«1

Ni,

 

f% m
e u
t...“ .w
fns c
08.“ m N
5cm n
mls .m
ea a
tnm w
shl cm
, can“ MN.
MN“ m1.
9—- m.%
w 0."
D.m.m g

 

 

 

"vvvvv_~777 “w W, ,

 

 

 

W 0— .02 J mmtmm o 35255 52mm: 5% 5“ch 6:252 a 833m r281 0:95 o mwo_>mmm Z<_>_DI QZ< I._.n_<wI m0 ._.Zm_>_._.m<n_mo .mj
v
w
w

 

 

 

Library of Congress Cataloging in Publication Date

National Center for Health Statistics (U.S.)
Data systems of the National Center for Health Statistics.

(Vital and health statistics. Series 1, Programs and collection pro-
cedures :no. 16) (DHHS publication ;no. (PHS) 82-1318)

Prepared by: Nancy D. Pearce.

Supt. of Docs. no.: HE 20.6209:1/16

1. National Center for Health Statistics (U.S.) 2. Public health—
United States—Statistical services. 3. Medical care—United States— Sta-
tistical services. 4. United States—Statistics, Medical. l. Pearce,
Nancy D. ll. Title. lll. Series. IV. Series: DHHS publication ; no.
(PHS) 82-1318. [DNLMz 1. Data collection. 2. Health surveys—-
United States. W2 A N148va no. 16]

RA409.U44 no. 16 [RA407.3] 312'.07235 81-607874
ISBN Dam-02324 AACR2

 

 

For sale by the Superintendent of Documents. U.S. Government Printing Office, Washington, DC. 20402

E [III SlllllSll

 

 

Data Systems of
the National Center
for Health Statistics

 

Each of the data collection programs of the
National Center for Health Statistics is de‘
scribed. For each program the background,
purpose, scope, sample design, data collec-
tion procedures, content of data collection
forms, method of data release, and uses of
data are presented.

 

Programs and Collection Procedures
Series 1, No. 16

 

DHHS Publication No. (PHS) 82-1318

 

US. Department of Health and Human
Services

Public Health Service

Office of Health Research, Statistics, and
Technology

National Center for Health Statistics

Hyattsville, Md.

December 1981

National Center for Health Statistics

DOROTHY P. RICE, Director

ROBERT A. ISRAEL, Deputy Director

JACOB J. FELDMAN, Ph .D ., A sso ciate Director for Analysis
and Epidemiology

GAIL F. FISHER, Ph.D., Associate Director for the
Cooperative Health Statistics System

GARRIE J. LOSEE, Associate Director for Data Processing
and Services

ALVAN O. ZARATE, Ph.D., Assistant Director for
International Statistics

E. EARL BRYANT, Associate Director for Interview and
Examination Statistics

ROBERT C. HUBER, Associate Director for Management
MONROE G. SIRKEN, Ph. D., Associate Director for
Research and Methodology

PETER L. HURLEY, Associate Director for Vita] and Health
Care Statistics

ALICE HAYWOOD,Information 0f tcer

Foreword

This brief, semitechnical description of the various data
systems of the National Center for Health Statistics has been
created to serve a variety of needs. The greatest need is for a
single publication to bring together different descriptive mate-
rials under one cover. For the user who needs more detailed,
technical knowledge of these systems, this is available in sepa-
rate publications for the individual programs. It is expected
that this summary publication will serve the needs of many
visitors to the Center, as well as the needs of those at the Fed-
eral, State, and local levels who wish to understand the struc-
ture and scope of Center programs.

One feature that should be carefully noted is the flexible
nature of many of these surveys. While subject to the need for
standard measurement over time, the surveys are also able to
change with the demands for new data. Systematic reviews
lead to careful revisions and additions as needs are identified.
When necessary, new survey activities are designed and fielded.
Thus, this report is a cross-sectional description of the current
data collection program of the Center.

Dorothy P. Rice
Director

iii

 

Contents

Foreword .............................................................................. iii
Introduction ............................................................................ 1
Background .......................................................................... 1
Confidentiality of data ................................................................... 1
Data collection mechanisms ................................................................ 2
Morbidity and mortality data coding .......................................................... 2
Data release mechanisms .................................................................. 2
Descriptive, methodological, and other related reports .............................................. 3
Vital statistics ........................................................................... 4
Basic vital statistics ..................................................................... 4
Vital statistics followback surveys ............................................................ 6
National Survey of Family Growth ........................................................... 8
General population surveys .................................................................. 11
National Health Interview Survey ............................................................ 11
National Medical Care Utilization and Expenditure Survey ........................................ -. . . 14
National Health and Nutrition Examination Survey ................................................ 16
Health resources utilization statistics ............................................................ 20
National Hospital Discharge Survey ................................ ’ ........................... 20
National Ambulatory Medical Care Survey ...................................................... 21
National Nursing Home Survey ............................................................. 23
Health resources statistics ................................................................... 26
National Master Facility Inventory ........................................................... 26
National health professions inventories and surveys ............................................... 27
Appendixes
Contents .............................................................................. 30
I. Definition of certain terms used in this report ........................................................... 31
ll. Listing of reports related to specific programs ........................................................... 34

List of text tables

A. Sample size and total response rate (percent) for each of the vital statistics followback surveys: United States, 1961-80. . 7
B. Response rates for National Nursing Home Surveys, by type of questionnaire used (percent): United States, 1973.74
and 1977 .......................................................................... 25

vi

 

0.0

Symbols

Data not available
Category not available
Quantity zero

Quantity more than zero but less than
0.05

Quantity more than zero but less than
500

Figure does not meet standards of
reliability or precision (more than
30-percent relative standard error)

Figure suppressed to comply with
confidentiality requirements

 

Data Systems of the National
Center for Health Statistics

by Nancy D. Pearce, Office of the Center Director

Introduction

Background

The National Center for Health Statistics is one of
the major Federal statistical organizations. It operates
a diverse survey and inventory program with legisla-
tive authorization to collect statistics on the fol-
lowing:

. The extent and nature of illness and disability of
the population of the United States, including life
expectancy, maternal morbidity, and mortality.

0 The impact of illness and disability of the
population on the economy of the United States
and on other aspects of the well-being of its
population.

Environmental, social, and other health hazards.
0 Determinants of health.

0 Health resources, including health professionals
by specialty and type of practice; and the supply
of services by hospitals, extended care facilities,
home health agencies, and other health insti-
tutions.

O Utilization of health care, including utilization of
ambulatory health services, services of hospitals,
extended care facilities, home health agencies,
and other institutions.

0 Health care costs and financing.
0 Family formation, growth, and dissolution.

This report provides brief descriptions of each of
the data systems operated by the National Center for
Health Statistics. Each project description contains a
statement of the purpose of the data system, the
contents of the data collection forms, an overview of
the data collection process, and an indication of the
availability of the data to the public. Although this
report is nontechnical in nature, it does include some

technical terms, a limited glossary of which is
included as appendix I. For the purposes of many of
the Center surveys, however, many terms have highly
specialized definitions. No effort has been made here
to provide a comprehensive series of such definitions.
Each analytic report in series 10 through 23 (see
“Data release mechanisms”) includes in its own
appendix the definitions of certain specialized terms
used.

Confidentiality of data

Data collection activities of the Center are subject
to the specific provisions of both the Privacy Act of
1974 (Public Law 93-579) and the Health Services
Research, Health Statistics, and Health Care Technol-
ogy Act of 1978 (Public Law 95-623). The Privacy
Act covers all federally sponsored and operated data
collection that involves creation of a system of
records containing unique personal identifiers, while
the latter constitutes legislative authorization for a
portion of the Public Health Service.

In keeping with the requirements of the Privacy
Act of 1974 and Center policy, each individual,
household, or establishment asked to provide data for
the Center is informed in writing of the following:

0 The authorization for soliciting the information.
0 That disclosure of such information is voluntary.

O The principal purpose or purposes (usually statis-
tical or research) for which the information is
intended to be used.

0 The routine uses (as published in the Federal
Register) which may be made of the information.

0 The effects on the recipient, if any, of not
providing all or any part of the requested infor-
mation.

The Center is also bound by the provision of its
authorizing legislation that the information obtained
in the course of its statistical activities may not be

used for any purpose other than that for which it was
supplied. Such information may not be published or
released in other form if the particular establishment
or person supplying the information or described in it
is identifiable, unless such establishment or person
has consented to its publication or release in other
form. Thus, the Center assurance to respondents that
data will be kept confidential has a legal basis.

Almost all data collected by the Center are
requested under a pledge to the respondent (either an
individual or an establishment) that the information
will be used only for statistical purposes and will not
be released in such a way that individual persons,
households, or facilities will be identifiable. The
Center never releases information that would identify
an individual or a household. Some information is
released that identifies individual facilities, however.
When it is planned to release such information,
facility administrators are informed at the time they
provide the data which, if any, items will not be
accorded confidential treatment.

The NCHS Policy Statement on Release of Data
for Individual Elementary Units and Special Tabula-
tions presents and explains in detail the confi-
dentiality policy.

Data collection mechanisms

The Center does not have a large data collection
staff of its own. It collects most of its data through
interagency agreements with the US Bureau of the
Census or through contracts with non-Federal organi-
zations.

It has, for a number of years, been Center policy
to increase the capacity at the State level to collect
and utilize data. As in the case of vital statistics data,
the Center also tries to identify instances when it is
possible to unduplicate the collection of data by
having it collected at the most appropriate point
(Federal, State, or local) and then making it available
to others who have need for it. While vital statistics is
currently the only area in which the Center relies on
this method of data collection and sharing for the
production of national data, it is hoped that in the
years to come other types of data may be collected
through a national data-gathering system operating on
the State level, the Cooperative Health Statistics
System.

Morbidity and mortality data coding

The Center collects a wide range of information
about illness and the health problems of the popula-
tion of the United States. In order to enhance the
comparability of international statistics and of mor-
bidity and mortality estimates from such diverse
sources as household respondents and health care
provider records, these data are coded to appropriate

2

adaptations of the International Classification of
Diseases (ICD). ICD is a system for classification of
diseases and injuries for statistical purposes and is
revised under the auspices of the World Health
Organization every 10 years.

All Center mortality and morbidity information is
classified according to the most recent revision of
ICD. Data for the approximately 10-year period prior
to January 1, 1979, are coded to the Eighth Revision
International Classification of Diseases, Adapted for
Use in the United States (ICDA—8). Mortality data for
the period beginning in January 1979 are classified
according to the ninth revision of the International
Classification of Diseases (ICD-9), and morbidity data
are classified using the clinical modification of ICD-9
(ICD-9-CM).

Data release mechanisms

The Center releases its data in several ways. The
Vital and Health Statistics series contain detailed
reports on the design of the various data collection
systems, findings of methodological studies, and
detailed cross-tabulations of data. Advance Data
reports are timely but brief (4—12 pages each)
summaries of newly available data on topics of special
interest. These reports have replaced supplements to
the Monthly Vital Statistics Report for all survey
programs except the natality and mortality follow-
back surveys. All these reports are distributed without
charge, as issued, to libraries, medical colleges,
schools of public health, and other selected inStitu-
tions that have requested to receive all issues in the
series. Other organizations and individuals receive
publication notices and information to purchase
copies from the Government Printing Office. The
annual Catalog of Publications 0 f the National Center
for Health Statistics contains a cumulative list of
these reports, and the quarterly Publication Note is
issued to provide the latest information.

The Center also releases an annual booklet that
describes public use data tapes available for purchase.
This program, begun in 1969, permits researchers to
perform their own analyses of data. Tape availability
and contents, along with information on ordering the
tapes, are detailed in each year’s Catalog of Public
Use Data Tapes From the National Center for Health
Statistics. The majority of these data tapes are now
sold by the National Technical Information Service
(NTIS) and are not available from the Center.
Eventually, all Center public use data tapes will be
available from NTIS. The order form in the back of
the catalog gives the address and instructions for
ordering from NTIS or, in a few instances, from the
National Center for Health Statistics.

Many requests for unpublished data are filled
regularly using tabulations that have been previously
compiled. In addition, many special tabulations are

prepared each year to meet data requests that cannot
be fulfilled in any other way. When special tabula—
tions are necessary, the requester is given a cost and
time estimate. Tabulations and public use data tapes
are carefully reviewed to ensure that confidentiality is
maintained.

Descriptive, methodological, and other related reports

In addition to the reports from each program that
are issued in the analytic series designated in the

“release of data” section in each program description,
there are reports in other Vital and Health Statistics
series that are directly related to the programs. These
include program descriptions, methodological reports,
and reports that are in-depth analysis or contain
policy recommendations. A listing of these reports,
by program, is presented in appendix II.

Requests for publications and information or
inquiries concerning data tapes, special tabulations,
and other assistance should be directed to the
Scientific and Technical Information Branch.

Vital statistics

Basic vital statistics

Background. —Basic vital statistics come from
records of live births, deaths, fetal deaths, induced
terminations of pregnancy, marriages, and divorces or
dissolutions of marriages. Registration of these events
is a local and State function, but uniform registration
practices and use of the records for national statistics
have been established over the years through coopera-
tive agreements between the States and the National
Center for Health Statistics and its predecessor
agencies.

Purpose and scope—The purpose of the basic
vital statistics program is to formulate and maintain a
cooperative and coordinated vital records and vital
statistics system, promoting high standards of per-
formance. The program is nationwide in scope,
covering the entire population of the United States.

Data collection procedures. —Both provisional and
final vital statistics are derived from the registration
system. The provisional data are obtained from
counts of vital records registered without reference to
the date the event occurred; the final data are
obtained from the record and its contents, processed
by date of occurrence of the event.

The civil laws of every State provide for a
continuous and permanent birth, death, and fetal
death registration system. In general, the local regis-
trar of a town, city, county, or other geographic place
collects the records of births and deaths occurring in
the area; inspects, queries, and corrects if necessary;
maintains a local copy, register, or index; and
transmits them to the State health department. There
the vital statistics office inspects the records for
promptness of filing and for completeness and con-
sistency of information; queries if necessary; num-
bers, indexes, and processes the statistical informa-
tion for State and local use; and binds the records for
permanent reference and safekeeping. Microfilm
copies of the individual records or machine-readable
data are transmitted to the National Center for

4

Health Statistics for use in compiling the final annual
national vital statistics volume.

The system for collecting national data on mar-
riages and divorces is not as well developed as the,
system for births and deaths is. All States have
marriage and divorce laws; but, as of January 1, 1981,
three States did not have a central file of marriage
records, and four States did not have a central file of
divorce records. Forty-two States and the District of
Columbia are in the designated marriage-registration
area, and 30 States are in the divorce-registration area.
These registration-area States provide the Center
microfilm copies of their marriage and divorce certif-
icates or machine-readable data from which the
Center derives data on characteristics of marriages
and divorces and the persons involved. States not in
the registration areas provide only counts of mar-
riages by month and counts of marriages and divorces
by county. For States without central registration,
final counts are collected from individual counties
either by the State vital statistics office or by the
Center. In these instances some counties report only
marriage licenses issued and divorce petitions filed
rather than marriages performed and divorces
granted.

Most States submit microfilm copies of vital
certificates or machine-readable data tapes prepared
from the certificates to the National Center for
Health Statistics within 90 days following the end of
each data month. Special arrangements for less
frequent shipments are made with smaller registration
areas. Relevant information derived from answers to
State queries about the original certificates and
received by the State after shipment of the certif-
icates to the Center are filmed and submitted with
the first shipment following their receipt by the
State.

Provisional vital statistics are collected and pub-
lished monthly and summarized annually. They are
derived from monthly reports from the States to the
Center giving the number of certificates accepted by

each State for filing between two dates a month
apart, withoutregard to actual date of occurrence.
These reports to the Center are mailed on or before
the 25th of the month following the data month.
They are the source of the provisional vital statistics
published in the Monthly Vital Statistics Report and
its Annual Summary. Provisional data also include a
10-percent sample of death certificates. the Current
Mortality Sample, which provides provisional cause-
of-death data on a monthly basis. The Current
Mortality Sample is selected by the Center from the
regular data file of deaths for those States submitting
their entire month’s file by the end of the following
month. Otherwise the State is asked to provide a
sample of records on a current basis. The sample is
selected by including each record with a given last
digit in the certificate number.

Collection forms. —To promote uniformity in the
statistical information collected from States and local
areas for national statistical purposes, the National
Center for Health Statistics recommends standard
certificates or reports for birth, death, fetal death,
induced termination of pregnancy, marriage, and
divorce. The standard certificates and reports are
developed cooperatively with the States and local
areas and the Center and its predecessor agencies,
taking into account the needs andproblems expressed
by the major providers and users of the data. They
are reviewed about every 10 years to ensure that they
meet to the fullest extent feasible current needs as
legal records and as sources of vital and health
statistics. Although the use of standard certificates
and reports by States is voluntary, and their form and
content may vary according to the laws and practices
of each State, the certificates and reports in most
States closely follow the standard.

The Series 4 report “The 1968 Revision of the
Standard Certificates” outlines the history and basic
principles of the standard certificates and the major
objectives of the 1968 revisions; the report also
contains copies of the standard certificates and
describes the principal additions, modifications, and
deletions of items. A similar report in preparation
describes the 1978 revision of the standard cer-
tificates.

Data processing methods. —Machine-readable data
are accepted from States participating in the Coopera-
tive Health Statistics System. Data from all other
States are coded and entered on magnetic tape.

All death and fetal death records are processed.
For States not participating in the Cooperative Health
Statistics System, a 50—percent sample of live births is
processed by selecting records with image numbers
terminating in an even number. For States in this
cooperative system the entire birth file is used. In
general, all States in the marriage and divorce
registration areas that do not submit these data under
the system send microfilm copies of all their marriage

or divorce certificates to the Center, where a sample
of the records is selected and processed.

Release of data. ——Vital statistics data reach the
public through the Monthly Vital Statistics Report
and its supplements, Vital Statistics of the United
States (bound volumes issued annually), public use
data tapes (issued annually), and Vital and Health
Statistics series reports and special reports. Data on
mortality are published in Series 20; and data on
natality, marriage, and divorce are published in Series
21. Unpublished tabulations may also be available
upon request.

The Monthly Vital Statistics Reports containing
provisional monthly counts are published within
about 2 months following the end of the data month.
Provisional cause-of-death information from the Cur-
rent Mortality Sample are published about 3 months
after the data month, and the Monthly Vital Statistics
Report advance supplements releasing final annual
statistics are published in about 18 months after the
data year. Public use data tapes are released to the
National Technical Information Service within 1
month after publication of the advance supplements;
series and special reports are issued as resources
permit; and the annual Vital Statistics of the United
States are soon expected to be distributed with about
a 36-month lag.

For vital records, names and addresses are not
coded and never appear on data tapes or tabulations.
Certificate numbers are never released without writ-,
ten permission from the States whose records are
involved, and data tapes including data from localities
having less than 250,000 population are classified to
reflect population size without revealing specific
geographic areas.

Uses of data. —When vital statistics data have been
collected, processed, and made available for public
use, they serve a variety of purposes. They are used in
making population estimates and projections, in
assessing the health of the US population, in
pinpointing health problems, and in measuring prog-
ress made by national health programs. They are also
used in epidemiological studies; in marketing re-
search; in demographic, sociological, and economic
studies; in studies of the family; and in other
numerous and varied aspects of research aimed at
understanding our society and its problems and
progress.

Among the principal users of the data are other
parts of the Public Health Service, such as the Centers
for Disease Control, the Food and Drug Administra-
tion, and the National Institutes of Health; other
Federal agencies, such as the Bureau of the Census,
Federal Reserve Board, Bureau of Labor Statistics,
and Departments of Agriculture, Defense, and Trans-
portation; members of the United States Congress;
State and city governments; public and private
research institutions; life insurance companies; the

faculty and students of universities; newspaper re—
porters and feature writers; and physicians and
workers in health information and education groups.

Vital statistics followback surveys

Background. —National natality and mortality sur-
veys are periodic data collections based on samples of
registered deaths and births occurring during a calen-
dar year. Mortality surveys were conducted annually
from 1961 through 1968, and natality surveys from
1963 through 1969 and again in 1972. A National
Infant Mortality Survey was also conducted from
1964 to 1966. For 1980, a National Natality Survey
was conducted that was expanded to include fetal
mortality.

Purpose and scope. —The national followback
surveys extend for statistical purposes the range of
items normally included on the vital records. They
provide national estimates of births and deaths by
characteristics not available from the vital registration
system. They also serve as a basis for evaluating the
quality of information reported on the vital records.

Sample design. —The birth or death record serves
as the sampling unit, and samples of these units are
selected from a frame of records representing births
or deaths registered during a given period, usually a
calendar year.

The sampling frame for the National Mortality
Survey is the Current Mortality Sample, the 10-
percent systematic sample of death certificates re-
ceived each month by the National Center for Health
Statistics from the registration areas in the United
States. The sample for the National Mortality Survey
is a sample subselected monthly from the Current
Mortality Sample. The sampling frame for the Na-
tional Natality Survey is the file of birth certificates
from each of the registration areas of the United
States. Each registration area assigns a file number to
each birth certificate, and these file numbers run
consecutively from the first to the last birth occurring
in that area during that year. The samples for the 1980
survey are based on a probability design that makes
use of these certificate numbers. For the sample of
live births each 500th birth certificate record was
systematically selected; for the sample of fetal deaths
each 4th fetal death certificate was systematically
selected. The sample size and the total response rate
for each of the surveys are shown in table A.

Data collection procedures. —Data for all the
followback surveys are collected primarily by mail. In
the natality surveys, from addresses given on the birth
certificates, questionnaires are sent to the mother, the
physician who delivered the baby, and the medical
facility where the baby was born. In addition, the
1980 surveys obtained data from the medical sources
who the woman named as having given her radiation

6

treatment or examination in the year prior to the
delivery.

In the mortality surveys, a questionnaire is sent to
the person who provided the funeral director with the
decedent’s personal information for recording on the
certificate. This questionnaire requests socioeconomic
information about the decedent as well as the names
and addresses of hospitals and institutions that might
have provided care to the decedent at any time during
his or her last year of life. If the death occurred in a
hospital or institution, a hospital questionnaire is sent
directly to the hospital or institution asking for
information about the care provided and for the
names and addresses of other medical facilities
providing care.

Content of questionnaires. —The questionnaires
for past National Mortality Surveys have contained
questions concerning the patient’s last year of life.
The 1961 survey included questions on hospital
utilization, diagnoses, operations performed, institu-
tions in which hospitalized, number of hospital stays,
place of death, income, and whether working or
retired during most of the last year of life. The
1962-63 questionnaire included a few questions on
hospital and institutional care in the last year of life,
household composition, education, and income, with
detailed questions on place of residence of the
decedent. The 1964 and 1965 questionnaires in-
cluded questions on health insurance, place of hos-
pital and institutional care, the charges of hospital
care and who paid them, operations performed, how
much of the surgeon’s bill was paid by insurance,
household composition, assets, and income. The
1966-68 survey also included questions on the smok—
ing habits of the deceased and questions regarding the
household and income.

The questionnaire for the 1964-66 National In-
fant Mortality Survey included questions on hospitali-
zation of the infant who died, information about
other children of the mother, household composition,
income, employment of mother, education of mother
and father, and health insurance.

The National Natality Surveys collect information
from mothers who had live births during a given year.
The 1963 survey gathered info‘rmation on the medical
and dental care and radiological treatment of the
mother, employment and education of mother and
father, and family income. The 1964-66 survey
gathered information on pregnancy history, expecta-
tions of having more children, household composi-
tion, income, whether this was a first or later
marriage, date of first marriage and date of this
marriage, whether mother was employed and when
during her pregnancy she stopped working, educa—
tion, and health insurance. The 1967-69 survey dealt
with the medical care and smoking habits of the
mother, breast feeding, information on earlier preg-
nancies, expectations of having more children,

 

Table A. Sample size and total response rate (percent) for each of the vital statistics followback surveys: United States, 1961-80

 

 

 

 

 

 

 

 

 

 

 

Survey Sample size Response
rate
National Mortality Survey, 1961:
Hospitals and institutions ..................................... 1 out of 330 deaths 98
lnformants providing names of hospitals and institutions ................... 1 out of 330 deaths 93
National Mortality Survey, 1964-65 ................................. 1 out of 330 deaths 91
National Infant Mortality SurVey, 1964-65 ............................. 1 out of 110 infant deaths 88
National Natality Survey, 1963:
Mothers ............................................... 1 out of 1,000 births 86
Physicians .............................................. 1 out of 1,000 births 93
Medical facilities .......................................... 1 out of 1,000 births 98
Dentists .................. ' ............................. 1 out of 1,000 births 97
National Natality Survey, 1964-66 .................................. 1 out of 1,000 births 89
National Natality Survey, 1967 .................................... 1 out of 1,000 births of 91
white infants and 1 out of
500 births of all other infants
National Natality Survey, 1968-69 .................................. 1 out of 1,000 births of 87
white infants and 1 out of
500 births of all other infants
National Mortality Survey, 1966-68 ................................. 1 out of 260 of all deaths of 92
persons 35-84 years of age
National Natality Survey, 1972:
Mothers ............................................... 1 out of 500 births 72
Physicians .............................................. 1 out of 500 births 72
Hospitals ............................................... 1 out of 500 births 85
National Natality Survey, 1980: 1 out of 350 births, including Provisional
Mothers ............................................... an oversampling of low birth rates are
Physicians ...................... ' ........................ weight infants and 2 out of 5 all 75-80

Hospitals and other medical sources providing X-ray

and ultrasound or nuclear medicine .....................

fetal deaths of 28 weeks
gestation or more

 

whether this was her first marriage, dates of first and
present marriages, education and religious preference
of husband and wife, information on household
composition, work history, and income. The 1972
survey covered sources of medical care, health status
of mother and infant, health insurance, outcomes of
all pregnancies, whether pregnancy was wanted and
expectations of having more children, whether the
mother had an operation to prevent future pregnan-
cies, whether this was the first marriage, dates of first
and present marriages, education of mother and
father, information about the household, and family
income.

The 1980 national natality survey was broadened
to include fetal mortality. It included many of the
same or similar questions as previous surveys in order
to allow trend studies in the areas of smoking habits,
marriage and pregnancy history, education, income,
health status of mother and infant, sterilization,
radiological treatment, employment, childbearing ex-
pectations, and breast feeding. Many new areas of
study were added, such as alcohol consumption,
electronic fetal monitoring, amniocentesis, additional

maternal and infant health indicators, occupation of
mother and father, and ethnicity. Furthermore, there
was an oversampling of low birth weight infants in
the natality survey, which will enable more in-depth
study of this high risk group.

Copies of the data collection forms for the
various surveys are included as an appendix in each
analytic report based on data from a particular
survey.

Data processing methods. —After all methods of
obtaining complete questionnaires have been ex—
hausted, the data are edited and coded. Basic range
edits are edits made to eliminate some coding errors.
Computer processing includes consistency checks,
estimation or assignment of weights, and imputation
of missing data.

Release of data. —As with the basic vital statistics
provided through the registration system, the follow-
back survey data are released through the Vital and
Health Statistics series reports, supplements to the
Monthly Vital Statistics Report, and public use data
tapes, as the data become available. Series reports
issued prior to 1975 were published in Series 22,

“Data from the National Natality and Mortality
Surveys”; since 1975 reports from these sample
surveys based on vital records have been included in
Series 20, “Data on Mortality,” and Series 21, “Data
on Natality, Marriage, and Divorce,” as appropriate.
The privacy and confidentiality of study participants
are carefully maintained.

Uses of data. —Data from the followback surveys
are in demand by all major users of vital statistics
data. They help meet the expanding needs for
natality and mortality data in public health, medical
research, epidemiology, demography, and other re-
lated fields. Data provided through the natality
surveys are used by agencies and individuals responsi—
ble for or studying maternal and child health care
programs, and by researchers and others studying
social and demographic issues.

National Survey of Family Growth

Background. —The National Survey of Family
Growth (NSFG) is a multipurpose statistical survey
that provides a wide range of information serving
needs of persons and organizations concerned with
the dynamics of population change, family planning,
and maternal and child health. Developmental funds
and necessary positions were provided for establishing
NSFG in the National Center for Health Statistics in
fiscal year 1971. Fieldwork for the first cycle of the
survey was begun in July 1973 and completed in
January 1974. The second cycle began in January
1976 and was completed in September 1976. A third
cycle is expected to begin in January 1982.

Purpose and scope. —NSFG is designed to produce
data on factors influencing trends and differentials in
fertility, family planning practices, sources from
which family planning advice and services are ob-
tained, the effectiveness and acceptability of the
various methods of family planning, and those aspects
of maternal and child health that are most directly
related to fertility and family planning. The survey is
based on personal interviews with women of child-
bearing ages selected from a nationwide area probabil-
ity sample of households and group quarters, ex—
clusive of the institutional population and women
living on military bases.

Sample design—Cycles I and II of the National
Survey of Family Growth were based on a cross-
sectional sample of women in the conterminous
United States, 15-44 years of age, who were married,
who had been married, or who had never been
married but who had offspring of their own living
with them in the household. Excluded from the
sample were women living in group quarters (that is,
five or more unrelated persons 18 years or older,
unrelated to the head of the household, who live and
eat together, or six or more unrelated adults who live
and eat together—as, for example, in college dormi-

tories, barracks, or long-term care institutions) and
never-married women without children.

NSFG is designed as a multistage area probability
sample. In the first cycle of this survey, the first-stage
primary sampling units included 101 standard metro-
politan statistical areas, counties, parts of counties,
and independent cities. Secondary sampling units
consisted of enumeration districts or block groups
within selected primary sampling units. Where feasi-
ble, secondary units were subdivided into third-stage
listing units with a probability of selection propor-
tional to estimated housing. In the fourth stage,
housing units within the third-stage listing units were
chosen by systematic sampling. A fifth level of
sampling among eligible women within a household
was required to obtain one and only one interview in
each household with at least one eligible respondent.
To ensure sufficient precision for presenting intra-
group comparisons on various fertility variables, the
black population was oversampled.

The overall sample design called for 10,000
completed interviews, including approximately 4,000
black women and 6,000 women of white and other
racial groups. The response rate was an overall 81
percent. The final number of interviews was 9,797,
consisting of 3,856 black respondents and 5,941
respondents of white and other races.

For Cycle II of NSFG a multistage area probabil-
ity design based on 79 primary sampling units was
developed. The remaining stages were broadly similar
to those of the first cycle, except that a stratum .for
new housing (housing built since 1970), sampled
from building permit listings, was added. Overall
expected sample size remained at 10,000, with
comparable racial composition. After a certain level
of effort was completed in the fieldwork, remaining
nonrespondents were subsampled at a rate of l in 2.
The response rate for Cycle II was an overall 83
percent. The actual number of completed extended
interviews was 8,611, composed of 3,009 black
respondents and 5,602 respondents of white and
other races.

For Cycle III the sample design is broadly the
same as for Cycle 11, but with adjustments to
accommodate the more inclusive universe of women
of any marital status, 15-44 years of age. The sample
of 7,300 women will include women in group
quarters such as college dormitories and boarding
houses with more than five unrelated persons. As in
the previous cycles of the survey, military bases and
the institutional population will not be sampled.

Data collection procedures. -—Data collection for
NSFG is conducted by private contractors according
to Center specifications. Personal, in-depth interviews
are conducted with women identified as eligible for
extended interviews by means of a household screen-
ing interview.

The questionnaires used for data collection in—

clude a household screener designed to obtain house-
hold composition data and to identify eligible
extended-interview respondents in the sample house-
holds. For Cycles I and 11, two different versions of
the questionnaire were used for the extended inter-
view—a “currently married” questionnaire for women
who were married at the time of the interview and a
“post-married” questionnaire for women who were
widowed, divorced, or separated or who had never
been married but had their own children living in the
household. The differences between the two versions
were a result primarily of rewording the questions and
of deleting questions related to husbands in order to
make the “post-married” questionnaire appropriate
for respondents not married at the time of the
interview. Throughout the questionnaires, interviewer
instructions were used to skip respondents over
questions, or entire questionnaire sections, that were
not applicable to their individual situations. Cycle III
of the survey involves a screener questionnaire similar
to that used in the previous cycles and two separate
questionnaires for the extended interviews, one for
women under 25 years of age and one for women 25
years of age and over. The construction and format of
the questionnaires are modeled on those of the
previous cycles.

Standardized interviewer training programs,
which provide in-depth training on the questionnaires
and on NSFG concepts and procedures, are con-
ducted in several different locations in the country.
In addition to successfully completing training, inter-
viewers are required to conduct, for review and
approval, several practice interviews prior to begin-
ning their field assignments. Only women are em-
ployed as interviewers and observers in NSFG. Field
supervisors receive the same in-depth training as the
interviewers, in addition to intensive training in field
procedures.

Several quality control procedures were devel-
oped in Cycle 1 and refined in Cycle 11 to ensure the
quality of the collected and processed data: a
systematic field edit of selected interview items and a
validation of a sample of each interviewer’s work (a
sample recheck) were performed throughout the
fieldwork. Data preparation was validated through a
5-percent sample recode of all questionnaires, in
addition to systematic verification of each coder’s or
keyer’s work. A comprehensive legal-code and con-
sistency cleaning program was developed and used for
the data tapes. Similar quality control procedures are
specified for Cycle 111.

Content of questionnaires. —In Cycles I and II, the
NSFG questionnaires covered marital history, a de-
tailed pregnancy history, fecundity and expected or
intended future births, pregnancy planning practices
and utilization of specific contraceptive methods, the
source and financing of family planning services, and
a broad range of socioeconomic and demographic

characteristics. Cycle III will collect information on
sex education and the sexually active population as
well as on the topics covered in Cycles I and II.

In the first cycle of the survey, the average length
of an interview was 73 minutes; in the second cycle
the average length of an interview was reduced to
about 56 minutes; in the third cycle interviews will
again be kept within an hour, on the average.

A copy of the Household Screener questionnaire
for Cycle I is contained in appendix III of Series 2,
No. 76; the screener for Cycle II is in appendix II of
the comparable report for Cycle 11, Series 2, No. 87.
Relevant portions of the extended interview question-
naires are reproduced in appendixes of Series 23
reports. The “currently married” questionnaire for
Cycles I and II is reproduced in appendix III of Series
4, No. 18.

Release of data. —The publication program for
NSFG includes advance reports issued in Advance
Data, methodological reports and in-depth substan-
tive analyses in Series 2 and 23 of Vital and Health
Statistics, and public use data tapes. Publications
from the first cycle were issued beginning in 1976;
the public use data tape for this cycle was released in
1977. Advance reports from the second cycle were
issued beginning in 1978, and later the public use
data tape became available in two parts: the first part,
containing characteristics of respondents, including
summary fertility characteristics, was released in
April 1979; and the second part, containing charac-
teristics relating to each birth, became available in
1980. A public use data tape for Cycle III is
expected to be available by the spring of 1984.

Uses of data. —The rapidly growing need for
collection of data on trends and differentials in the
birth rate, family planning, and related aspects of
maternal and child health has been recognized by the
President and Congress. Attention has been directed
toward studying population trends, assessing the
implications of future population growth, making
appropriate policy recommendations, and helping to
educate the Nation regarding the consequences of
population growth. More information is needed than
is currently available from the regular data collection
programs and occasional surveys of the US Bureau
of the Census and the National Center for Health
Statistics. NSFG has been developed to provide, on a
periodic basis, detailed data on factors influencing
fertility, such as desired family size, birth spacing
intentions, and family planning practices, in order to
interpret current trends in the birth rate and to
prepare more realistic projections of future popula-
tion growth. In addition, for more effective planning,
management, and evaluation of the expanded family
planning programs, new information is being provided
concerning the social and economic characteristics of
couples who are either unable to limit their pregnan—
cies, or who are unable to have all the pregnancies

that they want. The nature and severity of the
problems these couples face and the extent to which
efforts to help them are succeeding are important
aspects of the survey. A major limitation of the
survey has been the exclusion from Cycles I and II of
single women who do not have offspring living with
them. Single women make a significant contribution
to the nation’s annual fertility, and they will produce

10

approximately one-fourth of total births in the few
years immediately following Cycle III of the survey.
In expanding the sample for Cycle III to include all
women 15-44 years of age, the survey will provide
data on the sexually active population as a whole and
will fill a significant gap in the need for detailed data
on the problems of adolescent pregnancy.

General population surveys

National Health Interview Survey

Background. —The National Health Interview Sur-
vey (NHIS) is a principal source of information on
the health of the civilian, noninstitutionalized popula-
tion of the United States. The National Health Survey
Act of 1956 provided for a continuing survey and
special studies to secure on a voluntary basis accurate
and current statistical information on the amount,
distribution, and effects of illness and disability in the
United States and the services rendered for or because
of such conditions. The survey referred to in the Act,
now called the National Health Interview Survey, was
initiated in July 1957. In its early years the survey
was known to many as the National Health Survey,
the name now given to a broader program of surveys
in the National Center for Health Statistics.

Purpose and scope—The purpose of the survey is
to provide national data on the incidence of acute
illness and accidental injuries, the prevalence of
chronic conditions and impairments, the extent of
disability, the utilization of health care services, and
other health-related topics. A major strength of this
survey lies in the ability to display these health
characteristics by selected demographic and socioeco-
nomic characteristics of the US civilian, noninstitu-
tionalized population.

Because NHIS data are obtained during household
interviews from the people themselves, the statistics
measure health status and experiences and reflect the
social and economic dimensions of health as reported
by individuals—the extent and impact of illness and
disability and the resulting uses of health care services
by the people experiencing them.

Interviews are conducted each week throughout
the year in a probability sample of households. The
interviewing is performed by a permanent staff of
carefully trained and supervised interviewers of the
US. Bureau of the Census under detailed specifica-
tions established by the National Center for Health
Statistics. Data collected over the period of a year
form the basis for the development of annual

estimates of the health characteristics of the popula-
tion and for the analysis of trends in those character-
istics.

The survey covers the noninstitutionalized, civil-
ian population of the United States. Because of
technical and logistical problems, several segments of
the population are not included in the sample or in
the estimates from the survey. Persons excluded are
patients in long-term care facilities (data are obtained
on patients in some of these facilities through the Na-
tional Nursing Home Survey of the Center); persons
on active duty with the Armed Forces (though their
dependents are included); U.S. nationals living in
foreign countries; and persons who died during the
calendar year preceding the interview.

Sample design. ~The National Health Interview
Survey is a cross-sectional household interview sur-
vey. The sampling plan follows a multistage probabil-
ity design that permits the continuous sampling of
households. The first stage consists of a sample of
376 primary sampling units drawn from approxi—
mately 1,900 geographically defined primary sam-
pling units that cover the 50 States and the District of
Columbia. A primary sampling unit consists of a
county, a small group of contiguous counties, or a
standard metropolitan statistical area. Within primary
sampling units, smaller units called segments are
defined in such a manner that each segment contains
an expected four households. The sampling plan is
designed to yield national estimates, but some esti-
mates can be obtained separately for the four
geographic regions. \

The households selected for interview each week
are a probability sample representative of the target
population. Each calendar year, data are collected
from approximately 40,000 households, including
about 110,000 persons. The annual response rate of
NHIS is usually at least 95 percent of the eligible
households in the sample. The S-percent nonresponse
is divided equally between refusals and households
where no eligible respondent could be found at home
after repeated calls.

11

Data collection procedures. —Data are collected
through a personal household interview conducted by
interviewers employed and trained by the US.
Bureau of the Census according to procedures speci-
fied by the National Center for Health Statistics.

All adult members of the household 17 years of
age and over who are at home at the time of the
interview are invited to participate and to respond for
themselves. The mother is usually the respondent for
children. For individuals not at home during the
interview, information is provided by a responsible
adult family member (19 years of age or over)
residing in the household. Between 65 and 70 percent
of the adults 17 years of age or over are self-
respondents. On occasion, a random subsample of
adult household members is selected to respond to
questions on selected topics. There are also instances
in which followup supplements are completed either
for the entire household or for individuals identified
as having particular health problems. As required,
these supplements are either left for the appropriate
person to complete and return by mail, or the
interviewer calls again in person or by telephone to
secure the information directly.

Nationally there are approximately 1 10 interview-
ers, trained and directed by health survey supervisors
in each of the 12 Census Bureau regional offices. The
supervisors are career civil service employees whose
primary responsibility is the National Health Inter—
view Survey. The interviewers are part-time employ-
ees, selected through examination and testing. Inter-
viewers receive thorough training in basic interviewing
procedures and in the concepts and procedures
unique to NHIS.

On the average, the interviews require about 45
minutes in the household. Depending upon the family
size and the nature and extent of health conditions of
family members, the length of interview ranges
between 15 and 90 minutes.

Content of questionnaire. —The questionnaire
consists of two basic parts: a core set of health,
socioeconomic, and demographic items; and one or
more sets of supplementary health items. The core
items constitute approximately 70 percent of the
questionnaire and are repeated each year. The ar-
rangement'of rotating and single-time supplements
allows the survey to respond to changing needs for
data and to cover a wide variety of topics, while the
core items provide continuous information on funda-
mental topics. Series 1, No. 11, “Health Interview
Survey Procedures, 1957-74,” describes changes dur-
ing that period in survey concepts and questions.

The questionnaire now provides for the following
types of core data:

0 Basic demographic characteristics of household
members, including age, sex, race, education, and
family income.

0 Disability days, including restricted activity and

12

bed days, and work- and school-loss days occur-
ring during the 2-week period prior to the week
of interview.

0 Physician and dental visits occurring during the
same 2-week period.

0 Acute and chronic conditions responsible for
these disability days and doctor visits.

0 Long-term limitation of activity resulting from
chronic disease or impairment and the chronic
conditions associated with the disability.

0 Short-stay hospitalization data, including the
number of persons with hospital episodes during
the past year and the number of discharges from
short-stay hospitals.

0 The intervals since the last doctor and dental
visits.

The questionnaire also includes six lists of chronic
conditions. Each condition list concentrates on a
group of chronic conditions involving a specific
system of the body (for example, digestive, skin and
musculoskeletal, circulatory, and respiratory). The
body systems approach to chronic conditions was
adopted because it was found that organizing ques-
tions around a particular body system resulted in
more thorough reporting and increased the number of
conditions for which estimates of prevalence could be
made. Beginning in 1978, each of six representative
subsamples has been asked questions based on one of
the six lists of chronic conditions. In this way
national estimates on each of the six body systems
are obtained during the same interview year. Prior to
1978 only one condition list was asked each year.

The supplements to the questionnaire change in
response to current interest in special health topics.
Supplements for 1979 were included on immuniza-
tion, smoking, home health care, eye care, residential
mobility, and retirement income. In 1980 the supple-
ments on home health care, residential mobility, and
retirement income were continued and a supplement
on health insurance added. Throughout 1981 a child
health supplement was included.

Suggestions and requests for special supplements
are solicited and received from many sources. These
include university—based researchers, administrators of
national organizations and programs in the private
and public health sectors, and specialists from other
parts of the Department of Health and Human
Services (for example, the National Institutes of
Health and the Centers for Disease Control). Al-
though it is not possible to include all of the
suggested topics, every effort is made to be responsive
to the data needs of such groups. A leadtime of at
least 1 year is required to develop and pretest
questions for new topics to be included as special
supplements. For example, supplements selected in
1981 will be developed and pretested in 1982 and

implemented in 1983 or 1984. Relevant portions of
the questionnaire are included in an appendix to each
Series 10 report, and each year’s questionnaire is
reproduced in its entirety in the annual “Current
Estimates” report.

Data processing and quality control methods.—
Throughout the data collection and processing phases
there are extensive quality control activities. Each
interviewer edits all completed work before returning
it to the regional supervisor, and in the regional office
there is a preliminary edit of all questionnaires. As
part of the quality control program, interviewers are
provided feedback on their errors. In addition, a
sample of households is reinterviewed by a supervisor
or senior interviewer.

The Census Bureau regional offices forward the
questionnaires to the National Center for Health
Statistics for coding and data processing. Each ques-
tionnaire is checked again at the Center for complete—
ness of field coverage, omissions, and other errors.
~ Illnesses, diseases, and injuries reported by the house-
holds are coded to a modified version of the
International Classification of Diseases. All coding is
subject to recoding on a sample basis to ensure a high
level of accuracy.

Thus, potential nonsampling errors such as those
in reporting, processing, and nonresponse, which are
inherent in any sample survey, are kept to a minimum
by methods built into the survey procedures. In
addition, with regard to sampling error, standard
error charts are created and provided so that a user
can calculate the variations in the estimates that
might occur because only a sample of the population
is surveyed.

Release of data. —Data release occurs in several
forms, such as publications, tabulations by various
sociodemographic characteristics, and public use data
tapes. The earliest reports containing National Health
Interview Survey data were issued in Series B and C
of Health Statistics from the US. National Health
Survey. About 50 reports were published in these
series before they were replaced by the Vital and
Health Statistics series. More than 140 reports from
NHIS have been published in Series 10 of Vital and
Health Statistics. Publication of a year’s data begins
in about October of the year following completion of
data collection and extensive computer editing of the
data to ensure quality and reliability.

The first report, “Current Estimates,” is followed
by 8 to 10 other Series 10 publications. Since not all
possible cross-tabulations can be analyzed and pub-
lished in Series 10 reports, many unpublished tabula-
tions are routinely generated and made available upon
request. In addition, these data are published in a
number of related special analyses prepared by the
Analysis and Epidemiology Program of the Center.

Another medium for data release is public use
data tapes. A public use tape is usually available for

distribution within about 2 years after completion of
data collection. This time lag is due primarily to the
need for staff to work with the data in order to
ensure completeness and reliability of data obtained
from the questionnaire supplements and to permit
development of adequate documentation for the
tapes. The data tapes, available since the 1969 data
collection year, have been widely used by universities
and Federal, State, and local governmental agencies.

Uses of data. ~Generally, uses of NHIS data can
be broadly classified into the following areas: pro-
gram planning and evaluation, public health educa-
tion and health promotion, epidemiological research,
and market research.

NHIS data on morbidity (prevalence estimates)
and associated disability (limitation of activity, dis-
ability days, and so forth) are used in Federal, State,
and local program planning in the assessment of the
extent and impact of public health problems. NHIS
estimates on trends in incidence, prevalence, and
disability are also used extensively in the evaluation
of the effectiveness of therapeutic and preventive
programs.

Estimates of health problems and resultant access
to health care services by various age, sex, occupa-
tional, and income subpopulations help identify gaps
in the provision and receipt of needed services.
Statistics on the prevalence of chronic disease, inci-
dence of injuries, and disability are used in consid-
ering the cost of or extending the scope of medical
care insurance or compensatory insurance of various
types. Estimates of health problems and resultant
utilization of health care services were used in
formulating the legislation for both the Medicare and
Medicaid programs and for making estimates of
potential costs of the programs. NHIS data on
out-of-pocket health care expenditures and the extent
of coverage under private health insurance and public
health care programs have been used by the Congres-
sional Budget Office, the Office of Management and
Budget, the Office of the Assistant Secretary for
Health, and the Social Security Administration to
evaluate programs covering the civilian, noninstitu-
tionalized population.

NHIS statistics are used by public, private, and
voluntary agencies and organizations for the purposes
of public health education and health promotion. For
example, NHIS statistics on health characteristics of
smokers and nonsmokers appeared in the Surgeon
General’s reports on Smoking and Health in 1964,
1979, and 1980. In addition, widespread and fre-
quent use of NHIS statistics in newspaper articles,
television news spots, popular magazines, and profes-
sional journals contributes to the American public’s
awareness of their health status. This awareness
ultimately leads to further individual and organiza-
tional health promotion and prevention efforts.

NHIS morbidity statistics are used in the epidemi-

n' 13

ological analysis of associations between the preva-
lence of various diseases and sociodemographic char-
acteristics (such as age, sex, and occupation) and in
the identification of high-risk subgroups in the
population. Identification of geographic patterns in
the distribution of diseases can lead to clues for
persons researching etiological factors. Researchers
also use NHIS data in studies relating to disease to
estimate direct costs (for example, expenditures for
physician visits and hospitalizations) and indirect
costs (for example, economic loss due to premature
death, disability, and work loss) due to illness.

Estimates of the number of persons with particu-
lar conditions as well as basic sociodemographic
attributes are frequently used by private industry not
only for estimating potential and current markets for
new products or drugs but also for designing adver-
tising campaigns targeted to the affected groups.
Specifically, NHIS data on prescription drugs, special
aids, corrective lenses, and hearing aids have been
used in and of themselves and also in conjunction
with the NHIS data on injuries, chronic and acute
conditions, and impairments and their impact in
terms of limitation of activity and mobility.

About 2,000 data requests are answered annually
by the Division of Health Interview Statistics. More
than two-fifths of the data requests are from private
industry and foundations; almost two-fifths are from
Congress and Federal, State, and local government
agencies; and the remaining fifth is divided among
persons affiliated with educational institutions and
private citizens interested in health data.

National Medical Care Utilization and
Expenditure Survey

Background. —The National Medical Care Utiliza—
tion and Expenditure Survey (NMCUES) is a unique
source of detailed national estimates on the utiliza-
tion and expenditures for various types of medical
care. This survey builds on the experience of the
National Health Interview Survey, the National Medi-
cal Care Expenditure Survey, and the former Current
Medicare Survey. The first cycle of the survey was
conducted to cover calendar year 1980. A second
cycle of the survey is planned for the mid-1980’s.

Purpose and scope. —NMCUES is designed to be
directly responsive to the continuing need for statisti-
cal information on the health care expenditures
associated with health services utilization for the
entire US. population. Cycle I was designed and
conducted in collaboration with the Health Care
Financing Administration to provide detailed utiliza-
tion and expenditure data for persons in the Medicare
and Medicaid populations. NMCUES will produce
estimates over time for evaluation of the impact of
legislation and programs on health status, costs,
utilization and illness-related behavior in the medical
care delivery system.

14

Cycle I was composed of several related surveys.
The household portion of the survey consisted of a
national survey of the civilian noninstitutionalized
population and a separate survey of the Medicaid-
eligible populations of the States of New York,
California, Texas, and Michigan. These two surveys
each consisted of five interviews over a period of
about 15 months to obtain information on medical
care utilization, expenditures, and other health-
related information. A third survey, an administrative
records survey, was designed to verify the eligibility
status of the household survey respondents for the
Medicare and Medicaid programs. It also checked
insurance claims filed with the national Medicare
program and Medicaid programs in each of the four
States for persons in the sample of Medicaid eligibles.

Sample design. —The national Cycle I household
survey comprised persons residing in about 6,000
households. The sample for this survey was a multi-
stage area probability sample drawn from 106 pri-
mary sampling units representing the 50 States and
the District of Columbia. The State Medicaid house-
hold survey sample consisted of about 1,000 families
in each of the four States; these families were selected
with a known probability of selection from the State
Medicaid enrollment lists. Thus, the total sample for
the survey was about 10,000 households.

An overall response rate of 89.4 percent was
achieved in the first interview for both household
surveys in Cycle 1: for the national survey the
response rate was 91.4 percent, and for the State
Medicaid survey the rate was 86.7 percent. Attrition
over the course of interviewing resulted in final
response rates of 84.9 percent for the National
household survey and 76.1 percent for the State
Medicaid household survey.

Data collection procedures —NMCUES is a panel
study. Interviews in Cyclel were conducted with
each household at approximately 3-month intervals,
with interviewing beginning in February 1980 and
ending in March 1981. The first two interviews were
conducted by personal visit of the interviewer to the
household, the next two were conducted by tele-
phone (if a telephone was available and acceptable to
the household), and the final interview was con-
ducted in person. Each round of interviewing asked
about the period following the preceding interview
except the first round, which asked about the period
of time following January 1, 1980.

Collection of data from the households was
facilitated by the use of a calendar and a summary.
At the time of the first interview, the household
respondent was given a calendar on which to record
information about health problems and health serv-
ices utilization and to assemble physician and other
provider bills between interviews. Following each
household interview, information about health pro-
vider contacts and the payment of charges associated

with them was used to generate a computer summary
of information provided. This summary was then
printed out in a simple format and mailed to the
household for their review for accuracy and com-
pleteness prior to the next interview. At the subse—
quent interview, the interViewers reviewed this infor-
mation with the household respondent to ensure
accuracy and to obtain information not available
during a previous interview.

The administrative records survey of Cycle I was a
check of the eligibility and claims records of persons
reported as covered by Medicare or Medicaid. The
supplementary and confirmatory data on Medicaid
and Medicare enrollees provided information from
the administrative records of the programs for com—
parison with that reported by the household respond—
ents; the survey was designed specifically to meet the
needs of the Health Care Financing Administration
for programmatic data that is not otherwise available.
For those individuals identified in the surveys as being
enrolled in either the Medicare or Medicaid programs,
identifying information was collected sufficient to
allow the Health Care Financing Administration to
flag incoming claims for services paid by Medicare, to
allow the State Medicaid agencies in the four States in
the State Medicaid survey to abstract data from
claims, and to permit all State Medicaid agencies to
confirm eligibility and type of eligibility under their
State Medicaid program.

Content of questionnaires. ——Questionnaires for
the household surveys were designed to obtain some
information on a repeated basis throughout the
survey and some information only one time. The
repetitive core of questions for CycleI included
health insurance coverage, episodes of illness, the
number of bed days, restricted activity days, hospital
admissions, physician and dental visits, other medical
care encounters, and purchases of prescribed medi-
cines. For each contact with the medical care system,
data were obtained on the nature of the health
conditions, characteristics of the provider, services
provided, charges, sources, and amounts of payment.
Questions asked only once included data on access to
medical care services, limitation of activities, occupa-
tion, income, and other sociodemographic character-
istics.

Data processing and quality control methods.—
Data collection and processing phases incorporated a
number of quality control activities, including a
detailed edit by each field interviewer and, immedi-
ately upon receipt at the central office, a premachine
edit of all key linkage variables. Errors discovered in
this premachine edit and in further edit of a sample
of all items provided feedback to interviewers and
supervisors.

Illness conditions, operations performed, special-
ties of medical providers not precoded on the
questionnaire, and the industries in which persons

were employed were coded prior to data entry.
Quality control samples provided continuous moni-
toring of each coder, with retraining of interviewers
and recoding of data, as appropriate, to ensure high
levels of accuracy. Data entry through key to disk
minicomputers provided a further level of editing for
valid ranges and codes. Check digits embedded in key
identifying numbers kept miskeyed linkage numbers
to an absolute minimum.

Release of data. ~Data from Cycle I of NMCUES
will be released in several forms. The publications
program is a collaboration between the National
Center for Health Statistics and the Health Care
Financing Administration. Preliminary data will be
published in Preliminary Data reports, similar to the
Center Advance Data reports. More detailed series
reports based on the survey findings will be issued in
a joint series with the Center preparing reports
pertaining to the population as a whole and the
Health Care Financing Administration preparing re-
ports specific to the Medicare and Medicaid popu-
lations. A public use data tape is planned with release
targeted for late 1982. The tape will actually be
several sets of files including the national household
data and the State data for each of the four States in
the administrative records survey. Each set of files
will contain data on the family, the individual, and
the health care visit, pertaining to expenditures,
utilization of services, and measures of health. In
addition, unpublished tabulations will be made avail-
able; and, as resources, permit, special tabulations will
be prepared upon request.

Uses of data. —The National Center for Health
Statistics has the responsibility for meeting the data
needs of the Department and other users for expendi-
ture and utilization data. The NMCUES data provide
a statistical base for major health policy decisions on
key issues such as health care financing and cost
containment. Data are available for the Medicare and
Medicaid populations and non-Federal beneficiaries
so that comparisons can be made. Also, some data
were collected on the State level for Medicaid
beneficiaries because of the high degree of variability
of eligibility requirements and service coverage in the
State-managed Medicaid programs.

In 1979 national health care expenditures
amounted to $212.2 billion, about 9 percent of the
gross national product; and the figures have been
growing steadily for several decades. Of this $212.2
billion, 43.1 percent was paid from public sources in
1979. These figures, compiled by the Health Care
Financing Administration, are aggregates. NMCUES is
designed to provide statistics on the effects of health
care utilization and expenditures at the individual and
family level. In gross terms, who is using what types
of medical care, how much is it costing them, and
how is the bill being paid?

NMCUES provides a unique source of extremely

15

detailed health care utilization and financing informa-
tion, which will enable detailed measurement of
utilization and expenditures for different subgroups
of the population. Specific items of service are
identified that can be related to chronic and acute
illness, disability, work-loss days, and limitation of
activity. Such data are essential to the measurement
of the impact of legislation and health insurance
programs on the charges for and financing of health
services and insurance. This survey will be the
principal source of ongoing data needed to monitor
changes over time in the impact on individuals and
families of private and public programs. Specific uses
of the data include the following:

0 Providing a statistical base for the health care cost
monitoring efforts of the Department of Health
and Human Services.

0 Providing updated, comparable measures of uti-
lization and expenditures for the design and
monitoring of health insurance proposals.

0 Providing data on trends and costs of health care
services for different population subgroups over
time.

National Health and Nutrition
Examination Survey

Background—The first National Health and Nu-
trition Examination Survey, referred to as
NHANES I, was initiated in 1970, with data collec-
tion beginning in April 1971. NHANES I was a
modification and expansion of the Health Examina-
tion Survey (HES), which had been initiated a decade
earlier and had carried out three separate programs.
The restructuring and modification of HES reflected
the assignment to the National Center for Health
Statistics of an additional specific responsibility—the
measurement of the nutritional status of the popula-
tion and the subsequent monitoring of changes in
that status over time. Upon completion of
NHANES I, data were collected for a second National
Health and Nutrition Examination Survey
(NHANES 11) beginning in February 1976 and finish-
ing in February 1980. A special health and nutrition
examination survey directed to persons in families
having one or more members of Hispanic origin or
descent who live in areas with a high concentration of
Hispanics has been planned for the period between
completion of NHANES II and the beginning of
NHANES 111 (now planned to begin data collection in
1987). Pretesting for the Hispanic HANES began in
1981, and data collection will begin in 1982.

Purpose and scope. —NHANES and its predeces-
sor, HES, share a common purpose—the collection
and utilization of data that can be obtained best or

16

only by direct physical examination, clinical and
laboratory tests, and related measurement proce-
dures. This information, which cannot be furnished
by the people themselves or by the health profes-
sionals who provide their medical care, is of two
kinds. Prevalence data are collected for specifically
defined diseases or conditions of ill health; and
normative health-related measurement data are collec-
ted that show distributions of the total population
with respect to particular parameters such as blood
pressure, visual acuity, or serum cholesterol level.

The surveys use probability samples of the US.
population to provide representative national data
that are analyzed and made available in a series of re-
ports. Successive surveys in the HES and NHANES
programs have been directed to different segments of
the population and have had different sets of target
conditions. Thus, the first Health Examination Sur-
vey, or “cycle,” involved examining a sample of
adults with the focus primarily on selected chronic
diseases. The second and third cycles of HES were
directed respectively to children between the ages of
6 and 1 land to youths between the ages of 12 and 17.
Both of these surveys emphasized growth and devel-
opment data and sensory defects. The nutrition com-
ponent of the first NHANES program was directed to
a probability sample of people in the broad age range
1 to 74 years, while the detailed health examination
component focused on the population between ages
25 and 74. NHANES II was again directed to a broad
population aged 6 months to 74 years, and the data
on nutrition that were collected will be used in con—
junction with the earlier NHANES I data to monitor
changes in nutritional status over time.

Sample design. —The samples for all of the HES
and NHANES programs have been multistage,'highly
clustered probability samples. All of the samples were
stratified by broad geographical region and by popu-
lation density grouping. Within the strata, the sam-
pling stages employed have been the primary sam-
pling unit, the census enumeration district, the
segment, the household, and lastly the individual
person. Until the household stage was reached, all
sampling was carried out centrally in conjunction
with the US. Bureau of the Census.

The next stage of the sampling was conducted in
the field in the particular chosen area. It involved
interviewer visits and questionnaire completion at
each one of the selected households, with the final se-
lection of individuals included in the sample being de-
pendent upon information elicited by the household
interview questionnaire. The size of the sample in the
survey program has varied. In each of the three HES
programs, the sample size was approximately 7,500
persons. In NHANES I the sample selected for the
major nutrition components of the examination con-
tained approximately 28,000 people and yielded
about 21,000 examined persons. A sample compara-

ble in size for NHANES 11 again yielded approxi-
mately 21,000 examined persons.

Data collection procedures. —Household inter-
viewing during NHANES II was conducted by Bureau
of the Census personnel. This represented a departure
from previous surveys in which Center employees
served this function. However, Center employees did
the rest of the interviewing, as well as the history tak-
ing, examining, testing, and measuring in the mobile
examination centers. Data collection teams consisted
of specially trained interviewers and examiners, in-
cluding physicians, nurses, dentists, dietitians, and
medical, laboratory, and X-ray technicians.

The examinations take place in the survey’s spe-
cially constructed mobile examination centers, each
consisting of three truck-drawn trailers. These trailers
are interconnected and provide a standardized envi-
ronment with equipment for the performance of spe-
cific parts of the examination. This standardized envi-
ronment is necessary for such components of the
examination as audiometry, which requires hearing
chambers within which the ambient noise level con-
forms to the American Speech Association standards
for acoustical measurements.

The general pattern of data collection has meant
that each survey has been conducted over a period of
3 or 4 years. This is due to the limitations on the
number of persons examined in a given time span (for
example, the number of field teams and the number
of sample areas). The kinds of data to be collected are
also limited, since conditions that might show marked
year-to-year variation or seasonal patterns cannot be
included. However, many important chronic diseases
and health-related measurements are not subject to
such changes in prevalence within shortrun periods.
The distribution of the population according to
unassisted visual acuity levels and the prevalence of
such conditions as diabetes or hypertensive heart
disease may vary over long periods of time, but not
so rapidly as to prevent data collection over a 3- or
4-year period from giving a correct picture of the
population levels during or at the midpoint of that
period.

Voluntary sample surveys present a problem if no
data are collected on a large fraction of the selected
sample because individuals selected to participate in
the program are not willing to do so. In the HES and
NHANES program there has always been and contin-
ues to be much attention devoted to the question of
the response rate, the proportion of sample persons
who are actually examined. In NHANES there have
been, as anticipated, more problems in the area of re-
sponse than had been encountered in the earlier HES
programs. The difficulties faced have led to a variety
of innovative measures, including a policy of remu-
nerating examined persons. As a result, both
NHANES programs succeeded in obtaining household
interview data on about 99 percent of the sample

population. More detailed health data appear in the
medical history questionnaires; these were completed
for 88 percent of the selected sample persons in
NHANES I and for 91 percent of the selected sample
persons in NHANES 11. Finally, in NHANES I, 74
percent of the sample persons selected for the
nutrition component and 70 percent of the persons
selected for the detailed health component were given
the standard examinations and tests; in NHANES II
the overall response rate for the examination compo-
nent was 73 percent. There is considerable ancillary
information on most of the persons in the sample
population who were not examined, and it is possible
to make use of that data in the process of imputation
and analysis of nonresponse bias. There is, moreover,
some evidence that data obtained through examina-
tions, tests, and measurements such as those used in
these surveys are less susceptible to potential bias
from a given rate of nonresponse than are data
provided by the individuals themselves.

Content of questionnaires, tests, and examina—
tions—The kinds of information collected in the
NHANES and other examination survey programs are
so varied and extensive that they are only illustrated
here. With respect to nutrition, four types of data are
included:

0 Information concerning dietary intake—taken
from 24-hour recall interviews and food fre-
quency questionnaires, both administered by an
interviewer who has been a trained dietitian.

O Hematological and biochemical tests—a sizable
battery of such tests, with processing at the mo-
bile examination centers where necessary, but for
the most part at a central nutrition laboratory es-
tablished at the Centers for Disease Control.

0 Body measurements—an especially important bat-
tery in connection with infants, children, and
youths, where growth may be affected by nutri—
tional deficiencies.

0 Various signs of high risk of nutritional defiv
ciency—based on clinical examinations.

The health (as distinguished from nutrition) com-
ponent of the NHANES program includes detailed ex-
aminations, tests, and questionnaires, which have
been developed to obtain a measure of prevalence
levels of specific diseases and conditions. These vary
with the particular program and have included such
conditions as chronic rheumatoid arthritis and hyper-
tensive heart disease. Important normative health-
related measurements, such as height, weight, and
blood pressure, are also obtained.

A major element in the health component of
NHANES I was an assessment, using index condi-
tions, of unmet health needs. For the index condi-
tions the examination established whether or not
emphysema or another chronic respiratory disease

17

was present. At the same time, information was ob-
tained from the examined person with respect to
self-perceived health needs and actions taken in
seeking medical care. Analytic plans call for the
interrelating of these two kinds of information to
produce measures of unmet health care needs.

In NHANES II the nutrition component remained
nearly identical to that fielded previously. From an
early analysis of the NHANES I data, it was decided
to focus the nutritional examination elements on an
anemia-related assessment approach. This involved
the addition of certain medical history items and a
more tailored set of laboratory determinations. Less
was done in the area of health care needs. However,
the emphasis placed on the effects of the environ—
ment upon health will probably be continued in
future programs. Data were gathered to measure the
levels of pesticide exposure, the presence of certain
trace elements in the blood, and the amounts of car-
bon monoxide present in the blood. In the medical
area, primary emphasis in NHANES II was placed on
diabetes, kidney and liver functions, allergy, and
speech pathology.

The time required for the examination varies with
the content of the examination and the age of the ex-
aminee. The time constraint included among planning
factors has been that the total examination time not
exceed 2-2/2 hours. Much attention is given in the
planning process and in the actual pretests of a survey
to the actual flow of examinees through the examina-
tion center, and every effort is made to streamline
this process in order to reduce the time burden on
the sample person. Additional respondent burden
arises from the interview, from the completion of
forms and questionnaires in the household, and
from the varying time required to travel to the
examination site. Data collection forms used in
HES I, II, and III, NHANES I, the NHANES I
Augmentation Survey, and NHANES II are repro-
duced in Series 1, reports numbered 4, 5, 8, 10b,
14, and 15, respectively.

Data processing methods. —The data collected in
NHANES require a variety of data processing meth-
ods. X-rays must be interpreted; blood and other lab-
oratory specimens must be processed through the ap—
propriate laboratory operations. Certain data, such as
electrocardiographic tracings, record directly onto
magnetic tape and must undergo appropriate proc-
essing to be translated into digital tape form and sub-
sequently to be interpreted. Examination record
forms and interview and questionnaire data must be
coded and put onto magnetic tape.

Methods for handling specific examination ele-
ments are chosen on the basis of appropriateness. The
program makes use of precoded forms, marked-sense
record forms, self-administered forms, interviewer-
administered forms, automatic recording devices,
photographs, and so forth. Throughout whatever

18

process is chosen, constant emphasis is placed on
quality control measures such as editing, verifying,
and replicating. Finally, outside consultants fre-
quently collaborate in the analysis and reporting of
the resultant data.

Release of data. —Findings from NHANES pro-
grams traditionally are presented primarily through
publication of individual reports in the National Cen-
ter for Health Statistics Vital and Health Statistics,
Series 11. In order to make NHANES findings more
rapidly available to consumers, a program has been
initiated to release basic descriptive summary tables
in what are called “basic data publications.” These
publications are an adjunct to the other Series 11
reports for selected tOpics. Additionally, some
NHANES data appear in abbreviated preliminary
reports from Advance Data.

Data are also released through appropriate scien-
tific journals, separate monographs, special reports,
and data tapes. Because of the voluminous amount of
data collected in NHANES I and the need for dissem-
inating data as expediently as possible, a policy shift
was made that allows for the simultaneous release of
data tapes to in-house analysts and sale of these tapes
to the public as soon as final editing has been per-
formed and documentation prepared. Data tapes
from NHANES II are also being released upon com-
pletion of final editing and preparation of docu-
mentation.

Published reports are not issued on a set fre-
quency but rather made available as completed. The
reports are organized on a topical basis with, for ex-
ample, one report presenting data on periodontal dis-
ease, another on auditory acuity, and so forth.
Generally, the first reports produced are descriptive
in nature while later reports become more analytic
and maybe interdisciplinary in approach.

Uses of data. —The diverse nature of the data at-
tracts a wide variety of users. In addition to policy
makers and planners in Federal, State, and local
health agencies, users of NHANES data include large
numbers of health workers in the fields of nutrition,
child health, pediatrics, anthropology, medical educa-
tion, and health research, who have requested reports
on body measurements, growth, and development.
Data on body measurement have also been used by
many industrial groups, including automobile manu-
facturers. Since some of the primary users of
NHANES and HES data have been authors of text
and reference materials used in medical schools and
by health workers, these normative data reach a vast
audience.

The sale of data tapes from this program has in—
creased greatly during the past few years, with known
data users numbering in the hundreds. Several col-
leges and universities have purchased data tapes to
be included in their computer tape libraries. Some of
the users are medical research institutions and individ—

ual researchers who wish to explore in detail specific sumption of food items for marketing studies, small
elements of the data. Projects'currently underway scale or local health surveys that use the NHANES
using NHANES data tapesinclude research on the ef— data as norms, and epidemiological studies that ex-
fects of alcohol intake on health, research on the con- amine geographic patterns in morbidity.

19

Health resources utilization statistics,

National Hospital Discharge Survey

Background. —The National Hospital Discharge
Survey (NHDS) is the principal source of information
on inpatient utilization of short-stay hospitals. Data
collection began in 1964 and has been continuous
since then.

Purpose and scope. —The purpose of NHDS is to
produce statistics that are representative of the ex-
perience of the US. civilian population discharged
from short-stay hospitals. Specifically, the survey pro-
vides information on characteristics of patients,
lengths of stay, diagnoses, surgical operations, and
patterns of use of care in hospitals of different size
and ownership in the four major geographic regions
of the country.

The scope of NHDS is limited to discharges from
non-Federal hospitals in the 50 States and the District
of Columbia. Only hospitals with six or more beds
and an average length of stay for all patients of less
than 30 days are included in the sample.

Sample design. —The unit of enumeration in the
survey is a hospital discharge. The sample plan is ba-
sically a two-stage stratified design. The first stage is a
sample of about 10 percent of the short-stay hospi-
tals, excluding Federal hospitals, listed in the
National Master Facility Inventory (the sampling
frame—see section “National Master Facility Inven-
tory”). The primary stratification variables are bed
size and geographic region. Hospitals are selected in
direct proportion to size so that hospitals with 1,000
or more beds are selected with certainty, and hospi-
tals with less than 50 beds are sampled with a prob-
ability of approximately 1 in 40. Growth in the in-
ventory of hospitals is represented in the survey by
a sample of hospitals selected from a special universe
of new hospitals.

The second stage of the design is a systematic
sample of the discharges from the sampled hospitals.
‘The sampling frame in nearly all hospitals is the
daily listing of discharges. The size of the within-

20

 
 
  

hospital sample varies inversely with the ize of the
hospital, from about 1 per 100 discharge in hospi-
tals with 1,000 or more beds to 4 per 10 ischarges
in hospitals with less than 50 beds. The 0 erall sam-
pling rate for each bed-size group is about 1 per 100
discharges, the product of the first- and se ond-stage
sampling rates.

In 1979 the sample consisted of 54 hospitals
from a universe of approximately 8,000 short-stay
hospitals. Of the 496 hospitals in the 50 pe of the
survey, information was collected from 4 6 partici-
pating hospitals (an approximately 84-
sponse rate) on approximately 215,000 dis harges.

Data collection procedures. —The US. Bureau of
the Census, acting as the data-collecting ag nt for the
National Center for Health Statistics, indu ts sample
hospitals into NHDS. After induction, h spitals are
visited at least once a year by a represent tive of the
Bureau of the Census. At that time survey procedures
are reviewed, and information about the hospital is
updated.

Discharge data are collected throughout the year.
Sample discharges are systematically selected, usually
on the basis of the final digit or digits of the patient’s
medical record number. For each sample discharge,
an abstractor records personal, administrative, diag—
nostic, and surgical information from the face sheet
of the patient’s medical record onto a medical ab-
stract form. Data collection frequency depends on
the arrangement made with the hospital. In about 35
percent of the participating hospitals, a representative
of the Bureau of the Census visits the hospital every 2
months, completes the abstract forms for records se-
lected during the previous visit, and selects records
for abstracting at the next visit. This allows time for
records to be completed and properly filed (or pulled
from file) prior to the visit. In about 65 percent of
the hospitals, the same forms are 'completed by mem-
bers of the hospital medical record department. All
completed forms are forwarded to one of the Census
regional offices for review and then to the National

Center for Health Statistics for coding and data proc-
essing operations.

Content of medical abstract form. —The medical
abstract form contains items relating to the personal
characteristics of the patient, including date of birth,
sex, race, and marital status, but not name and ad-
dress; administrative information, including admission
and discharge dates, discharge status, and medical
record number; and medical information, including
diagnoses and surgical operations or procedures. It is
estimated that medical record personnel can sample
and complete each form, on the average, in about 5
minutes.

The contents of the medical abstract form did not
change from the inception of the survey until 1977,
when modifications were made so that it more nearly
parallels the Uniform Hospital Discharge Data Set.
The items added to the abstract at that time are resi-
dence of patient (zip code), expected source of pay-
ment, disposition of patient, and dates of procedures.
In 1968-70 actual hospital charges by service and pay-
ments by source were recorded on a ledger abstract
form for approximately one-third of the sample dis-
charges. A copy of the medical abstract form is
included as an appendix in Series 13 reports based on
data from NHDS.

Data processing methods. —After transmittal from
the hospital, abstract forms are subject to two re-
views, two machine edits, and two quality control
procedures. Forms are reviewed for completeness at
the Census regional offices and either forwarded to
the National Center for Health Statistics or returned
to the hospital for more information. On receipt at
the Center forms are again checked and, if necessary,
returned to the hospital. After review, up to seven
diagnoses and four operations are coded according to
the clinical modification of the International Classi-
fication of Diseases—ICD-9-CM. Coded information is
then keyed from abstract to disk, and a preliminary
machine-edit program checks for missing, invalid,
and inconsistent codes. The information is corrected,
if necessary, and then transferred from disk to com-
puter tape. For 1970-78 data, up to five diagnoses
and three operations were coded according to the
eighth revision of the International Classification of
Diseases—ICDA-8.

Release of data. —Annual data are published in
Advance Data and in the Center’s Vital and Health
Statistics Series 13. The publication program is, at a
minimum, to update nonmedical, medical, and surgi-
cal data for characteristics of patients and hospitals.
Special reports on average length of stay, patient
charges, geographic utilization, and methodology are
also published.

Unpublished data are available on request from
the Hospital Care Statistics Branch, which receives
about 1,000 requests per year, usually for specific
diagnostic and surgical listings in the International

Classification of Diseases. In addition, data for the
years from 1970 to the present are available on public
use data tapes.

The names of the participating hospitals and all
information related to individual patients are confi-
dential. No data are released in published, unpub-
lished, or tape form that could identify hospitals or
patients.

Uses of data. —The National Hospital Discharge
Survey is the principal source for national data on the
characteristics of patients discharged from short-stay
hospitals. The data are used for a variety of planning,
administrative, and evaluative activities by govern-
mental, professional, scientific, educational, and
commercial institutions, as well as by private citizens.
The users include Federal agencies such as the Na-
tional Institutes of Health, the Centers for Disease
Control, and the Indian Health Service; universities
and medical schools; professional organizations such
as the American Medical Association, American Hos-
pital Association, and World Health Organization;
hospitals; medical research laboratories; pharmaceutié
cal and medical supply manufacturers; publishing
houses; market research groups; and insurance
companies.

National Ambulatory Medical Care Survey

Background. —In May 1973 the National Center
for Health Statistics inaugurated the National Ambu-
latory Medical Care Survey (NAMCS) on a continuing
basis to gather and disseminate statistical data about
ambulatory medical care provided by office-based
physicians to the population of the United States.
The need for such a system had been recognized
many years before, having been pointed out as early
as 1953 by the Subcommittee on National Morbidity
Surveys of the US. National Committee on Vital and
Health Statistics. During a 5-year feasibility study
period beginning in 1967 three major national pilot
studies and several small area studies, combined with
extensive consultation with experts in various areas of
medical care delivery, were employed in developing
the current NAMCS instruments and procedures.
Through this process the current NAMCS materials
and procedures have been refined to request only in-
formation considered essential to describe adequately
the utilization of ambulatory services and to require a
minimal amount of time for the participating phy-
sicians.

Purpose and scope. —The purpose of NAMCS is to
meet the needs and demands for statistical informa-
tion about the provision of ambulatory medical care
services in the United States. Ambulatory services are
rendered in a wide variety of settings, including phy-
sicians’ offices, neighborhood health centers, and hos-
pital outpatient facilities. It is expected that NAMCS
will eventually encompass all of these settings and

21

that appropriate survey instruments and methodology
will be developed as resources permit. Currently, the
NAMCS target population consists of all office visits
within the conterminous United States made by am-
bulatory patients to non-Federal physicians who are
in office—based practice and engaged in direct patient
care. Excluded are visits to hospital-based physicians;
visits to specialists in anesthesiology, pathology, and
radiology; and visits to physicians who are principally
engaged in teaching, research, or administration. Tele-
phone contacts and visits other than office visits are
also excluded. Since about 70 percent of all direct
ambulatory medical care visits occur in physicians’
offices, the current NAMCS design provides data on
the majority of ambulatory care services.

Sample design—The most objective and reliable
sources of data about physicians’ services rendered to
ambulatory patients during office visits are the physi-
cians themselves and members of their office staff.
The sampling frame is a list of licensed physicians in
“office-based, patient care” practice compiled from
files that are classified and maintained by the Ameri-
can Medical Association and the American Osteo-
pathic Association. These files are continuously up-
dated by both associations, making them as current
and correct as possible at the time of sample se-
lection.

NAMCS uses a modified probability-proportional-
to-size sampling procedure using separate sampling
frames for standard metropolitan statistical areas and
for nonmetropolitan counties. After sorting and strat-
ifying by size, region, and demographic characteris-
tics, each frame is divided into sequential zones of 1
million residents, and a random number is drawn to
determine which primary sampling unit from each
zone is included in the sample. The NAMCS final
first-stage sample contains 87 primary sampling units,
corresponding to individual counties or small groups
of contiguous counties across the country.

The second-stage sample is selected from the list
of physicians located in the sample primary sampling
units, ordered by major specialty categories so that
theoverall probability for including any individual
physician is approximately constant. The present
annual sample consists of approximately 3,000 physi-
cians. The sample physicians are randomly distributed
across the 52 weeks of the year so that the resulting
data reflect any seasonal variations. Since the assign-
ment of the reporting week is an integral part of the
sample design, each physician is required to report
during his predetermined period, and no substitute
reporting periods are permitted. Approximately 75
percent of the eligible physicians in the sample par-
ticipate in the survey. From this size physician sam-
ple, information is secured from about 50,000 patient
visits a year. Samples for subsequent years exclude,
with certainty, sample physicians included in the pre-
vious 2 years of the study.

22

The final stage involves sampling patient visits
within a physician’s practice. The sampling rate,
which is determined at the time of the interviewer’s
appointment, is dependent on the number of days
during the reporting week that the physician is in
practice and the number of patients expected to be
seen. In actual practice, the sampling procedure is
handled through the use of a Patient Log (see “data
collection procedures”).

Field procedures. —To maximize participation
levels and minimize data reporting burden in the phy-
sician’s office, NAMCS field procedures have been de-
signed to accommodate the circumstances of individ-
ual physicians. Each physician is contacted by several
means, including mail, telephone, and personal inter-
view. Initially, each physician in the sample is sent an
introductory letter from the Director of the Center.
The physician is then telephoned by an informed and
trained interviewer who explains the survey briefly
and arranges a personal appointment to relate more
detailed instructions. During this appointment, the
interviewer verifies the physician’s eligibility for par-
ticipation in the survey, delivers survey materials with
printed instructions, provides detailed verbal instruc-
tions, and assigns a predetermined 7-day (Monday
through Sunday) reporting period. Also, during this
appointment data concerning basic practice character-
istics, such as the physician’s specialty and staff
makeup, are collected for use in analysis.

At the beginning and again during the physician’s
assigned reporting' week, the interviewer contacts the
sample physician to answer possible questions and to
ensure that procedures are going smoothly. At the
end of the week, the participating physician mails the
finished forms to the interviewer, who edits them for
completeness before transmitting them for central
data processing.

Data collection procedures. —The actual data col-
lection for the National Ambulatory Medical Care
Survey is carried out by the participating physician,
aided by office assistants when possible. The physi-
cian completes a Patient Record for a sample of his
patients seen during his assigned reporting week.
Based on the physician’s own estimate of patients ex-
pected to visit during the survey period, the physician
is assigned to use an “every-patient” or a “patient-
sampling” procedure. These sampling procedures are
designed so that Patient Records are completed each
day of practice for, at most, 10 patient visits. Physi-
cians expecting 10 or fewer visits per day record
data for all of them, while those expecting more than
10 visits per day record data after every second, third,
or fifth visit, using a random start and observing the
same predetermined sampling interval continuously.
These procedures minimize the workload of data
collection and maintain equal reporting levels among
sample physicians, regardless of the size of their
practice. Each form requires 1 to 2 minutes to

complete, so that approximately 15 minutes are
required on days when patients are attended in the
physician’s office.

Two data collection forms are employed by the
participating physician: the Patient Log and the Pa-
tient Record. The Patient Log is a sequential listing of
patients that serves as a sampling frame to indicate
which visits data should be recorded for. The Patient
Record varies slightly from year to year, but it gener-
ally contains items of information about the visit
such as its date and duration; the patient’s date of
birth, sex, race, ethnicity, and reason for visit;
whether the patient had been seen for the particular
problem before and whether the patient was referred
by another physician; the length of time since onset
of the problem; diagnoses; diagnostic and therapeutic
services; and disposition. Periodically, supplementary
items are added to the basic Patient Record to in-
vestigate specific health conditions or other aspects
of ambulatory care. For example, questions about
specific medications were added for 1980-81, and
questions about accident- or product-related illnesses
were added for 1979. A copy of the Patient Record is
included in the annual “Summary: NAMCS” Advance
Data report and as an appendix in Series 13 reports
based on NAMCS data.

Data processing methods—Initial edits for com-
pleteness of the Patient Record are done by the inter-
viewer as the forrns are received from the participat-
ing physician. These forms are then sent to central
processing, where they are clerically edited for con-
sistency. The two medical questions are then coded:
the patient’s reasons for visit are coded using the
NAMCS “Reason for Visit Classification” (described
in Series 2, No. 78) and the physician’s diagnoses are
coded using the International Classification of Dis-
eases. The data are then keyed onto magnetic tape, at
which time additional machine edits and consistency
checks are run.

Release of data. —NAMCS results in the form of
summary statistical tabulations of national and re—
gional estimates for number of visits, percent distribu-
tions, and population rates of use are published as
soon as possible after each annual cycle is complete.
More detailed tabulations and analyses follow, which
present visit characteristics by major physician spe-
cialty groups, patient groups, diagnostic categories,
treatment provided, and disposition. Cross-
tabulations of less common visit characteristics are
published when sufficient data about them are avail-
able to meet practical standards of precision. These
data are released through the Center’s Advance Data
and Vital and Health Statistics Series 13 reports.
Other publications, including the various journals and
newsletters of the NAMCS endorsing organizations,
also provide a means of releasing data to the medical
community. In addition, public use data tapes are
made available approximately 1 year after the end of

data collection, when documentation of the data is
complete and its validity is checked.

As part of the procedures designed to protect the
identity of the patients, the Patient Logs containing
the names of the patients are detachable from the Pa-
tient Records. At the end of the reporting period the
Patient Logs, and thus the names of the patients, are
retained by the physician. Sample physicians are
further assured that all information that would per-
mit identification of an individual, a practice, or an
establishment will be held confidential. All findings
are released in the form of summary statistics that
preclude any individual identification, and all identi-
fying data are destroyed as soon as data processing
operations are complete.

Uses of data. —NAMCS provides a wide range of
data on the characteristics of the users and providers
of ambulatory medical care. These baseline data,
together with trend data as the survey progresses, pro-
vide new insights into ambulatory medical care and
stimulate further research on the use, organization,
and delivery of ambulatory care.

The manner in which ambulatory care is actually
dispensed, as described by the survey, can be related
to arrangements for educating and training physicians
and other health personnel to provide ambulatory
medical care, particularly primary care. Such com-
parisons suggest useful new directions and emphases
that can be incorporated in the curricula, helping to
make medical education more responsive to the shift-
ing requirements and health problems of the popula-
tion. NAMCS information is useful to health planning
agencies; managers of health delivery systems; and
others concerned with planning, monitoring, and
managing health care resources. It is valuable to those
who develop and evaluate new and modified health
care systems and arrangements. The continuing na-
ture of the survey permits observation and measure-
ment over time of different modes for managing and
treating patients’ problems, and it provides general in-
formation on the etiology and epidemiology of se-
lected conditions. It also provides valuable informa-
tion about the speed and effectiveness with which
certain advances in medical practice are adopted by
the basic source of medical care, the office-based
physician.

National Nursing Home Survey

Background—Between 1963 and 1969 the Na-
tional Center for Health Statistics conducted surveys
of nursing homes and their residents on an ad hoc
basis. With the implementation of the Medicaid and
Medicare programs, the increased utilization of nurs-
ing homes, and the projected increases in the aged
population, those who set standards for, plan, provide,
and assess long-term care services needed comprehen-
sive national data on a continuing basis. To meet their

23

needs, the National Nursing Home Survey was devel-
oped in 1972, with the initial survey conducted in
1973-74 and the second conducted in 1977.

Purpose and scope. —This periodic data collection
system is a series of nationwide sample surveys of
nursing homes, their residents, and staff. The pur-
poses of the surveys are as follows:

0 To collect national baseline data on characteris-
tics of the nursing home, its services, residents,
and staff for all nursing homes in the Nation, re-
gardless of whether or not they are participating
in Federal programs such as Medicare or Medicaid.

0 To collect data on the costs incurred by the facil-
ity for providing care by the major groupings—
labor, fixed, operating, and miscellaneous costs.

0 To collect data on certification for participation
in the Medicare and Medicaid programs (such as
the utilization of certified beds and the health of
residents receiving program benefits) so that all
data can be analyzed by certification status.

0 To provide comparable data for valid trend analy-
ses on a variety of topics. Such analyses can, for
example, identify the impact of legislative changes
in standards and in reimbursement on the growth
of facilities, and of the impact of institutionali-
zation on the health of the aged.

0 To interrelate facility, staff, and resident data to
reveal the relationships that exist between utiliza-
tion, services offered, charges for care, and the
cost of providing care.

For the initial survey conducted in 1973—74, the
universe included only those nursing homes that pro-
vided some level of nursing care, regardless of
whether or not they were participating in the Medi~
care or Medicaid programs. Thus, homes providing
only personal or domiciliary care were excluded.
Beginning with the 1977 survey, the universe was
expanded to include all nursing, personal care, and
domiciliary care homes, regardless of their participa-
tion in Medicare or Medicaid. Homes that provide
room and board only are excluded. In both surveys,
homes in the universe included those that were
operated under proprietary, nonprofit, and govern-
ment auspices. The universe also included homes that
were units of a larger institution (usually a hospital or
retirement center).

Sample design—The National Master Facility In-
ventory (NMFI) listing is the universe from which the
sample homes are selected. The NMFI listing, main-

tained by the Center, contains the basic information ,

about each home (such as name, address, size, owner-
ship, number of residents, and number of staff) that
is needed to design efficient sampling plans.

Resident data are collected by reviewing medical
records and questioning the nurse who usually pro-
vides care for the resident. Residents are not inter-

24

viewed directly. The initial National Nursing Home
Survey, conducted from August 1973 to April 1974,
had a nationally representative sample of 2,100 nurs-
ing homes, with a subsample of 25,100 staff and
19,400 residents. The second survey, conducted from
May to December 1977, had a total sample of 1,700
nursing homes, with a subsample of 16,800 staff,
7,100 residents, and 5,300 discharged residents.

Data collection procedures and questionnaire
content. —The survey system uses several question-
naires. The facility questionnaire includes questions
on number of beds and residents, services provided,
certification status, and various utilization measures.
The expense questionnaire includes questions on the
facility’s expenses by the major groupings—labor,
fixed, operating, and miscellaneous. The staff ques-
tionnaire includes questions about the employee’s
demographic characteristics, work experience, educa-
tion, and salary. The residents questionnaire includes
questions about the resident’s demographic character-
istics, health status, functional status, participation in
social activities, monthly charge, and source of pay-
ment. Included in the 1977 survey was a question-
naire for discharged residents. This included some of
the same items as the questionnaire for current resi-
dents, items selected on the basis of their availability
in the medical record.

The survey has a number of respondents in each
nursing home and is a combination of personal inter-
view and self—administered questionnaires. Facility
information is secured through a 20-minute personal
interview with the administrator. Expense data are
collected on a self-administered questionnaire, requir-
ing about 30 minutes to answer, completed either by
the facility’s accountant under authorization from
the administrator or by the administrator. Sampled
staff members complete a brief form that requires
about 5 minutes to complete. Information on sample
current residents is secured by the interviewer in a
personal interview with the nurse who provides care
to the resident and who refers to information from
the medical record. About 15 minutes is required for
each sample resident. For the 1977 survey, informa-
tion on the sample discharged residents was secured
by the interviewer in a personal interview with the
nurse who was most familiar with the medical records
and who referred to them for replying to all
questions.

All survey questionnaires for the 1973-74 survey
are included in Series 13, No. 22; and the question-
naires used in the 1977 survey are contained in Series
13, No. 43. Selected questionnaires from the surveys
are found in other Series 13 reports based on the
National Nursing Home Survey.

The 1977 survey included several modifications
of the 1973-74 survey design and methodology.

0 It collected data on discharged residents, espe-
cially in the areas of health status, length of stay,

discharge status, and where the resident was to go
after discharge.

0 It collected data on the revenues of the facility as
well as on its expenses.

0 It produced estimates for the five States with the
largest pr0portion of nursing home residents.

0 It limited respondent reporting by limiting sub-
sarnples of residents and discharges to a maximum
of 8 per facility, and of staff to 23 per facility.

The response rates for the surveys differed
according to the type of questionnaire used, as pre-
sented in table B.

Data processing methods.~All data items are
edited to ensure that all responses are accurate,

logical, and complete. Those items left unanswered-

are imputed by assigning a value from a responding
unit with major characteristics identical to those of
the nonresponding unit.

Release of data. —National estimates are available
to requesters in the form of tabulations, Center publi-
cations, and data tapes. Tables are aggregated so that
no one respondent can be identified. Similarly, no
identifying information concerning the facility, staff,
or residents is contained on the data tapes, although a
pseudoidentification number allows linkage of data
among the various files.

A report analyzing provisional tabulations is avail-
able 6 months after data collection ends. Selected
final tabulations are released 6 months later. Data are
analyzed and presented in the Center Advance Data
reports and in Vital and Health Statistics Series 13
and 14. These reports present data on facilities (utili-

 

Table B. Response rates for National Nursing Home Surveys, by type
of questionnaire used (percent): United States, 1973-74 and 1977

 

Response rate

Questionnaire 1977 7973-74

 

survey survey
Facility .................... 95 97
Expense ................... 85 88
Staff ..................... 81 82
Residents ................... 99 98
Discharges .................. 97 . . .

 

zation, expenses, and services), residents (demo-
graphic characteristics, health status, services re-
ceived, and charges), discharged residents, and staff
(number, training, and experience). Once data have
been reviewed and analyzed, data tapes are available
for release to requesters.

Uses of data. —Nursing homes are the most rapidly
expanding sector of the health care industry. Federal
and State funds spent for nursing home care have
been increasing each year. Thus, national data on the
costs and characteristics of the nursing home, its resi-
dents, discharged residents, and staff, are needed on a
continuing basis by the Department of Health and
Human Services for program planning and the setting
of national policies. Congress needs information on
the interrelationships among facility-resident—cost
data in order to draft legislation on standards and
reimbursement. In addition, these data are needed by
professionals who conduct epidemiological research
and those who plan, provide, and assess long-term
care services at State and local levels.

25

Health resources statistics

National Master Facility Inventory

Background. —The National Master Facility Inven-
tory (NMFI) is a comprehensive file of the 33,000
facilities in the United States that provide medical,
nursing, personal, or custodial care to groups of
unrelated persons on an inpatient (at least overnight)
basis. Facilities in NMFI are categorized into three
broad types: hospitals, both short- and long-stay;
nursing and related care homes; and other custodial
or remedial care facilities, including homes or resident
schools for the deaf, blind, mentally retarded, emo-
tionally disturbed, and other neurologically impaired
or physically handicapped; resident treatment centers
for alcohol and drug abusers; orphanages or homes for
dependent children; and homes for unwed mothers. It
is the most comprehensive file of inpatient health
facilities available in the United States.

Purpose and scope—NMFI has two basic pur-
poses. It is an important national source of statistics
on the number, type, and geographic distribution of
inpatient facilities in the United States. In addition, it
serves as the universe from which probability samples
are selected for conducting sample surveys.

Data collection procedures. —NMFI was first
assembled in 1962-63 by collating the files of four
Federal agencies that contained the names and
addresses of facilities, the directories of national
associations and organizations, and State licensure
files.

Two mechanisms are used to keep the data in
NMFI as current as possible. For all facilities except
hospitals, the National Center for Health Statistics
conducts a series of mail surveys, first, to ensure that
the data on file on the basic characteristics of the
facilities are accurate and, second, to identify and
then delete those facilities that have gone out of busi-
ness or are no longer eligible for inclusion. These
surveys are conducted approximately once every 2
years.

In addition, at regular intervals State licensure
agencies, national voluntary associations, and other

26

appropriate sources send to the Center their most
recent directories or lists of new facilities. These lists
are then clerically matched with the most current
NMFI file, and facilities not already included are
added.

. For hospitals, data were gathered annually in a
joint survey of the American Hospital Association
and the Center. The contractual arrangement, which
in effect merged the hospital portion of NMFI with
the American Hospital Association’s annual survey of
hospitals, began in 1968. Through 1975 the American
Hospital Association performed the data collection
(from October through January each year) for its
member hospitals, while the Center performed the
data collection for the approximately 400 hospitals
that were not registered with the American Hospital
Association. The American Hospital Association proc-
essed both portions of the survey according to the
same edit and processing specifications and delivered
to the Center two edited tapes, one for American
Hospital Association hospitals and one for nonmem-
ber hospitals. In 1976 and 1977 the American
Hospital Association surveyed all hospitals as part of
its annual survey. The contractual relationship with
the American Hospital Association ceased in 1978.
Since that time the Center has purchased hospital
data tapes annually from the American Hospital
Association, tapes which are solely for the Center’s
internal statistical use.

During the period 1976-80, under the health
facilities component of the Cooperative Health Sta-
tistics System (CHSS), health facilities data for some
States were collected and processed by the State
CHSS contractor as a byproduct of the licensure
renewal process. Edited data were then provided on
computer tape by each CHSS State to the Center,
and these data were merged into NMFI. Data for all
facilities, except hospitals, in States not participating
in CHSS continued to be collected by the Center
every 2 years as part of NMFI. The next NMFI survey
is planned for 1982.

Content of data collection instruments. —The fol-

lowing types of data are collected for the three
categories of facilities found in the National Master
Facility Inventory—hospitals, nursing homes, and
other facilities:

Hospitals: Ownership; major type of service of-
fered; whether various facilities and services are
offered; number of beds, admissions, inpatient
days of care, and discharges; patient census;
number of bassinets, live births, and newborn
days of care; outpatient utilization; number of
surgical operations; revenue, expenses, and assets;
and staffing.

Nursing homes: Ownership; major type of service;
licensed and staffed beds; beds certified for
Medicare and Medicaid; admission policy with
regard to age, sex, and various conditions; patient
census by age and sex; inpatient days of care;
number of admissions, discharges, and deaths;
staffing; who is in charge of nursing care; number
of patients receiving nursing care; services rou-
tinely provided; basic monthly charge; and oper—
ating expenses.

Other facilities: Ownership; major type of service;
licensed and staffed beds; beds certified as inter-
mediate care beds; admission policy regarding age
and sex; patient census by age and sex; inpatient
days of care; number of admissions, discharges,
and deaths; staffing; basic monthly charge; and
operating expenses.

A copy of the data collection instruments is included
in Series 14 reports based on NMFI data.

Data processing and quality control. —As already
stated, NMFI is considered to be the most compre-
hensive file of inpatient health facilities available in
the United States. In order to measure statistically
the extent of its coverage, a “complement” survey
was developed. It is an application of a general
technique often called a “multiframe survey.” In this
application there are two frames—the National Master
Facility Inventory and a geographic area sample list.
From a probability area sample, all institutions found
in the sample areas are identified, and the probability
with which each comes into the sample is determined.
Those inpatient health facilities found in the area
sample survey (the complement survey) are matched
against the NMFI list of facilities for that area. Any
facility discovered within the scope of the sample
survey but missed by NMFI constitutes undercover-
age, and an appropriate weight is assigned to the
omitted facility.

Data for nonresponding facilities, as well as data
for questions not answered on the returned question-
naires, are obtained through various types of imputa-
tion procedures. Whenever possible, data from the
previous NMFI survey are used to replace missing
data. For those responding facilities that omitted data
that are not available from a previous survey, an

imputation method is used. In this method, all
facilities of the same type are stratified by predeter-
mined variables such as bed size groups, ownership
categories, and major types of service. Missing data
for a facility are then supplied by using the data from
a facility of like characteristics. When a responding
facility fails to answer any of the major questions (re-
garding, for. example, bed size, ownership, type of
facility, or number of patients), a followup question-
naire containing the questions omitted is sent to the
respondents and they are asked again to answer.

Release of data. —Data from NMFI surveys are
published in various Center reports. The biennial
editions of Health Resources Statistics contain the
most current available data (often preliminary esti-
mates) from these surveys. Final data are published in
Series 14 of Vital and Health Statistics and in two
special reports (one for hospitals and one for nursing
homes) showing data by county and standard metro-
politan statistical area. Directories of facilities for the
mentally retarded and of nursing homes have been
published from NMFI survey data. Methodology
reports appear in Series 1 and 2 of Vital and Health
Statistics.

NMFI data are also released in the form of special
tabulations prepared specifically for individual re-
quests, computer tapes available through the Center’s
data tape release program, and special printouts or
listings from the data tapes. Approximately 45-50
tape copies are purchased annually by such organiza-
tions as drug companies, State agencies, other Federal
agencies, private contractors working on projects for
other government agencies, universities, and national
health care organizations.

Only the financial data from the hospital data
tapes prior to 1977 are confidential; all other
information is available for individual hospitals. How-
ever, beginning with the 1977 hospital data tape, the
data release agreement between the American Hospi-
tal Association and the Center has limited the
availability of hospital data. Portions of the nursing
home and other health facility data are collected
under a pledge of confidentiality and can be released
only in statistical aggregates or with all identification
items stripped.

National health professions inventories and surveys

Background. —The National Center for Health
Statistics first embarked on a health professions
statistics program in 1964, when an ongoing program
involving the compilation of secondary data on eight
health occupations was transferred from the Division
of Public Health Methods to the Center. As an
adjunct to secondary source data, various surveys of
facilities and health occupations were conducted in
subsequent years. In 1974 the Center’s health profes-
sions program was further enhanced by the transfer

27

of various surveys from the Bureau of Health Man-
power (now the Bureau of Health Professions) to the
Center.

Purpose and scope. —The primary purpose of the
Center’s health professions data collection is to
produce reliable statistics on distribution; availability;
and demographic, educational, and professional char-
acteristics of persons employed in the health occupa-
tions. Under the health professions component of the
Cooperative Health Statistics System for the 13
health occupations that are licensed in all of the 50
States and the District of Columbia, data were
collected in the late 1970’s on an inventory basis
covering all licensees in participating States. The 13
health occupations covered were doctors of medicine,
doctors of osteopathy, pharmacists, chiropractors,
dentists, dental hygienists, registered nurses, licensed
practical nurses, nursing home administrators, optom-
etrists, physical therapists, podiatrists, and veterinari-
ans. Data can be obtained on the nonlicensed health
occupations largely through special surveys of facili-
ties or of other employers. Special surveys are
required because of the large number of employers of
health workers, the limited use of licensure in the
health field, the large and rapidly growing numbers
and types of health workers, and the inadequate data
base.

Data collection procedures—Recent health pro-
fessions inventories were conducted principally
through the mechanism of the Cooperative Health
Statistics System. Those States that participated in
CHSS collected and processed the data according to
specifications provided by the Center. They then
provided the Center with data tapes. For those States
not participating in CHSS, data for three professions
(pharmacists, optometrists, and registered nurses)
were collected by the national organizations repre-
senting these professions, under contract to the
Center. The Center then merged data from these
various sources to obtain national data for the three
professions.

Data collection was performed in conjunction
with each State’s licensure renewal mechanism, either
on an annual or on a biennial basis as dictated by the
licensing procedures in each State. The licensure
mechanism offers a number of distinct advantages for
the development of a data system for the production
of timely, accurate, and complete statistics on health
professionals. Licensure data provide accurate counts
and distributions by geographic areas of licensed
health personnel and contain the necessary informa-
tion for contacting all persons employed in a health
profession in order to obtain additional detail.

Data were obtained by means of a self-
administered questionnaire, which was either mailed
out as an attachment to the application for license or
enclosed as a separate form. Both mail and telephone
followup procedures were used with nonrespondents.

28

Approximately 10 minutes was required for the
completion of a questionnaire. Item completion rates
of at least 90 percent were generally achieved, and
coverage approached 100 percent of the universe.

The national health professions surveys con-
ducted by the Center have been of two types:
censuses of facilities (such as the National Master
Facility Inventory) to gather headcount data for
detailed lists of health occupations employed in such
facilities as hospitals, nursing homes, group practices,
and clinics; and surveys of facilities to gather detailed
information about specific types of health workers,
such as nursing personnel in hospitals, health care
administration personnel, and public health per-
sonnel.

The mail questionnaire represents the primary
method of data collection for most of these surveys.
Both mail and telephone followup procedures were
used. Time required for completion of the question-
naire varied, depending on the content of the
particular survey. Response to these surveys generally
exceeds 80 percent.

Content of questionnaire. —For the health profes—
sions inventories conducted through CHSS,
occupation—specific minimum data sets were estab-
lished for some of the 13 occupations. Each mini-
mum data se‘t consisted of standard core items basic
to all 13 health professions and additional data items
unique to each specific occupation. The following
items were generally included for all occupations:

Administrative data: Licensing State, occupation,
type of licensure action, name, and mailing
address of licensee.

Education and training: Location of school of
graduation, year of graduation, and total years of
education or highest degree.

Demographic characteristics: Date of birth, sex,
marital status, race and ethnicity.

Employment characteristics: Occupational spe-
cialty, years active in the occupation, weeks
worked in the past 12 months, current activity
status, geographic location of place of practice 1
year ago, geographic location of current primary
place of work, principal setting of primary activ-
ity, and number of hours usually spent per week
by type of activity.

The Center’s health professions surveys vary in
content and purpose. At a minimum, the censuses of
facilities obtain information on full- or part-time
status and credentialing status. The sample surveys
generally are designed to obtain data on personal
characteristics, educational background, specialty,
practice setting, and location.

Data processing procedures. —The data from the
inventories and surveys are generally processed by the
agency or organization responsible for the data

collection. Standardized coding, editing, and imputa-
tion procedures are developed by the Center and
provided to the contractor for purposes of ensuring
both the quality and comparability of the data. The
Center unduplicates the edited data in order to
account for persons licensed by more than one State.
In addition, the Center performs considerable verifi-
cation and cross-checking of the data against other
lists and sources of health personnel data.

Release of data. —The unduplicated data are sum-
marized and disseminated by the Center to all
participating agencies in a manner that is comparable
among State and local areas in terms of content,
definitions, and format. The data appear in the
following Center publications: Viral and Health Sta-
tistics Series 14; Health Resources Statistics; and

various State reports. Special tabulations are available
on request, and copies of many of the data tapes may
be purchased. The data tapes are stripped of all
identifiable information pertaining to individuals.

Uses of data. —Analysis of the data is intended to
fulfill the following purposes: to assess current health
personnel capabilities relative to the needs for health
services at the local, State, and national levels; to
project future demands on the supply of health
professionals; to increase the efficiency of such data
collection efforts; and to reduce the effort and costs
involved in disseminating and using health professions
information. The inventories can also be used as a
sample frame for special surveys designed to obtain
more detailed or specialized information.

29

Appendixes

Contents

I. Definition of certain terms used in this report ........................................................... 31
Terms relating to population and geography .................................................. 31
Terms relating to health ............................................................... 31
Terms relating to sampIing ............................................................. 32
Terms relating to data processing ......................................................... 33

||. Listing of reports related to specific programs ................................................... 34
Basic vital statistics .................................................................. 34
Vital statistics followback surveys ......................................................... 35
National Survey of Family Growth ........................................................ 35
National Health Interview Survey ......................................................... 35
National Health and Nutrition Examination Survey ............................................. 36
National Hospital Discharge Survey ........................................................ 37
National Ambulatory Medical Care Survey ................................................... 37
National Nursing Home Survey .......................................................... 37
National Master Facility Inventory ............................. . ........................... 37

Appendix figure

Regions and divisions of the United States as defined by the U.S. Bureau of the Census .......................... 32

30

Appendix |. Definition of certain

terms used in this report

Brief definitions of some of the technical or
specialized terms used in this report are provided.
These are simplified definitions of what are often
complex procedures, and the definitions are some-
what general in nature. For detailed definitions of the
terms as they are used in the context of specific
surveys, a reader should refer to the definitions
contained in the appendix of the appropriate reports
for each survey.

Terms relating to population and geography

Resident population. —The resident population is
the population living in the United States. This
includes members of the Armed Forces stationed in
the United States and their families as well as
foreigners working or studying here; it excludes
foreign military, naval, and diplomatic personnel and
their families located here and residing in embassies
or similar quarters as well as Americans living abroad.
The resident population is often the denominator
when calculating birth and death rates and incidence
of disease.

Civilian population. —The civilian population is
the resident population excluding members of the
Armed Forces. Families of members of the Armed
Forces are included, however.

Civilian noninstitutionalized population—The
civilian noninstitutionalized population is the civilian
population not residing in institutions. Institutions
include correctional institutions, detention homes,
and training schools for juvenile delinquents; homes
for the aged and dependent (for example, nursing
homes and convalescent homes); homes for depend-
ent and neglected children; homes and schools for the
mentally or physically handicapped; homes for un-
wed mothers; and psychiatric, tuberculosis, and
chronic disease hospitals and residential treatment
centers.

Household. —All persons occupying a single hous-
ing unit are referred to as a household for sampling
purposes.

Family. —A family is two or more persons living in
the same household who are related by blood,
marriage, or adoption. All persons living in a house-
hold who are related to each other are regarded as
one family.

Group quarters. —Group quarters are quarters
occupied by five or more persons unrelated to the
head of the household; quarters with no designated
head but with six or more unrelated persons are also
group quarters. Also included are some quarters
occupied by only one or two persons (for example,
one to five persons occupying a surgical ward of a
general hospital who have no usual residence else—
where); also included are quarters occupied by
persons in military barracks and college dormitories.

Registration area. —The United States has separate
registration areas for birth, death, marriage, and
divorce statistics, in which data are collected annually
from States whose registration data are at least
90-percent complete.

Region and division. —The 50 States and the
District of Columbia are grouped for statistical
purposes by the US. Bureau of the Census into nine
divisions within four regions as shown on the map
provided.

Terms relating to health

Condition—A health condition is a departure
from a state of physical or mental well—being.
Conditions are coded by the Center to the most
recent revision of the International Classification of
Diseases.

Incidence. —Incidence is the number of cases of
disease having their onset during a prescribed period
of time; it is often expressed as a rate (for example,
the incidence of measles per 1,000 children 5-15
years of age during a year). Incidence is a measure of
morbidity or other events that occur within a
specified period of time.

Prevalence. —Prevalence is the number of cases of
a disease, infected persons, or persons with some

31

West North Central

West South Central

NATIONAL CENTER FOR HEALTH STATISTICS

 

Regions and divisions of the United States as defined by the U.S. Bureau of the Census

other attribute present during a particular interval of
time. It is often expressed as a rate (for example, the
prevalence of diabetes per 1,000 persons during a
year).

Medicare (Title XVIII). —Medicare is a nationwide
health insurance program providing health insurance
protection to people 65 years of age and over, people
eligible for social security disability payments for
more than 2 years, and people with end-stage renal
disease, regardless of income. The program was
enacted July 30, 1965, as Title XVIII, Health Insur-
ance for the Aged, of the Social Security Act, and
became effective on July 1, 1966. It consists of two
separate but coordinated programs: hospital insur-
ance (Part A) and supplementary medical insurance
(Part B).

Medicaid (Title XIX).—Medicaid is a federally
aided but State operated and administered program.
It provides medical benefits for certain low-income
persons in need of medical care. The program,
authorized in 1965 by Title XIX of the Social
Security Act, categorically covers participants in the
Aid to Families with Dependent Children program as
well as some participants in the Supplemental Se-

32

curity Income program and other people deemed
medically needy in a participating State. States also
determine the benefits covered, rates of payment for
providers, and methods of administering the program.

Inpatient facility. —An inpatient facility is, in
general, one in which residents or patients are
admitted for overnight stay for care. Included are
hospitals, nursing homes, and various custodial and
rehabilitation facilities.

Outpatient facility. —An outpatient facility is, in
general, a place which treats patients without admit-
ting them for an overnight stay. Included are physi—
cians’ offices, neighborhood health centers, and
hospital outpatient departments.

Terms relating to sampling

Target population. —The target population is the
population about which information is wanted.

Sampling frame. —A sampling frame is a complete
list of empirical cases that constitute the universe (or
total population) of cases that correspond to the
conceptual universe to which findings of a sample
survey are generalized.

Probability sample. —The distinguishing character-
istic of this type of sample is that every unit in the
frame must have a known, but not necessarily equal,
probability of being included in the sample.

Stratified sample. ~All individual units in the
sampling frame of a stratified sample are divided into
groups or categories that are defined in such a way
that each unit appears in one and only one stratum.
Independent samples are selected within each group
or stratum. The object is to arrange the population

into strata that differ as widely as possible from each '

other but are as homogeneous as possible within each
stratum.

Area probability sample. —An area probability
sample is a procedure for demographic surveys of the
population in which the initial frame consists not of
individuals but of geographic areas. The sampling
progresses through a series of stages in which first
areas, then dwelling units, and finally the individuals
themselves are selected.

Primary sampling unit. —In multistage samples, a
primary sampling unit is the first level of sampling; in
area probability sampling the United States is often
divided into groups of counties, each group of which
is a primary sampling unit.

Sampling error. —Sampling error is that part of the
error of an estimate that is due to the fact that the
estimate is obtained from a sample rather than from a
100-percent enumeration using the same procedures.

Nonsampling error. —Nonsamp1ing error is the
error in an estimate arising at any stage in a survey
from such sources as varying interpretation of ques-
tions by enumerators, unwillingness or inability of

respondents to give correct answers, nonresponse,
improper coverage, and other sources exclusive of
sampling error.

Response rate. ——The response rate for a survey is
the percent of an eligible sample for whom informa-
tion is obtained. For an interview survey the numer-
ator of the formula is the number of completed
interviews. The denominator is the total sample size
minus noneligible respondents, that is, minus those
not meeting the criteria for a potential respondent as
defined for a particular study.

Terms relating to data processing

Coding. —Coding is a technical procedure for
converting verbal information into numbers or other
symbols that can be more easily counted and tabu-
lated.

Coding error. ——Coding error is the assignment of
an incorrect code to a survey response.

Editing. —Editing is a preliminary step in data
handling in which the responses are inspected, cor-
rected, and sometimes precoded according to a fixed
set of rules.

Imputation.—Imputation is the process of devel-
oping estimates for missing or inconsistent data in a
survey. Data obtained from other units in the survey
are usually used in developing the estimate.

Quality control. —Quality control is any observa-
tion or procedure used in any operation of a survey in
order to prevent or reduce the effect of nonsampling
errors. It includes many types of data editing and
recoding.

33

Appendix II. Listing of reports
related to specific programs

This list of reports from the National Center for
Health Statistics relating to each of the data collec-
tion programs is provided to facilitate identification
of reports about each program; the list excludes
analytic reports from the programs. Information
about the availability of these publications can be ob-
tained from the Center’s Scientific and Technical
Information Branch, Division of Data Services.

Copies of all Center reports are typically available
at large municipal libraries, medical libraries, and
libraries of schools of public health. Review of publi-
cations at these places would assist in determining
whether a particular report is of interest.

Basic vital statistics

Series 2

No.83 Comparability of Reporting Between the
Birth Certificate and the National Natality

Survey. 44 pp.

A Methodological Study of Quality Control
Procedures for Mortality Medical Coding.
40 pp.

1

Vital Registration , Systems in Five Devel-
oping Countries: Honduras, Mexico, Philip—
pines, Thailand, and Jamaica. 159 pp.

No.81

No.79

No.66 Comparability of 1Mortality Statistics for
the Seventh and Eighth Revisions of the
International Classification of Diseases,

United States. 93 pp.
Vital Signs Present at Birth. 19 pp.

Comparison of Marital Status, Race, Nativ-
ity, and Country of Origin on the Death
Certificate and Matching Census Record,
United States, May-August 1960. 47 pp.

Comparison of the Classification of Place
of Residence on Death Certificates and
Matching Census Records, United States,
May-August 1960. 60 pp.

No. 46

No.30

No.29

No.20

No. 4

Series 3
No. 19

No. 18

No. 16

No. 15

No. 6

No.4

No. 1

Series 4
No. 18

No. 12

No. 8

Comparability of Age on the Death Certif-
icate and Matching Census Record, United
States, May-August 1960. 53 pp.

Variance and Covariance of Life Table
Functions Estimated From a Sample of
Deaths. 8 pp.

Comparison of Two Methods of Construc-
ting Abridged Life Tables by Reference to
a “Standard” Table. l 1 pp.

‘National Estimates of Marriage Dissolution

and Survivorship. 31 pp.

Geographic Patterns in the Risk of Dying
and Associated Factors, Ages 35-74 Years,
United States, 1968-72. 120 pp.

Cohort Mortality and Survivorship: United
States Death Registration States, 1900-68.
36 pp.

Trends in “Prematurity,” United States,
1950—67. 51 pp.

International Comparisons of Perinatal and

Infant Mortality: The United States and
Six West European Countries. 97 pp.

Infant and Perinatal Mortality in the

United States. 87 pp.

The Change in Mortality Trend in the
United States. 43 pp.

Statistics Needed for National Policies Re-
lated to Fertility. 154 pp.

Needs for National Studies of Population
Dynamics: A Report of the US. National
Committee on Vital and Health Statistics.

31 pp.
The 1968 Revision of the Standard Certifi-
cates. 47 pp.

No. 3 Report of the International Conference on
the Perinatal and Infant Mortality Problem

of the United States. 21 pp.

National Vital Statistics Needs: A Report
of the US. National Committee on Vital
and Health Statistics. 51 pp.

Fertility Measurement: A Report of the
US. National Committee on Vital and
Health Statistics. 26 pp.

No. 2

No. 1

Vital Statistics of the United States Annual Volumes

Volume I, Natality

Volume II, Mortality, Part A
Volume II, Mortality, Part B
Volume III, Marriage and Divorce

Vital statistics followback surveys

Series 2

No.83 Comparability of Reporting Between the
Birth Certificate and the National Natality
Survey. 44 pp.

No.56 Completeness and Quality of Response in

the North Carolina Marriage Follow-Back
Survey. 42 pp.

National Survey of Family Growth

Series 2

No.87 National Survey of Family Growth, Cycle
11: Sample Design, Estimation Procedures,
and Variance Estimation. 36 pp.

No. 76 National Survey of Family Growth,Cycle 1:
Sample Design, Estimation Procedures, and
Variance Estimation. 32 pp.

Series 3

No. 17 Patterns of Aggregate and Individual
Changes in Contraceptive Practice, United
States, 1965-1975. 23 pp.

Series 4

No. 18 Statistics Needed for National Policies Re-
lated to Fertility. 154 pp.

No. 12 Needs for National Studies of Population

Dynamics: A Report of the US. National
Committee on Vital and Health Statistics.

31 pp.

National Health Interview Survey

Series 1

No. 11 Health Interview Survey Procedure,
1957-74. 153 pp.

No. 2 Health Survey Procedure: Concepts, Ques-

Series 2
No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

82

75

69

57

54

52

50

49

48

45

.41

38

37

26

18

16

tionnaire Development, and Definitions in
the Health Interview Survey. 66 pp.

Small Area Estimation: An Empirical Com-
parison of Conventional and Synthetic
Estimators for States. 19 pp.

Synthetic Estimation of State Health Char-
acteristics Based on the Health Interview
Survey. 22 pp.

A Summary of Studies of Interviewing
Methodology. 78 pp.

Net Differences in Interview Data on
Chronic Conditions and Information De-
rived From Medical Records. 58 pp.

Quality Control and Measurement of Non-.
sampling Error in the Health Interview
Survey. 53 pp.

Reliability of Estimates with Alternate
Cluster Sizes in the Health Interview Sur-

vey. 17 pp.

Optimum Recall Period for Reporting Per-
sons Injured in Motor Vehicle Accidents.

31 pp.

Reporting Health Events in Household In-
terviews: Effects of an Extensive Question-
naire and a Diary Procedure. 80 pp.

Interviewing Methods in the Health Inter-
view Survey. 86 pp.

Reporting Health Events in Household In-
terviews: Effects of Reinforcement, Ques-
tion Length, and Reinterviews. 70 pp.

Effect of Some Experimental Interviewing
Techniques on Reporting in the Health
Interview Survey. 62 pp.

Estimation and Sampling Variance in the
Health Interview Survey. 28 pp.

Development and Evaluation of an Ex-
panded Hearing Loss Scale Questionnaire.
42 pp.

The Influence of Interviewer and Respond-
ent Psychological and Behavioral Variables
on the Reporting in Household Interviews.
65 pp.

Interview Data on Chronic Conditions

Compared With Information Derived From
Medical Records. 84 pp.

Interview Response on Health Insurance
Compared With Insurance Records, United
States, 1960. 43 pp.

Identifying Problem Drinkers in a House-
hold Health Survey. 46 pp.

35

No. 12

No. 10

No.8

No.7

No.6

No. 2

Series 3
No. 7

Methodological Aspects of a Hearing Abil-
ity Interview Survey. 19 pp.

Hospital Utilization in the Last Year of
Life. 30 pp.

Comparison of Hospitalization Reporting
in Three Survey Procedures. 48 pp.

Health Interview Responses Compared
With Medical Records. 74 pp.

Reporting of Hospitalization in the Health
Interview Survey. 71 pp.

Measurement of Personal Health Expendi-
tures. 59 pp.

Selected Family Characteristics and Health
Measures Reported in the Health Interview
Survey. 26 pp.

National Health and Nutrition Examination Survey

Series 1
No. 15

No. 14

No. 10

No. 8

No. 5

No.4

Series 2
No. 86

No. 74

No. 73

No. 72

36

Plan and Operation of the Second National
Health and Nutrition Examination Survey,
1976-80. 144 pp.

Plan and Operation of the HANES I Aug-
mentation Survey of Adults 25-74 Years,
United States, 1974-75. 1 10 pp.

Plan and Operation of the Health and Nu-
trition Examination Survey, United States,
1971-73. Part A, 46 pp. Part B, 77 pp.

Plan and Operation of a Health Examina-
tion Survey of US. Youths 12-17 Years of
Age. 80 pp.

Plan, Operation, and Response Results of a
Program of Children’s Examinations. 56 pp.

Plan and Initial Program of the Health Ex-
amination Survey. 43 pp.

Computer Assisted Spirometry Data Analy-

vsis for the National Health and Nutrition

Examination Survey. 48 pp.

The Construction and Utility of Three
Indexes of Intellectual Achievement—An
Intellectual-Development (ID) Index, a
Socio-IntellectualStatus (SIS) Index, a
Differential-Intellectual-Development(DID)
Index—US. Children and Youths, 6-17
Years. 26 pp.

A Concurrent Validational Study of the
NCHS Well-Being Schedule. 53 pp.

Methodologic Programs in Children’s Spi-
rometry. 43 pp.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

No.

71

67

62

60

59

58

47

44

43

40

36

35

27

25

24

22

21

.15

9

3

Development of the Speech Reception
Test. 16 pp.

A Study of the Effect of Remuneration
Upon Response in the Health and Nutri-
tion Examination Survey, United States.
23 pp.

Language and Adjustment Scales for the
Thematic Apperception Tests for Youths
12-17 Years. 84 pp.

The Rationale, Development, and Stand-
ardization of a Basic Word Vocabulary
Test. 71 pp.

Vision Test Validation Study for the
Health Examination Survey Among Youths
12-17 Years. 36 pp.

Language and Adjustment Scales for the
Thematic Apperception Test for Children
6—11 Years. 70 pp.

Subtest Estimates of the WISC Full-Scale
IQ’s for Children. 42 pp.

Quality Control in a National Health Ex-
amination Survey. 22 pp.

Sample Design and Estimation Procedures
for a National Health Examination Survey
of Children. 40 pp.

Loudness Balance Study of Selected Audio-
meter Earphones. 37 pp.

Factors Related to Response in a Health
Examination Survey, United States, 1960-
62. 48 pp.

Comparison of Timed and Untimed Presen-
tation of the Goodenough-Harris Test of
Intellectual Maturity. 16 pp.

Development of the Brief Test of Literacy.
29 pp.
Orthodontic Treatment Priority
49 pp.

A Study of the Achievement Test Used in
the Health Examination Surveys of Persons
Aged 6-17 Years. 60 pp.

Three Views of Hypertension and Heart
Disease. 43 pp.

Index.

Calibration of Two Bicycle Ergometers
Used by the Health Examination Survey.
10 pp.

Evaluation of Psychological Measures Used
in the Health Examination Survey of Chil-
dren Ages 6—1 1. 67 pp.

Cooperation in Health Examination Sur-
veys. 38 pp.

The One-Hour Oral Glucose Tolerance
Test. 34 pp.

No. 1 Comparison of Two Vision-Testing Devices.
33 pp.

Series 3

No. 14 Height and Weight of Children in the

United States, India, and the United Arab
Republic. 51 pp.

National Hospital Discharge Survey

Series 2

No.88 The Status of Hospital Discharge Data in
Denmark, Scotland, West Germany, and
the United States. 58 pp.

No.80 The Status of Hospital Discharge Data in
Six Countries. 77 pp.

No.68 Quality Control in the Hospital Discharge
Survey. 41 pp.

No.55 The Prediction Approach to Finite Popu-
lation Theory: Application to the Hospital
Discharge Survey. 31 pp.

No.39 Development of the Design of the NCHS
Hospital Discharge Survey. 29 pp.

No.28 Pilot Study on Patient Charge Statistics.
59 pp.

No. 19 Participation of Hospitals in the Pilot
Study of the Hospital Discharge Survey.
55 pp.

No. 13 Computer Simulation of Hospital Dis-
charges. 34 pp.

Series 4

No. 14 Uniform Hospital Abstract, Minimum Basic
Data Set: A Report of the US. National
Committee on Vital and Health Statistics.
17 pp.

No. 1 1 Use of Hospital Data for Epidemiologic and

Medical-Care Research: A Report of the

US. National Committee on Vital and
Health Statistics. 9 pp.

National Ambulatory Medical Care Survey
Series 2

No. 78 A Reason for Visit Classification for Ambu-
latory Care. 63 pp.

No. 63 The National Ambulatory Medical Care
Survey: Symptom Classification. 35 pp.

No. 61 National Ambulatory Medical Care Survey:
Background and Methodology, United
States, 1967-72. 76 pp.

Series 4

No. 16 Ambulatory Medical Care Records, Uni-

form Minimum Basic Data Set: A Report
of the US. National Committee on Vital
and Health Statistics. 16 pp.

National Nursing Home Survey
Series 1

No.7 Design and Methodology for a National

Survey of Nursing Homes. 37 pp.

National Master Facility Inventory

Series I

No. 12 Development of the National Inventory of
Family Planning Services, United States.
43 pp.

No. 9 Design and Methodology of the 1967
Master Facility Inventory Survey. 30 pp.

No.6 The Agency Reporting System for Main-
taining the National Inventory of Hospitals
and Institutions. 19 pp.

No.3 Development and Maintenance of a Na-

tional Inventory of Hospitals and Institu-
tions. 25 pp.

37

.7.” 'an

U.C, BERKELEY LIBRARIES

* INHIIIIIIBIE

(021205333

Vital and Health Statistics series descriptions

SERIES 1.

SERIES 2.

SERIES 3.

SERIES 4.

SERIES 10.

SERIES”.

1

SERIES 12.

SERIES 13.

 

Programs and Collection Procedures—Reports describing
the general programs of the National Center for Health
Statistics and its offices and divisions and the data col-
lection methods used. They also include definitions and
other material necessary for understanding the data.

Data Evaluation and Methods Research.—Studies of new
statistical methodology including experimental tests of
new survey methods, studies of vital statistics collection
methods, new analytical techniques, objective evaluations
of reliability of collected data, and contributions to sta-
tistical theory.

Analytical and Epidemiological Studies—Reports pre-
senting analytical or interpretive studies based on vital
and health statistics, 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 and health sta-
tistics and documents such as recommended model vital
registration laws and revised birth and death certificates.

Data from the National Health Interview Survey.—Statis-
tics on illness, accidental injuries, disability, use of hos-
pital, medical, dental, and other services, and other
health-related topics, all based on data collected in the
continuing national household interview survey.

Data From the National Health Examination Survey and
the National Health and Nutrition Examination Survey.—
Data from direct examination, testing, and measurement
of national samples of the civilian noninstitutionalized
population provide the basis for (1) estimates of the
medically defined prevalence of specific diseases in the
United States and the distributions of the population with
respect to physical, physiological, and psychological
characteristics and (2) analysis of relationships among the
various measurements without reference to an explicit
finite universe of persons.

Data From the Institutionalized Population Surveys.—Dis-
continued in 1975. Reports from these surveys are in-
cluded in Series 13.

Data on Health Resources Utililation.—Statistics on the
utilization of health manpower and facilities providing

SERIES 14.

SERIES 15.

SERIES 20.

SERIES 21.

SERIES 22.

SERIES 23.

long-term care, amMfipspital care, and family
. e 1;

planning services.

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

Data From Special Surveys.—-Statistics on health and '
health-related topics collected in special surveys that are
not a part of the continuing data systems of the National
Center for Health Statistics.

Data on Mortality.—Various statistics on mortality other
than as included in regular annual or monthly reports.
Special analyses by cause of death, age, and other demo~
graphic variables; geographic and time series analyses; and
statistics on characteristics of deaths not available from
the vital records based on sample surveys of those records.

Data on Natality, Marriage, and Divorce—Various sta-
tistics on natality, marriage, and divorce other than 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 not available from the vital
records based on sample surveys of those records.

Data From the National Mortality and Netality Surveys.—
Discontinued in 1975. Reports from these sample surveys
based on vital records are included in Series 20 and 21,
respectively.

Data From the National Survey of Family Growth.—
Statistics on fertility, family formation and dissolution,
family planning, and related maternal and infant health
topics derived from a periodic survey of 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 Branch
National Center for Health Statistics

Public Health Service

Hyattsville, Md. 20782

 

 

 

 
  

' us, WWTOF HEALTHAND HUMAN SERVICES
V , of Health Research, Stafi'stics, and Technology
National Center for Health Statistics

37m East-West Highway

Hymme, ‘Maryiand 20782

OFFICIAL ‘Bu'sméss
PENALTY FOR PRIVATE use, $300

   
  

 

H mm, A

. 7

 

 

 

‘1

 

   

m

  

1-.TH LIBRA

 
    

from the Office of Health Research, Statistics, and Technology

RY

 

POSTAGE AND FEES PAID
US. DEPARTMENT OF HHS
HHS 3%

 

Third Class

   

 

DHHS Publication N9. '(PHS) 82-1318, Series 1, No. 16