noninstitutionalized population.The size of the sample permits some analysis of*the data by broad geographic region, populationdensity groups, or other major subgroups of thetotal sample, but it does not permit analysis bysmaller breakdowns, such as by State. The dataare analyzed and the .findings are made availableto interested persons primarily through thepublication of reports prepared in a form usableby large numbers of consumers of health statis-tics. The reports are limited to objective, scien-tific presentation of the particular findings,including estimated levels of prevalence andrelevant discussion of various observed relation-ships. They do not include discussion of pro-gram implications of the findings, nor do theypresent value judgments concerning their impli-cations to public health. The principal reportsare published by the National Center for HealthStatistics in Vital and Health Statistics, Series 2and 11.

Past HES Programs

During the period from 1959 to 1970, threeseparate survey programs or cycles were con-ducted. Cycle I, conducted between November1959 and December 1962, was directed towardthe civilian, noninstitutionalized U.S. populationbetween the ages of 18 and 79 years, inclusive.The examination was focused on certain chronicdiseases, cardiovascular diseases, arthritis andrheumatism, and diabetes. Also included were adental examination, tests for visual and auditory

acuity, X-rays,blood chemistry

electrocardiographic tracings,tests, and numerous body

measurements. The sample size of Cycle I was7,710 persons, of which 6,672 (86.5 percent)were examined. Details of the plan of thatprogram are described in an earlier report.*Reports of various methodological studies3-l 1and of the findings!*-49 are also available.

The target population of the second cycle ofthe HES consisted of children aged 6 through 11years. That cycle became operational in July1963 and was concluded in December 1965. Theexamination was focused primarily on variousparameters of growth and development, but italso screened for heart disease, congenital abnor-malities, ENT abnormalities, and neuromusculo-skeletal abnormalities. The size of the samplewas 7,417, of which 7,119 (96.0 percent) wereexamined. A detailed report of the plan, opera-tion, and response results of that cycle50 as wellas several methodological reports 51-56 and re-ports of findings57-81 have been published.

The third cycle, conducted between March1966 and March 1970, was concerned withyouths 12-17 years of age, inclusive. As inCycle II, the focus was on growth and develop-ment. A unique feature of the survey was thatthe same sample areas and housing units ofCycle II were used again. Thus, many of theCycle II sample children were also examined inCycle III, providing valuable longitudinal data.Of the total sample size of 7,518 youths, 6,773(90.1 percent) were examined. Of those exam-ined, 2,271 were examined in both cycles. Arenort of the x>lan and oneration of that cvcle8*and several methodological reports83-85 ’ havebeen published. Reports of the findings arebecoming available and will increase rapidly nowthat initial analyses of the data from the secondcycle are virtually completed.

PLANNING FOR THE HEALTHAND NUTRITION EXAMINATION

SURVEY

General

The decision to amalgamate HES and NNSSinto one dual-purpose survey and to distinguishit from the previous HES cycle concept led to itsinformal renaming as the Health and Nutrition

4

Examination Survey (HANES). The planning forthe two components of HANES did not occursimultaneously, although obviously in the laterstages the two pieces had to be merged into asingle workable plan. Prior to. the developmentor suggestion that the HES program be respon-sible for carrying out a nutrition survey, prelimi-nary planning had already begun for HES CycleIV. It can be judged from the beginning andconcluding dates of the three HES cycles dis-cussed in the preceding section that the opera-tion of HES has had to proceed simultaneouslyon three different ‘levels-planning, collection,and analysis.

There are a number of reasons for thisthree-level concept of operation, but a principalone is to avoid complete dismantling and re-building of the field organization between exam-ining phases of successive cycles. It also avoidsthe loss of highly trained field and headquarterspersonnel whose skills are unique and difficultto replace. Thus, while data were being collectedin HES Cycle III, analysis was being made ofCycle II data. At the same time, plans andpreparation were being made for Cycle IV, sothat upon completion of Cycle III, data collec-tion could begin immediately in the new cycle.

The planning which had taken place for CycleIV came to center around the problems ofcurrent and unmet health care needs of the U.S.adult population in response to the widespreadinterest in that aspect of health services. Thetarget population was to be adults, with heaviersampling of older persons and minority groups.Although Cycle I had also studied a sample ofadults, it was planned that the content wouldinclude some major areas of interest not in-cluded in the earlier cycle. Areas under consider-ation were diseases of the pulmonary andurinary systems, thyroid diseases, dermatology,determination of bone mineral density, variousantibody levels, tests of psychological func-tioning, and additional blood chemistries. Repe-tition of certain Cycle I examination proceduresusing modified techniques and obtaining addi-tional specific data was also considered; forexample, data on hearing as a byproduct of aspeech discrimination test, data on the cardio-vascular system obtained by electrocardiogramsrecorded directly onto magnetic tape, newermethods for obtaining blood pressures, and data

on vision from the measurement of intraoculartension and from testing with lenses for acuitycorrection.

Development of HANES General Objectives

The specifications around which the generalobjectives of HANES were developed were basedprimarily upon the principal points and recom-mendations of the NCHS task force with somemodifications, as follows:

1 .

2 .

3 .

4 .

5 .

6 .

7 l

Each cycle of HANES would cover ap-proximately a Z-year period based on asample of about 30,000 persons aged 1through 74 years. (Recommendations ofthe task force had been for the age range2-84 years; however, because of the prob-lem of response associated with the agegroup 75-84 years and the expected smallnumber of such persons that would fallinto the sample, the upper limit wasreduced to 74 years.)All sample persons would receive a specifi-cally designed nutrition examination, witha one-fifth subsample of those aged 25-74years also receiving a more detailed exami-nation based upon the HES component.The sample would be more heavilyweighted on the low-income groups, theolder age groups, preschool children, andwomen of childbearing age.Three mobile examination centers wouldbe used. (This number was reduced fromthat of an earlier plan for four for budget-ary reasons.)The examination time for persons receiv-ing the nutrition examination should- beunder 2 hours; for persons receiving thedetailed examination, under 4 hours.The nutrition component would consist ofa general physicallogical, ophthalmol

examination;.ogical, and den

dermato-tal exam-

inations; body measurements; biochemicalassessments; and dietary intake measures.The HES (detailed) component wouldcarry out the general objectives of HES,with emphasis on health care needs. Thedata collection period would encompasstwo HANES cycles because of a smalleryield of examinees due to subsampling.

5

,

8. Demographic, health history, health careneeds, and dietary data and’ data onparticipation in food programs would beobtained through the use of question-naires.

Development of Some Specific Areas ofthe Detailed Component

As stated earlier, much of the preliminaryplanning for the detailed component had alreadybeen performed in preparing for HES Cycle IV.It was decided that the central purpose shouldremain as planned-to obtain data concerningthe current and unmet health care needs ofpersons in the age group 25-74 years. It wasconsidered that these data could best be ob-tained by ascertaining the health needs asself-perceived by the individuals examined, andas professionally and scientifically determinedby the survey’s examination and tests. Informa-tion obtained through the use of questionnaireswould include data on what health care has beenreceived; and even though the examinationwould provide some indications of the healthcare needed, it would not identify specificallythe entire range of unmet medical needs. Besidesidentifying health needs, some account would betaken of the importance of these needs byconsidering what effect they have had or willhave on the individual’s functioning. The result-ant data on health needs would provide anenormous amount of information, heretoforeunavailable, that could be studied in relation tothe health care that had been, is being, or shouldbe received. It would not, however, provide anytotal systems-analysis-type assessment of thepresent overall functioning of the medical caresystem.

It was recognized that the detailed examina-tion component could not cover all aspects of anindividual’s health. The approach used was toselect a number of index conditions that couldbe targets of a single, time-limited examinationand that could be related to the symptoms andindividually felt health needs of the sampleperson. These target conditions were chosen toprovide needed prevalence information on con-ditions related to some of the commonestsymptoms experienced by patients. Thus, theinterview and questionnaire data would yield

information concerning the sample person’sexperiencing the following seven sets of symp-toms or complaints: shortness of breath, jointpain, chest pain, skin problems, dental diffi-culties,, trouble with hearing, and visual disturb-ance. In addition, the sample person would beasked about any other kinds of symptoms,complaints, or health troubles he may have.Through the detailed examination it would bepossible to establish the presence of chronicpulmonary disease; chronic disabling arthritis ofthe hip, knee, and other joints; specific derma-tological disease; dental and oral conditions;cardiovascular disease (including peripheral vas-cular disease) ; thyroid abnormality; auditoryacuity; correctable level of visual acuity; as wellas the presence of ocular hypertension and otherocular conditions. The above concert is dis-played as providing the information toenter thecells of the matrix shown in figure 1.

As a result of having to reshape the detailedcomponent for HANES, it was necessary to dropcertain elements planned for HES Cycle IV. Themost important of these elements was thealmost total elimination from the first HANESprogram of any significant separate psycholog-ical examination. It was decided to proceed withfurther developmental work in this area over the

Health needs from HANESexamination and test

9

a b C

Self-perceived needs and action

NoPathology Pathology

pathologynoted but found,

foundno treatment treatment

needed indicated

1. No problem or relevant complaint d d X- - - - - - - - - - - - - - - - - -

2. Condition not seen as needingany medical attention d d X

m - - - - - - - - - v - - - - - - -

3. Condition seen as needing medicalattention but not under treatment X

---------------e-m

4. Condition under treatment byM.D., D.D.S., etc.

d d

NOTE: Index conditions for the matrix.-Cardiovasculardisease; chronic respiratory disease; disabling arthritis of hip,back, knee; dental and oral conditions; dermatological disease;ophthalmologic conditions (including visual acuity); hearing loss;psychological problem.

KEY:

d = No unmet needX = Unmet need

Figure 1. Conceptualization of the Health and Nutrition Exam-ination Survey approach to measurement of unmet healthneeds.

6

next 2 years in order to carry out necessaryvalidation, calibration, and developmental workon the full battery of instruments developed bythe psychological advisor and on other possiblemodifications of a psychological battery. In thiswork it was planned to collaborate with personsin the National Institute of Mental Health andwith selected outside experts. One instrument,developed from a small portion of the proposedbattery, was to be used in HANES. The final formdeveloped for use in the detailed portion of thesurvey was the General Well-Being Question-naire. The questionnaire is intended to fillseveral purposes: to serve as an indicator ofoverall adjustment; to provide subscales ofadjustment, such as emotional stability andcontrol, depressive mood, worry or concernabout health, and tension; to collect informationon psychological services needed and prevalenceof use of some services; and to serve as a“moderator variable” or control in the statisticalassessment of unmet medical needs.

Considerable interest in further national andregional information on hearing sensitivity topure tone and to speech among adults had beenindicated by the staffs of the National Instituteof Neurological Diseases and Stroke (NINDS)and the National Bureau of Standards, bymembers of the American Academy of Ophthal-mology and Otolaryngology, and by otherexperts in related fields. Data were needed in thedevelopment of standards for bone conductionthresholds and for more precise determinationof the relationship of bone to air-conductionthresholds and to speech discrimination. Thesewould provide a more valid base or normalstandard than is now available for use in thediagnosis of specific conditions and for assessingthe functional implication of hearing impair-

.ment. It was later decided, however, that thespeech-testing portion be postponed until thesecond HANES program because of difficultiesin preparing a reliable, valid test on tape thatcould be administered within certain allowabletime limits of the examination and becausenecessary pretesting requirements, could not besatisfactorily completed in time for implementa-tion into the first HANES program.

In planning the specific components of theaudiometry testing, including the instrumenta-tion and methods to be used, the following

experts were consulted individually and in adhoc meetings: Dr. Eldon L. Eagles, AssociateDirector, NINDS; Dr. Hallowell Davis, CentralInstitute for the Deaf; Dr. Ralph Naunton andDr. Stanley Zerlin, University of Chicago; Dr.John W. Black, Ohio State University; Dr. PaulLaBenz, NINDS; Dr. Leo Doerfler, University ofPittsburgh;.Dr. Edith L. R. Corliss and Mrs. PearlWeissler, National Bureau of Standards; Dr.Hayes Newby and Dr. Donald Causey, Univer-sity of Maryland; Dr. Sadanand Singh, HowardUniversity; and Mr. Kenneth Stewart, Universityof Pittsburgh.

A s p e c i f i c p r o t o c o l f o r a i r - a n d bone-conduction testing for use in the survey wasdeveloped by Dr. LaBenz, who also advised onall aspects of this part of the examination andtrained the audiometric technicians. Initial train-ing of the HANES examiners was given by Dr.Mark Doudna, University of Maryland. The fieldprotocol related to the instrumentation andacoustical environment, including the weeklyfield calibration of the instrument used, themonthly environmental noise surveys, and thedaily physical checks of the instruments,* wasdeveloped by Mr. Kenneth Stewart, in charge ofthe Acoustics Laboratories, University of Pitts-burgh, who also advised on all related aspectsduring the survey.

The vision examination of adults in HESCycle I had consisted of testing with andwithout glasses for near and distance visualacuity. In addition, there had been a limitedfunduscopic. examination of the eyes, but pri-marily for the determination of hypertension. Inthe planning for HES Cycle IV, it was recom-mended that a glaucoma screening procedure,probably involving tonometry (Schi#z-Sklar)and visual field testing, be added to the examina-tion. With the increased interest in more defini-tive information on visual problems and eyeconditions among the general population, how-ever, Dr. Carl Kupfer, Director, National EyeInstitute (NEI), ‘proposed collaboration and pro-vision for logistical and technical backing for amuch more ambitious program. NE1 was partic-ularly interested in determining the prevalenceand distribution of specific eye diseases andrelated conditions throughout the United Statesfor use in setting goals and priorities for futureemphasis in the field of ophthalmology. Consist-

ent with the overall objectives of the survey, an an oscillatory method of determining respiratoryevaluation of treatment needs was incorporated resistance. In the pilot work on these variousinto the examination. tests, it was determined that the body box and

Two ophthalmologists from NEI, Drs. James the oscillatory method were unsuitable under1 P. Ganley and Arthur F. Garcia, developed the the operational conditions of the survey.

examination form and standardized protocol for Specific body-size measurements were to bethe ophthalmic examination and were respon- included for a variety of reasons: the measure-sible for training and recruiting the examining ments would provide a minimum of informationophthalmologists. to be compared with some similar measurements

The main impetus for inclusion of an exten- taken in Cycle I, height and weight data to besive dermatology component for HES Cycle IV related to pulmonary function, and other meas-came from within the profession itself through urements to correlate with joint disease.the National Program for Dermatology. The The arthritis examination of Cycle I hadexamination protocol, including an assessment included determinations of the prevalence ofof treatment needs, was worked out by Dr. rheumatoid arthritis and osteoarthritis basedMarie-Louise Johnson, Division of Dermatology, primarily on X-rays of the hands and feet. It wasDartmouth Medical School, and Director of the decided in the Cycle IV planning that , inData Collection Unit for the National Program. keeping with the overall objectives, more empha-She also assumed responsibility for recruiting sis should be placed on the determination of theand training the examining dermatologists re- amount of disability and on evaluating thequired to carry out this highly specialized part medical care that had been received or wasof the examination. needed. Arthritis of the hip, knee, and lower

Because of the considerable interest expressed back was to be stressed since they are the causeby representatives of the American Heart Asso- of much disability from this disease. Two X-raysciation, the National Heart and Lung Institute, were to be taken, one of the hips and sacroiliacseveral areas of the Health Services and Mental region, and the other of the knees. In addition,Health Administration, and others, it was deter-mined that Cycle IV should include plans toobtain data for the determination of the totalprevalence and the distribution of cardiovasculardisease, primarily hypertension and heartdisease. Although collection of such data hadbeen a maj or part 0f the Cycle I adult surveyprogram, it was felt that there was a need forfurther and more up-to-date information in thisarea, particularly with respect to the extent ofneed for medical care and normative electro-cardiographic data.

The Cycle I survey of adults did not includean evaluation of pulmonary function. Sincechronic pulmonary disease is second only toheart disease as a cause of disability in the adultpopulation, it was recommended that the prev-alence and distribution of this disease be deter-

the range of motion of the hips and knees was tobe determined by using goniometers. Develop-mental work w& performed to determine the-necessary modifications of X-ray equipment andsettings required to produce the proper qualityof hip X-ray for a measurement of leg length at6 feet instead of the usual 3 feet. This measure-ment was to be obtained to provide moreinformation on its relationship - to unilateralosteoarthritis of the hip and the possible allevia-tion of disability from this condition.

Much of the work in the development of anarthritis history and of the content of theexamination was performed in collaborationwith Dr. John Decker, Chief, Arthritis andRheumatism Branch, National Institute ofArthritis and Metabolic Diseases (NIAMD).

Osteoporosis, a fairly common condition,mined. For this purpose, a large battery of tests especially in postmenopausal women, was one ofwas considered originally in the Cycle IV plan- the other skeletal conditions considered forning. Tests considered included spirometry, inclusion in Cycle IV. The initial choice forsingle-breath carbon monoxide tests for pulmo- measuring bone density was the photon absorp-nary diffusion, PA and lateral X-rays of the tion method of Cameron. This method, althoughchest to determinemethod, the body

lungbox

volume by a planimeteror plethysmograph, and

quite accurate, required the use of a radio-isotope as a radiation source. After thorough

8

investigation, however, it was found thatarrangements for State and national licensing forradioisotopes posed extremely difficult barriersto the use of this method. On the recommenda-tion of Dr. G. Donald Whedon, Director,NIAMD, it was decided to use Xiray densi-tometry instead. The X-ray to be used was oneof the hand-wrist with the density determinedon the fifth finger and the radius. A contractwas arranged with Dr. George Vose of TexasWoman’s University to process the X-ray films.Bone density was to be determined bY amicrodensitometer coupled to a computer.

Previous cycles had’ included some bio-chemical and hematological determinations, e.g.,cholesterol, uric acid, protein-bound iodine,hemoglobin, and serologic tests for syphilis. Thefinal selection of all tests to be included in thedetailed component of the examination was notmade until the tests required for the nutritioncomponent had been considered since many ofthe tests could serve both.

In the selection of tests for the total HANES,a large number of individual hematologists,biochemists, nutritionists, and clinical pathol-ogists were consulted. Particularly involved werethe committee of clinical pathologists of theNational Academy of Sciences and the labora-tory division of the Center for Disease Control(CDC). The final selection of tests to be in-cluded in HANES was based not only uponmeeting the requirements of the two compo-nents but also upon other criteria. These criteriawere that the test be in.general widespread use,adaptable to shipping over long distances, thatthe number of abnormal values expected in ageneral population would not be extremelysmall, and that a period of fasting would not’berequired before taking a sample of blood fromthe examinee. Some recommended tests thatmet the above criteria, such as tests for immuno-globulins, had to be eliminated because of cost.

The hematological determinations finallydecided upon for all sample persons were hema-tocrit, hemoglobin, red cell count, white cellcount, and sedimentation rate. The nutritionalbiochemistry component would be performedon appropriate specimens of serum or plasmaand would consist of determinations for vita-mins A and C, magnesium, serum iron, iron-binding capacity , serum folates, total pr‘otein

and albumin, and cholesterol. In addition,further biochemical determinations were to bemade on blood samples from examinees in thedetailed component. These were: total bilirubin,SGOT, alkaline phosphatase, uric acid, calcium,phosphorus, a T-3 index by resin uptake, a T-4by column, and serum antibody titers for polioI, II, and III, measles, rubella, diphtheria, teta-nus, and amebiasis. Also, a differential white cellcount was to be made on blood sme s ofdetailed examinees. i/

In addition to providing advice and dsistancein the planning operation, the CDC, Dr. DavidSencer, Director, established a Nut ition Lab-ioratory where, under a reimbursable arrange-ment, essentially all the HANES laboratorywork would be performed. CDC was also instru-mental in developing procedures for obtainingand shipping the specimens and quality controlprocedures to be used in the field.

A casual urine specimen for testing pH,albumin, glucose, and hematuria using reagentstrips was included in Cycle IV plans. The samespecimen would be acidified, frozen, and sent toCDC for determinations of creatinine, thiamine,riboflavin, and iodine. The last three would berelated to creatinine (thiamine per gram ofcreatinine). In addition, collection of a timedurine specimen was seriously considered for thepurpose of a creatinine clearance test, but laterpilot work demonstrated that this procedure wastoo difficult to administer in conjunction withother parts of the proposed examination.

Examples of other procedures that had beenconsidered for inclusion in Cycle IV but thatwere not finally included in later pilot testingwere the use of a special X-ray procedure to

-demonstrate coronary artery calcification, ballis-tocardiogram for evaluating cardiac function,use of ultrasound to determine liver size, semi-automated procedures for taking blood pressure,and the use of tonography for glaucoma evalua-tion. Factors such as cost and difficulty ofadaptation to field operation, including time foradministration and stage of development amongother considerations, were instrumental in theirexclusion.

In keeping with the overall objective of HESCycle IV, health care needs, it was planned toinclude in the dental examination an assessmentof the needs for dental care. The assessment was

’ II.

. . .

to be accomplished by having the examiningdentist apply his clinical judgment above thecustomary index assessments of oral health andarrive at an estimate of unmet treatment needs.A large portion of the proposed dental examina-tion would also provide data for comparisonwith that of Cycle I.

A final phase of planning for the dentalexamination resulted in the inclusion of anenamel biopsy procedure. This newly developedprocedure is a simple and rapid technique forremoving a microscopic layer of enamel from asmall area of a tooth for laboratory analysis forfluoride content. The result expressed in partsper million serves as an estimate of the individ-ual’s exposure to and absorption of fluoride.Besides the obvious interest in this finding fromthe large national sample, comparisons will bemade with the number of cavities and fillings on

_ selected surfaces where it is thought that fluo-ride has its most striking impact.

Early plans for HES Cycle IV led to construc-tion of a rather detailed medical history ques-tionnaire that was to.be complemented by sevensupplements designed to elicit additional infor-mation on positive responses related to certaintarget chronic diseases. With the inclusion of thenutrition component, drastic cuts in both thenumber and size of the questionnaires produceda package that consisted of the Medical HistoryQuestionnaire, ages 12-74, and the General Medi-cal History Supplement, ages 25-74, to be com-pleted by all of the detailed examination subsetof sample persons, along with a possible threesupplem ents-Supplement A, Arthritis; Supple-ment B, Respiratory; and Supplement C, Cardio-vascular-to be completed only as indicated bYpositive response to screening questions on theGeneral Medical History Supplement. Theseforms were the result of extremely widespreadconsultation in connection with the planning ofthe content of the detailed examination. Theydrew extensively on the experience of othersurveys including, for example, the respiratorystudies carried out by the British Council onMedical Research. Specific advice on matters ofdirect concern to them came from a number ofinstitutes within the National Institutes ofHealth (e.g., NIAMD), from various outsideprofessional groups, and from some individualexperts in the areas. The development and

consultation work was carried out largely by themedical staff within the advisory group of theDivision of Health Examination Statist ics(DHES).

The Health Care Needs Questionnaire wasalso constructed to obtain information on theindividual’s perception of his own health careneeds along with information concerning actionsrelated to obtaining health care. Medical-advisorswithin DHES worked closely with experts at theSchool of Public Health at Johns HopkinsUniversity Medical School in developing thequestionnaire. Consultation with staff membersof the Medical Sciences Division of the NationalAcademy of Sciences-National Research Councilprovided corroboration of the approach used.Members of other divisions within NCHS,notably the Health Interview Statistics Divisionand the Health Resources Statistics Division,were very helpful in advising on questionnairewording.

The household questionnaire developed for-Cycle IV had to undergo very little changeexcept for the addition of several questionsregarding housing facilities. This questionnaire,which was developed jointly by members of theDHES staff and the Bureau of the Censuspersonnel, is the basic source document fordemographic and socioeconomic data of thepopulation sample and also serves in the finalstage of sample selection. In addition to infor-mation on the age, race, and sex of all householdmembers, a variety of other data is obtained-family income, marital status, ethnic back-ground, education, work status, occupation andindustry, and a series of questions concerninghousing characteristics.

Development of Specific Areas of theNutrition Component

As in planning for the detailed component ofHANES,- the final content of the -nutritioncomponent was made -only after extensive con-sultation with manv agencies and individuals.Personnel of CDC provided very valuable adviceand assistance, particularly in the developmentof the nature of the blood analyses to beperformed and in planning for the necessaryfacilities for the extensive laboratory work. Totake maximum benefit of the experience of the

Ten-State Nutrition Survey, several especiallyknowledgeable individuals on the staff of theNutrition Program within CDC who had playedan active role in that earlier survey, along withother staff members from the Nutrition Pro-gram, participated with NCHS staff in a workconference directed specifically to the problemof planning HANES. A number of consultationswere held with personnel of the U.S. Depart-ment of Agriculture, in both the AgriculturalResearch Service and the Food and NutritionService, who provided valuable input and experthelp, as did nutrition specialists within theMaternal and Child Health Service and theIndian Health Service of HSMHA. Assistancewas also provided by the Office of EconomicOpportunity, the Office of Education, and theFood and Drug Administration, to list but a fewof the major contributors.

The Food Programs Questionnaire, developedin collaboration with the Department of Agricul-ture, was designed to elicit information aboutfamily participation in food stamp and commod-ity programs and the participation of youngsample persons in school lunch, breakfast, andmilk programs. Data from the questionnaire willbe related to the various examination findings aswell as to the hematological and blood chem-istry results. Information about participation infood programs such as school lunch programswill be related to dietary adequacy of partici-pants.

The medical, dental, and nutrition advisorswithin DHES played an important role in thedevelopment of the medical history question-naires for the nutrition component. Many of theoutside consultants and experts in the area wereasked to and did review drafts of the forms.Because of the wide age span (l-74) of thepopulation being surveyed, three different ques-tionnaires were developed relative to threedifferent age segments-l-5, 6-l 1, and 12-74.The first two questionnaires, while essentiallyalike, differ in content primarily because of thefactors of question-recall validity and prev-alence. The questionnaire for age group 12-74 isoriented more toward information associatedwith the dental and medical conditions that aremore prevalent in that age group.

One of the three essential parts in obtaining afull nutritional profile of individuals or groups is

some assessment of food intake. While foodconsumption data alone are not a valid measureof nutrition, such data help to interpret clinicaland biochemical findings. Information aboutdietary intakes is useful also for such purposes ascharacterizing food preparation practices, identi-fying sources of nutrients, and determining thetypes of food consumed at different seasons andin different geographic locations.

Although a variety of methods have beendeveloped during the past 40 years to estimatefood intakes as part of nutritional status orepidemiological studies, a number of practicalconsiderations influenced the selection of the24.hour recall and food frequency methods overother methods for HANES. Principal amongthese considerations were the nature of thedata-collecting process and simplicity of the twomethods, the fact that data would be analyzedby groups and not by individuals, the limitationsof interviewing time, the availability of staff andtraining facilities, and the recruitment potentialfor interviewers. In addition, the 24-hour recallmethod and 7-day food record have.been com-pared by several researchers who concluded thatfor estimating the intakes of population groups,the two methods tend to be interchangeable.

Because of the large sample size (30,000), it isanticipated that subgroups, such - as age, sex,income, education, family size, health status,and geographic area, will be large enough foranalysis to indicate groups of persons where it isobvious that steps need to be taken to improvetheir diets.

Another necessary part in assessing thecurrent nutritional status of an individual is aclinical appraisal that includes general assess-ment by a trained physician looking especiallyfor stigmata of malnutrition and includes takingcertain anthropometric measurements. The clini-cal examination was prepared by the NutritionAdvisor to DHES in conformance with acceptedcriteria for such examinations. The assessmentconsists essentially of an inspection by thephysician of the head and neck for the presenceor absence of various signs which are foundassociated with possible nutritional deficiencies.In addition, it was considered essential that theneck be inspected and palpated for any visible orpalpable enlargement of the thyroid gland; thatthe abdomen be inspected and palpated, and the

11 ’

1.,,

liver size be determined by percussion in allpersons over 25; that the deep tendon reflexesbe checked; that the musculoskeletal system beobserved for any marked deformities due topossible rickets; and that there be an inspectionand palpation of the skin for possible signssuggesting nutritional problems.

It was felt that while height and weight werethe most simple measurements for assessingnutritional status, the information derived fromthem would be rather crude. Additional meas-urements were needed for more re finedaccurate information on nutritional statu

and.s in

relation to body build and composition. Amongthe large variety of measurements considered,eight were finally selected in addition to heightand weight. These were triceps and subscapularskinfolds to provide a measure of the presenceor absence of obesity; triceps skinfold that,

* when subtracted from the upper arm girth,provides an approximate measure of musclemass; elbow and bitrochanteric breadth to pro-vide more information on body build, partic-ularly on the bony structure; sitting height toprovide a comparison of trunk length in variousage-sex groups of different ethnic and socio-economic backgrounds; and head and chestcircumferences of children 1 to 7 years of ageonly, as a source of useful information throughtheir interrelation as possible indicators of earlyprotein-calorie deficiency in that age group.

The dental examination planned for thedetailed component was expanded in severalways to meet the objectives of the nutritioncomponent. Among these were a more detailedassessment of the gums for manifestations ofsystemic nutritional deficiencies and diseases,and a series of questions about chewing foods todetermine the relationship between dietary in-take and dental conditions.

Pilot Testing

The first of the pilot test operations wasperformed in Georgetown, Delaware, from April27 through May 23, 1970, immediately follow-ing the completion of Cycle III . The mainemphasis of this test was on the detailedexamination component of the survey, whichhad been in the planning stages for 2 years.Testing was intended to determine, among other

things, the feasibility and acceptability of newexamination procedures such as goniom.etry,pulmonary function tests, tuberculin testing,knee X-rays, thyroid grading, and examinationrecording forms. Other equally important por-tions of the survey evaluated in this work werequestionnaires, interviewing techniques, adminis-trative areas, and. the time factors involved in allaspects of the work. A total of 70 persons aged25-74 were examined during the Delaware work.

Much was learned from the first test. Manyrevisions were required in the coding andsequencing of the questionnaires, and a generalappraisal was made of the reliability and rele-vance of items contained in the questionnaires.Problems associated with interviewing wereidentified, and the amount of time required forhousehold interviewing was obtained. In theexamination portion, the procedures that weretried proved feasible and workable with fewexceptions. Examples of the exceptions includeevaluation of the two types of thyroid gradingwhich had to be deferred until further testingcould be performed because of the small numberof persons with enlarged thyroid glands. A newpiece of equipment to measure respiratory resist-ance presented difficulties in obtaining validreadings. Further trials were also indicated in thearea of tuberculin skin testing.

Further pretest w o r k f o l l o w e d a t t h eResearch Triangle Park, North Carolina, betweenJune 10 and October 6, 1970. With the exce-p-tion of a small number of persons, all partici-pants were recruited fi;om Government offices inthe Research Triangle Park area. A total of 428persons were examined. As in Delaware, thispretest work was focused on the detailed com-ponent of the survey. With approval from theOffice of Management and Budget for thenutritioIn component which occurred during thelast days of the test work at this location, it waspossible to examine 25 individuals using theprocedures and forms developed for that portionof the survey. These examinations were veryimportant since the examinees were children inthe age group 1-6 years, and they providedexperience for the first time in the history of theHealth Examination Survey with examinees ofthis age. The detailed examination componentwas conducted employing certain refinements ofprocedures initially tried out in Delaware.

12

Several new procedures were tested, however,including laboratory techniques for red andwhite cell counts, hemoglobin, smears, and adermatology examination. During the time ofthis pretest, special training in laboratory proce-dures was provided to the technicians by expertsfrom the Center for Disease Control.

The third phase of the pretest work wasconducted at the same location, but using aprobability sample of persons from DurhamCounty, North Carolina. Several important partsof the overall survey were tested for the-firsttime during this phase. The household question-naire was one of these. This questionnaire,administered by Bureau of the Census inter-viewers, is the first contact with a samplehousehold. It establishes the household composi-tion and obtains certain demographic infor-mation about the households and the individualswho live in them. It is also essential in theselection of persons to be included in thesample. Sample selection .procedures, which arecomplicated by the dual concept of the surveyand the use of different sampling ratios forvarious age-sex groups, were also performed forthe first time during this test. -

The sample selection provided individuals forboth the nutrition and detailed examinationcomponen ts. The examination sessions thus in-cluded both types of examinees, permitting thetesting of procedures and questionnaires of thetwo camp onents taken together. The examina-tion also included several procedures not triedpreviously, such as the ophthalmological exam-ination, collection of urine specimens, and theuse of a lung analyzer machine. The equipmentfor measuring respiratory resistance continuedto present difficulties in operation and was laterexcluded from the examination plan. One of thefour medical history supplements to the detailedexamination, Supplement D, Gastrointestinal,was also dropped to reduce examination timebecause it was relatively less important to thesurvey than were the others.

In addition to the examination and question-naire portions of the survey, administrativeprocedures, consisting principally of recordkeeping, scheduling and the rescheduling ofexaminees with broken appointments, schoolcontacts, and transportation of examinees, werealso tested.

A total of 274 sample persons were identifiedfor this pretest. Of these, 204 were to receivethe nutrition portion and the remaining 70 thedetailed examinations. At the end of operations,71 percent of those in the nutrition componentand 74 percent in the detailed component hadbeen examined.

The fourth phase of pretest work was con-cluded December 17, 1970, in Winston-Salem,North Carolina. Prior to this phase, all forms andquestionnaires were reviewed, and many changeswere made in wording, sequence, and so forth.Questions not felt to have sufficient validity orrelevance to the survey were deleted. TheBureau of the Census questionnaire also under-went some changes, primarily on the question ofsources of income where less detail was requiredof nonpoverty persons. No substantive additionswere made to any of the questionnaires or to theexamination procedures. The test of visual fieldsin the ophthalmology examination was excludedbecause of problems encountered in its adminis-tration. Various sequencing procedures for moreefficient examinee flow in the examinationcenter were also tested.

During these pretests, new personnel to staff asecond caravan were hired for the varyingaspects ofnurses, co

the survey (interviewers, technicians,ordinators, and administrative per-

sonnel), and the operation of these pretests wasa part of their training. The nutrition inter-viewers used during the preceding test had beenhired as a temporary arrangement pending therecruitment of permanent personnel. Permanentnutritionists were recruited in time for the lasttest, and extensive training in this area was givenduring the Winston-Salem pretest.

Following the Durham pretest, a completepackage of the HANES material, including adescription of the program and sample design,was distributed to the panel of advisors to theNational Center for Health Statistics and others,asking for any comments they might care tomake concerning the program or any of thespecific procedures and forms involved. Replieswere received fromadvisors, and their

morecorn

than a score of thesements were carefully

considered and, in many cases, taken intoaccount in the final version of the HANES plan.

The final “dress rehearsal” took place in theBaltimore metropolitan area. The date for the

13

opening phase, household interviewing by theBureau of the Census, was February 8, 197 1,with examinations between March 1 and April16. Prior to this, however, field employeesparticipated in further formal and informaltraining sessions in such areas or procedures asinterviewing techniques, questionnaire adminis-tration, laboratory, audiometry, dietary inter-viewing, and coding amounts and types of food.They also received an orientation by key staffmembers of NCHS and DHES. A total of 573sample persons were included in this final test.Of these, 460 were to receive the nutrition

Recipien ts Questionnaires

All households in the sample . . . . . . . . . . . . . .

All households containing one or moresample persons . . . . . . . . . . . . . . . . . . . .

portion only, and 113 the detailed examination.The overall examination rate was 66 percent,with 68 percent of those in the nutrit ioncomponent and 56 percent in the detailed partbeing examined.

Summary

The following summary shows the question-naires, procedures, and measurements of thesurvey, by recipient, as it proceeded through thefirst 35 of the total 65 primary sampling units:

Household Questionnaire

Food Programs Questionnaire

Recipien ts

‘General Medical History, Ages l-5General Medical History, Ages 6-11

All sample persons . . . . . . . . . . . . . . . . . . . { General Medical History, Ages 12-74Dietary Intake, 24-Hour RecallDietary Intake, Food Frequency

General Medical History Supplement, Ages 25-74Supplement A, Arthritis; Supplement B, Respira-

Additional for all sample persons in thedetailed component . . . . . . . . . . . . . . . . . (

tory; Supplement C, Cardiovascular. Supple-ments A, B, and C depend on certain positiveresponses in other history questionnaires

Health Care Needs QuestionnaireGeneral Well-Being Questionnaire

Examination procedures andmeasurements

IGeneral medical examinationDental examinationDermatological examinationOphthalmic examinationAnthropometric measurementsHand-wrist X-rays (ages l-l 7 only)

All sample persons . . . . . . . . . . . . . . . . . . . < Laboratory determinations:Hemoglobin Serum ironHematocrit Iron binding capacityRed cell count Serum folatesWhite cell count CholesterolSedimentation rate Glucose qualitative (urine)MCV Albumin qualitative

\ M C H (urine)

Recipients

All sample persons-Con. . . . . . . . . . .

Additional for all sample persons in thedetailed component . . . . . . . . . . ,

SAMPLE DESIGN

General Plan

. . . . .

The design of the sample, which is expectedto yield approximately -30,000 sample personsfor HANES, is quite similar in a number of waysto the designs used in the first three HES cycles.Genera l descr ip t ive repor ts o f those de-signs 2,50,82 are available, as is a more detailedreport of the Cycle II sample design.55 NCHSset specifications for the sample design andcarried out some of the steps of drawing thesample. Other steps in the design and sampleselection were performed by the Bureau of theCensus under a contractual arrangement.

The primary similarity of the design to that ofthe HES cyc les i s that it is a multistage,stratified, probability sample of loose clusters ofpersons in land-based segments. The successiveelements dealt with in the process of samplingare primary sampling unit (PSU), census enumer-ation district (ED), segment (a cluster of house-

Examination procedures andmeasurements

MCHC Occult blood qualitativeVitamin A (urine)Vitamin C Creatinine (urine)Magnesium Thiamine (urine)Total protein Riboflavin (urine)

, Albumin Iodine (urine)

(Extended medical examination

I X-rays of chest and major joints (hand-wrist,knee, hip)

Audiometry (air and bone)ElectrocardiographyGoniometrySpirometryPulmonary diffusionTuberculin testLaboratory determinations:

Bilirubin PhosphorusSGOT W.B.C. differential countAlkaline phosphatase Serological tests forUric acid amebiasis, measles,Calcium tetanus, diphtheria,

b Thyroid (T-3, T-4) rubella, polio

holds), household, eligible person, and, finally,sample person.

The HANES design was further complicated,however, by the fact that unlike the precedingcycles it had two distinct examination compo-nents-nutrition and detailed-to be consideredinstead of only one. Similarly, the age range inHANES covers more than one specific agegroup, and emphasis is placed on the low-income -groups, preschool children, women of child-bearing age, and the elderly, because these arethe groups liable to be affected most often bymalnutrition and for which detailed informationis most needed. Therefore, the design had totake into consideration the sample size require-ments for the population subgroups to obtain anoptimum mix for reliability of estimates.

Design Specifications

The sample design of HANES was developedessentially /from a set of specifications that took

’ ‘,”

,

. I,’

into consideration the requirements and limita-tions placed upon it. It was important that therequirements be consistent with survey objec-tives and that the limitations not be so serious asto materially distort the objectives. Specifi-cations considered to be of primary importancewere as follows:

1. The target population would be thecivilian, noninstitutionalized population1-74 years of age residing in the coter-minous United States, with one excep-tion. Because of operational difficultiesexperienced in Cycle I, #all people residingupon any of the reservation lands setaside for the use of American Indianswould be excluded.

2. For the nutrit ion component, broadnational estimates must be made annual-ly, with more detailed estimates pub-lished upon the completion of a 2-yearcycle. For the detailed examination,broad national estimates would be basedon data collected during the 2-year cycle,with more detailed estimates being madeafter the completion of the two succes-sive 2-year cycles.

3. Three mobile examination centers similarto the ones used in earlier cycles of HESwould be used. Thus, with appropriatemodifications, the survey would be basedon administrative and logistical proce-dures that have been developed andproved over a period of more than 10years. A team could examine about 20persons per day; of these, all wouldreceive the nutrition examination, andfour would receive the detailed examina-tion. Other time limitations were a 5-dayworkweek, a loss of 5 weeks per year dueto vacations and holidays, and a loss of 7days per move from one examining loca-tion to another.

4. Operationally, the three caravans couldvisit a maximum of about 65 PSU’s overa period of approximately 2 years.

5. A team must stay at least 3 weeks at astand because of the expense of movingand the need to allow enough time in anarea to give sample persons adequate time ’to be examined. A team cannot stay in an

area longer than 6 weeks because of therequirement to finish the. survey in 2years.

6. Because of the considerations in items3-5 above, there would be a minimumnumber of 300 and a maximum numberof 600 sample persons for each stand.

7. To the extent possible, the schedule ofexamining locations must take account ofclimate.

8. About 20 percent of the sample shouldbe selected from the population classifiedat or below the poverty level. Othergroups of special interest are preschoolchildren, women of childbearing age, andthe aged.

9. The estimates from the survey would beof two kinds: (1) distributions of thepopulation by specified characteristicssuch as height, weight, blood pressure,and selected biochemical determinations;and (2) prevalence in the population ofselected chronic conditions, particularlythose in the arthritic, respiratory, andcardiovascular groups.

10. Maximum target tolerances for samplingvariability would be set for several keystatistics, permitting a general analysis bybroad geographic regions, population sizegroups, and other major subgroups suchas income, race, age, and sex.

11. Data from the 1960 Decennial Censuswould have to be used in the samplingprocedures unt i l 1970 da ta becomeavailable.

Stratification and Selection ofPrimary Sampling Units

The first-stage sample consists of 65 geo-graphic areas, or PSU’s. These have been selectedfrom among approximately 1,900 PSU’s intowhich the geographical territory of the mainlandhas been divided. Each PSU consists of a countyor a small group of contiguous counties. For thepurposes of the design of the Health InterviewSurvey, one of the other major NCHS daMcollection programs, PSU’s are stratified $nto357 groups, and one PSU is selected froJn eachstratum with a probability proportional to its

16

size. For the design of HANES, these 35 7 stratawere collapsed into 40 superstrata.

Fifteen of the superstrata contain only onevery large metropolitan area of more than2,000,OOO population, and thus were choseninto the sample with certainty. The others weregrouped into 25 superstrata on the basis ofgeographic region and population density class,as shown in table 1. Then, using a controlledselection technique to assure representation ofspecified State groups and of classes by rate ofpopulation change, two PSU’s were chosen fromeach of the 25 strata with probability propor-

tional to the PSU’s 1960 population. Thus thesample contains 65 PSU’s. A listing of the PSU’sis given in appendix I, along with definitions ofgeographic region, State groups, and classes byrate of population change,.

To provide the ability to make early nationalestimates, PSU’s were divided randomly intotwo parts. Thus, the survey cycle of 2 years willbe conducted in two rounds. The first roundinvolves 35 locations, including 10 of the largemetropolitan areas and 25 of the smaller, non-certainty PSU’s.

Table 1. Number of self-representing and nonself-representing superstrata for the Health and Nutrition Examination Survey design, by regionand population density class, with average size of superstrata and definitions of population density classes

Region and populationdensity class

l-Total

All regions . . . . . . . . .

Northeast . . . . . . . . . . . .

40-

13

Largest SMSA’s . . . . . . . . . . . . 9Other large SMSA’s . . . . . . . . . . 1Other SMSA’s . . . . . . . . . . . . . 1Non-SMSA, urban . . . . . . . . . . . 1Non-SMSA, rural . . . . . . . . . . . 1

Midwest . . . . . . . . . . . . . IO

Largest SMSA’s . . . . . . . . . . . .Other large SMSA’s . . . . . . . . . .Other SMSA’s . . . . . . . . . . . . .Non-SMSA, urban . . . . . . . . . . .Non-SMSA, rural . . . . . . . . . . .

South . . . . . . . . . . . . . . 8

Largest SMSA’s . . . . . . . . . . . .Other large SMSA’s . . . . . . . . . .Other SMSA’s . . . . . . . . . . . . .Non-SMSA, urban . . . . . . . . . . .Non-SMSA, rural . . . . . . . . . . .

West . . . . . . . . . . . . . . .

Largest SMSA’s . . . . . . . . . . . .Other large SMSA’s . . . . . . . . . . .Other SMSA’s . . . . . . . . . . . . .Non-SMSA, urban . . . . . . . . . . .NonSMSA, rural . . . . . . . . . . .

21122

9

32121

Number of superstrata

Self-representing

Nonself-representing

Average size ofsuperstrata,

1960 populationin millions

Definitions of population density classes

25 4.5 . . .

4 3.4

- 2.7 SMSA population greater than 2.4 million.1 5.2 70% or more of SMSA’s population was urban.1 6.5 Less than 70% of SMSA’s population was urban.1 4.1 40% or more of the population was urban.1 4.8 Less than 40% of the population was urban.

7 4.8

I 3.5 SMSA’s population greater than 3 million.2 4.3 90% or more of SMSA’s population was urban.2 5.3 Less than 90% of SMSA’s population was urban.2 5.9 34% or more of the population was urban.1 6.2 Less than 34% of the population was urban.

8 5.4

2 4.81 5.11 4.52 6.02 5.9

90% or more of SMSA’s population was urban.

Less than 90% of SMSA’s population was urban.30% or more of the population was urban.Less than 30% of the population was urban.

6 4.8

m 3.2 SMSA population greater than 2.5 million.2 6.0 72% or more of SMSA’s population was urban.1 6.2 Less than 72% of SMSA’s population was, urban.2 5.0 36% or more of the population was urban.1 5.0 Less than 36% of the population was urban.

T

I

17

Within PSU Design

For the first 44 HANES stands, only 1960, census data were available for the purpose of

sampling within PSU’s. The remaining 21 standswill use 1970 census data, resulting in a differentprocedure for within-PSU sampling. A principalreason for this change is that socioeconomicchanges within ED’s from the time of the 1960census until the start of HANES precluded asatisfactory method to classify ED’s efficientlyinto poverty and nonpoverty groups. This classi-fication can now be made using the 1970 data.

For the stands using the 1960 data, ED’s ineach PSU were divided into segments of anexpected six housing units each. In urban areaswhere listing units were well defined in 1960,this division was quite accurate since the sam-pling frame was composed of listings that resultedfrom the a1960 census. For ED’s not covered bythe listing books, area sampling was employedand, consequently,ment size occurred.

some variation in the seg-To make the sample repre-

sentative of the current population of theUnited States, the list segments were supple-mented by a sample of housing units that hadbeen constructed since 1960.

Then a systematic sample of segments in eachPSU was selected. The ED’s that fell into thesample were identified and coded into twoeconomic classes. One of the classes, identifiedas the “poverty stratum,” was composed of“current poverty areas” that had been identifiedby the Bureau of the Census in 1970 (pre-1970census) plus other ED’s in the PSU with 1959mean income less than $3,000 (based on the 1960census). The other economic class, identified asthe “nonpoverty stratum,” included all otherED’s not designated as belonging to the povertystratum. A description of how the currentpoverty areas were determined is given inappendix I.

For those sample segments in poverty stratumED’s, all segments were retained in sample. Forthose sample segments in nonpoverty stratumED’s the segments were divided into eightrandom subsamples, and one of the subsampleswas chosen to remain in sample for HANES.One advantagestratum in thisreserve segments in case the sample of persons ina PSU is less than the specified minimum of 300.

from samplingway related to

the nonpovertythe need to have

For the remaining 21 stands, 1970 censusdata will be used. ED’s in each PSU will bedivided into segments of an expected eighthousing units each. As in 1960, the 1970 urbansegments will be more stable in size than areasegments. For each PSU using 1970 materials,ED’s will first be sorted into poverty and ’nonpoverty strata. The proportion of persons inpoverty will be used to determine the povertyand nonpoverty status of each ED. The desig-nated proportion will vary from stand to stand.The *poverty indices will be based on 1969income (1970 census), size of family, sex ofhead of family, age (65 years or under) of headof family, and farm-nonfarm status. The sam-pling rate for selection of segments from the 21s t a n d s w i l l b e c h a n g e d f r o m a poverty-nonpoverty ratio of 8: 1 to a ratio of 2 : 1. Thischange is being made as a result of a study using1970 data by the Bureau of the Census thatindicated a significant decrease in the samplingvariance could be obtained by employing the2: 1 ratio.

Then a systematic sample of segments will bedrawn from each poverty-no npoverty stratum atdifferent rates. As in using 1960 materials,reserve segments from 1970 materials will alsobe provided to meet the specified minimum of300 sample persons per stand.

Selection of Sample Persons

After the sample segments have been identi-fied, a list of all current addresses within thesegment boundaries is made, and the householdsare interviewed to determine the age and sex ofeach household member, as well as other demo-graphic and socioeconomic information requiredfor the survey . If no one can be found at homeafter repeated calls or if the household membersrefuse to be interviewed, the interviewer tries todetermine the household composition fromneighbors.

To identify the sample of people to receivethe nutrition examination, the householdmembers aged l-74 in each segment are listed onthe Sample Selection Worksheet as illustrated infigure 2, with each household in a segmentnumbered serial ly from 1 through K, thenumber of households in the segment. Thehousehold members are listed on the worksheet

18

lntervi ewer - Enter Person Number (Cols. d-i) I I It I AGES

HHSER.NO.

NON-INTERVIEWS

DAYCOM-

PLETED 65-74 45-6425-44

(020-24

(g)

FORM HES-7 (Cycle IV) U.S. DEPARTMENT OF COMMERCE12-22-7 1) BUREAU OF THE CENSUS I 1. H E S

I2, HES STAND N A M E

I3. SEGMENT NO.

STANDNO.

SAMPLE SELECTION WORKSHEET4. INTERVIEWER’S NAME 5. CONTROL NO.

(a)

l-5

(0(=I

6-19

(W(d) MALE !- FEMALE MALE ! FEMALE0)21 I

I2A - WithJEP’s

No.

A - WithoutEP’s

BC

I-A

-I-PersonNo. o fnon EP’s

1 I-

+-----

2

A - WithJEP’s

N o .

A - WithoutEP’s

BC

I-

+----PersonNo. o fnon EP’s

I -

-I-I1Ii -

A - WithJEP’s

N o . -

A - WithoutEP’s

BC

7+--

- I - I-PersonNo. o fnon EP’s

- I -3 - I -I

I

-A---

I

I

1

- -

- *

-I-

-t-

--A - WithJ EP’s

N-o.

A - WithoutEP’s

I-----I-------- ----.

PersonNo. o fnon EP’s

4 -.

- ; -i- I - i-0 A- ,dJt Nb. -.

EP’s In o n EP’s - - - - - -I

B f- -I-C I I Il/4 l/2

WASHINGTON USE ONLY - TOTALS ,TOTAL INTERVIEWED PERSONSEFER 6 5 - 7 4 4 5 - 6 4 M 2 5 - 4 4 ; F 2 5 - 4 4 M 20.24 1

I

F 20-24 6-19 l-5 _

! .

INCL. EXTRAS

HOUSEHOLDSH H’S NONINTERVIEWED HOUSEHOLDSINTERVIEWED

TOTAL TYPE A TYPE B TYPE CNONINTER-VIEWS TOTAL WITH EP’S ! NUMBER EP’S WITHOUT EP’S

II

USCOMM-DC

Figure 2. Sample Selection Worksheet.

in the order of that serial number. The entry any, is usually listed in column 2, while themade on the worksheet corresponds to the children and other household members are listedperson’s column number on the household in succeeding columns of the questionnaire.questionnaire. For example, column 1 is re- Suppose the f irst three households in aserved for the household head. The spouse, if segment have the following age-sex composition:

19

Household serial number

These household members would be recordedon the worksheet as shown in figure 2. Note thattwo persons are not listed; one is 75 and theother is only 6 months of age and, therefore,they are not part of the target population ofpersons l-74 years of age. After the SampleSelection Worksheets are put in order by seg-ment number, a systematic random sample ofeach age-sex group is selected, using the sam-pling rates shown at the bottom of the work-sheets.

There still remains one sampling operation-selection of adults to receive the detailed healthexamination. Overall, about 20 percent of thetotal sample receive the detailed health examina-tion, producing a subsample of about 6,000

1 2

Male, Female,age 45 age 42years years

Male,age 34years

Female,age 27years

Male,age 75years

Female,age 70years

Column number on questionnaire

3

Female,age 13years

Male,age 7years

4

Female,age 16years

Female,age 5years

5

Male,age 3years

6

Female,age 6months

persons. This group IS a subset of the nutritionsample aged 25-74, inclusive; the sampling frameis the nutrition sample designated on the SampleSelection Worksheet. The subsample is chosensystematically after a random start, using thesampling rates shown in table 2.

The sample size varies from one PSU toanother, depending on the PSU population andthe number of persons living in the low-incomeED’s. For the reason stated in the designspecifications, the design provides for a probabil-ity sample of reserve segments to insure thenumber of expected sample persons per PSU. Adeletion procedure is employed as necessary toreduce the number of sample persons per PSU tothe number expected.

Table 2. Subsampling rates and expected sample size by age and sex for the detailed health examination

Both sexes Males Females

Age Expectedsample size

RateExpected

sample sizeRate Expected

t

sample size

. . . 3,150Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I 6,000

t I25-44 years ................................

.:2,000

45-64 years ............... ............... 2,70065-74 years ................................ 1,300

I

I. . . I 2,850

215 1,000 l/5 1,000315 1,300 315 1,400l/4 550 l/4 750

20

OTHER SAMPLING ASPECTS

Sampling Features of the Examination

The sampling aspects of the survey are notrestricted to choosing the sample persons andhaving them participate in the examination. Theconduct of the examination itself has numeroussampling features that should be mentioned.

Examinations will be conducted in 65 differ-ent locations throughout the United States bythree different teams of examination staff. Eachteam for any one location consists of a physi-cian, dermatologist, ophthalmologist, dentist,two health technicians, laboratory technician,and two dietary interviewers. Because of normalpersonnel turnover and the lack of availability ofdermatologists and ophthalmologists for ex-tended periods of time over 1 month or evenshorter duration, the number of different exam-ining staff members employed through the 65locations will be quite large. At the time ofpreparation of this report, it was estimated thatthe total number of individuals in each of theabove positions for all locations would beapproximately 30 physicians, 65 dermatologists,100 ophthalmologists (for 35 stands only, seep. 30), 9 dentists, 15 health technicians, 10laboratory technicians, and 20 dietary inter-viewers. Ideally, assignment of each examinee tothe particular parts of the examination shouldbe random with respect to time, place, andexaminer. Operationally, such assignment isimpossible. Therefore, any peculiarities in theconduct of a part or parts of the examinationprocedures, difficulties with equipment, orchanges in the standards of the laboratoriesdoing blood chemistry analysis may be reflectedin the examinat ionpeculiarity.

f ind ings as a p lace

Stand Sequencing and Scheduling

ofAs in previous cycles of HES, the scheduling

stands for HANES has been deliberatelyarranged so that the North is avoided in winterand the South in summer. Such scheduling is afairly obvious operational necessity as it wouldbe quite impractical to conduct a mobile exami-nation survey of this kind in the Northern States

in the middle of the winter. The schedule ofstands for HANES is shown in table 3.

While this type of scheduling is desirable froman operational point of view, it can producecertain limitations on the examination data. Anycharacteristic under study which may have aseasonal variation will be difficult to interpretby geographic region. For example, to theextent that if persons in all parts of the countryweigh more in winter than in summer, the meanweight of northerners would be underestimatedand that of southerners overestimated. Anotherarea of concern is the effect of season on thequality of data. For example, relatively morepoor diets were reported in the spring than inother seasons in the recent U.S. Department ofAgriculture Food Consumption Suwey. Possi-bilities such as these must be taken into accountin analysis of the data. The limitations resultingfrom such a scheduling arrangement, however,were not considered to be too serious in eitherCycle II or Cycle III. Most of the characteristicsof the examination in the age group 6-17 did notexhibit any marked seasonal variation. Even inCycle I, where the focus of the examination wason chronic conditions in the adult population,seasonal variation was not considered to be aserious problem. This would not be true if theexamination, in any of the cycles, attempted toobtain estimates of conditions such as acuterespiratory disorders.

An important consideration in sequencingstands is economy of operation. Efforts aremade to follow the seasonal pattern describedwith a minimal amount of travel necessary inmoving from one stand to the next by sequenc-ing with regard to geographic proximity. Individ- -ual stand time schedules, featuring the variousoperational aspects involved in conducting theexaminations at a ptrticular stand, are alsorequired in the development of the sequencing.Time allowances are based on the distancebetween stands and, therefore, the time requiredfor movement of the trailers and personnelbetween stands, the time required for censusinterviewing, followup by the health examina-tion representative (HER), trailer setup, staffsetup and dry runs, staff vacation periods, andexaminations. The number of days allotted forexaminations is dependent upon the expectedsample size at a particular stand and is deter-

21

: I. ,

.‘I ’.-

Table 3. Schedule of stand operations by caravan, Health and Nutrition Examination Survey: 1971-73

Dates of field operations

1971:April-May . . . . . . . . . . . .May-June . . . . . . . . . . . . .June-July . . . . . . . . . . . . .July-August . . . . . . . . . . . IAugust-September . . . . . . .September-October . . . . . .October-November . . . . . .November-December . . . . . 1

1972:January-February . . . . . . .February-March . . . . . . . .March-April . . . . . . . . . . .April-May . . . . . . . . . . . .May-June . . . . . . . . . . . . . IJune-July . . . . . . . . . . . . .

- July-August . . . . . . . . . . .August-September . . . . . . .September-October . . . . . .October-November . . . . . .November-December . . . . . IDecember-January . . . . . . .

1973:3January-February . . . . . . .February-March . . . . . . . .March-April . . . . . . . . . . . IApril-May . . . . . . . . . . . .May-June . . . . . . . . . . . . .June-July . . . . . . . . . . . . .July-August . . . . . . . . . . . IAugust-September . . . . . . .September-October . . . . . .October-November . . . . . .November-December . . . . . I

Caravan I

Location

Philadelphia, Pa.Albany, N.Y.

Boston, Mass.

Springfield, Mass.New York, NY.

Cabarrus, N.C.

West Palm Beach, Fla.Barbour, Ala.Columbia, S.C.

New York, N.Y.

Hartford, Conn.Sussex, Del.Milwaukee, Wis.Omaha, Nebr.

Chillicothe, Ohio

Tampa, F la.

Morristown, Tenn.

St. Louis, MO.Fillmore, Minn., Howard, Iowa

Chicago II, Ill.

Columbus, OhioBedford, Pa.

Roanoke, Va.

TStand

number’

13

1 5(5)8

111 14, (11)

(14)

1720

23, (20)1 26, (23)

(26)29, (26)32, (29)35, (32)38, (35)

i 37, (38)(37)

45i 46, (45)

M-6)4952

i 55, (52)(55)5861

i 64, (61)(64)

Caravan I I

Location

Pittsburgh, Pa.Mercer, Pa.

Detroit, Mich.

Bay City, Mich.La Porte, Ind.

Los Angeles, Calif.

Tucson, Ariz.Fresno, Calif.San Francisco, Calif.Ciallum, Wash.Grant, Wash.Boone, IowaWashington, D.C.Oak Hill, W. Va.

Standnumber’

24, (2)

t6, (4)

(6)1012

i

15(15)

182124

27, (24)30, (27)

3334

39, (34)(41, (39)

Dallas, Tex.I

(41)(41)

Globe, Ariz. I 44 New Orleans, La. 43

San Diego, Calif.I {

47(47)

Minneapolis, Minn. 62, (47)Ottertail, Minn. 59Fargo, N. Dak. 56

St. Joseph, Mich. i65,661

(65)Monterey, Cal if. 53, (65)

Los Angeles II, Calif.1

50(50)

T Caravan I I I

Location

2 . . .2 . . .2 . . .Newark, N.J.

Angola, Ind.

Savannah, Ga.

San Antonio, Tex.

Avoyel les, La.

Lamar, Miss.St. Joseph, MO.

Chicago, I I I,

Cleveland, OhioKnoxville, Tenn.

Natchitoches, La.

Statesboro, Ga.

Philadelphia I I, Pa.Chemung-Tioga, N.Y.Scranton, Pa.Providence, R.I.New York V, N.Y.

New York IV, N.Y.

Standnumber ’

2 . . .2 . . .2 . . .1 7

(7)13, (7)

1 16(16)

191 22, (19)

(22)25 .

28, (25)1 31, (28)

(31)9, (31)42, (9)

40, (42)(40)(40)

51365754

63, (541 60, (63)

(60)

‘Stand locations are counties, cities, or towns in which the examination center is located. Sample areas from which examinees are drawn for the standconsist of the PSU’s, which may include several counties. Numbers in parentheses indicate a carryover for the stand number into the last month of themonth group.

2 Not in operation.3Schedule for 1973 is tentative.

mined on the basis of approximately 19 samplepersons per day. Schedules for two stands foreach caravan are shown in table 4.

ADVANCE ARRANGEMENTS

Professional Relations

Before the interviewing or examination proce-dures can be started in a sample area, advancearrangements involving professional relations,public relations, and arrangements for the logis-tical requirements of the survey are necessary.

22

T-he conduct of the survey in any specific areais the responsibility of the Public Health Service(PHS), as distinct from the State or local healthauthorities or others in the area. In addition tonotifying various directors within the regionaloffices of the Department of Health, Education,and Welfare, it is the policy of the. survey tofully acquaint the State and local health author-ities and the medical, dental, and osteopathicprofessional organizations in the States and inthe communities with the HANES objectivesand method of operation. Since school childrenare involved in the survey, the State and local

Table 4. Excerpt from HANES schedule of stands

Operation

Stand number ..................Location ......................Sample size ....................Office setup .....................Census interviewing ...............HER followup ..................Examination center arrival ..........Examination center setup ...........Staff setup and training ............Dry runs ......................Examinations ...................Dismantle-transit ................

Stand number ..................Location ......................Sample size ....................Office setup ....................Census interviewing ...............HER followup ..................Examination center arrival ..........Examination center setup ...........Staff setup and training ............Dry runs ......................Examinations ...................Dismantle-transit ................

-

TCaravan I Caravan I I Caravan I I I

20 21 22Barbour, Ala. Fresno, Calif. Avoyel les, La.600 350 590Feb. 4, Fri. Jan. 28, Fri. Feb. 25, Fri.Feb. 7, Mon. Jan. 31, Mon. Feb. 28, Mon.Feb. 15, Tues. Feb. 8, Tues. Mar. 7, Tues.Feb. 29, Tues. Feb. 15, Tues. Mar. 14, Tues.Mar. 1, Wed. Feb. 16, Wed. Mar. 15, Wed.Mar. 2, Thurs. Feb. 17, Thurs. Mar. 16, Thurs.Mar. 3, Fri. Feb. 18, Fri. Mar. 17, Fri.Mar. 4, Sat.-Apr. 18, Tues. Feb. 19, Sat.-Mar. 17, Fri. Mar. 18, Sat.-Apr. 29, Sat.Apr. 19, Wed. Mar. 20, Mon. May 1, Mon.

23 24 25Columbia, S.C. San Francisco, Calif. Lamar, Miss.300 570 440Mar. 31, Fri. Feb. 25, Fri.. Apr. 7, Fri.Apr. 3, Mon. Feb. 28, Mon. Apr. IO, Mon.Apr. II/ Tues. Mar. 7, Tues. Apr. 25, Tues.Apr. 20, Thurs. Mar. 20, Mon. May 2, Tues.Apr. 21, Fri. Mar. 21, Tues. May 3, Wed.Apr. 22, Sat. Mar. 22, Wed. May 4, Thurs.Apr. 24, Mon. Mar. 23, Thurs. May 5, Fri.Apr. 25, Tues.-May 16, Tues. Mar. 24, Fri.-May 4, Thurs. May 6, Sat.-June 8, Thurs.May 17, Wed. May 5, Fri. June 9, Fri.

officials concerned with public schools are alsoinformed, as are the appropriate local anddiocesan officials of the parochial‘ schools.

.

A letter announcing the survey, the local areasto be sampled, and the dates of survey opera-tions and a brochure describing the survey aremailed 3 to 4 months before examinations arescheduled to begin, to the Health, Education,and Welfare regconal offices, State medical andosteopathic societies, local medical societies, andState and local health departments. A request ismade of the State and local medical andosteopathic societies that an enclosed profes-sional release be printed in their respectiveprofessional journals. The letter to local healthauthorities includes a request to provide HANESwith a listing of local and State health agencies,clinics, and medical services to whom HANESexaminees without present medical resourcesand requiring medical care may be referred, orto whom a report of their examination findingsmay be sent.

regional dental program director for the area.Following the regional director’s recommenda-tions, telephone calls are made and then lettersare sent to the State dental director, whoinforms the State and local dental groups aboutthe survey plans. Occasionally, letters are sentby the HANES Dental Advisor to State and localhealth dental groups on the advice of theregional director.

-Three to 4 weeks after the mailing of the

About 2 months before examinations begin,the HANES Dental Advisor consults the PHS

-.

Caravan schedule

initial letters, the local health authorities arecalled by telephone, and any further questionsabout the survey are answered. Personal visits byHANES medical and dental advisory staff aremade to any health agency or society makingsuch a request.

Public Relations

A general news release explaining the programis prepared for each sample area and is distrib-uted to local news media. The release is timed tocoincide with the start of interviewing by theBureau. of the Census. As a result, local news-

23

,’ .

papers at most of the locations publish itemsconcerning the program. No special effort ismade to have radio and television stationspublicize the survey, but at some locationsmembers of the staff have been interviewed bythese media and film has been taken to betelevised. Under no circumstances, however, arepictures or films taken of any sample examineesince this would be a breach of the promise ofconfidentiality.

Sample households having a mailable address(house or post office box number) are sent an“advance” post card by the Bureau of theCensus several days before their personnel begininterviewing. This card informs the householdmembers that a Bureau of the Census inter-viewer will be calling at their home within thenext few days in connection with a survey being

- conducted in the area for the Public HealthService.

Logistical Arrangements

Four to 6 weeks before the start of a stand, amember of the HANES field staff, the FieldOperations Manager (FOM), visits the samplearea to make physicial arrangements for themobile examination center and the administra-tive office, to meet personally with local healthand school officials, and to initiate the manylogistical actions required for the survey. Selec-tion of a site for the Health Examination Centeris extremely important to the success of thesurvey. The following items are considered:

1 . .

2 .3 .

4 .

5 .

6 .

7 .

Location of sample households and trans-portation arteriesCommunity attitude toward the locationProximity to power, water, and sewerconnectionsReasonable freedom from noise and/orexcessive vibrationAvailability of living accommodations forthe staff within a reasonable distanceAdequate space to accommodate trailersand cars of staffAvailability of office space near the exam-ination site for the administrative office

If the household does contain persons eligiblefor inclusion in the survey, the remaining ques-tions may ‘be asked of any responsible adultmember of the household. A callback is made bythe Census interviewer if a responsible adult isnot present initially. At the end of the interview,the interviewer leaves a thank-you letter signedby the Surgeon General. The interviewer ex-plains that i f anyone in the household isselected, a representative of the Public HealthService will be calling again within a week or soto explain the survey. The interviewer alsoinquires as to the best time of the day for therepresentative to visit the household.

During this visit to the sample area, the FOM The role of the Census interviewers ends afteralso arranges for electricity, water, sewerage, all household questionnaires have been edited by

telephone, and transportation services. Anyother logistical arrangements required before thearrival of the mobile examination center and thestaff are also taken care of at this time. Withinthe time allowed, the FOM makes a courtesyvisit to the local health department and contactsthe superintendents of the larger school districtsto explain the program.

HOUSEHOLD INTERVIEWINGPROCEDURES

Census Interviewing

Trained Bureau of the Census personnel callon all housing units contained in the segments ofthe sample area to determine their householdcomposition and to obtain demographic andother data if the household contains any eligiblepersons aged l-74 years, inclusive. They pave theway for the HES interviewers who subsequentlyvisit the household. Each of the householdsshould have received the advance post card fromthe Bureau of the Census informing them of thevisit. The front of the household questionnaire,shown as appendix IIA in part B of this repprt,contains standard Census identification entriesrelated to the housing unit and space forrecording information on calls. On the inside ofthe questionnaire, questions l-3 identify allpersons living in the household, according torelationship to the head of the household, age,race, and sex. If the household does not containany persons in the age range l-74, inclusive, theinterview is concluded.

the Census supervisor for omissions or incon-sistencies and turned over to the HANES fieldmanagement office.

HANES Interviewing

The FOM and the Field Management Assist-ant (FMA) draw the nutrition sample daily asthe household questionnaires are turned over tothem. The sample for the detailed examinationis drawn at the end of the week. A master list isprepared giving the name, age, race, sex, andhousehold identification of each person selected.In the event that the number of persons on themaster list significantly exceeds a predesignatedexpected number for a particular stand, asubsampling pattern is provided to reduce thesample size to the maximum number that can behandled according to the schedule. All personsremaining on the master list then receive asample number. Those receiving only the nutri-tion examination are given numbers in the001-599 series; the detailed examinees are givennumbers in the 600-799 series.

After Census interviewing is completed andthe master list prepared, HANES representatives(HER’s) visit all households containing samplepersons. The main purpose of this visit is to getthe sample person(s) to make an appointment tocome in for the examination. During the inter-view, the HER administers one of the medicalhistory questionnaires (appendixes IIC-IIE, part B)as appropriate for the age of the sample person,a Food Programs Questionnaire (appendix IIB,part B), and the General Medical History Supple-ment, Ages 25-74 (appendix IIF, part B), if thesample person is to receive the detailed examina-tion. The HERexamination of

obtai.ns written consent for theminors and gets a written

authorization to obtain additional informationfrom the records of physicians, dentists, hos-pitals, schools, and State registrars. The HERindicates to the sample person that the PublicHealth Service will be glad to send a report ofsignificant findings to his physician (or clinic)and dentist if he so wishes.

In the course of the interview, the HER mustbe able to explain the program fully and toanswer many questions, such as how the samplewas selected, examination content, and value ofthe examination to the individual. They must

also be alert to signs of noncooperation and tryto overcome 1t.

APPOINTMENT AND TRANSPORTATIONPROCEDURES

The HER carr ies a copy o f the mas terappointment schedule, the original of which iskept in the f ield management off ice. Thisschedule calls for four morning sessions (in-cluding Saturday), four afternoon sessions (in-cluding Saturday), and two evening sessions. Tenpersons (two detailed examinees and eight nutri-tion examinees) can be seen in any one session.The HER’s schedule two detailed and fivenutrition examinees for the beginning of thesession and three additional nutrition examinees1% hours later. Once the sample person hasagreed to come in for the examination, aconvenient time is worked out, and the informa-tion is telephoned to the office from thehousehold. The sample number for that personis then entered on the master appointmentschedule.

Finally, an appointment slip is left with thesample person indicating the day and date onwhich he is to be examined and the time thatthe taxi will call. The use of a taxi, for whicharrangements have already been made, is encour-aged because it reduces the chance that thesample person will not appear.

Some sample persons elect to drive themselves3to the examination center and be reimbursed atthe rate of 10 cents a mile. If the HER’s thinkan appointment is “shaky,” they may offer topick up the sample person themselves. At least 3days before the date of examination, a remindernotice (a duplicate of the one left in the home)is mailed to the person. On the day before theexamination, a list of names and addresses ofexaminees is furnished the taxi company. Forthose sample persons of school age, a writtenexcuse is obtained from the parent or guardianduring the interview; this excuse is given to thetaxi company, HER, or sample person, depend-ing on where and when he is being picked up.

There are always a number of persons who,for one reason or another, cancel their appoint-ments or are not available at the time they are tobe brought to the center. Those who cancel are ,

I I

r.

fairly easily rescheduled for another time. Thosewho fail to appear without any notice of theirintention to do so or who change their minds

’ about participating are followed up as soon aspossible, preferably the same day by the same

before examinations at a preceding area arecompleted. With three teams performing exam-inations, four and sometimes five locations areoperating simultaneously. The third element isthe examining staff operating within the mobile

HER. Immediate followup of these personshelps to reinforce in the sample person’s mindthe importance placed on his participation. Inmany cases, the person can thus be brought tothe examination center only a little later than

- -examination center, which includes a physician,a nurse, a dermatologist, an ophthalmologist, adentist, two dietary interviewers, two healthtechnicians, one laboratory technician, and acoordinator.

originally scheduled.

EXAMINATION CENTER ANDFIELD STAFF

With the exception of the dermatologist andophthalmologist, all other members of the fieldstaff are civil service employees or commissionedofficers of the Public Health Service. The derma-tologist and ophthalmologist are usually senior

As in the preceding three cycles, examinationsare carried out in a specially constructed mobileexamination center (MEC). For the HANES

residents who generally are employed only for asingle sample area.

program, nine new trailers, 45 feet long and 8feet wide,, were constructed. The individual EXAMINATION CENTER PROCEDURES

trailers are drawn by detachable truck tractorswhen making moves from one area to another.Three trailers ’ are set up side by side andconnected by enclosed passageways to makeeach examination center. Figure 3 shows thethree trailers included in each MEC and the floor

General

As discussed earlier, the content of theexamination was developed after extensive plan-ning, consultation, and methodologic and pilot

plan of each. A minimum space of 50 feet by 50 studies. Thus, it is a special examination tailoredfeet is required to accommodate the MEC. The to meet the objectives and limitations of thesite on which the MEC is located must be hard survey and its two components and is notsurfaced and as level as possible to avoid anyeffect on certain examination proce dures and tobe accessible to the truck tractors. Heating andair-conditioning units are installed to help pro-vide a standardized environment for conductingthe examinations.

The field staff necessary to carry out the

intended to be a complete medical examination.The fact that the examination is not a substitutefor a visit to the examinee’s own physician anddentist is explained to the sample person or toth.e parents or guardians of sample children. Areport of medical findings for each examineereceiving the detailed examination is sent to the

three-team operation of the survev-

may be examinee’s physician or clinic. This report in-considered to consist of three elements. The firstelement is the team of Census interviewers(usually 8 to 16 persons) and a supervisor. Thesecond element consists of the administrative

eludes any new signif icant medical , derma-tological, and ophthalmological findings; data onheight, weight, visual acuity, hearing levels; andthe results of urinalysis, hematology, bloodchemistries, and the tuberculin skin test. En-closed with the report are a copy of the chest

staff and HES interviewers. The administrativestaff (the field operations manager and oneassistant) arrive at the location and set up their X-ray and a tracing and computer printout ofoffice on the Friday before Census interviewing, the electrocardiogram.with from four to six HER’s arriving 1 week Reports of medical findings of nutritionlater. The total administrative staff consists of examinees are sent only if there are any newfive field operations managers, five field manage- significant medical, dermatological, ophthal-ment assistants, and 12 HER’s. The adminis- . mological, urinalysis, hematological, or bloodtrative staff includes extra positions because chemistry findings. A complete report is sent fortheir operations at a new sample area begin all of these areas if the results in one or any part

TRAILERS

t

-m----m

I

1

Body Imeasurements

Cardio-pulmonary Wash ]room

Audiometry

I I

Staff I I.room I I

I, - -

4

4 - - - yL---A

Staffentrance .

--a- -

Physical Dermatologyexamination examination

Washroom

Darkr o o m

X-ray

I----’ 1 1

Laboratory

DietaryI interview

Dental OphthalmologyWaiting

areaexamination examination Dietary

interview

Examinee entrance

FLOOR PLAN

Figure 3. Mobile examination center.

27

of one, such as a biochemical test, are found tobe abnormal.

Since the reports of detailed and nutritionf examinees are necessarily delayed because of the

processing of blood chemistries, any conditionfound that, in the opinion of the examiningphysician, requires early medical attention isreported immediately by phone to the personalphysician or medical care facility identifiedearlier by the examinee.

A number of examinees are unable to providethe name of a regular physician or medicalresource to whom they wish to have theirfindings reported. In such cases, the approval ofthe examinee is obtained during the householdinterview to have the findings referred to asource, such as a county health department, thathad been obtained as a result of the advanceprofessional relations described in an earliersection. ‘These sources are aware of the HANESprogram and have been alerted to the possibilityof receiving reports of findings.

Reports of dental. findings are mailed by thedentist in the field for all examinees requesting areport. Conditions that require immediateattention are handled individually, usually byphone. For an examinee who does not have aregular dentist and for whom immediate atten-tion is’ required, after obtaining the patient’sapproval the referral service of the local dentalsociety, the personal physician of the examinee,or other medical-dental source is informed.

All forms used in the conduct of the examina-tion procedures as well as the questionnairesadministered within the examination center areshown as appendixes IIG-IIQ in part B.

Flow of Examinees

In HANES, the wide age range of samplepersons, the large number of examinees sched-uled per day, and the fact that two differentexaminations are being carried out simul-taneously, necessitate an examinee flow schemedifferent from those used in earlier surveys. InCycle I, schedules were staggered, with twoexaminees scheduled to begin at each half-hourinterval. All went through the same fixed se-quence of examination elements. In Cycles IIand III, six sample persons were scheduled forthe beginning of each session and went through

one of three fixed sequences of examinationelements according to a flow chart.

The flow scheme used in HANES containselements of both earlier systems. The usualworkday consists of two sessions, with up to 10examinations per session, eight nutrition exam-inees and two detailed examinees. The twodetailed sample persons and five of the nutritionsample persons are scheduled for the beginningof a session. Three additional persons are sched-uled M2 hours later for the nutrition examina-tion. Scheduling is somewhat flexible to accom- 1modate situations when it is desirable that morethan 10 persons be scheduled for a session andthat persons be scheduled at times differentfrom those normally used. This flexibility allowsscheduling to be more responsive to the specialproblems of individual sample persons as well asto the conditions created by high or lowresponse at a particular stand.

A primary objective of this flow system is toreduce the time examinees are in the examina-tion center while using the examination staff inas efficient a manner as possible. The schemegives a set of priorities by which examinees areassigned to the examination elements. However,these priorities do not require that an examineereceive one part of the examination beforeanother if it means that he must wait for thefirst part because the examiner is busy. Thisbasic scheme of assigning examinees was modi-fied by restrictions designed to meet operationalrequirements such as getting blood samples tothe examination center laboratory in time tocomplete the laboratory work before the end ofa session and insuring that examinees are seen bythe ophthalmologist during a certain interval oftime after receiving drops to dilate the pupils oftheir eyes. This system also incorporates recom-mended and maximum times for elements of theexamination. For some elements of the examina-tion, two examiners are trained to routinelygather data. The assignment of examinees tothese elements is controlled in such a way thatthe two examiners’ data can be comparedstatistically.

When examinees arrive at the examinationcenter, they are greeted by the nurse and thecoordinator; the latter is a staff member withspecial responsibilities in the area of examineeflow and records preparation and review. As

28

indicated by the flow system, soon after theirarrival, examinees change from their streetclothes into disposable examination uniformsdesigned to facilitate and standardize variouselements ofcian’s examX-rays.

the examination such as the pinati.ons, body measurements,

Physician and Nurse Examination

.lhysl-

and

The general physician’s examination isoriented toward gathering data on physicalconditions pertinent to nutrition and certainchronic diseases, in contrast to the concept of ageneral clinical examination performed in themanner most familiar to the examining physi-cian. Before beginning examinations, each newphysician spends from 1 to 3 days being trainedby the HANES Nutrition Medical Advisor torecognize symptoms or conditions associatedwith nutritional deficiencies. This training isusually performed with ongoing children andyouth projects supported by the Office ofEconomic Opportunity and in a Maryland Statehospital with patients suffering from nutritionaldeficiencies secondary to underlying chronicconditions such as alcoholism. Additional train-ing is provided at the examination center withrespect to the objectives of the detailed compo-nent of the physician’s examination just beforethe start of the regular examination.

Each medical history questionnaire is re-viewed by the examining physician on the daybefore the scheduled examination. Special atten-tion is paid to any entries that suggest restric-tions on the examinee’s ability to participate inany tests or procedures, particularly the spirom-etry, single breath‘diffusing capacity, or X-rays,and to items that may require further followupin the course of the examination.

All examinees receive a physical examinationwith emphasis on nutritional aspects. Aftermonitoring the examinee’s sitting blood pressureand pulse, the physician examines the ears forany abnormalities and then the head, eyes,m.outh, and neck ‘(including the thyroid, figure4), looking especially for lesions associated withnutritional deficiencies of vitamins A, B com-plex, and C, and minerals such as iodine andiron. While examining the chest (heart andlungs), an inspection is made of the chest and

Figure 4. Physician’s examination of the thyroid.

back for signs of possible deficiencies of vita-mins A and D. The physician then palpates theabdomen, and in examinees over age 25, per-cusses the liver. The neurological and musculo-skeletal systems are evaluated by testing thedeep tendon reflexes and neuromuscular excita-bility for stigmata of thiamine or mineral defi-ciencies, and by palpating and inspecting theskeleton for lesions associated with vitamin D orC deficiencies. The skin of the extremities isthen inspected and palpated, particularly that ofthe thighs and upper outer arms, for lesions thatmight be associated with deficiencies of vitaminsA, C, B,, or essential fatty acids. Findings arerecorded in two categories, those related andthose not related to nutrition. A subjectiveimpression is recorded on the nutritional statusof the examinee based on the examination andthe medical history.

Venipuncture is done on all examinees by thenurse, either at the beginning of the derma-tologist’s examination or at some other con-venient time during the course of the examina-tion. It is performed in the dermatologist’sexamining area by the nurse with the assistanceof the dermatologist or, in certain instances,with the assistance of the examining physician.

The examinees in the detailed componentreceive a more comprehensive cardiovascularevaluation and musculoskeletal examination.The cardiovascular evaluation includes a routineauscultation of the heart. If abnormal conditionsare found, a tentative diagnosis is made alongwith an evaluation of the degree of severity and 1

29

.

, / I

6 - I’

the certainty of the diagnosis using a scale present glasses, slit lamp examination (figure 5),ranging from 0 to 9. In addition to the initial and retinoscopy for detailed examinees onlyblood pressure reading taken at the beginning of with acuity less than 20/40; applanation tonom-the examination on all examinees, two more etry on examinees age 20 years and over;readings are taken by the nurse at the end of the maxillary sinus transillumination for detailedphysician’s examination-one with the examinee examinees only; and examination of the pupilssupine, and the other immediately after with the as well as examination of the l ids, globe,examinee sitting on the edge of the examination conjunctiva, sclera, cornea, anterior chamber,table. The musculoskeletal examination involves iris, and lens. The pupils are dilated in mostthe recording of findings of abnormalities and instances for evaluation of the vitreous andvarious manifestations of the knees, hips, retina. Diagnoses are recorded for the six mostshoulders, elbows, wrists, phalanges, ankles, feet, serious eye conditions found, with an indicationand back. for each of whether it affects vision and whether

Detailed examinees also receive an examina-tion of the ears, nares, reticuloendothelial sys-tern, an arterial evaluation, and a tuberculin skintest. The ear examination is of special interestbecause of its relevance to the audiometric data.

- It consists of a general inspection of the externalear, and a routine otoscopic examination of theexternal auditory canals and tympanic mem-branes. The tuberculin skin test, which is admin-istered by the nurse, is read by her or anotherspecially trained staff member between 48 and72 hours later by having the examinee return tothe examination center or by visiting his home.The test will be discontinued after the 35thHANES stand because of the burden imposed onthe examinee and the field staff by the necessityof a second visit.

At the end of the detailed examination by the

treatment is being given or needed. Ophthal-mologists from the National Eye Institute areresponsible for verifying the resultant diagnosesand for other aspects of quality control in thisarea.

The ophthalmology examination will be dis-continued after the first 35 HANES stands havebeen completed. This decision was arrived at bythe National Eye Institute as a result of theproblems encountered in recruiting ophthal-mologists to conduct the examinations and ofthe insufficient number of staff within theInstitute to carry out the program adequately. Itis felt, however, that the data collected from thefirst 35 stands will provide a basis for analysis ofthe data for the original purposes of the exami-nation.

physician, he administers appropriate supple-mental medical history questionnaires as re-quired, based on positive responses to certainitems on the medical history questionnairesadministered earlier in the home. These supple-ments are identified as Supplement A, Arthritis;Supplement B, Respiratory; and Supplement C,Cardiovascular.

Dermatology Examination

The dermatological examination is a completeclinical examination of the skin and its append-ages that considers normal variations in texture

Ophthalmology Examination

The ophthalmologic examination, with a fewexceptions, is essentially the same for all exam-inees. It includes an ocular history regardingpreviously known eye disease or previoussurgery; for examinees age 4 years and over adetermination of monocular distance visual acu-ity with usual correction, if any, and with apinhole test to determine correctability for thosewith acuity less than 20/20, prescription of Figure 5. Ophthalmology examination.

30

and color, certain manifestations of aging, andall pathological changes, documenting significantdiagnoses by biopsy or culture whenever possi-ble. Estimates ‘are made of actinic exposureexperienced as well as actinic damage sustained,and of occupational risk from irritant andallergic contractants. For an examinee with asignificant hand, foot, or generalized problem, ajudgment is made about the burden to theexaminee in terms of discomfort or disability,about the care sought, and about the effectexpected from current best care. A lesion isphotographed if there is some question aboutthe diagnosis, if the lesion is in any way unique,or if it is to be biopsied.

Dental Examination

The dental examiners derive their findingsuniformly by following a written set of objectivestandards in which they have been thoroughlytrained. The standards are guidelines that, ineffect, narrow the range of examiner variabilityby eliminating many of the borderline or ques-tionable conditions that are frequently a sourceof disagreement. To avoid other sources thatmight result in systematic bias, the dentist doesnot dry or isolate teeth, remove oral debris andcalculus, or probe any tooth surface that doesnot have an overt sign of decay.

The dentist dictates the condition of eachtooth present to a trained recorder (healthtechnician). The teeth are classified as sound,filled, decayed, filled-defective, and nonfunc-tional. Missing permanent teeth are classifiedunder one of the following four categories:unerupted, carious extraction, accidental loss,and orthodontic extraction. When missing teethare replaced on a fixed or partial denture, thetissue- under the replacement as well as thereplacement itself, is rated. When no naturalteeth remain in the jaw, the condition of the jawand the status of the artificial replacement,when present, are recorded.

The next step of the examination is anassessment of the periodontal structures and thestatus of oral hygiene. The Periodontal Index isemployed to assess the presence or absence ofperiodontal disease. By this system of classifi-cation, scores are assigned according to theextent of gingival inflammation, the presence or

absence of periodontal pockets, and the firmnessof a tooth in its socket. To assess oral hygiene,scores are recorded for al l or any of sixpredesignated teeth that are present. The scoresindicate the amount of debris-and the amount ofcalculus on selected surfaces. Fluoride and non-fluoride opacities and other conditions such asbleeding gums, diffuse marginal inflammation,swollen red papillae, and recession are alsorecorded.

The occlusion of all persons age 6-21 years isappraised by a series of counts and measure-ments. Anteroposterior position of the lowerjaw in relation to the upper jaw is recorded.Counts are made of teeth in crossbite and teeththat are malaligned. Measurements are made inthe anterior area of the jaws of overjet, man-dibular protrusion, overbite, and openbite.

An enamel biopsy is taken on persons whohave a natural upper incisor present with a frontsurface free of cavities and fillings. The enamelsample is “polished off” from an area aboutone-eighth of an inch in diameter and to a depthof approximately .0002 inch. This is about -asmuch enamel as is removed during a routineprophylaxis by a dentist or dental hygienist. Thesample is analyzed to determine the fluoridecontent of the tooth from which it was re-moved. The result, expressed in parts permillion, will be compared with the number ofcavities and fillings on selected tooth surfaces toassess the relationship between fluoride contentand the occurrence of dental caries.

The dental ‘examiner, using his best clinicaljudgment, estimates the dental treatment re-quired for every sample person. In so doing 9 hetakes into consideration the status of oralhygiene and periodontal disease, the quantityand quality of past dental care, the responses toquestions asked at the beginning of the examina-tion about chewing and eating difficulties, theage of the individual, and the probable benefitof each specific treatment plan to the individ-ual’s health and nutrition. The treatment recom-mendation may include any procedure rangingfrom a simple filling to extraction of all remain-ing teeth and denture construction.

At the close of the examination, the dentalexaminer makes a brief oral report to theexaminee about the status of his oral health. It isalways stressed that the survey examination

31

’ ,1

should not be considered as a substitute for aregular dental checkup. A report of findings ismailed, as described in a previous section.

Dietary Interview

The dietary interview is conducted byHANES personnel with minimum qualificationsof a Bachelor’s degree in food and nutrition.Most, however, are registered dietitians withexperience in dietary interviewing. With someexceptions, interviews are conducted in smallprivate rooms in the MEC. A small number ofpersons scheduled for the detailed examinationmay be visited in their homes. Home visits aresometimes required if the mother or otherperson responsible for a child’s regular feedingdoes not accompany him to the examinationcenter. The 24-hour Dietary Recall Question-naire is administered for the total day before theday of examination (figure 6). Fifty-one three-dimensional food portion models are used as aguide in conducting the interview to help thesample person to estimate the amounts ofvarious foods consumed. This questionnaire isfollowed by the Dietary Frequency Question-naire that obtains information about how oftencertain foods have been eaten during the pre-ceding 3 months. Foods reported in the 24-hourDietary Recall Questionnaires are later coded by

Figure 6. Dietary interview.

the interviewers using nutrient information fromthe U.S. Department of Agriculture HandbookNo. 8. Other food codes used are from theTulane University’s master dietant list, fromBowes and Church’s Food Values of PortionsCommonly Used,86 or from USDA House andGarden Bul le t in No . 72 , and commerc ia lsources. All dietary data will be analyzed by acomputer program based on nutrient data forlOO-gram portions of foods.

Following administration of the dietary ques-tionnaires, the interviewers are also responsiblefor the completion of the Health Care NeedsQuestionnaire and the General Well Being Ques-tionnaire for all examinees in the detailedcomponent. While the first of these is inter-viewer administered, the second is intended tobe essentially self-administered.

Laboratory Procedures

The laboratory technician is responsible forscreening a urine specimen from each examineefor sugar, albumin, and blood; for performingthe basic hematology tests; and for preparingand packaging all blood and urine samples to besent to the CDC for analysis (figure 7). The basichematology performed in the MEC for eachexaminee, if sufficient specimen is available, ishemoglobin, hematocrit, and red and white cellcounts. Sedimentation rates are determined anda smear for differential W.B.C. count is alsoprepared. All hematology tests are performed induplicate, and all results are recorded on a dailyworksheet. All clinically borderline results arerepeated immediately. Once it is ascertained thata particular result is abnormal according to CDCguidelines, it is reported directly to the HANESexamining physician for any necessary followup.With the exception of the T-3 and T-4 deter-minations, which are performed by a privatecontractor, the remaining laboratory determina-tions, listed in an earlier section, are performedby CDC (figure 8).

Health Technician Procedures

Two technicians conduct the following partsof the examination on all examinees: measure-ment of height and weight, a series of body and

32

Figure 7. Examination center laboratory.

F,%- we 8. CDC Nutrition Laboratory.

33

skinfold measurements, and X-rays of the handand wrist of examinees 1-17 years of age. Inaddition, these technicians are also responsiblefor conducting the following on all examinees inthe detailed component of the survey: anaudiometric test, spirometry, electrocardiogram,single breath diffusing capacity, goniometry, andX-rays of the chest, hand and wrist, hips, andknees.

Audiometric testing of detailed examinees isdone in a specially constructed, acousticallytreated room built into one of the trailers ineach of the mobile examining units. The room isdesigned to provide sufficient ‘sound attenuationfor pure tone testing at frequencies of 250-6,000Hz to at least as low as -20-dB relative toaudiometric zero (International StandardizationOrganization, 1964) in the presence of thedegree of external noise usually present during

- the course of the examinations at the variouslocations.

Each adult is tested at the following fourfrequencies: 500, 1,000, 2,000, and 4,000 Hz,with the l,OOO-Hz frequency repeated a secondtime. Air-conduction tests for both ears arecompleted first, then the bone-conduction testsin the order indicated on the recording form.Alternation of presentation to each ear variesamong examinees with the testing started in theright ear when the examinee?s sample number isodd and in the left ear when even. The thresholdrecorded for each frequency is the lowestdecibel level at which 50 percent or more of theresponses are obtained, that is, in two out ofthree or three out of five trials. Masking for thenontest ear is done in air-conduction testingonly on retest when there is a 40-dB differenceor more in the thresholds for the two ears. Inbone-conduction testing, masking is done rou-tinely in the nontest ear at 30, 40, and 50 dBabove threshold for that ear. Standardized test-ing procedures are used to insure as consistenttest results as possible throughout the survey.Any condition such as earache, cold, or otherproblem that might affect the test results is alsorecorded.

All detailed examinees are given a 12.leadelectrocardiogram and spirogram with resultsrecorded on magnetic tape using a BeckmanDigicorder. Under terms of .an agreement withthe George Washington University School of

Medicine, Washington, D.C., the tapes are for-warded to their facilities for processing. Foreach examinee, HANES is provided with tabularprintouts and digital computer tapes of all basicdata. For the electrocardiogram, these dataconsist of the amplitudes and durations ofvarious waves in each of the 12 leads, as well assuch data as QRS and T axes, and rates. Basicdata for the spirogram consist of measurementsfrom five trials of maximal forced expiratoryvolume, the forced expiratory volumes at 1/2, 3/4,1, 2, and 3 seconds, and flow rates including themaximum expiratory, the maximum midexpira-tory, and the maximum terminal (figure 9).

The single breath carbon monoxide diffusionstudies are performed on detailed examinees ona Collins modular lung analyzer machine with adigital readout module. Basic data are providedfrom which computations can be made of thediffusion of gases across the pulmonary mem-branes. In addition to compiling much-needednormative data on carbon monoxide diffusion,the test also identifies individuals with pul-monary deficiencies. These data will comple-ment other HANES pulmonary data such aschest X-rays, spirometry, and respiratoryhistory.

Examinees in the age range of l-l 7 years aregiven routinely an X-ray of the hand-wrist forbone age and density (figure 10). This X-ray isan 8 X 10 film taken at a distance of 36 inches.No other X-rays are made of nutrition exam-inees except in those few instances where anX-ray of the chest is indicated as an aid to adiagnosis by the physician. Detailed examinees,

Figure 9. Spirometry.

34

Figure 10. Hand and wrist X-rays.

however, do receive the chest X-ray routinely.Two 14 X 17 films are taken at a distance of 72inches-one a PA film and the other a lateralfilm. In addition, detailed examinees also receiveX-rays of the hips, knees, and hand-wrist. Thehip and the knee X-rays are both 14 X 17 films;the former is taken at a distance of 72 inchesand the latter is an anteroposterior film ofboth knees at 40 inches. The hand-wrist is an8 X 10 film taken at a distance of 36 inches.Certain precautions are taken to protect exami-nees as well as technicians. Females between theages of 12 and 45 are carefully screened so thatthose pregnant will not be X-rayed, and nofemale under the age of 50 is given an X-ray ofthe hip. To minimize radiation hazard, use ismade of a special “no scatter” cone, of lead-rubber apron shields, and of radiation badgesthat are provided- at the beginning of testing ineach location for each technician to wear.Periodic dosimetry field surveys are conductedby the Radiological Health Division of the PHS.All films, except the hand-wrist X-ray of the

detailed examinees, are developed and reviewedso that unsatisfactory films can be repeatedbefore the examinees leave the examinationcenter.

Goniometry measurements are taken on alldetailed examinees to determine the range ofmotion of certain joints, the hip, and the knee.Specifically, 16 measurements are taken involv-ing the extension, flexion, abduction, adduction,both internal and external rotation of both hips,and the extension and flexion of the knees. Thisprocedure will be discontinued after the 35thHANES stand to shorten the length of theexamination time and because of problemsencountered in the reproducibility of the data.

In addition to height and weight measure-ments and a determination of handedness, sixother body measurements are made of allexaminees. These include elbow breadth, upperarm girth, triceps and subscapular skinfolds(figure 11); bitrochanteric breadth, and sitting

Figure 11. Taking skinfold measurements.

35

height. Children 1-7 years of age are alsomeasured for head and chest circumferences.Chest circumference measurements of full inspi-ration and expiration are also made on allexaminees in the detailed part of the examina-tion.

QUALITY CONTROL

The efforts of the quality control programextend to all phases of the operation-from thebeginning of the Census interview until allcollected data have been coded, edited, andplaced on magnetic tape for computer use. Thegoal of the program is to assure that the nationalestimates of the various characteristics collectedby the survey represent data of the highestattainable accuracy and precision within thelimitation imposed by reasonable procedures

- and costs. A report concerning quality controlactivities in previous HES cycles has beenpublished!’

In HANES, as in all sample surveys, there aretwo sources of error to be considered-samplingerror and nonsampling error. Sampling error,that is, error due to making measurements on asample rather than on the entire population, canbe quantified and is the concern of all statisti-cians in sample survey design and in analysis.During the data-collection phase, problems dueto this type of error are minimal. The non-sampling error is of constant concern during thedata-collection phase and considerable attention,time, and effort of the HANES personnel aredevoted to minimizing and measuring this typeof error.

One type of nonsampling error that can occurin voluntary surveys such as HANES is the biasintroduced by nonresponse if the nonrespond-ents differ from the respondents with respect tothe measurements being made. In past HEScycles and in the present HANES, a sampleperson is not considered a respondent unless heis actually examined, even though he may havecompleted several questionnaires during initialinterviews. Past HES samples and the presentHANES sample are defined at the time of thefirst household interview. Consequently, there isa certain amount of built-in nonresponse sincepersons who move, go on vacation, become ill,or for other reasons are not physically available

cannot be examined. If nonrespondents differfrom respondents for a given measurement, theamount of nonresponse bias introduced into anestimate generally would be expected to varywith the amount of nonresponse. Therefore,response rates for a survey such as HANES areimportant as indicators of possible nonresponsebiases. Response rates for Cycles I, II, and IIIwere 87, 96, and 90 percent, respectively. Thesehigh response rates may be attributed to variousmethodological studies, to advance planning andpublicity, to much diligent work by the healthexamination representatives, and to properhandling of examinees by the entire staff as wellas, in Cycles II and III, to the age of thepopulation segment sampled.

The response rate in HANES at the time ofpreparation of this report had not been as highas in the earlier cycles, and the final rate will belower than those obtained previously.

Concern over the lower response rate and itspossible implications resulted in a study con-ducted in conjunction with the ongoing surveyin the San Antonio, Texas, stand to determinethe effect, of remuneration on response. Half ofthe 600 sample persons were told during theHER interview that they would receive $10 afterthe examination, while no mention of remunera-tion was made to the other half. All who wereexamined, however, received payment after theexamination. The study design also controlledon family income, number of housing units persegment, and the HER assignments. The findingsof the study showed- a response rate 12 percenthigher in the group of sample persons offeredthe remuneration. As a result, payment of $10 isnow being offered routinely to all samplepersons if they participate in the examination. Areport describing the methodology and findingsof the study is currently being prepared forpublication.

Another type of nonsampling error that is ofgreat concern in the quality control program isthe measurement error that inevitably occursduring the examination procedure. Its impor-tance is easily recognized when it is consideredthat, in the present survey, each sample personrepresents approximately 7,500 persons; there-fore, any blemish on the survey findings for aparticular person is greatly enlarged in the finalanalysis of the larger universe. Not only is it

important to control and minimize this error,but it is equally important to measure, whereverpossible, the amount of error.

In HANES, several procedures are relied uponto accomplish these objectives. Before the col-lection of data, it was necessary to defineprecisely what is to be measured and to obtaininstruction as to how the measurement shouldbe performed. Advisors, both from within thestaff of HANES and from outside sources, wereinstrumental in constructing the necessary defi-nitions and instructions, which were later incor-porated into a staff instruction manual coveringall procedures. Intensive specialized training isgiven to each examination staff member in thespecific procedures performed by them in thesurvey. The special advisors within HANESprovide training in their respective areas whileadditional training in other areas is obtainedfrom various outside sources.

Although precise definitions and good initialtraining are necessary, they are generally notsufficient in a lengthy survey as HANES. Thetime factor creates a problem that does notoccur when data are gathered in a short periodof time. It is important to be consistent through-out the entire survey. In order to achieveconsistency, in addition to providing the de-tailed written instructions on all aspects of theexamination, the forms are structured, andperiodic retraining is provided. Retraining timemay range from a few minutes for a single itemup to several days for an entire area, such asbodv measurements.

In further efforts to attack measurementerror, mechanical equipment is used whereverfeasible to obtain a “hard document.” Em-ployed for this purpose are such devices as taperecorde rs, automatic recording of weight, photo-graphs of height, X-rays, and the recording ofspirometry and electrocardi ograms on magnetictape. The reading and interpretation of therecords so obtained can be done independentlymore than once. The use of instruments formeasuring as well as for recording introducesanother source of possible variation; thus, sys-tematic calibration is necessary. All instrumentsare calibrated at the beginning of each stand andalso periodically throughout the stand, somebefore each examination. In some instances-forexample, audiometers-resources are not avail-

able in the examining center and machines mustbe sent away for calibration. Other instrumentsalso receive periodic maintenance and servicethrough special contract arrangements with themanufacturers.

Environment is another important factor inachieving valid and standardized data. Goodlighting, heating, and air conditioning are essen-tial. For example, it is very important to be ableto standardize the light conditions under whichthe ophthalmology examination is given. Simi-larly, it is essential that the room in which thehearing test is given be soundproof.

The subject being examined can also intro-duce error into the measurement. If the exami-nee fails to stand up straight for a heightmeasurement, is uncooperative during thespirometry examination, or does not understandthe directions given for the audiometry test-togive only a few examples-error will occur. It is,therefore, very important that staff members beaware of such possibilities and see that theexaminee fully understands what he is to do andthat his fullest cooperation is obtained.

Despite precautions, there are biases andvariable measurement errors that cannot be orare not judged important enough to be elim-inated. Another objective of the quality controlprogram, therefore, is the determination of thetotal effect of these errors. For certain parts ofthe examination performed by the health techni-cian, the assignment of examinees is controlledso that the relative bias of individual technicianscan be monitored. The collection of replicatedata provides another means for evaluatingmeasurement errors. R.eplicate data are obtainedbasically in two ways: by reevaluating or reread--ing a hard document or by reproducing an actualmeasurement, either by the usual procedure orbv another standard procedure. Although harddocuments such as the weight and height meas-urements are reevaluated, the replicate programis primarily concerned with reproducing actualmeasurements.

During the actual operation of the survey , theprimary use of replicate data i s in indicatingareas where retraining or reevaluation of proce-dures is needed. When the reports of findings ofthe survey are published, data from the repli-cat es willextent to

be used to apprise the reader of thewhich the data may be affected by

37

measurement error and to call his attention tothis problem.

pendent contractor experienced in the art ofdietary interviewing.

Replicate data are gathered in many specificareas of the examination with varying degrees offrequency. For example, replicate measurementsare made on every examinee for measurementssuch as spirometry and hematocrit. Ophthal-mologists from the NE1 independently replicatethe complete ophthalmic examination on all thefirst day’s examinees for each test stand afterthe examiner is trained in the survey technique.The dental advisors systematically replicate thefield examiners on a subsample of examinees forthe purpose ‘of surveillance and on-the-spotretraining. They also periodically replicate oneanother. Although replicates are performed for adifferent purpose, the data are preserved and inprevious HES surveys have proved useful forindicating the extent of error in final evalua-tions. Additional blood is drawn from a system-atic subsample of detailed examinees for thepurpose of replicating thyroid hormone determi-nations. Comparisons are made upon receipt ofresults and have been valuable as the basis forcorrective action. This type of replicate will alsobe used in the final evaluation of measurementprocess error.

PLANS FOR ANALYSIS ANDPUBLICATION OF DATA

Because the data colle ction operation of thenutrition componen t in the HANES programwill take about 2% years to complete, thesample and schedule designs have been arrangedso that it will be possible to do some preliminary‘analysis of

-certain portions

before that time.componentof the nutritionOf the tot41 65

PSU’s in the survey, a subset of 35 was selectedcarefully so as to be representative of the whole.These are scheduled for completion in October1972, and the first preliminary analyses will beavailable in early calendar year 19 73.

All data from the subset will be statisticallyweighted and nonresponse adjusted so as torepresent closely the total U.S. population withrespect tovariables . B

age, race, se‘X9 and several 0ecause of the small sample size,

therthe

Several measures are taken to assure com-pleteness and consistency in the recording proc-ess. All questionnaires are reviewed for omis-sions and inconsistencies. With the exception ofthe questionnaires completed in the examiningcenter, all are reviewed by personnel in the fieldmanagement office. If errors are noted, correctinformation is obtained by phone or from theexaminee when he comes in for the examina-tion. Errors in recording body measurements,goniometry, and results of the dental examina-tion are reduced by having a second person actas a recorder. In addition, all data gathered intheoexamining center are reviewed by the exam-ination staff before the examinees leave. AllDietary, 24.Hour Recall and Food FrequencyQuestionnaires are coded by the interviewer inthe field and are checked by another interviewerbefore being forwarded to headquarters. At eachlocation, each dietary interviewer records twocomplete interviews on randomly selectedsample persons. The resulting taped interviewsare critiqued for adherence to established guide-lines, procedures, and policies by an inde-

preliminary reports will be unable- to providesome of the detailed subclassifications of thedata, such as region or urbanization. Moredetailed breakdowns of the data will be availablein the analysis of the 65 PSU’s. The first of thesereports as related to nutrition is scheduled formid-1974.

The data for the first 35 sampling units willbe reportenecessarily

d in folimited

.es with aur primary categoricontent as follows:

Report Category Contents

Die tary intake data . . . . . . . Distribution of selected nutri-ents reported on the Dietary24-hour Recall Questionnaireand percent of individuals notmeeting recommended nutri-ent allowances.

Hematological and biochemicaltest results . . . . . . . . . .

Distributions of selected nutri-tionally related test results andpercent of individuals belowspecified levels.

Anthropometric data . . . . . . Distributions of selected nutri-tionally related measurementdata by population groupswith some comparative norma-tive data from earlier DHESprograms.

38

Report Category Contents

Nutritional findings from the Prevalence of selected condi-physician’s examination . . tions as stigmata of nutritional

deficiency by populationgroups.

A final completion date for the collection ofdata of the detailed component has not been

firmly established at the time of this writing,primarily because the unknown sample sizes offuture PSU’s affect scheduling. Based on thedesign of the survey, a complete cycle (for atleast many parts .of the detailed component)encompasses two of the nutrition cycles, orabout 130 PSU’s.

REFEREtNCES

‘National Center for Health Statistics: Origin, program, andoperation of the U.S. National Health Survey. Vital and HealthStatistics. PHS Pub. No. lOOO-Series l-No. 1. Public Health

d Service. Washington. U.S. Government Printing Office, Aug.’ 1963.

*National Center for Health Statistics: Plan and initialprogram of the Health Examination Survey. Vital and HealthStatistics. PHS Pub. No. 1000Series l-No. 4. Public HealthService. Washington. U.S. Government Printing Office, July1965.

3U.S. National Health Survey: A study of special purposemedical-history techniques. Health Statistics. PHS Pub. No.5840Dl. Public Health Service. Washington. U.S. GovernmentPrinting Office, Jan. 1960.

4U.S. National Health Survey: Attitudes toward cooperationin a health examination survey. Health Statistics. PHS Pub. No.5840D6. Publics Health Service. Washington. U.S. GovernmentPrinting Office, July 1961.

5U.S. National Health Survey: Evaluation of a single-visitcardiovascular examination. Health Statistics. PHS Pub. No.584-D’. Public Health Service. Washington. U.S. GovernmentPrinting Office, Dec. 1961.

‘National Center for Health Statistics: Comparison of twovision-testing devices. Vital and Health Statistics. PHS Pub. No.1000~Series Z-No. 1. Public Health Service. Washington. U.S.Government Printing Office, June 1963.

‘National Center for Health Statistics: The one-hour oralglucose tolerance test. Vital and Health Statistics. PHS Pub. No.1000~Series Z-No. 3. Public Health Service. Washington. U.S.Government Printing Office, July 1963.

*National Center for Health Statistics: Cooperation in healthexamination surveys. Vital and Health Statistics. PHS Pub. No.100OSeries Z-No. 9. Public Health Service. Washington. U.S.Government Printing Office, July 1965.

9Nationa.l Center for H_ealth Statistics: Replication, anapproach to the analysis of data from complex surveys. Vital andHealth Statistics. PHS Pub. No. 1000Series Z-No. 14. PublicHealth Service. Washington. U.S. Government Printing Office,Apr. 1966.

lONational Center for Health Statistics: Three views ofhypertension and heart dkease. Vital and Health Statistics. PHSPub. No. 1000Series ~-NO. 22. Public Health Service. Washing-ton. U.S. Government Printing Office, Mar. 1967.

1lNationa.l Center for Health Statistics: Factors related toresponse in a health examination survey, United States,1960-1962. Vital and Health Statistics. PHS Pub. No. lOOO-Series ~-NO. 36. Public Health Service. Washington. U.S. Govem-ment Printing Office, Aug. 1969.

12National Center for Health Statistics: Cycle I of the HealthExamination Survey sample and response, United States,1960-1962. Vital and Health Statistics. PHS Pub. No. lOOO-Series 1 l-No. 1. Public Health Service. Washington. U.S. Govem-ment Printing Office, Apr. 1964.

13National Center for Health Statistics: Glucose tolerance ofadults, United States, 1960-1962. Vital and Health Statistics.PHS Pub. No. 1000Series 1 l-No. 2. Public Health Service.Washington. U.S. Government Printing Office, May 1964.

14National Center for Health Statistics: Binocular visualacuity of adults, United States, 1960-1962. Vital and HealthStatistics. PHS Pub. No. 1000Series 1 l-No. 3. Public HealthService. Washington. U.S. Government Printing Office, June1964.

15National Center for Health Statistics: Blood pressure ofadults by age and sex, United States, 1960-1962. Vital andHealth Statistics. PHS Pub. No. 1000~Series 1 l-No. 4. PublicHealth Service. Washington. U.S. Government Printing Office,June 1964.

16Nationa.l Center for Health Statistics: Blood pressure ofadults by race and area, United States, 1960-1962. Vital andHealth Statistics. PHS Pub. No. 1000~Series 1 l-No. 5. PublicHealth Service. Washington. U.S. Government Printing Office,July 1964.

“National Center for Health Statistics: Heart disease inadults, United States, 1960-1962. Vital and Health Statistics.PHS Pub. No. 1000Series 1 l-No. 6. Public Health Service.Washington. U.S. Government Printing Office, Sept. 1964.

‘*National Center for Health Statistics: Selected dentalfindings in adults by age, race, and sex, United States,1960-1962. Vital and Health Statistics. PHS Pub. No. lOOO-Series 1 l-No. 7. Public Health Service. Washington, U.S. Govem-ment Printing Office, Feb. 1965.

lgNational Center for Health Statistics: Weight, height, andselected body dimensions of adults, United States, 1960-1962.Vital and Health Stat&tics. PHS Pub. No. 1000Series 1 l-No. 8.Public Health Service. Washington. U.S. Government PrintingOffice, June 1965.

*ONational Center for Health Statistics: Findings on the 33National Center for Health Statistics: Serum cholesterolserologic test for syphilis in adults, United States, 1960-1962. levels of adults, United States, 1960-1962. Vital and HealthVital and Health Statistics. PHS Pub. No. 1000~Series 1 l-No. 9. Statistics. PHS Pub. No. lOOO-Series 1 l-No. 22. Public HealthPublic Health Service. Washington. U.S. Government Printing Service. Washington. U.S. Government Printing Office, Mar.Office, June 1965. 1967.

*lNational Center for Health Statistics: Coronary heartdisease in adults, United States, 1960-1962. Vital and HealthStatistics. PHS Pub. No. 1000~Series 1 l-No. 10. Public HealthService. Washington. U.S. Government Printing Office, Sept.1965.

**National Center for Health Statistics : Hearing levels ofadults by age and sex, United States, 1960-1962. Vital andHealth Statistics. PHS Pub. No. 1000~Series 1 l-No. 11. PublicHealth Service. Washington. U.S. Government Printing Office,

oc~~1g65e

34National Center for Health Statistics: Decayed, missing, andfilled teeth in adults, United States, 1960-1962. Vital and HealthStatistics. PHS Pub. No. 1 OOO-Series 1 l-No. 23. Public HealthService. Washington. U.S. Government Printing Office, Feb.1967.

35National Center for Health Statistics: Mean blood hemato-crit of adults, United States, 1960-1962. Vital and HealthStatistics. PHS Pub. No. 1 OOO-Series 1 l-No. 24. Public HealthService. Washington. U.S. Government Printing Office, Apr.1967.

National Center for Health Statistics: Periodontal disease inadults, United States, 1960-1962. Vital and Health Statistics.PHS Pub. No. 1000Series 1 l-No. 12. Public Health Service.Washington. U.S. Government Printing Office, Nov. 1965.

24National Center for Health Statistics: Hypertension andhypertensive heart disease in adults, United States, 1960-1962.Vital and Health Statistics. PHS Pub. No. 1000~Series 1 l-No. 13.Public Heahh Service. Washington. U.S. Government PrintingOffice, May 1966.

36National Center for Health Statistics: Binocular visualacuity of adults by region and selected demographic charac-teristics, United States, 1960-1962. Vital and Health Statistics,PHS Pub. No. 1000~Series 1 l-No. 25. Public Health Service.Washington. U.S. Government Printing Office, June 1967.

25National Center for Health Statistics: Weight by height andage of adults, United States, 1960-1962. Vital and HealthStatistics. PHS Pub. No. 1 OOO-Series 1 l-No. 14. Public HealthService. Washington. U.S.. Government Printing Office, May1966.

37National Center for Health Stakistics: Hearing levels ofadults, by race, region, and area of residence, United States,1960-1962. Vital and Health Statistics. PHS Pub. No. lOOO-Series 11-No. 26. Public Health Service. Washington. U.S.Government Printing Office, Sept. 1967.

26National Center for Health Statistics: Prevalence of osteo-arthritis in adults by age, sex, race, and geographic area, UnitedStates, 1960-1962. Vital and Health Statistics. PHS Pub. No.1000Series 1 l-No. 15. Public Health Service. Washington. U.S.Government Printing Office, June 1966.

27National Center for Health Statistics: Oral hygiene inadults, United States, 1960-1962. Vital and Health Statistics.PHS Pub. No. 1 OOO-Series 1 l-No. 16. Public Health Service.Washington. U.S. Government Printing Office, June 1966.

28National Center for Health Statistics: Rheumatoid arthritisin adults, United States, 1960-1962. Vital and Health Statistics.PHS Pub. No. 1 OOO-Series 1 l-No. 17. Public Health Service.Washington. U.S. Government Printing Office, Sept. 1966.

*‘National Center for Health Statistics: Blood glucose levelsin adults, United.States, 1960-1962. Vital and Health Statistics.PHS Pub. No. 1 OOO-Series 1 l-No. 18. Public Health Service.Washington. U.S. Government Printing Office, Sept. 1966.

3oNational Center for Health Statistics: Age at menopause,United States, 1960-1962. Vital and Health Statistics. PHS Pub.No. 1000~Series l&No. 19. Public Health Service. Washington.U.S. Government Printing Office, Oct. 1966.

3 ‘National Center for Health Statistics: Osteoarthritis inadults by selected demographic characteristics, United States,1960-1962. Vital and Health Statistics. PHS Pub. No. lOOO-Series 11-No. 20. Public Health Service. Washington. U.S.Government Printing Office, Nov. 1966.

32National Center for Health Statistics: Childbearing anddiabetes mellitus, United States, 1960-1962. Vital and HealthStatistics. PHS Pub. No. 1 OOO-Series 1 l-No. 2 1. Public HealthService. Washington. U.S. Government Printing Office, Nov.1966.

38National Center for Health Statistics: Total loss of teeth inadults, United States, 1960-1962. Vital and Health Statistics.PHS Pub. No. 1000Series 1 l-No. 27. Public Health Service.Washington. U.S. Government Printing Office, Oct. 1967.

3gNational Center for Health Statistics: History and examina-tion findings related to visual acuity among adults, UnitedStates, 1960-1962. Vital and Health Statistics. PHS Pub. No.1000~Series 1 l-No. 28. Public Health Service. Washington. U.S.Government Printing Office, Mar. 1968.

40National Center for Health Statistics: Osteoarthritis andbody measurements. Vital and Health Statistics. PHS Pub. No.1000~Series 1 l-No. 29. Public Health Service. Washington. U.S.Government Printing Office, Apr. 1968.

41National Center for Health Statistics: Monocular-binocularvisual acuity of adults, United States, 1960-1962. Vital andHealth Statistics. PHS Pub. No. 1000~Series 11-No. 30. PublicHealth Service. Washington. U.S. Government Printing Office,Apr. 1968.

42National Center for Health Statistics: Hearing levels ofadults, by education, income, and occupation, United States,1960-1962. Vital and Health Statistics. PHS Pub. No. lOOO-Series 1 l-No. 3 1. Public Health Service. Washington. U.S.Government Printing Office, May 1968.

43National Center for Health Statistics: Hearing status and earexamination: findings among adults, United States, 1960-1962,Vital and Health Statistics. PHS Pub. No. 1000~Series 1 l-No. 32.Public Health Service. Washington. U.S. Government PrintingOffice, Nov. 1968. c

44National Center for Health Statistics: Selected examinationfindings related to periodontal disease among adults, UnitedStates, 1960-1962. Vital and Health Statistics.. PHS Pub. No.1000~Series 1 l-No. 33. Public Health Service. Washington. U.S.Government Printing Office, Sept. 1969.

45National Center for Health Statistics: Blood pressure as itrelates to physique, blood glucose, and serum cholesterol, United

States, 1960-1962. Vital and Health Statistics. PHS Pub. No.lOOO=Series 1 l-No. 34. Public Health Service. Washington. U.S.Government Printing Office, Oct. 1969.

46National Center for Health Statistics: Skinfolds, bodygirths, biacromial diameter, and selected anthropometric indicesof adults, United States, 1960-1962. Vital and Health Statistics.PHS Pub. No. 1000Series 1 l-No. 35. Public Health Service.Washington. U.S. Government Printing Office, Feb. 1970.

47National Center for Health Statistics: Need for dental careamong adults, United States, 1960-1962. Vital and HealthStatistics. PHS Pub. No. 1000Series 1 l-No. 36. Public HealthService. Washington. U.S. Government Printing Office, Mar.1970.

48National Center for Health Statistics: Selected symptoms ofpsychological distress, United States. Vital and Health Statistics.PHS Pub. No. 1000Series 1 l-No. 37. Public Health Service.Washington. U.S. Government Printing Office, Aug. 1970.

4gNational Center for Health Statistics: Parity and hyper-tension. Vital and Health Statistics. DHEW Pub. No. (HSM)72-l 024Series 1 l-No. 38. Public Health Service. Washington.U.S. Government Printing Office, Mar. 1972.

50National Center for Health Statistics: Plan, operation, andresponse results of a program of children’s examinations. Vitaland Health Statistics. PHS Pub. No. 1000Series l-No. 5. PublicHealth Service. Washington. U.S. Government Printing Office,Oct. 1967.

51National Center for Health Statistics: Evaluation of psycho-logical measures used in the Health Examination Survey ofchildren ages 6-11. Vital and Health Statistics. PHS Pub. No.1000Series ~-NO. 15. Public Health Service. Washington. U.S.Government Printing Office, Mar. 1966.

52National Center for Health Statistics: Calibration of twol

bicycle ergometers used by the Health Examination Survey.Vital and Health Statistics. PHS Pub. No. 1000~Series ~-NO. 21.Public Health Service. Washington. U.S. Government PrintingOffice, Feb. 1967.

53National Center for Health Statistics: A study of theachievement test used in the Health Examination Surveys ofpersons aged 6-17 yeaxs. Vital and Health Statistics. PHS Pub.No. 1000~Series ~-NO. 24. Public Health Service. Washington.U.S. Government Printing Office, June 1967.

54National Center for Health Statistics: Orthodontic treat-ment priority index.. Vital and Health Statistics. PHS Pub. No.lOOO-Series ~-NO. 25. Public Health Service. Washington. U.S.Government Printing Office, Dec. 1967.

55National Center for Health Statistics: Sample design andestimation procedures for a national health examination surveyof children. Vital and Health Statistics. DHEW Pub. No.(HSM) 72-1005Series ~-NO. 43. Public Health Service. Wash-ington. U.S. Government Printing Office, Aug. 1971.

56National Center for Health Statistics: Subtest estimates ofthe WISC full scale IQ’s for children. Vital and Health Statistics.DHEW Pub. No. (HSM) 72-1047Series ~-NO. 47. Public HealthService. Washington. U.S. Government Printing Office, Mar.1972.

57National Center for Health Statistics: Visual acuity ofchildren, United States. Vital and Health Statistics. PHS Pub.No. lOOO-Series 1 l-No. 101. Public Health Service. Washington.U.S. Government Printing Office, Feb. 19 70.

58National Center for Health Statistics: Hearing levels ofchildren by age and sex, United States. Vital and Health

Statistics. PHS Pub. No. 1000~Series 110No. 102. Public HealthService. Washington. U.S. Government Printing Office, Feb.1970.

5gNational Center for Health Statistics: School achievementof children 6-11 years as measured by the reading and arithmeticsubtest of the Wide Range Achievement Test, United States.Vital and Health Statistics. PHS Pub. No. 1000~Series 11-No,103. Public Health Service. Washington. U.S. GovernmentPrinting Office, June 1970.

60National Center for Health Statistics: Height and weight ofchildren, United States. Vital and Health Statistics. PHS Pub.No. 1000~Series 1 l-No. 104. Public Health Service. Washington.U.S. Government Printing Office, Sept. 1970.

‘lNati0na.l Center for Health Statistics: Intellectual maturityof children as measured by the Goodenough-Harris drawing test,United States. Vital and Health Statistics. PHS Pub. No.1 OOO-Series 11 -No. 105. Public Health Service. Washington. U.S.Government Printing Office, Dec. 1970.

62National Center for Health Statistics: Decayed, missing, andfilled teeth among children, United States. Vital and HealthStatistics. DHEW Pub. No. (HSM) 7%lOOSSeries 1 l-No. 106.Public Health Service. Washington. U.S. Government PrintingOffice, Aug. 197 1.

63National Center for Health Statistics: Intellectual develop-ment of children as measured by the Wechsler Intelligence Scale,United States. Vital and Health Statistics. DHEW Pub. No.(HSM) 7201004Series ll-No. 107. Public Health Service. Wash-ington. U.S. Government Printing Office, Aug. 197 1 e

64National Center for Health Statistics: Parent ratings ofbehavior patterns of children, United States. Vital and HealthStatistics. DHEW Pub. No. (HSM) 720101OSeries 11-No. 108.Public Health Service. Washington. U.S. Government PrintingOffice, Nov. 197 1 e

65National Center for Health Statistics: School achievementof children by demographic and socioeconomic factors, UnitedStates. Vital and Health Statistics. DHEW Pub: No. (HSM)72-l 01 l-Series 1 l-No. 109. Public Health Service. Washington.U.S. Government Printing Office, Nov. 19 7 1.

66National Center for Health Statistics: Intellectual develop-ment of children by demographic and socioeconomic factors,United States. Vital and Health Statistics. DHEW Pub. No.(HSM) 72-101 *-Series 1 l-No. 110. Public Health Service. Wash-ington. U.S. Government Printing Office, Dec. 197 1.

67National Center for Health Statistics: Hearing levels ofchildren by demographic and socioeconomic characteristics,United States. Vital and Health Statistics. DHEW Pub. No.(HSM) 7201025Series 1 l-No. 111. Public Health Service. Wash-ington. U.S. Government Printing Office, Feb. 1972.

68National Center for Health Statistics: Binocular visualacuity of children: demographic and socioeconomic charac-teristics, United States. Vital and Health Statistics. DHEW Pub.No. (HSM) 72-1031 -Series 1 l-No. 112. Public Health Service.Washington. U.S. Government Printing Office, Feb. 1972.

6gNational Center for Health Statistics: Behavior patterns ofchildren in school, United States. Vital and Health Statistics.DHEW Pub. No. (HSM) 72-1042Series 1 l-No, 113. PublicHealth Service. Washington. U.S. Government Printing Office,Feb. 1972.

7oNational Center for Health Statistics: Hearing sensitivityand related medical findings among children, United States. Vitaland Health Statistics. DHEW Pub. No. (HSM) 7201046Series

41

1 l-No. 114. Public Health Service. Washington. U.S. Govem-me;; Printing Office, Mar. 1972.

National Center for Health Statistics: Eye examinationfindings among children, United States. Vital and HealthStatistics. DHEW Pub. No. (HSM) 72-l 057Series 11 -No. 115.Public Health Service. Washington. U.S. Government PrintingOffice, June 1972.

72National Center for Health Statistics: Intellectual maturityof children: demographic and socioeconomic factors, UnitedStates. Vital and Health Statistics. DHEW Pub. No. (HSM)7%1059Series 1 l-No. 116. Public Health Service. Washington.U.S. Government Printing Office, June 1972.

73National Center for Health Statistics: Periodontal diseaseand oral hygiene among children, United States. Vital and HealthStatistics. DHEW Pub. No. (HSM) 7%1060-Series 1 l-No. 117.Public Health Service. Washington. U.S. Government PrintingOffice, June 1972.

74National Center for Health Statistics: Color vision defi-ciencies in children, United States. Vital and Health Statistics.DHEW Pub. No. (HSM) 73016OOSeries 1 l-No. 118. PublicHealth Service. Washington. U.S. Government Printing Office,Aug. 1972.- 75National Center for Health Statistics: Height and weight ofchildren: socioeconomic status, United States. Vital and HealthStatistics. DHEW Pub. No. (HSM) 73-1601Series 11-No. 119.Public Health Service. Washington. U.S. Government PrintingOffice, Oct. 1972. .

76National Center for Health Statistics: Skinfold thickness ofchildren 6-11 years, United States. Vital and Health Statistics.DHEW Pub. No. (HSM) 7301602Series l&No. 120. PublicHealth Service. Washington. U.S. Government Printing Office,

oc:j1g72*National Center for Health Statistics: Relationship amongparent ratings of behavioral characteristics of children, UnitedStates. Vital and Health Statistics. DHEW Pub. No. (HSM)73.1603Series 1 l-No. 121. Public Health Service. Washington.U.S. Government Printing Office, Oct. 1972.

78National Center for Health Statistics: Hearing and related amedical findings among children: race, area, socioeconomicdifferentials, United States. Vital and Health Statistics. DHEWPub. No. (HSM) 73-16040Series 1 l-No. 122. Public HealthService. Washington. U.S. Government Printing Office, Sept.1972.

7gNational Center for Health Statistics: Selected body meas-urements of children 6-11 years, United States, 1963-65. Vital

and Health Statistics. DHEW Pub. No. (HSM) 7301605~Series1 ~-NO. 123. Public Health Service. Washington. U.S. Govem-ment Printing Office, Feb. 1973.

80National Center for Health Statistics: Height and weight ofyouths 12-l 7 years, United States. Vital and Health Statistics.DHEW Pub. No. (HSM) 73-1606Series ll-No. 124. PublicHealth Service. Washington. U.S. Government Printing Office,Feb. 1973.

8 1 National Center for Health Statistics: Prenatal-postnatalhealth needs and medical care of children, United States. Vitalarid Health Statistics. DHEW Pub. No. (HSM) 7301607Series 1 l-NO. 125. Public Health Service. Rockville, Md. In preparation.

82National Center for Health Statistics: Plan and operation ofa health examination survey of U.S. youths, 12-17 years of age.Vital and Health Statistics. PHS Pub. No. 1000~Series l-No. 8.Public Health Service. Washington. U.S. Government PrintingOffice, Sept. 1969.

83National Center for Health Statistics: Development of thebrief test of literacy. Vital and Health Statistics. PHS Pub. No.1000~Series ~-NO. 27. Public Health Service. Washington. U.S.Government Printing Office, Mar. 1968.

84National Center for Health Statistics: Comparison of timedand untimed presentation of the Goodenough-Harris Test ofIntellectual Maturity. Vital and Health Statistics. PHS Pub. No.lOOO-Series ~-NO. 35. Public Health Service. Washington. U.S.Government Printing Office, June 1969.

85National Center for Health Statistics: Loudness balancestudy of selected audiometer earphones. Vital and HealthStatistics. PHS Pub. No. 1000~Series ~-NO. 40. Public HealthService. Washington. U.S. Government Printing Office, Dec.1970.

86Church,C. F., and Bowes, H. N.: Bowes and Church’s FoodValues of Portions Commonly Used. Philadelphia. Lippincott,1966.

87National Center for Health Statistics: Quality Control in anational health examination survey. Vital and Health Statistics.DHEW Pub. No. (HSM) 7201023Series ~-NO. 44. Public HealthService. Washington. U.S. Government Printing Office, Feb.1972.

88Goodman, R., and Kish, L.: Controlled selection-a tech-nique in probability sampling. J.Am.Statist.A. 45(251):350-373,Sept. 1950.

42

Definition of Terms

APPENDIX I

TECHNICAL NOTES ON THE SAMPLE DESIGN

Standard metropoli tan stat is t ical area(SMSA).--SMSA consists of a county or groupof contiguous counties (except in New England)that contains at least one central city of‘50,OOOpeople or more, or “twin cities” with a com-bined population of at least 50,000 population.In addition, other contiguous counties are in-cluded in SMSA if, according to certain criteria,they are socially and economically integratedwith the central city. Definitions of SMSA’s thatidentify the composition and structure of eachare given in an Office of Management andBudget publication, Standard Metropolitan Sta-tistical Areas, 1967 edition.

Geographic regions. -For purposes of theHealth Examination Survey, the 48 contiguousStates and the District of Columbia are dividedinto four regions of about the same populationsize. The regions and their composition are asfollows:

Region

Northeast . .

States Included

Midwest . .

South . . . .

Pennsylvania, New Jersey, Con-necticut, Rhode Island, Massa-chusetts, New York, Vermont,New Hampshire, MaineOhio, Michigan, Indiana, Illinois,Wisconsin, Minnesota, Iowa,MissouriDelaware, Maryland,, Virginia,West Virginia, Kentucky, Arkan-sas, Tennessee, North Carolina,South Carolina, Georgia, Florida,Alabama, Mississippi, Louisiana,District of Columbia

West . . . . . Washington, Oregon, Idaho,Montana, Wyoming, Colorado,Utah, Nevada, California, Ari-zona, New Mexico, Texas, Okla-homa, Kansas, Nebraska, SouthDakota, North Dakota

Con trolled selection. -This term refers to ascheme that permits some element of subjectivedetermination in obtaining a “better balanced”or “more representative” sample, while retainingall the elements of true probability sampling.The procedure is described in a number ofpublications.55@ The control variables usedfor this sample design are “State groups” and“rate of population change,” that are defined asfollows:

State groups.Separate groups were formed within geo-graphic regions, as shown in table I. Toform the State groups, the Health InterviewSurvey (HIS) design strata were classified asbelonging to the State in which the HISsample PSU was located. If a sample PSUwas within two States, it was put in theState with the greater proportion of thepopulation.

Rate of population change.Groups were defined differently for eachregion as indicated in table II. In theNortheast Region, for example, PSU’s withless than a 3-percent increase in populationbetween 1950 and 1960 were classified ingroup 1, while this class in the MidwestRegion included only those PSU’s with aloss or with no gain in population.

t ’ .

Table I. State groups by geographic region

Region

Northeast . .

Midwest . . .

. . .South . .

. . . . . . . . . . .

. . . ..I..... . .............

West.................................

State group number States in group

New York.Pennsylvania and New Jersey.Maine, New Hampshire, Vermont, Massachusetts, Con-

necticut, and Rhode Island.

Ohio.Michigan.Indiana and Illinois.Missouri.Kansas, Nebraska, Iowa, and North Dakota.Wisconsin and Minnesota.

Maryland, Delaware, and District of Columbia.Virginia and West Virginia.Kentucky and Tennessee.North Carolina and South Carolina.Georgia.Alabama and Mississippi.Florida.Arkansas, Louisiana, and Texas.

California and Nevada.Texas.Washington, Oregon, Idaho, and Montana.Oklahoma, Arkansas, and Louisiana.Wyoming, Utah, Colorado, New Mexico, and Arizona.North Dakota, South Dakota, Nebraska, Kansas,

Minnesota, and Missouri.

Table I I. Ranges for rate-of-population-change control groups by geographic region

Rate-of-population-change group number

1 .............................................2 .............................................3 .............................................4 .............................................5 .............................................

.............................................

7 .............................................8 .............................................

Region

NortheastI

Midwest South West

3 and under5-I 112-232558

wIm

Percent population change, 1950-60

0 and underI-1516-2324-3034-8 1

-10 ;ind under-9-ol-89-1619-2627-3637-4750-30 1

-5 and under-2-o4-2124-3940-5973-I 67

w

44

Population density groups. -In general, thisterm refers to the proportion of the populationthat lives in urban areas. The density groups aredefined somewhat differently for each geo-graphic region as shown in table 1 in the text ofthis report. For the very large SMSA’s, exceptthose in the South Region, the criterion forinclusion was population size; these SMSA’swere chosen for the sample with certainty. Inthe South Region, the largest SMSA’s weredefined in the same way as “other largeSMSA’s,” but were put in a different stratum forsampling purposes.

Current poverty areas.a-Poverty areas wereoriginally defined on the basis of 1960 censusdata in the 100 largest metropolitan areas. Theywere determined by ranking census tracts inplaces with a 1960 population of 250,000 ormore, according to the, relative presence of eachof the following equally weighted poverty-linkedcharacteristics: (1) family income below $3,000,(2) children in broken homes, (3) persons withlow educational attainment, (4) males in un-skilled jobs, and (5) substandard housing. Thosetracts falling in the lowest quartile of theranking were defined as poor tracts and furtheradjusted for contiguity and minimum size inorder to approximate area1 concentrations ofpoverty. The “current poverty areas” are de-fined similarly, based on a detailed investigationmade by local metropolitan officials and theCensus Bureau in 1970 in these places withpopulation of 250,000 people or more.

In general, census tracts were deleted fromthe list of poverty areas if local officials sug-gested deletion and if (1) the combined (five-factor) ranking of the tract fell in the highestquartile of “poor” tracts in 1960, (2) the com-bined ranking of the tract fell in the other threequartiles of poor tracts in 1960 but the incomerank fell in the highest quartile, or (3) the 1960population of the tract was less than 1,000regardless of its rank, on the assumption thatmajor changes could have taken place within itsince 1960. Tracts originally classified as povertyareas were not included as current poverty areaswhen substantial urban renewal or other majorimprovements in housing conditions had taken

aArnoCensus.

I. Winard, unpublished paper, U.S. Bureau of the

place within them. Also, any “nonpoor” tractthat was originally included in the 1960 povertyarea because it was completely surrounded bypoor tracts was deleted. f

Census tracts were added as poverty areas assuggested by local officials and if (1) 1959median family income of the tract was below$6,000, (2) the 1959 median family income ofthe tract was between $6,000 and $7,000 and itsmost recent welfare recipient or illegitimacyrates ranked in the lowest two quintiles withinthe city, or (3) when the 1959 median familyincome of the tract was $7,000 or more and itranked in the lowest quintile of the charac-teristics cited above or if a specific writtenexplanation was provided stating the reasonswhy the tract should be added in terms ofchanges that had taken place since 1960. Notract was added unless it was contiguous to agroup of poor tracts and the resulting area had acombined population of 16,000 or more.

Location of the 65 Health and NutritionExamination Survey Stands by Region

Region Stand

Northeast . . New York Standard Consoli-dated Area (five stands)Philadelphia, Pa. (two stands)Boston, Mass.Pittsburgh, Pa.Albany-Schenectady-Troy, N.Y.Scranton, Pa.Springfield-Chicopee-Holyoke,Mass.Providence-Pawtucket, R.I.-Mass.Hartford-Tolland, Conn.Chemung-Tioga-Tompkins, N.Y.Mercer, Pa.Bedford-F&on, Pa.

Midwest . . Chicago Standard ConsolidatedArea (two stands)Detroit, Mich.Milwaukee, Wis.Minneapolis-St. Paul, Minn.Cleveland, OhioColumbus, OhioSt. Joseph, MO.F a r g o - M o o r h e a d , N.Dak.-Minn. I

Region

South . . . .

Stand

St. Louis, MO .-Ill.Bay City, Mich.DeKalb-Steuben, Ind.; Branch,Mich.Cass-St. Joseph, Mich.Fayette-Ross, OhioLa Porte-Marshall-Starke, Ind.Boone-Greene, IowaFillmore, Minn.; Howard, Iowa

New Orleans, La.Washington, D.C., Md., Va.Columbia, S.C.Knoxville, Tenn.Roanoke, Va.Savannah, Ga.Tampa-St. Petersburg, Fla.West Palm Beach, Fla.Natchitoches, La.Lamar-Marion, Miss.Cabarrus-Stanley-Union, N.C.

Region Stand

Clarborne-Hamblen-Hancock-Hawkins, Tenn.Barbour, Ala.Bullock-Jenkins, Ga.Sessex, Del.-Worcester, Md.Fayette, W. Va.

West . . . . . Los Angeles, Calif. (two stands)San Francisco, Calif.Dallas, Tex.San Antonio, Tex.Tucson, Ariz.Omaha, Nebr.-IowaSan Diego, Calif.Fresno, Calif.Monterey, Calif.Clallam-San Juan, Wash.Grant, Wash.Gila, Ariz.Avoyelles, La.Ottertail, Minn.

46

*U.S. GOVERNMENT PRINTING OFFICE : 1979 O-281-259/19

Series 1.

Series 2.

Series 3.

Series 4.

Series 10.

SerirJs 11.

Series 12.

Series 13.

Series 14.

Series 20.

Series 21.

Series 22.

Series 23.

VITAL AND HEALTH STATISTICS Series

Programs and Collection Procedures. -Reports which describe the general programs of the NationalCenter for Health Statistics and its offices and divisions and data collection methods used and includedefinitions and other material necessary for understanding the data.

Data Evaluation and Methods Research. -Studies of new statistical methodology including experi-mental tests of new survey methods, studies of vital statistics collection methods, new analyticaltechniques, objective evaluations of reliability of collected data, and contributions to statistical theory.

Analy t&al Studies. -Reports presenting analytical or interpretive studies based on vital and healthstatistics, carrying the analysis further than the expository types of reports in the other series.

Documents and Committee Reports. -Final reports of major committees concerned with vital andhealth statistics and documents such as recommended model vital registration laws and revised birthand death certificates.

Data From the Health Interview Survey. -Statistics on illness, accidental injuries, disability, use ofhospital, medical, dental, and other services, and other health-related topics, all based on data collectedin a continuing national household interview survey.

Data From the Health Examination Survey and the Health and Nutrition Examination Survey.-Datafrom direct examination, testing, and measurement of national samples of the civilian noninstitu-tionalized population provide the basis for two types of reports: (1) estimates of the medically definedprevalence of specific diseases in the United States and the distributions of the population with respectto physical, physiological, and psychological characteristics and (2) analysis of relationships among thevarious measurements without reference to an explicit finite universe of persons.

Data From the Institutionalized Population Surveys. -Discontinued effective 1975. Future reports fromthese surveys will be in Series 13.

Data on Health Resources Utilization. -Statistics on the utilization of health manpower and facilitiesproviding long-term care, ambulatory care, hospital care, and family planning services.

Data on Health Resources: Manpower and Facilities.-Statistics on the numbers, geographic distri-bution, and characteristics of health resources including physicians, dentists, nurses, other healthoccupations, hospitals, nursing homes, and outpatient facilities.

Data on Mortality. -Various statistics on mortality other than as included in regular annual or monthlyreports. Special analyses by cause of death, age, and other demographic variables; geographic and timeseries analyses; and statistics on characteristics of deaths not available from the vital records based onsample surveys of those records.

Data on Natality, Marriage, and Divorce. -Various statistics on natality, marriage, and divorce otherthan as included in regular annual or monthly reports. Special analyses by demographic variables;geographic and time series analyses; studies of fertility; and statistics on characteristics of births notavailable from the vital records based on sample surveys of those records.

Data From the National Mortality and Natality Surveys. -Discontinued effective 1975. Future reportsfrom these sample surveys based on vital records will be intruded in Series 20 and 21, respectively.

Data From the National Survey of Family Growth. -Statistics on fertility, family formation and dis-solution, family planning, and related maternal and infant health topics derived from a biennial surveyof a nationwide probability sample of ever-married women 15-44 years of age.

For a list of titles of reports published in these series, write to: Scientific and Technical Information BranchNational Center for Health StatisticsPublic Health Service Q4SHHyattsville, Md. 20782