Health, United States, 1999 With Health and Aging …Health, United States, 1999 Health and Aging Chartbook U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Centers for Disease Control

http://www.cdc.gov/nchs/hus.htm

Copyright Information

Permission has been obtained from the copyrightholders to reproduce certain quoted material in thisreport. Further reproduction of this material isprohibited without specific permission of the copyrightholder. All other material contained in this report is inthe public domain and may be used and reprintedwithout special permission; citation as to source,however, is appreciated.

Suggested Citation

National Center for Health Statistics.Health, United States, 1999With Health and Aging Chartbook.Hyattsville, Maryland: 1999.

Kramarow E, Lentzner H, Rooks R, Weeks J,Saydah S. Health and Aging Chartbook. Health,United States, 1999. Hyattsville, Maryland: NationalCenter for Health Statistics. 1999.

Library of Congress Catalog Number 76–641496

For sale by Superintendent of DocumentsU.S. Government Printing OfficeWashington, DC 20402

Health, United States, 1999Health and Aging Chartbook

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICESCenters for Disease Control and PreventionNational Center for Health Statistics6525 Belcrest RoadHyattsville, Maryland 20782-2003

September 1999DHHS Publication number (PHS) 99-1232-1

U.S. Department of Health and Human Services(DHHS)

Donna E. ShalalaSecretary

Office of Public Health and Science, HHS

David Satcher, M.D., Ph.D.Assistant Secretary for Health and Surgeon General

Centers for Disease Control and Prevention (CDC)

Jeffrey P. Koplan, M.D., M.P.H.Director

National Center for Health Statistics, CDC

Edward J. Sondik, Ph.D.Director

...................................................................................................... Preface

Health, United States, 1999 is the 23d report onthe health status of the Nation submitted by theSecretary of the Department of Health and HumanServices to the President and Congress of the UnitedStates in compliance with Section 308 of the PublicHealth Service Act. This report was compiled by theCenters for Disease Control and Prevention (CDC),National Center for Health Statistics (NCHS). TheNational Committee on Vital and Health Statisticsserved in a review capacity.

Health, United States presents national trends inhealth statistics. Major findings are presented in thehighlights. The report includes achartbook on healthand aging and detailed tables on trends.

Health and Aging Chartbook

In each edition of Health, United States, achartbook focuses on a major health topic. This yearhealth and aging was selected as the subject of thechartbook because older people are major consumersof health care and their numbers are increasing. TheUnited Nations’ General Assembly proclaimed 1999the ‘‘International Year of Older Persons.’’ The healthand aging chartbook consists of 34 figures andaccompanying text.

Detailed Tables

The chartbook is followed by 146 detailed tableson trends organized around four major subject areas:health status and determinants, utilization of healthresources, health care resources, and health careexpenditures. A major criterion used in selecting thedetailed tables is the availability of comparablenational data over a period of several years. Thedetailed tables report data for selected years tohighlight major trends in health statistics. Earliereditions of Health, United States may present data foradditional years that are not included in the currentprinted report. Where possible, these additional dataare available in Lotus 1–2–3 spreadsheet files as listedin Appendix III .

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

Racial and Ethnic Data

Several tables in Health, United States presentdata according to race and Hispanic origin consistentwith Department-wide emphasis on expanding racialand ethnic detail in presenting health data. Thepresentation of data on race and ethnicity in thedetailed tables is usually in the greatest detail possible,after taking into account the quality of data, theamount of missing data, and the number ofobservations. The large differences in health status byrace and Hispanic origin that are documented in thisreport may be explained by several factors includingsocioeconomic status, health practices, psychosocialstress and resources, environmental exposures,discrimination, and access to health care.

Changes in This Edition

Similar tables appear in each volume of Health,United States to enhance the use of this publication asa standard reference source. However, some changes inthe content of the tables are made each year toenhance their usefulness and to reflect emerging topicsin public health. New to Health, United States, 1999are data on death rates for selected causes of death byeducational attainment (table35); additional notifiablediseases (table53); the percent of children withuntreated dental caries (table72); the percent of adultswith no usual source of care (table81); studentenrollment and number of schools of public health(table107); and the percent of persons with privatehealth insurance through health maintenanceorganizations (table131).

Data for racial and ethnic groups have beenexpanded in tables showing the percentlow-birthweight live births by State (tables 13 and 14),the percent of persons with fair or poor health(table60), the percent of persons who currently smokecigarettes (table63), and the percent of childrenwithout a physician contact in the past year (table 79)and without a usual source of care (table 80). Inaddition new tables 72, 81, and 131 also present datafor racial and ethnic groups.

.......................................................................... iii

m

The Health and Aging Chartbook is reprintedfrom Health, United States, 1999 and includeshighlights of the detailed tables and theappendixes from the complete report.

iv ..........

Preface ....................................................................................................................

Trends in overweight among adults, presented intable70, have been revised to reflect current definitionsand to include the proportion of persons with healthyweight and those with obesity. Data on procedurespresented in tables 94 and 95 now include ambulatoryprocedures from the National Survey of AmbulatorySurgery and inpatient procedures from the NationalHospital Discharge Survey. Some of the tables in thehealth care expenditures section (tables 116, 117, 120,and 127) were reformatted to simplify presentation ofthe data.

AppendixesAppendix I describes each data source used in the

report and the limitations of the data and providesreferences for further information about the sources.Appendix II is an alphabetical listing of terms used inthe report. It also contains standard populations usedfor age adjustment and International Classification ofDiseases codes for cause of death and diagnostic andprocedure categories. Appendix II I lists tables withadditional years of trend data that are availableelectronically in Lotus 1–2–3 spreadsheet files on theNCHS homepage and CD-ROM.

Electronic AccessHealth, United States can be accessed

electronically in four formats. First, the entire Health,United States, 1999 is available, along with otherNCHS reports, on a CD-ROM entitled ‘‘Publicationsfrom the National Center for Health Statistics,’’featuring Health, United States, 1999, vol 1 no 4,1999. These publications can be viewed, searched,printed, and saved using Adobe Acrobat software onthe CD-ROM. The CD-ROM may be purchased fromthe Government Printing Office or the NationalTechnical Information Service.

Second, the complete Health, United States, 1999is available as an Acrobat .pdf file on the Internetthrough the NCHS home page on the World Wide

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

Web. The direct Uniform Resource Locator (URL)address is:

www.cdc.gov/nchswww/products/pubs/pubd/hus/hus.ht.

Third, the 146 detailed tables in Health, UnitedStates, 1999 are available on the FTP server as Lotus1–2–3 spreadsheet files and can be downloaded. TheURL address for the FTP server is:

www.cdc.gov/nchswww/datawh/ftpserv/ftpserv.htm.

The detailed tables are also included as Lotus1–2–3 spreadsheet files on the CD-ROM mentionedabove.

Fourth, for users who do not have access to theInternet or to a CD-ROM reader, the 146 detailedtables can be made available on diskette as Lotus1–2–3 spreadsheet files for use with IBM compatiblepersonal computers. To obtain a copy of the diskette,contact the NCHS Data Dissemination Branch.

QuestionsFor answers to questions about this report, contact:Data Dissemination BranchNational Center for Health StatisticsCenters for Disease Control and Prevention6525 Belcrest Road, Room 1064Hyattsville, Maryland 20782-2003phone: 301-436-8500E-mail: [email protected]

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

http://www.cdc.gov/nchswww/products/pubs/pubd/hus/hus.htm

http://www.cdc.gov/nchswww/datawh/ftpserv/ftpserv.htm

http://www.cdc.gov/nchswww/mail/mail.htm

http://www.cdc.gov/nchswww/data/hus99.pdf

n

.

c

h

k

oy

c

n

,

D.

n

J..

of

se

al

............................................................................................. Acknowledgments

The chartbook was prepared by Ellen A.Kramarow, Harold R. Lentzner, Ronica N. Rooks,Julie D. Weeks, and Sharon H. Saydah. Data andanalysis for specific charts were provided by ChristiS. Cox, Mayur M. Desai, Thomas A. Hodgson,Nadine R. Sahyoun, and Clemencia M. Vargas.

The Office of the Demography of Aging,Behavioral and Social Research Program, NationalInstitute on Aging, under the direction of RichardSuzman, provided support for the chartbook. Adviceon the content of the chartbook was provided byDonna L. Hoyert, Raynard S. Kington, and Harry MRosenberg of NCHS; Suzanne M. Smith of theNational Center for Chronic Disease Prevention andHealth Promotion, CDC; and Robert Clark of theOffice of Assistant Secretary for Planning andEvaluation, Department of Health and Human Servi(DHHS).

Overall responsibility for planning andcoordinating the content of this volume rested with tOffice of Analysis, Epidemiology, and HealthPromotion, National Center for Health Statistics(NCHS), under the general direction of Diane M.Makuc and Jennifer H. Madans.

Health, United States, 1999highlights, detailedtables, and appendixes were prepared under thesupervision of Kate Prager. Detailed tables wereprepared by Alan J. Cohen, Margaret A. Cooke,Virginia M. Freid, Andrea P. MacKay, Michael E.Mussolino, Mitchell B. Pierre, Jr., Rebecca A. PlaceAnita L. Powell, and Kate Prager with assistance froLa-Tonya Curl, Patricia A. Knapp, Mark F. Pioli,Sharon H. Saydah, and Catherine Duran of TRW,Information Services Division and Henry Xia ofNOVA Research Company. The appendixes, index tdetailed tables, and pocket edition were prepared bAnita L. Powell. Production planning and coordinatiowere managed by Rebecca A. Placek with assistanfrom Carole J. Hunt and Camille Miller.

Publications managementand editorial reviewwere provided by Thelma W. Sanders and Rolfe W.Larson. The designer was Sarah M. Hinkle. Graphicwere supervised by Stephen L. Sloan. Production wdone by Jacqueline M. Davis and Annette F. Holma

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

e

es

e

,m

ne

sas.

Printing was managed by Patricia L. Wilson andJoan D. Burton.

Electronic access through the CD-ROM andNCHS internet site was provided by June R. Gable,Gail V. Johnson, Thelma W. Sanders, Julia A.Sothoron, and Tammy M. Stewart-Prather.

Data and technical assistancewere provided byCharles A. Adams, Robert N. Anderson, VeronicaBenson, Linda E. Biggar, Kate M. Brett, Ronette R.Briefel, Catharine W. Burt, Margaret D. Carroll,Robin A. Cohen, Achintya N. Dey, Thomas D. DunnSylvia A. Ellison, Katherine M. Flegal, Nancy G.Gagne, Cordell Golden, Malcolm C. Graham,Edmund J. Graves, Barbara J. Haupt, Katherine E.Heck, Rosemarie Hirsch, Donna L. Hoyert, DeborahIngram, Susan S. Jack, Elizabeth W. Jackson,Clifford L. Johnson, Kenneth D. Kochanek, Lola JeaKozak, Robert J. Kuczmarski, Linda S. Lawrence,Karen L. Lipkind, Anne C. Looker, Marian F.MacDorman, Joyce A. Martin, Jeffrey D. Maurer,Linda F. McCaig, William D. Mosher, Sherry L.Murphy, Cheryl R. Nelson, Francis C. Notzon,Parivash Nourjah, Maria F. Owings, Elsie R. Pamuk,Gail A. Parr, Kimberly D. Peters, Linda S. Peterson,Linda J. Piccinino, Cheryl V. Rose, Harry M.Rosenberg, Colleen M. Ryan, Susan M. Schappert,Fred Seitz, Manju Sharma, Alvin J. Sirrocco, Betty LSmith, Genevieve W. Strahan, Luong Tonthat,Clemencia M. Vargas, Stephanie J. Ventura, andDavid A. Woodwell of NCHS; Jeffrey Y. Liu of TRW,Information Services Division; Carolyn M. ShermanSherman and Holmes Associates; Tim Bush andMelinda L. Flock of the National Center for HIV,STD, and TB Prevention, CDC; Samuel L. Grosecloand Myra A. Montalbano of the EpidemiologyProgram Office, CDC; Lisa M. Koonin and Myra A.Montalbano of the National Center for ChronicDisease Prevention and Health Promotion, CDC;Monina Klevens and Edmond F. Maes of the NationImmunization Program, CDC; Suzanne M. Kisner ofthe National Institute of Occupational Safety andHealth, CDC; Evelyn Christian of the HealthResources and Services Administration; MitchellGoldstein of the Office of the Secretary, DHHS;

.......................................................................... v

vi .........

Acknowledgments .............................................................................................

Joseph Gfroerer, Janet Greenblatt, Andrea Kopstein,Patricia Royston, Michael Witkin, Richard Thoreson,and Deborah Trunzo of the Substance Abuse andMental Health Services Administration; Ken Allison,Lynn A. G. Ries, and Arthur Hughes of the NationalInstitutes of Health; Cathy A. Cowan, Janice D.Drexler, Leslie Greenwald, Paula Higger, Roger E.Keene, Helen C. Lazenby, Katharine R. Levit, AnnaLong, Edward F. Mortimore, Anthony C. Parker,Madie W. Stewart, and Joanne Weller of the HealthCare Financing Administration; Loretta Bass, JosephDalaker, and Terry Lugaila of the Census Bureau;James Barnhardt, Alan Blostin, Daniel Ginsburg, andKay Ford of the Bureau of Labor Statistics; ElizabethAhuja of the Department of Veterans Affairs; SusanTew of the Alan Guttmacher Institute; Wendy Katz ofthe Association of Schools of Public Health; RichardHamer of InterStudy; and Patrick O’Malley of theUniversity of Michigan.

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

.............................................................................................................. Dedication

Jacob J. Feldman, Ph.D.

This volume ofHealth, United Statesis dedicatedto our colleague and friend, Dr. Jack Feldman, whoserved as Associate Director for Analysis,Epidemiology, and Health Promotion at the NationalCenter for Health Statistics (NCHS) from themid-1970’s until he retired from Federal service in1998. The development ofHealth, United Stateswasone part of the analytic program that Jack oversawduring his tenure at NCHS.

Jack made innumerable contributions toHealth,United Statesover a 23-year period, providinginsightful direction and wise guidance on the contentof the report. Jack’s comprehensive knowledge of thepublic health literature as well as his strong grasp ofemerging health issues have been key to ensuring thatthis report provides data on the most important healthtopics each year. Jack also brought a vast knowledgeof and keen interest in survey methodology, dataquality, and statistical analysis to his work on thisreport. He is equally knowledgeable about all of thewide range of topics and data sources that are includedin this publication. More amazing is Jack’s uncannyability to absorb the large volume of statisticspresented in the detailed trend tables of this report andto identify interesting and important trends that need tobe brought to the attention of the health community.On the other hand, Jack’s healthy skepticism ofchanges in trends and unusual patterns in the data hasensured that highlighted trends reflect true differencesin health rather than changes in data collectionmethods or other data artifacts.

Although Jack is no longer in the office with us ona daily basis, the example he provided still serves toguide and inspire the work we do. We are challengedto continue the high standards that he set not only forthis report but for all aspects of the collection,analysis, and dissemination of health data. A gratefulstaff acknowledges his unique contributions to thisprofile of the Nation’s health and wishes Jack all thebest in his new endeavors.

................................................................................................................................... vii

Preface iii

Acknowledgments v

Dedication vii

List of Figures on Health and Aging xi

Geographic Regions and Divisions of the United States xiii

Highlights

Health and Aging Chartbook 3Detailed Tables 5

Health and Aging Chartbook

Introduction 17

PopulationDemographic Characteristics 22Living Arrangements 24Nursing Home Residence 26Poverty 28

Health StatusLife Expectancy 30Life Expectancy by Race 32Deaths From All Causes 34Selected Leading Causes of Death 36Self-Reported Health 38Chronic Conditions 40Visual and Hearing Impairments 42Osteoporosis 44Physical Functioning and Disability 46Chronic Conditions Associated With Disability 48Overweight 50Oral Health 52Social Activities 54Exercise 56

Health Care Access and UtilizationCaregivers 58Unmet Needs 60Assistive Devices 62Physician Contacts 64Inpatient Health Care 66Influenza and Pneumococcal Vaccinations 68Home Health Care 70Health Insurance 72Medicare Health Maintenance Organization Enrollment 74Cost of Heart Disease 76Cost of Diabetes 78

Technical Notes 80Data Tables for Figures 1–34 86

.................................................................................................................................... ix

.................................................................................................................. Contents

Appendixes

Contents 101I. Sources and Limitations of Data 104II. Glossary 138III. Additional Years of Data Available 165

x ....................................................................................................................................

Contents .................................................................................................................

Population

1. Population 65 years of age and over: United States,1950–2030 23

2. Living arrangements of persons 65 years of age andover by age and sex: United States, 1997 25

3. Nursing home residents among persons 65 years ofage and over by age, sex, and race: United States,1997 27

4. Percent in poverty among persons 65 years of ageand over by sex, race, and Hispanic origin:United States, 1997 29

Health Status

5. Life expectancy at birth, age 65, and age 85 by sex:United States, 1950–97 31

6. Life expectancy at birth, age 65, and age 85 by sexand race: United States, 1997 33

7. Death rates for all causes among persons 65 years ofage and over by age and sex: United States, 1997 35

8. Death rates for all causes among persons 65 years ofage and over by age, sex, race, and Hispanic origin:United States, 1997 35

9. Death rates for selected leading causes amongpersons 65 years of age and over by age and sex:United States, 1997 36

10. Fair or poor health among persons 65 years of ageand over by age, sex, race, and Hispanic origin:United States, 1994–96 39

11. Percent of persons 70 years of age and over whoreported selected chronic conditions by sex:United States, 1995 41

12. Prevalence of visual impairment among persons 70years of age and over by age, sex, and race:United States, 1995 43

13. Prevalence of hearing impairment among persons 70years of age and over by age, sex, and race:United States, 1995 43

14. Prevalence of reduced hip bone density amongpersons 65 years of age and over by age, sex, andseverity: United States, 1988–94 45

15. Percent of persons 70 years of age and over whohave difficulty performing 1 or more physical activities,activities of daily living, and instrumental activities ofdaily living by age and sex: United States, 1995 47

16. Percent of persons 70 years of age and over whoreport specific conditions as a cause of limitation inactivities of daily living: United States, 1995 49

17. Distribution of weight among persons 65–74 yearsof age by sex: United States, 1988–94 51

18. Prevalence of obesity among persons 65–74 years ofage by sex: United States, 1960–94 51

19. Percent with untreated dental caries among dentatepersons 65 years of age and over by age and sex:United States, 1988–94 53

20. Prevalence of total tooth loss (edentulism) amongpersons 65 years of age and over by age: United States,1983 and 1993 53

21. Number of social activities in a 2-week periodamong persons 70 years of age and over by age andsex: United States, 1995 55

22. Percent who exercise and selected type of exerciseamong persons 65 years of age and over by sex:United States, 1995 57

Health Care Access and Utilization

23. Number of caregivers providing assistance withactivities of daily living or instrumental activities ofdaily living to persons 70 years of age and over by ageand sex: United States, 1995 59

24. Percent with unmet needs among persons 70 yearsof age and over who need help with 1 or more activitiesof daily living or instrumental activities of daily livingby age and sex: United States, 1995 61

25. Assistive devices used among persons 70 years ofage and over by age and sex: United States, 1995 63

26. Place of ambulatory physician contacts amongpersons 65 years of age and over by age and sex:United States, 1994–96 65

27. Hospital discharge rates in non-Federal short-stayhospitals for selected first-listed diagnoses amongpersons 65 years of age and over by age and sex:United States, 1996 67

28. Percent vaccinated against influenza andpneumococcal disease among persons 65 years of ageand over by race and Hispanic origin: United States,1993–95 69

29. Home health care patients among persons 65 yearsof age and over by age and sex: United States, 1996 70

30. Home health care services received by currentpatients 65 years of age and over: United States, 199671

31. Health insurance coverage among persons 65 yearsof age and over by age, race, Hispanic origin, and typeof insurance: United States, 1994–96 73

32. Percent of Medicare enrollees in health maintenanceorganizations by State: United States, 1997 75

................................................................................................................................... xi

........................................................ List of Figures on Health and Aging

33. Estimated amount of personal health careexpenditures attributed to heart disease among persons65 years of age and over by age, sex, and type of healthservice: United States, 1995 77

34. Estimated amount of personal health careexpenditures attributed to diabetes among persons 65years of age and over by age, sex, and type of healthservice: United States, 1995 79

xii ..................................................................................................................................

List of Figures on Health and Aging ........................................................

West

MountainMountain

West SouthCentral

SouthAtlanticSouth

AtlanticSouth

Atlantic

MiddleAtlanticMiddleAtlanticMiddleAtlanticWest North

CentralWest North

CentralWest North

Central

Midwest

Northeast

South

EastSouth

Central

EastSouth

Central

EastSouth

Central

East NorthCentral

NewEngland

Pacific

AK

.................................................................................................................................. xiii

..............Geographic Regions and Divisions of the United States

e

fr

0ef

ly

sn

it

ve

ers

re

t

nt

se

................................................................................................................ HighlightsHealth

andAging

Population

The older population of the United States is largand growing and will be more diverse in thetwenty-first century. Women constitute the majority othe older population. Only a small proportion of oldepersons reside in institutions, and a significantproportion of community-dwelling elderly persons(particularly women) live alone.

In 1997, 13 percent of the U.S.population was 65years of age and over. It is estimated that in 2030, 2percent of Americans will be 65 years of age and ovIn 1997 older persons made up a larger proportion othe non-Hispanic white population compared withother racial and ethnic groups. However, the oldernon-Hispanic white population is growing more slowcompared with other groups (figure 1).

The living arrangements of older persons varygreatly by age, sex, race, and marital status. While amajority of noninstitutionalized persons 65 years ofage and over lived with family members in 1997,nearly one-third lived alone. Women in every agegroup were more likely than men to live alone(figure 2).

In 1997 approximately 4 percent of the olderpopulation lived in nursing homes. The rate ofnursinghome residencerises sharply with age. Approximately1 percent of persons 65–74 years of age lived innursing homes compared with almost 20 percent ofpersons 85 years of age and over. Women at all agehad higher rates of nursing home residence than me(figure 3).

Although poverty rates among the elderly havedeclined significantly since the 1960’s, 1 out of 10persons 65 years of age and over in 1997 was livinga family with income below the Federal povertythreshold. The poverty rate was higher among olderblack and Hispanic persons compared with older whpersons (figure 4).

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

r.

in

e

Health Status

Americans have longer lives than ever before.Persons who survive to age 65 today can expect to lion average nearly 18 more years. The health of theolder population varies greatly. Rates of illness anddisability increase sharply among the ‘‘oldest-old,’’persons 85 years of age and over, compared withyounger persons. Nearly all measures reflect thisvariation by age.

Life expectancyat age 65 and age 85 increasedover the past 50 years. Women have on average longlives than men. In 1997 life expectancy at age 65 wahigher for white persons than for black persons.However, at age 85 life expectancy for black personswas slightly higher than for white persons (figures 5and6).

Chronic diseases such as heart disease, cancer,stroke, and chronic obstructive pulmonary diseases athe leading causes of deathamong the olderpopulation, although pneumonia and influenza wereresponsible for approximately 7 percent of deathsamong persons 85 years of age and over in 1997(figure 9).

Chronic conditions are prevalent among olderpersons. In 1995 among noninstitutionalized persons70 years of age and over, 79 percent reported at leasone of seven chronic conditions common among theelderly. The majority of persons 70 years of age andover reported arthritis, and approximately one-thirdreported hypertension. Diabetes was reported by11 percent (figure 11).

Visual and hearing impairments among olderpersons increase sharply with age. In 1995, 13 perceof persons 70–74 years of age were visually impairedcompared with 31 percent of persons 85 years of ageand over. For hearing impairments, the prevalence rofrom 26 percent of persons 70–74 years of age to 49percent of persons 85 years of age and over (figures 12and13).

......................................................................... 3

n

bov

g

gtie

n

n6tn

5te

0

ye

d

l

4 ..........

Highlights ...............................................................................................................H

ealth

and

Agin

g

Osteoporosisis common among older persons ais a strong predictor of subsequent fractures. In1988–94 just over one-half of noninstitutionalizedpersons 65 years of age and over had reduced hipdensity, either osteoporosis or osteopenia (a less seform of bone loss than osteoporosis). The proportionof older persons with osteoporosis was higher amonwomen than men and rose with age for both womenand men (figure 14).

Physical functioning and disability rates amongthe older population vary by age and sex. Nearly 9percent of noninstitutionalized persons 70 years of aand over were unable to perform one or more activiof daily living such as bathing, dressing, using thetoilet, and getting in and out of bed or chairs. Womein every age group were more likely to be disabledthan men, and the proportion disabled rose with age(figure 15).

Oral health indicators among the older populatioare improving over time. Yet, 30 percent of personsyears of age and over in 1993 were edentulous, thathey had no natural teeth. Non-Hispanic white persohad lower levels of total tooth loss compared withnon-Hispanic black persons and Hispanic persons(figure 20). In 1988–94 nearly one-third of persons 6years of age and over with natural teeth had untreadental caries in the crown or the root of their teeth(figure 19).

In 1995 nearly all noninstitutionalized persons 7years of age and over participated in somesocialactivities in a 2-week period. The most commonactivity was contact with family, either in person or btelephone. Persons who were disabled were less likthan nondisabled persons to participate in activitiesoutside of their house (figure 21).

In 1995, 71 percent of nondisabled persons 65years of age and over participated in some form ofexerciseat least once in a recent two-week period.Most older persons who exercise engage in light anmoderate activities such as walking, gardening, andstretching. However, only about one-third of personswho exercised achieved recommended levels of 30

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

d

neere

es

5is,s

d

ly

minutes each time on most days of the week(figure 22).


Changes in the health care system have affectedthe older population. Use of in-home contacts withmedical providers has increased, and length of hospitastays has decreased. Approximately 12 percent ofMedicare beneficiaries 65 years of age and over wereenrolled in a managed care plan in 1997, although thepercent varies widely by region. In general, persons 85years of age and over use health care services morethan those 65–84 years of age.

In 1995 approximately one-third ofnoninstitutionalized persons 70 years of age and overreceived help from acaregiver with daily activitiessuch as dressing, bathing, shopping, housework, andmanaging money. The number of caregivers providinghelp to an older person increased with age (figure 23).

Thirty-nine percent of noninstitutionalized persons70 years of age and over in 1995 usedassistivedevicessuch as hearing aids, diabetic and respiratoryequipment, and canes and walkers during the previous12 months. Rates of device use were twice as highamong persons 85 years of age and over comparedwith persons 70–74 years of age (figure 25).

In 1994–96 the mean number ofambulatoryphysician contactsamong persons 65 years of age andover was 11.4 per year. The number of contacts withphysicians or with other personnel working under aphysician’s supervision increased with age. From 1990to 1996 the proportion of contacts in the homeincreased by 63 percent (figure 26).

Older persons are major consumers ofinpatienthealth care.Older men had higher rates ofhospitalization than older women. Heart disease wasthe most common cause for hospitalization. Theaverage length of hospital stay in 1996 was 6.5 daysfor persons 65 years of age and over, about two daysless than in 1986 (figure 27).

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

with

ere

e

fn5.

d

begtal

nesr


andAging

Influenza and pneumococcal vaccinationsarerecommended for older adults. During 1993–95 anaverage of 55 percent of noninstitutionalized persons65 years of age and over reported receiving a flu showithin the previous 12 months. Twenty-nine percentreported ever having received a pneumoniavaccination. Vaccination coverage for both influenzaand pneumococcal disease was higher amongnon-Hispanic white persons than non-Hispanic blackpersons or Hispanic persons (figure 28).

On an average day in 1996, approximately 1.7million persons 65 years of age and over, roughly 51per 1,000 population, werehome health carepatients.In every age group women had higher rates of homehealth care usage than men, and the rate increasedage for both women and men (figures 29and30).

In 1994–96 persons 85 years of age and over wmore likely to rely on Medicare alone or on Medicarecombined with Medicaid for theirhealth insurancecoverage than persons under 85 years of age.Non-Hispanic black and Hispanic persons were lesslikely than non-Hispanic white persons to have privatinsurance to supplement their Medicare coverage(figure 31).

Participation inMedicare health maintenanceorganizations (HMO’s) is increasing among the olderpopulation. In 1997 over 4 million persons 65 years oage and over who received Medicare were enrolled ia managed care plan, a four-fold increase since 198The highest levels of Medicare HMO participation arein the West. Several states had no Medicare managecare plans in 1997 (figure 32).

In 1995 the overallcost of heart diseaseamongpersons 65 years of age and over was estimated tomore than 58 billion dollars. Hospital care and nursinhome care accounted for over three-fourths of the topersonal health care expenditures for heart diseaseamong the older population (figure 33).

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

t

In 1995 thecost of diabetesamong persons 65years of age and over was estimated to be 26 billiodollars. The largest personal health care expenditurattributed to diabetes, including chronic complicationand comorbidities asssociated with diabetes, was fohospital care. Nursing home care accounted forone-fifth of expenditures (figure 34).

......................................................................... 5

6 ..........

Highlights ...............................................................................................................H

ealth

Stat

us
Detailed Tables
Health Status and Determinants

Mortality

In 1997 life expectancy at birth increased to anall-time high and infant mortality fell to a record low.Life expectancy for black males increased for thefourth consecutive year.

In 1997 lif e expectancy at birth reached anall-time high of 76.5 years and infant mortalit y fell toa record low of 7.2 deaths per 1,000 live births(tables 22 and 28).

Between 1995 and 1997 lif e expectancy at birthfor black males increased 2 years to a record high of67.2 years, due in large part to declines in mortalityfrom HIV infection and homicide. However, lifeexpectancy was still 7.1 years shorter for black malesthan for white males in 1997 (table28).

The death rate for HIV infection declined by almostone-half. Death rates for heart disease, cancer,unintentional injuries, and homicide also decreased.Although death rates for two leading causes of death,stroke and suicide, were lower in 1997 than in 1996,the longer-term trend shows littl e change.

Mortality from heart disease, the leading cause ofdeath, declined 3 percent in 1997, continuing along-term downward trend in mortality. The 1997age-adjusted death rate for heart disease was almostone-half the rate in 1970 (tables 30 and 32).

Mortality from cancer, the second leading causeof death, decreased 2 percent in 1997, continuing thedecline that began in 1990. Over the preceding 20-yearperiod, 1970 to 1990, age-adjusted cancer death rateshad steadily increased (tables 30 and 32).

Mortality from HI V infection declined 48 percentin 1997 following a 29-percent decline in 1996. This2-year decline contrasts sharply with the period1987–94, when HIV mortality increased at an averagerate of 16 percent per year. In 1997 HIV infection fellfrom 8th to 14th in the ranking of leading causes ofdeath (table43).

Mortality from unintentional injuries , the fifthleading cause of death, declined 1 percent in 1997,

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

continuing the generally downward trend in injurymortality since the 1980’s (tables 30 and 32).

The age-adjusted homicide rate declined 6 percentin 1997. This decline continued a trend that began inthe early 1990’s (table46).

Mortality from stroke, the third leading cause ofdeath, was fairly stable between 1992 and 1997.Between 1980 and 1992 stroke mortality declined at anaverage rate of 3.6 percent per year (tables 30, 32, and38).

The age-adjusted death rate for suicide, the eighthleading cause of death, fell 2 percent between 1996and 1997, to 10.6 deaths per 100,000 population.Between 1980 and 1997 age-adjusted suicide ratesranged between 11 and 12 per 100,000 (tables 30, 32,and 47).

Despite overall declines in mortality, disparities amongracial and ethnic groups in mortality for many causesof death are substantial. Disparities among persons ofdifferent education levels continue. Persons with lessthan a high school education have death rates at leastdouble those with education beyond high school.

In 1996 infant mortalit y rates were highestamong infants of non-Hispanic black and AmericanIndian mothers (14.2 and 10.0 deaths per 1,000 livebirths). Infant mortality was lowest for infants ofChinese American mothers (3.2). Mortality rates forinfants of Hispanic mothers and non-Hispanic whitemothers were virtually the same (6.1 and 6.0)(table19).

Infant mortalit y decreases as the mother’s levelof education increases. In 1996 mortality for infants ofblack, white, and Asian American mothers with lessthan 12 years of education was 42–60 percent higherthan for infants whose mothers had 13 or more yearsof education. The disparity in infant mortality bymother’s education was smaller for Hispanic mothers,ranging from 6 percent for Mexican American mothersto 32 percent for Puerto Rican mothers (table20).

The firearm-related death rate for young blackmales 15–24 years of age declined 10 percent per yearon average between 1993 and 1997. The rate for 1997

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


Status

(119.9 deaths per 100,000) was still nearly 5 times therate for young white males (table48).

In 1997 the homicide rate for young Hispanicmales 15–24 years of age was almost 7 times the ratefor non-Hispanic white males. Among those 25–44years of age the homicide rate for Hispanic males wasmore than 3 times as high, and the HI V infectiondeath rate for Hispanic males was more than twice ashigh as for non-Hispanic white males (tables 43 and46).

In 1997 among American Indians theage-adjusted death rates for unintentional injuries(58.5 deaths per 100,000 population) and diabetes(30.4) were at least double the rates for white personsand the death rate for cirrhosis (20.6) was nearly 3times the rate for white persons. Death rates for theAmerican Indian population are known to beunderestimated (table30).

In 1997 overall mortality was 55 percent higherfor black Americans than for white Americans. In1997 the age-adjusted death rates for the blackpopulation exceeded those for the white population by77 percent for stroke, 47 percent for heart disease,34 percent for cancer, and 655 percent for HIVinfection (table30).

In 1997 the overall age-adjusted death rate forAsian-American males was 39 percent lower than therate for white males. However the homicide rate forAsian males was only 6 percent lower than for whitemales and the death rate for stroke was 10 percenthigher for Asian males than for white males. Deathrates for Asian Americans are known to beunderestimated somewhat (tables 36, 38, and 46).

In 1997 the age-adjusted death rate for chronicobstructive pulmonary diseases (COPD), the fourthleading cause of death, was 47 percent higher formales than females. Between 1990 and 1997age-adjusted death rates for males were relativelystable while death rates for females increased at anaverage annual rate of nearly 3 percent (tables 32 and42).

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

Death rates increase as educational attainmentdecreases. In 1997 the age-adjusted death rate forchronic diseases was more than twice as high amongadults with fewer than 12 years of education as amongthose with more than 12 years of education. The deathrate for injuries was 3 times as high for the leasteducated as for the most educated adults (table35).

NatalityThe overall fertility rate declined to a record low in1997, continuing the decline that began in 1990. Birthrates for teens, especially younger teens, and birthrates for unmarried women also continued to declinein 1997. The proportion of babies born with lowbirthweight continued to edge upward.

In 1997 the birt h rate for teenagers declined forthe sixth consecutive year, to 52.3 births per 1,000women aged 15–19 years. Between 1991 and 1997 theteen birth rate declined more for 15–17 year olds thanfor 18–19 year olds (17 percent compared with11 percent) (table3).

Between 1994 and 1997 the birt h rate forunmarried women declined almost 11 percent forblack mothers, to 73.4 births per 1,000 unmarriedblack women aged 15–44 years. The birth ratedeclined almost 10 percent for unmarried Hispanicmothers, to 91.4 per 1,000 (table8).

Low birthweight is associated with elevated riskof death and disability in infants. In 1997 the rate oflow birthweight (infants weighing less than 2,500grams at birth) increased to 7.5 percent overall, upfrom 7.0 percent in 1990. Since 1990 the lowbirthweight rate increased for most racial and ethnicgroups. However among black infants low birthweightdeclined slightly from 13.3 percent in 1990 to13.0 percent in 1997 (table11).

Cigarette smoking durin g pregnancy is a riskfactor for poor birth outcomes such as low birthweightand infant death. In 1997 the proportion of motherswho smoked cigarettes during pregnancy declined to arecord low of 13.2 percent, down from 19.5 percent in1989. However the percent of teenage mothers whosmoked increased between 1994 and 1997 (table10).

.......................................................................... 7

8 ..........

Highlights ...............................................................................................................H

ealth

Stat

us

MorbidityThe two overall measures of morbidity presented inthis report show littl e change over time. The percent ofpersons with activity limitation due to a chroniccondition has remained stable from 1990 to 1996 ashas the percent of persons who report fair or poorhealth status. As family income decreases the percentof persons reporting fair or poor health or reportingan activity limitation increases. Better summarymeasures of health for assessment of trends are neededand are under development. Trends in the incidence ofspecific diseases are additional measures of morbiditytrends.

In 1996 the percent of persons reporting fair orpoor health was four times as high for persons livingbelow the poverty level as for those with familyincome at least twice the poverty level (22.2 percentand 5.5 percent, age adjusted) (table60).

The number of AID S cases newly reported in1997 was 12 percent lower than in 1996. The numberof newly reported AIDS cases decreased 14 percent formales and 5 percent for females in 1997. AIDSincidence continues to be more common among malesthan females. The incidence rate for males 13 years ofage and over (38.5 cases per 100,000 population) wasnearly 4 times the rate for females during July1997–June 1998 (table54).

Between 1995 and 1997 the number of hospitalinpatient discharges with a diagnosis of humanimmunodeficiency viru s (HIV ) decreased 29 percentto 178,000 discharges, and average length of staydeclined by 1.2 days to 8.1 days (table 91).

In 1997 tuberculosis (TB) incidence declined forthe fifth consecutive year to 7.4 cases per 100,000population. In 1997, 39 percent of TB cases occurredamong foreign-born persons in the United States. Thisproportion has been increasing since the mid-1980’s, inpart attributable to changes in immigration patterns(table53).

Between 1990 and 1997 the incidence of primaryand secondary syphilis declined 84 percent to 3.2 cases

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

per 100,000. The incidence of gonorrhea declined56 percent to 122.5 per 100,000 (table53).

Overall cancer incidence has been declining inthe 1990’s, more so for males than for females.Between 1991 and 1995 overall cancer incidence ratesdeclined 13 percent for white males, 6 percent forblack males, 4 percent for black females, and 2 percentfor white females (table57).

Prostate cancer and lung cancer are the twomost frequently diagnosed cancers among men.Between 1991 and 1995 the age-adjusted incidencerate for prostate cancer declined 23 percent for whitemales and 5 percent for black males. During thisperiod lung cancer incidence declined by 9–11 percentfor white and black males (table57).

In 1995 breast cancer incidence was 12 percentlower for black females than for white females.However the 5-year relative survival rate for blackfemales with breast cancer diagnosed in 1989–94 was16 percentage points lower than for white females (71and 87 percent). In 1997 breast cancer mortality was41 percent higher for black women than white women(tables 41, 57, and 58).

Between 1990 and 1997 the injurie s with lostworkdays rate decreased 21 percent to 3.1 per 100full-time equivalents (FTE’s) in the private sector(table74).

Health Risk FactorsElevated blood pressure, high levels of serumcholesterol, and overweight are important risk factorsfor cardiovascular and other chronic diseases. Recenttrends show improvements in the prevalence ofhypertension and high cholesterol. However, theprevalence of overweight has increased. Overweightamong children and adolescents has doubled since theearly 1970’s, raising concerns for long-term healtheffects.

Between 1976–80 and 1988–94 the age-adjustedprevalence of hypertension among adults 20–74 yearsof age declined sharply from 39 percent to 23 percent,

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


Status

after remaining relatively stable over the previous 20years (table68).

Between 1960–62 and 1988–94 the age-adjustedmean serum total cholesterol level for adults 20–74years of age declined from 220 to 203 mg/dL. Theage-adjusted percent of adults with cholesterol greaterthan or equal to 240 mg/dL declined from 32 percentto 19 percent (table69).

Between 1960–62 and 1988–94, the prevalence ofoverweight (body mass index (BMI) greater than orequal to 25) among adults 20–74 years of ageincreased by one- quarter, from 44 to 55 percent.Almost one-half of overweight adults are obese (BMIgreater than or equal to 30), and obesity increased bymore than three-quarters from 13 to 23 percent duringthis time period (percents are age adjusted) (table70).

Between 1971–74 and 1988–94 the prevalence ofoverweight among 6–11 year-old children increasedfrom 6 to 14 percent. Among adolescents 12–17 yearsof age, overweight increased from 6 to 11 percentduring the same period (percents are age adjusted)(table71).

Cigarette smoking is the single leading preventablecause of death in the United States. It increases therisk of lung cancer, heart disease, emphysema, andother respiratory diseases. Cigarette smoking by adultshas remained stable at about 25 percent since 1990.Heavy and chronic use of alcohol and use of illicitdrugs increase the risk of disease and injuries.

Cigarette smoking is more prevalent among theAmerican Indian population than among other groups.In 1993–95, 40 percent of American Indian males and33 percent of American Indian females were currentsmokers compared with 27 percent of white males and24 percent of white females (percents are age adjustedand are for persons 18 years of age and over) (table63).

In 1998 cigarette smoking in the past month byhigh school seniors declined slightly, following 5consecutive years of increase. In 1998 the proportionof white seniors who smoked cigarettes, 41 percent,

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

was nearly three times the proportion of black seniorswho smoked, 15 percent (table65).

In 1998, 23 percent of high school seniors reportedusing marijuan a in the past month, nearly double theprevalence in 1992. Use among eighth graders nearlytripled to 10 percent during that time period (table65).

Between 1993 and 1998 the proportion of highschool seniors reporting alcohol use in the past monthincreased from 49 to 52 percent after declining from72 percent in 1980 (table65).

Heavy alcohol use, having five or more drinks onat least one occasion in the past month, is morecommon among young people 18–25 years of age thanamong younger or older persons. In 1997 among18–25 year olds, heavy drinking was 1.5–2.5 times aslikely for non-Hispanic white persons (33 percent) asfor Hispanic and non-Hispanic black persons (22 and13 percent) (table64).

In 1996 there were more than 152,000cocaine-related emergency room visits, almost twiceas many as in 1990. The greatest increases occurredfor persons 35 years and over, reflecting an agingpopulation of drug abusers being treated in emergencydepartments. However, the proportion of adults age 35years and over who reported using cocaine in the pastmonth has remained stable during this period at lessthan 1 percent (tables 64 and 66).

Environmental factors are important determinants ofhealth and disease. An environmental health objectivefor the year 2000 is that at least 85 percent of the U.S.population should be living in counties that meet theEnvironmental Protection Agency’s National AmbientAir Quality Standards.

In 1996, 81 percent of Americans lived in countiesthat met standards for all pollutants. However, therewere disparities among racial and ethnic groups. In1996, 56–64 percent of the Hispanic and AsianAmerican population lived in counties that met airquality standards for all pollutants compared with81–83 percent of the white, black and American Indianpopulations (table73).

........................................................................ 9

10 ..........

Highlights ...............................................................................................................H

ealth

Car

eU

tiliz

atio

n

Health Care Utilization and Resources

Ambulatory CareUse of preventive health services has substantialpositive effects on the long-term health status of thosewho receive the services. The use of several differenttypes of preventive services has been increasing.However, disparities in use of preventive health careby family income and by race and ethnicity remain inevidence.

Between 1990 and 1997 the percent of mothersreceiving prenatal care in the first trimester ofpregnancy increased from 76 to 83 percent. The largestincreases in receipt of early prenatal care haveoccurred for racial and ethnic groups with the lowestlevels of use, thereby reducing disparities in use ofearly care. However in 1997 the percent of motherswith early prenatal care still varied substantially amongracial and ethnic groups from 68 percent for AmericanIndian mothers to 90 percent for Cuban mothers(table6).

In 1997, 76 percent of children 19–35 months ofage received the combined vaccination series of 4doses of DTP (diphtheria-tetanus-pertussis) vaccine, 3doses of polio vaccine, 1 dose of measles-containingvaccine, and 3 doses of Hib (Haemophilus influenzaetype b) vaccine, up from 69 percent in 1994. Childrenliving below the poverty threshold were less likely tohave received the combined vaccination series thanwere children living at or above poverty (71 comparedwith 79 percent) (table51).

In 1997 only 138 cases of measles were reported,down from 28,000 cases in 1990, providing evidenceof the success of vaccination efforts to increasepopulation immunity to measles (table53).

Regular mammography screening for womenaged 50 years and over has been shown to be effectivein reducing deaths from breast cancer. In 1994,61 percent of women aged 50 years and over reportedmammography screening in the previous 2-year period,up from 27 percent in 1987. Women living below thepoverty threshold were one-third less likely than their

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

nonpoor counterparts to report recent screening in1994 (table82).

Some indicators of children’s access to health careservices include having health insurance coverage,having a usual source of health care, having a recentphysician contact, and treatment of health problemssuch as dental caries. Access to health care amongchildren varies by family income, race, and ethnicity.

In 1997, 14 percent of children under 18 years ofage had no health insurance coverage. More thanone-quarter of children with family income just abovethe poverty level were without coverage comparedwith only 6 percent of those with income above twicethe poverty level (table129).

In 1995–96, 9.2 percent of children under 6 yearsof age did not have a physician contact within theprevious 12-month period. Uninsured children were 2.5times as likely as those with health insurance to bewithout a recent visit (18.5 percent compared with7.3 percent) (table79).

In 1995–96, 7.2 percent of children 6–17 years ofage and 4.3 percent of children under age 6 had nousual source of health care. About one-quarter ofolder children without health insurance coverage hadno usual source of health care (table80).

In 1988–94, 23.1 percent of children 6–17 years ofage had at least one untreated dental cavity, downfrom 55.0 percent in 1971–74. Although substantialdeclines in untreated dental cavities have occurred forchildren at all income levels, poor children were 2.5times as likely as nonpoor children to have anuntreated cavity in 1988–94 (36.3 percent comparedwith 14.5 percent) (table72).

Inpatient CareMajor changes are occurring in the delivery of healthcare in the United States, driven in large part by theneed to rein in rising costs. One important change hasbeen a decline in use of inpatient services and anincrease in outpatient services. About 60 percent ofsurgical operations in community hospitals wereperformed on an outpatient basis in 1997.

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


Care

Expenditures

Between 1985 and 1996 the inpatient dischargerate declined by one-quarter from 138 discharges per1,000 population to 102 per 1,000, while averagelength of stay declined by more than a full day, from6.3 to 5.1 days (data are age adjusted) (table90).

Use of inpatient hospital care increases as familyincome declines. In 1996 the age-adjusted hospitaldischarge rate for persons with low family income(less than $16,000) was almost 3 times the rate forthose with high family income ($50,000 or more) andthe average length of hospital stay was nearly 2 dayslonger (6.6 days and 4.8 days) (table89).

In 1997, 61 percent of all surgical operations incommunity hospitals were performed on outpatients,up from 51 percent in 1990, 35 percent in 1985, and16 percent in 1980 (table96).

Between 1985 and 1997 the number ofcommunity hospital beds declined from 1 million to853,000 and during the same period occupancy rates incommunity hospitals declined from 65 to 62 percent(table110).

Between 1984 and 1994 the supply of beds ininpatient and residential mental health organizationsdeclined 14 percent to 98 beds per 100,000 population.The decline was greatest for state and county mentalhospitals with a reduction of 45 percent to 31 beds per100,000 population (table111).

In 1997 there were almost 1.5 million elderlynursing home residents 65 years of age and over.One-half of elderly nursing home residents were 85years of age and over and three-quarters were women.(table97)

In 1997 there were 1.7 million nursing homebeds in facilities certified for use by medicare andmedicaid beneficiaries. Nursing home bed occupancyin those facilities was estimated at 82 percent(table114).

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

Health Care Expenditures

National Health ExpendituresAfter 25 years of double-digit annual growth innational health expenditures, the rate of growth hasslowed during the 1990’s. However the United Statescontinues to spend more on health than any otherindustrialized country.

In 1997 national health care expenditures in theUnited States totaled almost $1.1 trillion, increasingless than 5 percent from the previous year andcontinuing the slowdown in growth of the 1990’s.During the 1980’s national health expenditures hadgrown at an average annual rate of 11 percent(table116).

This slowdown in growth is also reflected in theConsumer Price Index (CPI). The rate of increase inthe medical care component of the CPI declined from7.5 percent in 1985–90 to 3.0 percent in 1996–98(table117).

The combination of strong economic growth andthe slowdown in the rate of increase in health spendingover the last few years has stabilized healthexpenditures as a percent of the gross domesticproduct at 13.5–13.7 percent from 1993 to 1997, afterincreasing steadily from 8.9 percent in 1980(table116).

Despite the slowdown in the growth of healthspending, the United States continues to spend a largershare of gross domestic product (GDP) on healththan any other major industrialized country. TheUnited States devoted 13.5 percent of GDP to health in1997 compared with about 10 percent each inGermany, Switzerland, and France, the countries withthe next highest shares. (table115).

......................................................................... 11

12 .........

Highlights ...............................................................................................................H

ealth

Car

eEx

pend

iture

s

Expenditures by Type of Care and Source ofFundsExpenditures for hospital care as a percent of nationalhealth expenditures continue to decline. The sources offunds for medical care differ substantially according tothe type of medical care being provided.

Expenditures for hospital care continued todecline as apercent of national health expendituresfrom 42 percent in 1980 to 34 percent in 1997.Physician services accounted for 20 percent of the totalin 1997 and nursing home care and drugs for 8 and10 percent each (table119).

Between 1993 and 1997 the average annualincrease in total expenses in community hospitalswas 3.5 percent, following a period of higher growththat averaged 9.3 percent per year from 1985 to 1993(table123).

In 1997, 35 percent of personal health careexpenditures were paid by the Federal Governmentand 10 percent by State and local government; privatehealth insurance paid 32 percent, and 19 percent waspaid out-of-pocket. Between 1990 and 1997 the sharepaid by the Federal Government increased 6 percentagepoints, while the share paid out-of-pocket decreased4 percentage points (table120).

In 1997 the major sources of funds for hospitalcare were Medicare (33 percent) and private healthinsurance (31 percent). Physician services were alsoprimarily funded by private health insurance(50 percent) and Medicare (21 percent). In contrast,nursing home care was financed primarily by Medicaid(48 percent) and out-of-pocket payments (31 percent)(table120).

In 1995 funding for health research anddevelopment increased by 7 percent to $36 billion.Between 1990 and 1995 industry’s share of fundingfor health research increased from 46 to 52 percentwhile the Federal Government’s share decreased from42 to 37 percent (table127).

The National Institutes of Health (NIH ) accountfor about 80 percent of Federal funding for researchand development. In 1997 the National Cancer Institute

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

accounted for 20 percent of NIH’ s research anddevelopment budget, the National Heart, Lung andBlood Institute for 12 percent, and the NationalInstitute of Allergy and Infectious Diseases for10 percent (table127).

In 1998 Federal expenditures for HI V-relatedactivities increased 7 percent to $8.9 billion, aslowdown from an average annual increase of11 percent between 1995 and 1997. Of the totalFederal spending in 1998, 57 percent was for medicalcare, 21 percent for research, and 8 percent foreducation and prevention (table128).

Publicly Funded Health ProgramsThe two major publicly-funded health programs areMedicare and Medicaid. Medicare is funded by theFederal government and reimburses the elderly fortheir health care. Medicaid is funded jointly by theFederal and State governments to provide health carefor the poor. Medicaid benefits and eligibility vary byState. Medicare and Medicaid health care utilizationand costs vary considerably by State.

In 1997 the Medicare program had 38.4 millionenrollees and expenditures of $214 billion. The totalnumber of enrollees increased less than 1 percent overthe previous year while expenditures increased by7 percent (table134).

In 1997 hospital insurance (HI ) accounted for65 percent of Medicare expenditures. Expenditures forhome health agency care increased to 14.4 percent ofHI expenditures in 1997 up from 5.5 percent in 1990.Expenditures for skilled nursing facilities more thandoubled to 9.0 percent of the HI expenditures over thesame period (table134).

In 1997 supplementary medical insurance (SMI)accounted for 35 percent of Medicare expenditures.Group practice prepayment increased from 6.4 percentof the SMI expenditures in 1990 to 14.8 percent in1997 (table134).

Of the 33.4 million elderly Medicare enrollees in1996, 12 percent were 85 years of age and over and11 percent were 65–66 years of age. Medicarepayments increase with age from an average of $2,574

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


Care

Expenditures

per Medicare enrollee for those aged 65–66 years to$6,666 for those 85 years and over (table135).

In 1996 Medicare payments per enrolleeaveraged $5,048 in the United States, ranging from$3,500 in Nebraska, South Dakota, and Montana tomore than $6,200 in Massachusetts, Louisiana, and theDistrict of Columbia (table143).

In 1997 Medicaid vendor payments totaled $124billion, a 2-percent increase from the previous year.Recipients declined from 36.1 million in 1996 to 33.6million in 1997, a 7-percent decrease (table136).

In 1997 children under the age of 21 yearscomprised 46 percent of Medicaid recipients butaccounted for only 13 percent of expenditures. Theaged, blind, and disabled accounted for 30 percent ofrecipients and 74 percent of expenditures (table136).

In 1997 one-quarter of Medicaid payments wentto nursing facilities and 19 percent to general hospitals.Home health care accounted for 10 percent ofMedicaid payments in 1997, up from 5 percent in 1990(table137).

In 1997 almost 6 percent of Medicaid recipientsreceived home health care at a cost averaging $6,575per recipient. Early and periodic screening, rural healthclinics, and family planning services combinedreceived less than 2 percent of Medicaid funds in1997, with the cost per recipient averaging between$200 and $251 for each service (table137).

In 1997, 48 percent of Medicaid recipients wereenrolled in managed care, up from 40 percent theprevious year. In 1997 the percent of Medicaidrecipients enrolled in managed care varied substantiallyamong the States from 0 in Alaska and Wyoming to100 percent in Washington and Tennessee (table144).

Between 1996 and 1997 spending on health careby the Department of Veterans Affair s increased byless than 5 percent to $17.1 billion. In 1997, 43 percentof the total was for inpatient hospital care, down from58 percent in 1990, 37 percent for outpatient care, upfrom 25 percent in 1990, and 10 percent for nursinghome care, unchanged since 1990 (table138).

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

Privately Funded Health CareAbout 70 percent of the population has private healthinsurance, most of which is obtained through theworkplace. The share of employees’ total compensationdevoted to health insurance has declined in recentyears. The health insurance market is changing rapidlyas new types of managed care products are introduced.The use of traditional fee-for-service medical carecontinues to decline.

Between 1993 and 1997 the age-adjustedproportion of the population under 65 years of agewith privat e health insurance has remained stable at70–71 percent after declining from 76 percent in 1989.Some 92 percent of private coverage was obtainedthrough the workplace (a current or former employeror union) in 1997 (table129).

Nearly all persons 65 years of age and over areeligible for Medicare and most have additional healthcare coverage. However the percent with additionalcoverage has been declining. Between 1994 and 1997the age-adjusted percent of the elderly with privatehealth insurance declined from 78 to 70 percent whilethe percent with only Medicare coverage increasedfrom 13 to 21 percent (table130).

Between 1994 and 1998 privat e employers’health insurance costs per employee-hour workeddeclined from $1.14 to $1.00 per hour after increasingby 24 percent between 1991 and 1994. Among privateemployers the share of total compensation devoted tohealth insurance declined from 6.7 percent in 1994 to5.4 percent in 1998 (table122).

The average monthly contribution by full-timeemployees for family medical care benefits was morethan 50 percent higher in small companies ($182 in1996) than in medium and large companies ($118 in1995) (table133).

During the 1990’s the use of traditionalfee-for-service medical care benefits by full-timeemployees in private companies declined sharply. In1996 in small companies, 36 percent of full-timeemployees who participated in medical care benefitswere in fee-for-service plans, down from 74 percent in

........................................................................ 13

14 ..........

Highlights ...............................................................................................................H

ealth

Car

eEx

pend

iture

s

1990. In 1995 in medium and large companies,37 percent of participating full-time employees were infee-for-service plans, down from 67 percent in 1991(table133).

In 1998, 29 percent of the U.S. population wasenrolled in health maintenance organizations(HMO’s) , ranging from 21–23 percent in the Southand Midwest to 38–39 percent in the Northeast andWest. HMO enrollment has been steadily increasing.Enrollment in 1998 was 77 million persons, double theenrollment in 1993 (table132).

In 1997 non-Hispanic black and Hispanic personswere less likely to have private health insurance thannon-Hispanic white persons. However among thosewith private health insurance coverage, non-Hispanicblack and Hispanic persons were more likely than theirnon-Hispanic white counterparts to enroll in HMO’s .The elderly were less likely to be enrolled in privateHMO’s than younger adults and children (table131).

In 1998 the percent of the population enrolled inHMO’ s varied among the States from 0 in Alaska andVermont to 54 percent in Massachusetts. Other Stateswith more than 40 percent of the population enrolled inHMO’s in 1998 include Connecticut, Delaware,Maryland, Oregon, and California (table145).

In 1997 the proportion of the population withouthealth care coverage (either public or private) was16.1 percent, compared with 15.6 percent the previousyear and 12.9 percent in 1987. In 1997 the proportionof the population without health care coverage variedfrom less than 10 percent in Hawaii, Wisconsin,Minnesota, and Vermont to more than 20 percent inArkansas, Mississippi, Texas, New Mexico, Arizona,and California (table146).

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

e

a

te

ndiso

,y

lee

ho

l

e,ds

aet

s

............................................................................................... Health and AgingIntroduction

The older population in the United States is largand growing. The post-World War II baby boomgeneration is entering middle age, and in the early pof the twenty-first century, this group will swell theranks of the older population both in the United Staand in other Western industrialized countries.

The year 1999 has been proclaimed theInternational Year of Older Persons by the UnitedNations to draw attention to the aging of societies ato the contributions and needs of older persons. Thchartbook on health and aging describes the healtholder persons in the United States at the end of thetwentieth century.

The health of older Americans affects everyoneeither directly or indirectly. For older persons, qualitof life in later years is directly influenced by theirhealth and functional status. Persons who are disabby chronic conditions or by injuries such as falls havdifficulty living independently and managing theirpersonal affairs. Young and middle-aged persons wcare for aging parents, grandparents, relatives, andfriends know first hand the challenges, both financiaand emotional, of declining health in old age. Forsociety as a whole, the financing of health careservices for the elderly, particularly through Medicarthe Federal health insurance program for elderly andisabled persons, is a significant outlay of resource(1).

A long and healthy life is a universal goal. In thetwentieth century great progress has been made towincreasing the years of life for most Americans. In thUnited States today, most persons can look forwarda significant number of years spent in old age.Whether these will be healthy years, with high levelof physical and cognitive functioning, the ability tolive independently in the community, and access toaffordable health care, is of concern to all.

Organization of the Chartbook

This chartbook focuses on the group that hastraditionally been defined as elderly in the UnitedStates, persons 65 years of age and over. Thedefinition of old age is social as well as biological.Certain roles (for example, being retired or being a

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

rt

s

f

d

rd

o

grandparent) usually characterize old age, althoughthese life events may occur at many differentchronological ages. Within the population 65 years ofage and over, there is much variation in health andlevels of activity.

Age and sex differences are emphasized in thischartbook. Many of the health status and utilizationmeasures are shown by three or four age groups todraw attention to the heterogeneity in health among theolder population and to highlight the ‘‘oldest-old,’’persons 85 years of age and over, the fastest growingsegment of the elderly population. Data for women arepresented first, as they are the majority of the olderpopulation and represent 7 out of 10 persons 85 yearsof age and over. Race and ethnic variation in healthamong the older population, a topic of increasinginterest among researchers, is discussed when the datasources allow for such analysis (2).

Characterizing the health of older persons requiresnot only measuring mortality and morbidity but alsodescribing their living arrangements, their levels ofactivity, who assists them, and how they utilize thehealth care system.

This chartbook is divided into sections onpopulation, health status, and health care access andutilization. The emphasis is on current measures ofhealth and health care utilization among the olderpopulation. Important trends in health and health careare mentioned in the bullets accompanying the figures,and references are made to related tables inHealth,United States. Highlights are presented first. The 34figures and accompanying text are then followed bytechnical notes and data tables for each figure.

Here is a summary of each section:

Population

The first section of the chartbook describes somesociodemographic characteristics of older persons. Themost notable characteristic is the increasing size of theolder population (figure 1). Today, approximately 13out of every 100 Americans are 65 years of age andover. It is estimated that in 2030, 20 out of 100persons will be 65 years of age and over, and 2 out of100 will be 85 years of age and over. There are more

........................................................................ 17

g

rs

t

s

y0

r

y

s,

18 .........

Health and Aging ...............................................................................................In

trodu

ctio

n

women than men at every age among the elderlypopulation.

In 1997 approximately one-third of allnoninstitutionalized older persons lived alone. Amonwomen 85 years of age and over, 60 percent livedalone (figure 2). The proportion of all persons 85 yeaof age and over living alone rose from 39 percent in1980 to 49 percent in 1997. Approximately 4 percenof persons 65 years of age and over were in nursinghomes in 1997, and women had higher rates of nurhome residence than men (figure 3).

While poverty rates among the older populationhave declined since the 1960’s, 1 out of 10 personsyears of age and over in 1997 lived in families withincome below the poverty line (figure 4).

Health Status

The second section of the chartbook presentsmeasures of health status. Figures on life expectanc(figures 5and6) show gains in years of life from 195to the present and differentials in life expectancy byrace and sex. In 1997 life expectancy in the UnitedStates was 79.4 years for women and 73.6 years fomen. Life expectancy at birth, as well as lifeexpectancy at ages 65 and 85 years of age, hasincreased over time as death rates for many causesdeath have declined.

The biggest decreases in mortality have been indeath rates for heart disease and stroke. However,death rates for some causes of death among theelderly, for example pneumonia and influenza, haveincreased in the last two decades.

Many factors have contributed to mortalitydeclines in the last 50 years: changes in healthbehaviors, for example, declines in smoking andimprovements in nutrition, increases in the overalleducational level of the older population, andinnovations in medical technology.

Will life expectancy continue to increase? Theabove factors will likely also influence life expectancin the future. For example, the percent of elderlypersons who have completed high school will increafrom almost 66 percent in 1997 to an estimated 83percent in 2030. Nearly one-fourth of the elderly in

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

ing

65

of

e

2030 will be college graduates (3). Governmentprojections suggest that life expectancy in the UnitedStates will reach 84.3 years for women and 79.7 yearsfor men by 2050 (4). Japan currently has the longestlife expectancy in the world, 82.9 years for womenand 76.4 years for men in 1995. At age 65 Japanesewomen have a life expectancy of 20.9 years.Considerable research is underway to determine whatmaximum life span is, how life expectancy may beenhanced, and what are the characteristics oflong-lived families.

Next the chartbook presents measures of healthand disability, including the prevalence of chronicconditions (figure 11), visual and hearing impairments(figures 12and13), osteoporosis (figure 14), physicalfunctioning and disability (figure 15), conditionsassociated with disability (figure 16), overweight(figures 17and18), oral health (figures 19and20), andsocial activity and exercise (figures 21–22).

The wide differences by age in the health of theolder population are clearly seen in nearly allmeasures. Rates of illness and disability increasesharply among persons 85 years of age and overcompared with persons 65–74 years or 75–84 years ofage. For example, 35 percent of white men 70–74years of age in 1995 were hearing impaired comparedwith 56 percent of white men 85 years of age and over(figure 13). Nineteen percent of women 65–74 years ofage had osteoporosis compared with 51 percent ofwomen 85 years of age and over (figure 14). Fivepercent of women 70–74 years of age were unable todo one or more activities of daily living comparedwith 23 percent of women 85 years of age and over(figure 15). Increases in illness and disability areaccompanied by decreases in social activity. Amongwomen 70–74 years of age, 65 percent participated inat least five different social activities in a 2-weekperiod compared with 39 percent of women 85 yearsof age and over (figure 21).


The last section of the chartbook focuses on healthcare access and utilization. These measures show thatin general, persons 85 years of age and over have

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

a

f

oo

t


higher rates of health care utilization than youngerpersons. Women 85 years of age and over were twicas likely to use assistive devices such as canes,walkers, and hearing aids as women 70–74 years ofage (figure 25). Hospitalizations for fractures were 5times as high among women 85 years of age and ovas for women 65–74 years of age (figure 27). Rates ofhome health care use were over 4 times as high foroldest women as for women 65–74 years of age(figure 29).

At the same time, persons 85 years of age andover were less likely than younger persons to becovered by private insurance in addition to MedicareLess than one-half of non-Hispanic black persons anHispanic persons 85 years of age and over had privinsurance to supplement their Medicare coverage(figure 31). While the total costs of heart disease anddiabetes were lower among the population 85 yearsage and over compared with the population 75–84years of age, per capita costs of health care for thesillnesses were highest among the oldest members othe population (figures 33and34).

Data

The data presented in the charts are fromnationally representative health surveys or vitalstatistics. One of the data sources (the SecondSupplement on Aging to the 1994 National HealthInterview Survey) is a survey of persons 70 years ofage and over. Consequently, some figures present dfor the population 70 years of age and over insteadthe population 65 years of age and over. For data frthe National Health Interview Survey (except for theSecond Supplement on Aging and supplements on ohealth in 1983 and 1993 and exercise in 1995), survyears are combined (1994–96 or 1993–95) to createlarge enough sample for analysis.

Measures of health are based on thenoninstitutionalized population that excludes residenof nursing homes, except where noted, for examplerates of nursing home residence (figure 3), lifeexpectancy and mortality (figures 5–9), or hospitaldischarge rates (figure 27). Consequently, the measureof health in the chartbook in general are biased sligh

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

e

er

the

.dte

of

e

atafm

raleya

s

stly

upward; that is, the noninstitutionalized olderpopulation is healthier than the older population as awhole.

In national surveys that are not specificallydesigned to study the elderly, the number ofobservations may not be large enough to analyzedifferences among all age, sex, and race/ethnicitygroups. For certain topics, data are presented for allraces combined in the chart, and significant racedifferences (if they exist) are discussed in theaccompanying text.

Data Gaps

Although the chartbook focuses on the elderly, thehealth of persons in old age is related to their healthstatus and health behaviors throughout life. Those intheir middle ages are of particular interest toresearchers today in planning for the health care needsof the elderly of the twenty-first century. The large sizeof the baby boom cohorts ensures that their health,health care utilization, and financial status will have alarge impact on society. The figures in this chartbookdo not cover the population 50–64 years of age.However, many of the data sources used, for examplethe National Health Interview Survey, the NationalHealth and Nutrition Examination Survey, the NationalHospital Discharge Survey, and vital statistics, containinformation on these age groups. In addition, manytables inHealth, United Statespresent data for middleaged persons.

This chartbook does not include measures ofhealth by socioeconomic status. Although differencesin health status and health care utilization bysocioeconomic status exist, they are generally smallerfor older persons compared with younger persons.

Most surveys have only income-based measures ofsocioeconomic status and do not capture theaccumulated wealth and assets on which many olderpersons rely. In addition, many older surveyrespondents do not know their incomes or are notwilling to share this information, resulting in largeproportions of respondents with missing data. Forexample, in the Second Supplement on Aging,

......................................................................... 19

sut

a1

to

teheo

ine

e

n

t

nd

n

t

ht

d

ity

r

e

d

20 .........

Health and Aging ...............................................................................................In

trodu

ctio

n

approximately one-fourth of the sample are missingdata on the family income question.

Important work is in progress in this area tocollect better information on the socioeconomic statuof older adults. Two surveys conducted by the Institfor Social Research of the University of Michiganhave collected wealth and asset information: the Heand Retirement Study that focuses on persons 51–6years of age and a survey called Asset and HealthDynamics Among the Oldest Old, which studiespersons 70 years of age and over (5).

Cognitive and emotional functioning are crucialgood health but are difficult to measure in surveysusing traditional data collection tools. There is debaregarding the prevalence of Alzheimer’s disease in tolder population, and estimates of the number of oldAmericans suffering from the disease range from ab2 million to 4 million persons (6).

The prevalence of major depression among thenoninstitutionalized elderly is estimated by somestudies to be less than 3 percent, although theprevalence of depressive symptoms is higher. Inaddition, community-dwelling older persons havelower rates of depression than persons in nursinghomes or care facilities (7). New approaches are bedeveloped to provide better national, population-basestimates of cognitive functioning and mental healthamong the elderly.

Nutritional status is another area important to thhealth of older persons but difficult to measure innational surveys. What people eat and how well thefood is absorbed, digested, and utilized are crucial idetermining the health status of individuals.

Although the topic of nutrition is not presenteddirectly in the chartbook, many of the risk factors thacontribute singly or interactively to nutritionalproblems are covered here. For example, poverty aeconomic uncertainties are important contributors tomalnutrition. Limited income decreases the variety aquantity of food purchased and consumed. Livingalone and eating alone contribute to reduced foodintake, while sadness and depression may exacerbathis situation and lead to social isolation with itspotential for changes in appetite, energy level, weigand well-being.

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

e

lth

erut

gd

d

e

The suppressing effect of certain medications ontaste, smell, and appetite can also lead to reduced foointake. Physical disabilities such as difficulty walking,grocery shopping, and preparing food further restrictaccess to adequate amounts and types of food. Inabilto carry heavy things during shopping can limit theselection of food products such as fresh fruits andvegetables, and, therefore, limit variety and completenutrient intake. Similarly, missing, loose, or decayedteeth or ill-fitting dentures make it hard for elders toeat well.

In addition, altered mental status such asconfusion and memory loss make it hard to remembewhat, when, and if one has eaten and limit the abilityto modify diets in response to chronic diseases. To thextent that one has these risk factors, foodinsufficiency and/or malnutrition may be a problem.

The prevalence of food insufficiency in 1988–94has been estimated from the Third National Health anNutrition Examination Survey to be 1.7 percent amongall persons 60 years of age and over and 5.9 percentamong low-income persons in that age group (8).Forthcoming analyses from this survey will providenew data on the nutritional status of the olderpopulation.

Conclusion

The older population throughout the world isgrowing. The International Year of Older Personsproclaimed by the United Nations provides anopportunity to evaluate the health of older persons atthe end of a century of remarkable advances in healthand longevity.

Americans live longer than ever before. Personswho survive to age 65 in the United States today canexpect to live on average nearly 18 more years.

This chartbook examines a variety of currentmeasures of health and health care utilization fromnational data sources. The health of individuals in oldage reflects the cumulative effect of health behaviorsand health care over a lifetime as well as advances inmedical technology and the biological process ofaging. These factors cannot be disentangled bymeasuring health status at one point in time.

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


The health of the older population varies greatly.The largest differences are age related. Persons 85years of age and over, the majority of whom(71 percent) are women, have noticeably higher ratesof illness, disability, and utilization of health careservices than older persons who are less than 85 yearsof age. Ensuring good health and quality of life in oldage requires attention to differences in the populationby race and ethnicity, sex, and age.

References

1. Waldo DR, Sonnefeld ST, McKusick DR, Arnett RH. Health expenditures by agegroup, 1977 and 1987. Health Care Financing Review 10(4):111–20. 1989.

2. Martin LG, Soldo BJ, eds. Racial and ethnic differences in the health of olderAmericans. Washington, DC: National Academy Press. 1997.

3. U.S. Bureau of the Census. Current population reports. Special studies, P23-190.65+ in the United States. U.S. Government Printing Office, Washington. 1996.

4. Day JC. Population projections of the United States by age, sex, race, andHispanic origin: 1995 to 2050. U.S. Bureau of the Census. Current population reports;P25–1130. Washington: U.S. Department of Commerce. 1996.

5. Soldo BJ, Hurd MD, Rodgers WL, Wallace RB. Asset and health dynamics amongthe oldest old: An overview of the AHEAD study. J Gerontol 52B(special issue):1–20.1997.

6. U.S. General Accounting Office. Alzheimer’s disease: Estimates of prevalence in theUnited States. GAO/HEHS-98–16. 1998.

7. Diagnosis and treatment of depression in late life. NIH Consens Statement Online1991 Nov4–6. 9(3):1–27.<http://www.nih.gov>. November 1998.

8. Alaimo K, Briefel RR, Frongillo, Jr EA, Olson CM. Food insufficiency exists in theUnited States: Results from the Third National Health and Nutrition ExaminationSurvey (NHANES III). AJPH 88(3):419–26. 1998.

................................................................................................................................... 21

http://www.nih.gov

s

r

e

s

22 ..........

Health and Aging ...............................................................................................Po

pula

tion

Demographic Characteristics

In 1997, 13 percent of the U.S. population was 65years of age and over. Among these 34 millionpersons, nearly 4 million were 85 years of age andover. The population of the United States is aging; theelderly population is growing at a faster rate than thepopulation as a whole. In addition, the proportion ofthe population 85 years of age and over is growingfaster than the elderly population as a whole.Projections of the population indicate that 70 millionpersons will be 65 years of age and over in the year2030, representing 20 percent of the total U.S.population. It is estimated that the population 85 yearof age and over will more than double toapproximately 8.5 million persons.

There are more women than men among the oldepopulation. Among persons 65 years of age and overin 1997, 59 percent were women. At the oldest ages,the sex ratio is even higher; 71 percent of persons 85years of age and over were women.

In 1997 a larger proportion of the non-Hispanicwhite population was over the age of 65 comparedwith other racial and ethnic groups. Fifteen percent ofthe non-Hispanic white population was 65 years of agand over compared with 8 percent of the blackpopulation, 7 percent of the Asian or Pacific Islanderpopulation, and 7 percent of the American Indian orAlaska Native population. Among persons of Hispanicorigin, 6 percent were 65 years of age and over.However, the older non-Hispanic white population isgrowing more slowly compared with other groups.From 1990 to 1997, the proportion of the population65 years of age and over grew more than five times afast among Hispanic persons as among non-Hispanicwhite persons.

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

Figure 1. Population 65 years of age and over: United States, 1950–2030

1950 1960 1970 1980 1990 2000 2010 2020 2030

Year

0

10

20

30

40

50

60

70

Number in millions

85 years and over

................................................................................................................................... 23

............................................................................................... Health and Aging

NOTES: Figures for 1950–90 are based on decennial censuses. Figures for 2000–30 are middle series population projections of the U.S. Bureau of theCensus.

SOURCES: See Health,United States, 1999, table 1 for data years 1950–90. For data years 2000–30, see U.S. Bureau of the Census. Day JC.Population projections of the United States by age, sex, race, and Hispanic origin: 1995 to 2050. Current population reports; P25–1130. Washington:U.S. Department of Commerce. 1996.

Population

24 ......


pula

tion

Living Arrangements

The living arrangements of elderly persons reflecttheir health status as well as family and cultural ties.Older nonmarried persons who live alone (the majorityof whom are widowed) in general are in better healththan nonmarried persons who do not live alone.

The majority of noninstitutionalized older personslive with family members; however, the livingarrangements of the elderly vary greatly by age, sex,race, and marital status. In 1997 nearly one-third ofnoninstitutionalized persons 65 years of age and overlived alone. The proportion living alone was higheramong persons 85 years of age and over comparedwith persons 65–74 and 75–84 years of age. In eachage group, women were at least twice as likely as mento live alone. Six out of ten women 85 years of ageand over lived alone.

Older men are more likely to be married thanolder women, in large part because women outlivemen. Among persons 75–84 years of age, men weremore than twice as likely as women to live with aspouse. Among persons 85 years of age and over, menwere more than 4 times as likely as women to livewith a spouse.

In every age group, black and Hispanic womenwere more likely to live with other relatives comparedwith non-Hispanic white women. At ages 75 years andover, non-Hispanic white women were 1.2 times aslikely as black women and 1.7 times as likely asHispanic women to live alone. Compared withnon-Hispanic white or Hispanic men, older black menwere more likely to live alone and less likely to livewith a spouse.

Only a small proportion of the noninstitutionalizedolder population lived with nonrelatives. Amongwomen 65 years of age and over, 2 percent lived withnonrelatives. Among men, the percent living withnonrelatives was 3 percent among persons 65–84 yearsof age and 7 percent among persons 85 years of ageand over.

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

Figure 2. Living arrangements of persons 65 years of age and over by age and sex: United States, 1997

65-74years

65-74years

75-84years

75-84years

85 yearsand over

85 yearsand over

Women

Men

Alone With spouse With other relatives With nonrelatives

20 40 60 80 100

Percent

0

................................................................................................................................... 25

............................................................................................... Health and Aging

NOTES: Figures are based on the noninstitutionalized population. See Technical Notes for definitions of categories of living arrangements.

SOURCE: Lugaila TA. U.S. Bureau of the Census. Marital Status and Living Arrangements: March 1997 (Update), Series P20–506.

Population

t

sg

e,

d

.

ng

t

tsto

f

,

,

ld

26 .......


pula

tion

Nursing Home Residence

Older persons live in nursing homes whendisability, chronic illness, or mental incapacity preventhem from living on their own or being cared for inthe community. In 1997, 1.5 million persons 65 yearof age and over were in nursing homes, representin4 percent of the older population.

The rates of nursing home residence vary by agsex, and race. In 1997, 11 in 1,000 persons 65–74years of age were nursing home residents comparedwith 46 out of 1,000 persons 75–84 years of age an192 out of 1,000 persons 85 years of age and over.Women had higher rates of nursing home residencethan men, and the sex difference increased with ageBlack persons 65–74 years of age and black men75–84 years of age were more likely than their whitecounterparts to be nursing home residents.

One-half of the current elderly residents of nursihomes were 85 years of age and over, andthree-fourths were women. Research has shown thaunmarried elderly persons have a higher risk ofnursing home admission than married persons (1).Nearly two-thirds of all current nursing home residenwere widowed, with female residents twice as likelybe widowed as male residents.

There is variation in the health of nursing homeresidents. Among nursing home residents 65 years oage and over in 1997, 48 percent were receivingfull-time skilled nursing care under a physician’ssupervision. Twenty-nine percent had difficulty seeingand 26 percent had difficulty hearing. Nearly all(96 percent) required help with bathing or showeringwhile 45 percent needed assistance with eating.Seventy-nine percent of nursing home residents counot use the telephone on their own, and 65 percent

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

could not care for their personal possessions withouthelp. The levels of disability and functional status weresimilar among women and men.

Reference1. Freedman VA. Family structure and the risk of nursing home admission. J Gerontol (51B):S61–9. 1996.

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

Figure 3. Nursing home residents among persons 65 years of age and over by age, sex, and race: United States,1997

White WhiteBlack Black

Women Men

Residents per 1,000 population

00

50

100

150

200

250

65-74 years 75-84 years 85 years and over

................................................................................................................................... 27

............................................................................................... Health and Aging

NOTES: Nursing home residents exclude residents in personal care or domiciliary care homes. Age refers to age at time of interview. Rates are basedon the resident population as of July 1, 1997, adjusted for net underenumeration using the 1990 National Population Adjustment Matrix from the U.S.Bureau of the Census.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Nursing Home Survey. See related Health, UnitedStates, 1999, tables 97 and 98.

Population

leicith

d

er

as

ty.he

ial

ns

srs

o(l(yw

cmdA twba

28 ..........


pula

tion

Poverty

Socioeconomic status affects the health of peopof all ages (1). Differences in health by socioeconomstatus are smaller among older persons compared wyounger adults, yet they are still notable. Whilepoverty rates among the elderly have declinedsignificantly since the 1960’s, 1 out of 10 persons 65years of age and over in 1997 was living in a familywith income below the Federal poverty threshold.

The poverty rate was higher among older blackand Hispanic persons compared with older whitepersons. In 1997 among persons 65 years of age anover, black persons were 2.9 times as likely andHispanic persons were 2.7 times as likely to live inpoverty as white persons.

More older women than men live in poverty. Thepoverty rate among persons 65 years of age and ovwas higher for women than for men (13 percentcompared with 7 percent), although sex differencesvaried among racial and ethnic groups. Among blackand Hispanic older persons, women were 1.3 timeslikely to be in poverty as men. Among older whitepersons, women were twice as likely to live in poverPart of the sex difference in poverty rates is due to teffects of widowhood. Older women are particularlyvulnerable to declines in economic status after thedeath of their spouse (2).

Poverty rates among the elderly have beendeclining as older persons have benefitted from SocSecurity payments and health insurance throughMedicare and Medicaid. In 1959, 35 percent of perso65 years of age and over lived in poverty comparedwith nearly 11 percent in 1997. In 1959 older personhad higher poverty rates than children (under 18 yea

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

f age) or younger adults (18–64 years of age)27 percent and 17 percent, respectively). In 1997evels of poverty were highest among children19 percent) while poverty rates among persons 65ears of age and over and persons 18–64 years of ageere not significantly different from each other (3).

Poverty measures based on income do not fullyapture the accumulated wealth and assets on whichany older persons rely. One study has shown largeisparities in wealth by race among elderly Americans.mong persons 70 years of age and over, the mean neorth of households of white persons was estimated toe nearly 4 times the net worth of black householdsnd 3 times that of Hispanic households (4).

References1. Pamuk E, Makuc D, Heck K, Reuben C, Lochner K. Socioeconomic status and health chartbook. Health, United States, 1998. Hyattsville, Maryland: National Center forHealth Statistics. 1998.

2. Bound J, Duncan GJ, Laren DS, Oleinick L. Poverty dynamics in widowhood. J Gerontol (46): S115–124. 1991.

3. U.S. Census Bureau. ‘‘Historical Poverty Tables - People, (Table) 3. Poverty Status of People, by Age, Race, and Hispanic Origin: 1959–1997.’’ Last revised September 24,1998. <http://www.census.gov./hhes/poverty/histpov/hstpov3.html>.

4. Smith JP. Wealth inequality among older Americans. J Gerontol 52B (Special Issue): 74-81. 1997.

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

http://www.census.gov./hhes/poverty/histpov/hstpov3.html

Figure 4. Percent in poverty among persons 65 years of age and over by sex, race, and Hispanic origin:United States, 1997

Women

Men

0 5 10 15 20 25 30

Percent

White

White

Black

Black

Hispanic

Hispanic

................................................................................................................................... 29

............................................................................................... Health and Aging

NOTES: Figures are based on the civilian noninstitutionalized population. The race groups white and black include persons of both Hispanic andnon-Hispanic origin. Persons of Hispanic origin may be of any race. See Appendix II for poverty level definition.

SOURCE: Dalaker J, Naifeh M. U. S. Bureau of the Census. Poverty in the United States: 1997. Current population reports; Series P60–201.Washington: U.S. Government Printing Office. 1998.

Population

.4

d

s

stee

30 .........

Health and Aging ...............................................................................................H

ealth

Stat

us

Life Expectancy

Life expectancy at birth increased in the UnitedStates by approximately 8 years in the second half ofthe twentieth century. In 1997 life expectancy was 79years for women and 73.6 years for men. Taking intoaccount the tremendous growth in life expectancy(nearly 20 years) that occurred in the first half of thecentury, the expected number of years of life increaseby approximately 60 percent since 1900. Less thanone-half of all children born at the turn of the centurycould expect to live to age 65. About 80 percent borntoday can expect to survive to age 65 and roughlyone-third to age 85.

Life expectancy at ages 65 and 85 also increasedover the past 50 years. Under current mortalityconditions, women who survive to age 65 can onaverage expect to live to age 84, and women whosurvive to age 85 can on average anticipate livingalmost to age 92. Men can expect to have shorter liveon average.

Life expectancy has increased over time as deathrates have declined. In 1997 the death rate for person65–74 years of age was 16 percent lower than the rain 1980. Among persons 85 years of age and over, thdeath rate declined by 4 percent in the same timeperiod. However, not all causes of death havecontributed to this downward trend. The majorreductions in mortality during the past two decadeshave occurred for heart disease and stroke. Bycontrast, pneumonia and influenza mortality increasedamong elderly women and men.

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

Figure 5. Life expectancy at birth, age 65, and age 85 by sex: United States, 1950–97

Years of life

1950 1960 1970 1980 1990 1997

0

10

20

60

70

80

At age 65, men

At age 85, men

At age 85, women

At age 65, women

At birth, men

At birth, women

................................................................................................................................... 31

............................................................................................... Health and Aging

NOTE: See Technical Notes on life expectancy estimation.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System. See related Health,United States, 1999, table 28.

Health

Status

r

t

k

32 .........


ealth

Stat

us

Life expectancy varies by race at older ages aswell as at birth. In 1997 life expectancy at birth wasapproximately 5 years longer for white women thanfor black women and 7 years longer for white menthan for black men. At age 65, differences by racenarrowed and life expectancy was 1.7 years longer fowhite women than for black women and 1.8 yearslonger for white men than for black men. However, atage 85 life expectancy for black persons was slightlyhigher than for white persons.

The declining race differences in life expectancy aolder ages are a subject of debate. Some researchshows that age misreporting may have artificiallyincreased life expectancy for black persons,particularly when birth certificates were not available(1). However, other research has suggested that blacpersons who survive to the oldest ages may behealthier on average than white persons and havelower mortality rates (2).

References1. Preston SH, Elo IT, Rosenwaike I, Hill M. African-American mortality at older ages:Results of a matching study. Demography 33(2): 193–209. 1996.

2. Manton KC, Stallard E, Wing S. Analyses of black and white differentials in the agetrajectory of mortality in two closed cohort studies. Stat Med 10: 1043–59. 1991.

Life Expectancy by Race

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

Figure 6. Life expectancy at birth, age 65, and age 85 by sex and race: United States, 1997

At birthAt birth At age 65At age 65 At age 85At age 85

Years of life

Women Men

0

10

20

30

40

50

60

70

80

White Black

................................................................................................................................... 33

............................................................................................... Health and Aging

NOTE: See Technical Notes on life expectancy estimation.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System. See related Health, UnitedStates, 1999, table 28.

Health

Status

te

r

icdd

s

nic

iald

se

34 .........


ealth

Stat

us

Deaths From All Causes

In 1997 the overall death rate among the olderpopulation was higher for men than women in eachage group; however, death rates converged at theoldest ages. Men 65–69 years of age had a death ra1.7 times the rate for women of the same age. Forpersons 95 years of age and over, the death rates fomen and women were nearly equal.

Among persons 65–74 years of age and 75–84years of age, death rates were highest among blackmen and women compared with other racial and ethngroups. However, among persons 85 years of age anover, white men and women had the highest recordedeath rates compared with other groups. Among allpersons 65 years of age and over, Asians or PacificIslanders and Hispanics had lower death rates thanwhite and black persons.

Care must be taken in comparing mortality levelsamong older persons for different racial and ethnicgroups. The accuracy of death rates for black personat the oldest ages is a subject of debate. Death ratesfor American Indians or Alaska Natives, Asians orPacific Islanders, and Hispanics are regarded asunderstated. The Asian or Pacific Islander and Hispapopulations grew dramatically over the past twodecades, largely due to immigration. These broad racand ethnic categories include native-born persons animmigrants from many different countries and diversebackgrounds. The overall death rate may obscuredifferences in health and mortality between subgroupof these populations. See the Technical Notes on racand ethnicity for a discussion of data quality issues.

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

Figure 7. Death rates for all causes among persons 65 years of age and over by age and sex: United States, 1997

Deaths per 100,000 population

65-69 70-74 75-79 80-84 85-89 90-94 95 and overYears of age

0

5,000

10,000

15,000

20,000

25,000

30,000

Men

Women

65-74 years 85 years and over75-84 years

Women

65-74 years 85 years and over75-84 years

Men

0

5,000

10,000

15,000

20,000Deaths per 100,000 population

White Black HispanicAsian orPacific Islander

American Indianor Alaska Native

................................................................................................................................... 35

............................................................................................... Health and Aging

NOTE: Persons of Hispanic origin may be of any race.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System. See related Health, UnitedStates, 1999, table 36.

Figure 8. Death rates for all causes among persons 65 years of age and over by age, sex, race, and Hispanicorigin: United States, 1997

Health

Status

ee,

e of

ento

g

art

ren

nd

nds

e

se

s


0 200 400 600 800 1000 1200

Heartdisease

Heartdisease

Cancer

Cancer

Stroke

Stroke

COPD

COPD

Pneumonia/influenza

Pneumonia/influenza

Women

Men

65-74 years

36 .........

Health and Aging ...............................................................................................

Figure 9. Death rates for selected leading causesamong persons 65 years of age and over by age andsex: United States, 1997

See notes on page 37.

Hea

lthSt

atus

Selected Leading Causes of Death

Among all persons 65 years of age and over, the fivleading causes of death are heart disease, cancer, strokchronic obstructive pulmonary diseases, and pneumoniaand influenza. In 1997 heart disease alone was the caus35 percent of all deaths among men 65 years of age andover and for 40 percent of deaths among women and m85 years of age and over. The proportion of deaths duestroke increased with age, accounting for 11 percent ofdeaths among women and for 9 percent of deaths amonmen 85 years of age and over.

Cancer is also a major cause of death. Amongpersons 65–74 years of age, cancer was the leadingcause of death among women and roughly equal to hedisease among men. Lung cancer, followed by breastcancer, were the leading causes of cancer deaths forwomen in this age group. The proportion of deathsattributed to cancer declined with age for women andmen. Among persons 85 years of age and over, cancewas responsible for 10 percent of deaths among womand for 16 percent of deaths among men.

Other major causes of death among the olderpopulation include pneumonia and influenza, which wereresponsible for approximately 7 percent of all deaths forpersons 85 years of age and over. Chronic obstructivepulmonary diseases (COPD), bronchitis, emphysema,asthma, and other allied conditions, accounted for about6–7 percent of deaths for persons 65–84 years of age, aa slightly lower percent for persons 85 years of age andover. Unintentional injuries such as automobile crashes afalls were the seventh leading cause of death for person65 years of age and over.

The relative importance of certain other causes ofdeath varied according to race, ethnicity, and sex. In1997 diabetes was the third leading cause of deathamong American Indians 65 years of age and over, thfourth leading cause of death among older Hispanicpersons and black persons, and ranked sixth for olderwhite persons and Asian Americans. Alzheimer’s diseawas the sixth leading cause of death among whitewomen 85 years of age and over; however, it was lescommon among black women of the same age or menof either race.

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

Figure 9. Death rates for selected leading causes among persons 65 years of age and over by age and sex:United States, 1997—Continued


0 500 1000 1500 2000 2500 3000

Heartdisease

Heartdisease

Cancer

Cancer

Stroke

Stroke

COPD

COPD

Pneumonia/influenza

Pneumonia/influenza

Women

Men

75-84 years

0 1000 2000 3000 4000 5000 6000 7000


Heartdisease

Heartdisease

Cancer

Cancer

Stroke

Stroke

COPD

COPD

Pneumonia/influenza

Pneumonia/influenza

Women

Men

85 years and over

................................................................................................................................... 37

............................................................................................... Health and Aging

NOTES: COPD is chronic obstructive pulmonary diseases. For a description of International Classification of Diseases code numbers for causes ofdeath, see Appendix II.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System. See related Health,United States, 1999, tables 33, 37–39, and 42.

Health

Status

38 .........


ealth

Stat

us

Self-Reported Health

Self-assessed health, the reporting of health asexcellent, very good, good, fair, or poor, is a summarymeasure that represents physical, emotional, and socialaspects of health and well-being. Self-reported healthcorrelates highly with mortality (1). Research has alsodemonstrated that elderly persons who report theirhealth as poor are at increased risk for declines inphysical functioning, independent of the severity ofother medical conditions (2).

Older adults consider themselves to be in worsehealth than do young or middle-aged adults. In1994–96, 28 percent of persons 65 years of age andover reported their health status as fair or poorcompared with 17 percent of persons 45–64 years ofage. The percent of older adults in fair or poor healthincreased steadily with age, from one-fourth of persons65–74 years of age to over one-third of persons 85years of age and over. The age pattern and levels offair and poor health were similar among women andmen.

At every age and for both men and women,non-Hispanic black and Hispanic persons reportedworse health than non-Hispanic white persons. Amongpersons 65–74 years of age, non-Hispanic blackpersons were 1.7 times as likely and Hispanics were1.4 times as likely to be in fair or poor health asnon-Hispanic white persons. Racial and ethnicdifferences in self-reported health reflect objectivedifferences in health status and physical functioning aswell as cultural and socioeconomic differences in theassessment of health and in the interpretation of healthstatus questions (3,4).

References1. Idler EL, Benyamini Y. Self-reported health and mortality: A review of twenty-sevencommunity studies. J Health Soc Behav 38:21–37. 1997.

2. Idler EL, Kasl SV. Self-ratings of health: Do they also predict change in functionalability? J Gerontol 50B(6):S344–53. 1995.

3. Coward RT, Peek CW, Henretta JC, et al. Race differences in the health of elderswho live alone. J Aging Health 9(2): 147–70. 1997.

4. Krause NM, Jay GM. What do global self-rated health items measure? Med Care32(9): 930–42. 1994.

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

Figure 10. Fair or poor health among persons 65 years of age and over by age, sex, race, and Hispanic origin:United States, 1994–96

Percent

White,non-Hispanic

White,non-Hispanic

Hispanic HispanicBlack,non-Hispanic

Black,non-Hispanic

0

10

20

30

40

50

60

Women Men


................................................................................................................................... 39

............................................................................................... Health and Aging

NOTE: Figures are based on the noninstitutionalized population.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Interview Survey. See related Health,United States, 1999, table 60.

Health

Status

t

e

onf

a)

neartghf

ge

tedf

rs

nce

;

k

40 ..........


ealth

Stat

us

Chronic Conditions

Chronic diseases are prolonged illnesses that arerarely cured completely. Some examples are arthritisand hypertension. While not all chronic diseases arelife threatening, they are a substantial burden on thehealth and economic status of individuals, theirfamilies, and the nation as a whole. Chronic conditionaffect the quality of life of older persons andcontribute to disability and the decline of independenliving (1). In 1995 among noninstitutionalized persons70 years of age and over, 79 percent had at least onof seven chronic conditions common among theelderly.

In 1995 the majority of persons 70 years of ageand over had arthritis. Women reported higher levelsof arthritis than men (63 percent compared with50 percent). Hypertension is also a prominent conditiamong the elderly, affecting approximately one-third opersons 70 years of age and over. Older womenreported more hypertension than men. Respiratoryillnesses (asthma, chronic bronchitis, and emphysemaffected 11 percent of the older population in 1995,and levels were similar among women and men.

Other chronic diseases suffered by older personsinclude heart disease, diabetes, stroke, and cancer. I1995 more than one-fourth of persons 70 years of agand over reported having heart disease. Levels of hedisease were higher among men than women, althouthese differences declined with age. Eleven percent opersons 70 years of age and over reported that theycurrently had diabetes, with women and men reportinsimilar levels. Nine percent of persons 70 years of agand over had ever had a stroke, and 4 percent reporthat they currently had some form of cancer. Levels ostroke and cancer were higher among men thanwomen.

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

s

The prevalence of certain chronic conditionsvaried by race and ethnicity. Among persons 70 yeaof age and over, non-Hispanic black and Hispanicpersons had higher levels of diabetes thannon-Hispanic white persons; for women the prevaleof diabetes was twice as high. Non-Hispanic blackolder persons were 1.5 times as likely to reporthypertension as non-Hispanic white persons. Amongwomen, levels of heart disease did not vary by raceamong men, however, non-Hispanic white menreported more heart disease than non-Hispanic blacmen or Hispanic men. Non-Hispanic black womenreported higher levels of stroke than non-Hispanicwhite women.

Reference1. Centers for Disease Control and Prevention. Unrealized prevention opportunities: Reducing the health and economic burden of chronic disease. Centers for Disease Control andPrevention, National Center for Chronic Disease Prevention and Health Promotion. 1997.

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

Figure 11. Percent of persons 70 years of age and over who reported selected chronic conditions by sex:United States, 1995

Arthritis

Hypertension

Heartdisease

Diabetes

Respiratorydiseases

Stroke

Cancer

0 20 40 60 80

Women

Men

Percent

................................................................................................................................... 41

............................................................................................... Health and Aging

NOTES: Based on interviews conducted between October 1994 and March 1996 with noninstitutionalized persons. Percents are age adjusted. SeeTechnical Notes for definitions of conditions and age adjustment procedures.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, 1994 National Health Interview Survey, Second Supplementon Aging. See related figures 33 and 34 on cost of heart disease and cost of diabetes.

Health

Status

ti

ns

ed

ver.

arered

r

d

4and

e

ss

42 ..........


ealth

Stat

us

Visual and Hearing Impairments

Visual and hearing impairments affect a substanproportion of the elderly population. Adverseconsequences of these sensory impairments includedisability, dependency, falls, communicationdysfunction, and depression (1). Effective treatmentsexist for many types of visual and hearingimpairments.

In 1995, 18 percent of noninstitutionalized perso70 years of age and over were visually impaired,defined as full or partial blindness or other troubleseeing. The prevalence of visual impairment increaswith age from 13 percent of persons 70–74 years ofage to 31 percent of persons 85 years of age and o

Cataracts, glaucoma, and macular degenerationprimary causes of visual impairment among the oldepopulation. Just over 25 percent of noninstitutionalizpersons 70 years of age and over reported havingcataracts, and 8 percent reported having glaucoma.

Women were 1.2 times as likely to be visuallyimpaired as men, due in part to higher rates ofcataracts and glaucoma. The prevalence of visualimpairment among persons 70 years of age and ovewas similar for white persons and black persons.

In 1995 one-third of persons 70 years of age anover were hearing impaired. As with visualimpairment, the prevalence of hearing impairmentincreased with age from one-fourth of persons 70–7years of age to one-half of persons 85 years of ageover. Men were 1.5 times as likely to report hearingimpairment as women, although the sex differencedecreased with age. The sex difference in hearing

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

alimpairment may be attributed in part to differences incumulative occupational exposure to noise (2).

Older white persons were 1.8 times as likely to bhearing impaired as older black persons. This findingis consistent with studies that suggest black personsmay be less susceptible to noise-induced hearing lothan white persons (3,4).

References1. Lichtenstein MJ. Hearing and visual impairments. Clin Geriatr Med 8:173–82. 1992.

2. Wallhagen MI, Strawbridge WJ, Cohen RD, Kaplan GA. An increasing prevalence of hearing impairment and associated risk factors over three decades of the Alameda Countystudy. AJPH 87(3):440-2. 1997.

3. Jerger J, Jerger S, Pepe P, Miller R. Race differences in susceptibility to noise-induced hearing loss. Am J Otol 7:425–9. 1986.

4. Henselman LW, Henderson D, Shadoan J, et al. Effects of noise exposure, race, and years of service on hearing in U.S. Army soldiers. Ear Hear 16:382–91. 1995.

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

Figure 12. Prevalence of visual impairment among persons 70 years of age and over by age, sex, and race:United States, 1995

Percent

Women MenWhite Black White Black

0

10

20

30

40

50

60

70

70-74 years 75-79 years 80-84 years 85 years and over

Percent

Women MenWhite Black White Black

0

10

20

30

40

50

60

70

70-74 years 75-79 years 80-84 years 85 years and over

................................................................................................................................... 43

............................................................................................... Health and Aging

NOTES: Based on interviews conducted between October 1994 and March 1996 with noninstitutionalized persons. See Technical Notes for definitions ofvisual impairment and hearing impairment.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, 1994 National Health Interview Survey, Second Supplementon Aging.

Figure 13. Prevalence of hearing impairment among persons 70 years of age and over by age, sex, and race:United States, 1995

Health

Status

44 .........


ealth

Stat

us

Osteoporosis

Reduced bone density of the hip is a strongpredictor of subsequent fractures, particularly hipfractures (1). The physical limitations resulting fromosteoporosis and fractures contribute to disabilityamong the older population. Hip fractures alone wereresponsible for over 300,000 hospitalizations amongpersons 65 years of age and over in 1996, of which80 percent were women (2).

Loss of bone density (osteopenia and osteoporosis)is common among older persons. In 1988–94 just overone-half of noninstitutionalized persons 65 years ofage and over had reduced hip bone density. Theproportion with osteoporosis, a more severe form ofbone loss than osteopenia, was higher among womenthan men and rose with age for both women and men.Among persons 85 years of age and over, 90 percentof women and 54 percent of men had measurablereduced hip bone density.

The prevalence of osteoporosis in the hip increaseswith age for both women and men. The percent withosteoporosis increases more steeply with age for menthan women, although the overall percent is lower.Women 85 years of age and over were 2.7 times aslikely to have osteoporosis as women 65–74 years ofage. Men 85 years of age and over were 6.9 times aslikely to have osteoporosis as men 65–74 years of age.

Among women 65 years of age and over, theprevalence of osteoporosis was twice as high amongnon-Hispanic white persons compared withnon-Hispanic black persons.

References1. Cummings SR, Black DM, Nevitt MC, et al. Bone density at various sites forprediction of hip fractures. Lancet 341:72–5. 1993.

2. Centers for Disease Control and Prevention, National Center for Health Statistics,National Hospital Discharge Survey. 1996.

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

Figure 14. Prevalence of reduced hip bone density among persons 65 years of age and over by age, sex, andseverity: United States, 1988–94

Percent

65-74 years 65-74 years75-84 years 75-84 years85 years and over 85 years and over

Women Men

0

20

40

60

80

100

Osteoporosis: More severely reduced bone density

Osteopenia: Reduced bone density

................................................................................................................................... 45

............................................................................................... Health and Aging

NOTES: Figures are based on the noninstitutionalized population. Osteopenia is defined as bone mineral density 1–2.5 standard deviations below themean of non-Hispanic white women 20–29 years of age as measured in NHANES III; osteoporosis is defined as bone mineral density more than 2.5standard deviations below the mean of non-Hispanic white women 20–29 years of age as measured in NHANES III. See Technical Notes for furtherdiscussion.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, Third National Health and Nutrition Examination Survey.

Health

Status

if

s

g

irs

at

renss

t

enel

:

x

gand

46 .........


ealth

Stat

us

Physical Functioning and Disability

Quality of life in later years may be diminishedillness, chronic conditions, or injuries limit the abilityto care for oneself without assistance. Older personmaintain their independence and eliminate costlycaregiving services by, among other things, shoppinon their own, cooking their meals, bathing anddressing themselves, and walking and climbing stawithout assistance.

In 1995 among noninstitutionalized persons 70years of age and over, 32 percent had difficultyperforming and 25 percent were unable to performleast one of nine physical activities. Activitylimitations increased with age, and women were molikely than men to have a physical limitation. Perso85 years of age and over were 2.6 times as likely apersons 70–74 years of age to be unable to performphysical activities. Approximately 18 percent ofwomen and 12 percent of men 70 years of age andover were unable to walk a quarter of a mile withouassistance. Similarly for other important physicalactivities, older women were more likely than oldermen to be unable to climb a flight of steps (11 perccompared with 6 percent), or stoop, crouch, or kne(15 percent compared with 8 percent).

An indication of functional well-being is theability to perform certain tasks of daily living.Researchers group these tasks into two categoriesessential activities of daily living (ADL), such asbathing, eating, and dressing; and the more compleinstrumental activities of daily living (IADL), such asmaking meals, shopping, or cleaning. In 1995 amonthe noninstitutionalized population 70 years of age

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

t

over, 20 percent had difficulty performing at least oneADL, and 10 percent had difficulty performing at leastone IADL. Approximately 10 percent of women and 7percent of men were unable to do one or more ADL’s,and about 23 percent of women and 13 percent of mencould not do IADL’s without help. Women were morelikely than men to be disabled, and older persons hadhigher levels of disability than younger persons.

There are differences in physical functioning anddisability by race among the older population. Blackpersons reported higher levels of disability than whitepersons. In 1995 among noninstitutionalized persons70 years of age and over, black persons were 1.3 timesas likely as white persons to be unable to do certainphysical activities and 1.5 times as likely as whitepersons to be unable to perform one or more ADL’s.

Between the mid-1980’s and mid-1990’s theproportion of noninstitutionalized older women andmen who were unable to do one or more physicalactivities and unable to perform one or moreinstrumental activities of daily living declined.Moreover, disability appears to be declining moreamong women than men. This trend may provideimportant evidence of healthy aging, a result alsosupported by several other studies (1–3). However, theproportion of older persons unable to perform activitiesof daily living appears to have increased between 1984and 1995, although the level remains quite low (4).

References1. Crimmins E, Saito Y, Reynolds S. Further evidence on recent trends in the prevalence and incidence of disability among older Americans from two sources: The LSOA and theNHIS. J Gerontol 52B(2):S59–71. 1997.

2. Manton K, Corder L, Stallard E. Chronic disability trends in elderly United States populations: 1982–1994. Proceedings of the National Academy of Sciences: Medical Sciences,USA 94:2593–8. 1997.

3. Freedman V, Martin L. Changing patterns of functional limitation among the older American population. AJPH 88:1457-62. 1998.

4. Lentzner HR, Weeks JD, Feldman JJ. Changes in disability in the elderly population: Preliminary results from the Second Supplement on Aging. Paper presented at the annualmeetings of the Population Association of America. Chicago: April 1998.

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

Figure 15. Percent of persons 70 years of age and over who have difficulty performing 1 or more physicalactivities, activities of daily living, and instrumental activities of daily living by age and sex: United States, 1995

Percent

and sex: United States, 1995and sex: United States, 1995

Physical Activity IADLADL Physical Activity IADLADL0

20

40

60

80

100

Perform with difficulty

Unable to perform

Women Men

70-74 years 75-79 years 80-84 years 85 yearsand over

................................................................................................................................... 47

............................................................................................... Health and Aging

NOTES: Based on interviews conducted between October 1994 and March 1996 with noninstitutionalized persons. See Technical Notes for definitions ofphysical activities, activities of daily living (ADL) and instrumental activities of daily living (IADL).


Health

Status

48 .........


ealth

Stat

us

Conditions Associated With Disability

Disability can reduce an older person’sindependence and quality of life and may lead to theneed for formal or informal caregiving services.Chronic conditions are leading causes of disabilityamong the elderly and result in many older personsbeing limited in their daily activities of life.

Arthritis is the most commonly reported chroniccondition among older persons and the leading causeof disability. In 1995 among noninstitutionalizedpersons 70 years of age and over, 11 percentmentioned arthritis as one of the causes of theirdifficulty in performing activities of daily living (ADL)such as bathing, eating, dressing, and getting aroundthe house. Women were more likely than men to reportarthritis. Four percent of older persons listed heartdisease as one of the conditions leading to theirlimitation, and approximately 2.5 percent listed strokeand respiratory diseases.

Nonspecific conditions or procedures aresometimes reported as disablers. Nearly 2 percent ofolder persons mentioned ‘‘old age’’ as a cause of theirdisability. Surgery was also reported by almost2 percent as a cause of ADL limitations.

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

Figure 16. Percent of persons 70 years of age and over who report specific conditions as a cause of limitation inactivities of daily living: United States, 1995

Percent

Arthritis Heart disease Stroke Respiratory diseases Diabetes

0

2

4

6

8

10

12

................................................................................................................................... 49

............................................................................................... Health and Aging

NOTES: Based on interviews conducted between October 1994 and March 1996 with noninstitutionalized persons. Conditions are reported by personswho had any difficulty performing one or more activities of daily living (ADL). Multiple conditions may be reported. See Technical Notes for definitions ofrespiratory diseases and ADL.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, 1994 National Health Interview Survey, Second Supplementon Aging. See related figure 11 on chronic conditions and figure 15 on physical functioning and disability.

Health

Status

nu

ei

e

t

l to

el

.

y

50 ..........


ealth

Stat

us

Overweight

There is no consensus regarding optimal weightfor the older population. Some studies show little orassociation between high body mass index (a measof weight for height) and mortality (1,2), and someevidence indicates that higher weight may conferbeneficial effects such as providing nutritional reservin case of trauma and protection against osteoporos(2). Conversely, other studies show higher mortalityamong older overweight persons (3–5), and evidencindicates that obesity even among the older adultpopulation is associated with higher cardiovasculardisease risk factors (6). In addition, it is unknownwhether intentional weight loss among overweightolder adults is advisable (7). Currently it is believedthat a lean body weight throughout life is optimal, buthat stability in weight after age 50 is recommendedinstead of weight gain or loss (6).

In 1988–94 among noninstitutionalized persons65–74 years of age, 60 percent of women and68 percent of men were considered overweight orobese, with a body mass index greater than or equa25. Twice as many women as men in this age groupwere severely obese.

In 1988–94 the prevalence of obesity amongnon-Hispanic black women 65–74 years of age was63 percent higher than among non-Hispanic whitewomen. The level of obesity among Mexican-American women was 28 percent higher than the levamong non-Hispanic white women. Among men theprevalence of obesity was similar in the three groups

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

ore

ss

Among women 65–74 years of age, the overallpercent overweight has remained fairly constant overtime. However, the distribution of weight has shifted tohigher levels and increased especially in the severelyobese category. Between 1960–62 and 1988–94, thepercent of women who were considered obeseincreased by 16 percent. Among men, however, theincrease in overweight was more substantial. Theprevalence of overweight among men rose 43 percent,and the prevalence of obesity among men increased b131 percent.

References1. Stevens J, Cai J, Pamuk ER, et al. The effect of age on the association between body-mass index and mortality. N Engl J Med 338:1–7. 1998.

2. Diehr P, Bild DE, Harris TB, et al. Body mass index and mortality in nonsmoking older adults: The Cardiovascular Health Study. Am J Public Health 88(4):623–9. 1998.

3. Rumpel C, Harris TB, Madans J. Modification of the relationship between the Quetelet Index and mortality by weight-loss history among older women. Ann Epidemiol3(4):343–50. 1993.

4. Harris TB, Ballard-Barbasch R, Madans J, et al. Overweight, weight loss and risk of coronary heart disease in older women. The NHANES I Epidemiologic Follow-up Study. AmJ Epidemiol 137:1318–27. 1993.

5. Harris T, Cook EF, Garrison R, et al. Body mass index and mortality among nonsmoking older persons. The Framingham Heart Study. J Am Med Assoc 259(10):1520–4. 1988.

6. Harris TB, Savage PJ, Tell GS, et al. Carrying the burden of cardiovascular risk in old age: Associations of weight and weight change with prevalent cardiovascular disease, riskfactors, and health status in the Cardiovascular Health Study. Am J Clin Nutr 66:837–44. 1997.

7. Lee I-M, Paffenbarger RS Jr. Is weight loss hazardous? Nutr Rev 54(4 Pt 2):S116–24. 1996.

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

Figure 17. Distribution of weight among persons 65–74 years of age by sex: United States, 1988–94

Women Men0

10

20

30

40

50Percent

Low Healthy Overweight Moderately obese Severely obese

1960-62 1971-74 1976-80 1988-940

5

10

15

20

25

30Percent

Women

Men

................................................................................................................................... 51

............................................................................................... Health and Aging

NOTES: Figures are based on the noninstitutionalized population. See Technical Notes for definitions of weight categories. Obesity is defined as a bodymass index greater than or equal to 30 kilograms per meter squared.

SOURCES: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Examination Survey (1960–62), FirstNational Health and Nutrition Examination Survey (1971–74), Second National Health and Nutrition Examination Survey (1976–80), and Third NationalHealth and Nutrition Examination Survey (1988–94). See related Health, United States, 1999, table 70.

Figure 18. Prevalence of obesity among persons 65–74 years of age by sex: United States, 1960–94

Health

Status

tof

seyare

cls

al

ill

52 ........


ealth

Stat

us

Oral Health

Oral health is an important and often overlookedcomponent of an older person’s overall health andwell-being. Oral health problems may hinder aperson’s ability to be free of pain and discomfort, tomaintain proper nutrition, and to enjoy interpersonalrelationships and a positive self-image.

In 1988–94 nearly one-third of persons 65 yearsage and over with natural teeth had untreated dentalcaries in the crown or the root of their teeth. A higherpercent of older men than older women had at leastone untreated dental caries (35 percent compared wi27 percent). Dental caries is one of the main causestooth loss among the older population (see figure 19).

In 1993, 30 percent of noninstitutionalized person65 years of age and over were edentulous, that is, thhad no natural teeth. Levels of edentulism were similamong older women and men in each age group. Thprevalence of total tooth loss was higher amongnon-Hispanic black persons than among non-Hispaniwhite persons and Hispanic persons. In addition, leveof edentulism in the older population were higheramong those with lower socioeconomic status.

The rates of edentulism have been declining. Tottooth loss among persons 65 years of age and overdecreased by 23 percent from 1983 to 1993.Edentulism declined for all racial and socioeconomicgroups. However, rates of total tooth loss still exceedthe Healthy People 2000target that no more than 20percent of the population 65 years of age and over wbe edentulous.

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

of

h

As the proportion of older persons keeping theirteeth their entire life increases, so has the utilization ofdental care. In 1983, 39 percent of persons 65 years ofage and over reported at least one dental visit in theprevious 12 months, while in 1993 the proportion roseto 52 percent. This proportion is still below theHealthy People 2000target of 60 percent. Thisincrease in dental visits occurred even though mostpersons 65 years of age and over do not have dentalinsurance (1) and thus most of their dentalexpenditures are paid out-of-pocket (2).

References1. Manski, RJ. Dental care coverage among older Americans. J Am Coll Dent 62(3):41–44. 1995.

2. Moeller J, Levy H. Dental services: A comparison of use, expenditures, and sources of payment, 1977 and 1987. Research Findings 26. AHCPR Pub. No. 96–0005. 1996.

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

Figure 20. Prevalence of total tooth loss (edentulism)among persons 65 years of age and over by age:United States, 1983 and 1993

Percent

0

10

20

30

40

50

60

70

1983 1993


Women Men

0

10

20

30

40

50

60

70

Percent


................................................................................................................................... 53

............................................................................................... Health and Aging

NOTE: Figures are based on the noninstitutionalized population.

SOURCE: Centers for Disease Control and Prevention, National Centerfor Health Statistics, National Health Interview Survey.

Figure 19. Percent with untreated dental caries amongdentate persons 65 years of age and over by age andsex: United States, 1988–94

NOTES: Dental caries includes coronal and root caries. Dentate personshave at least one natural tooth. Figures are based on thenoninstitutionalized population.

SOURCE: Centers for Disease Control and Prevention, National Centerfor Health Statistics, Third National Health and Nutrition ExaminationSurvey.

Health

Status

n

s

ca

0lre

f ag

ot

r.

ocial

70atfident

t)

le

rly

54 ........


ealth

Stat

us

Social Activities

Social activity in old age has a positive effect ohealth. Interaction with friends and family membersoffers emotional and practical support that increasethe ability of an older person to remain in thecommunity and decreases the use of formal healthservices.

In 1995 nearly all noninstitutionalized persons 7years of age and over reported some form of sociaactivity in a 2-week period. Older persons were molikely to report no social activities compared withyounger persons. Even among persons 85 years oand over, however, only 4 percent reported engaginnone of seven common social activities.

Older persons engage in fewer types of socialactivities as they age. Among persons 70–74 yearsage, 64 percent participated in five to seven differensocial activities in a 2-week period compared with38 percent among persons 85 years of age and ove

Among persons who engaged in at least one sactivity in a 2-week period, contact with family wasthe most common. Eighty-seven percent of personsyears of age and over had talked on the telephoneleast once with family members who lived outside otheir household, and 76 percent had seen noncoresrelatives. Social contact with friends and neighborswas also prevalent: 72 percent got together withfriends or neighbors, and 81 percent talked on thetelephone with a friend or neighbor. Other commonactivities included eating at a restaurant (65 percenand attending a religious service (51 percent).

Overall levels of social activity were similaramong women and men. Women were more likelythan men to talk on the telephone with friends andneighbors, but women and men reported comparablevels of getting together socially.

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

re

gein

f

Disability limits social interaction. Persons whohad difficulty with at least one of seven activities ofdaily living (ADL), such as eating, dressing, orbathing, were less likely to participate in socialactivities than persons who were not limited in theirbasic daily activities. Contact with family members inperson or by phone was only slightly less commonamong persons with ADL limitations than amongpersons with no ADL limitations. However,participation in activities outside the house was muchless common among persons with ADL limitations. Foexample, nondisabled persons were 1.7 times as liketo have attended a religious service and 1.5 times aslikely to have eaten in a restaurant in the previous 2weeks as disabled persons.

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

Figure 21. Number of social activities in a 2-week period among persons 70 years of age and over by age andsex: United States, 1995

Percent

MenWomen

75-79years

75-79years

85 yearsand over

85 yearsand over

70-74years

70-74years

80-84years

80-84years

0

20

40

60

80

100

0 activities 1-2 activities 3-4 activities 5-7 activities

................................................................................................................................... 55

............................................................................................... Health and Aging

NOTES: Based on interviews conducted between October 1994 and March 1996 with noninstitutionalized persons. See Technical Notes for definitions ofsocial activities.


Health

Status

s

2in

ith

e

nd

l

ef

t

to

56 .........


ealth

Stat

us

Exercise

Research has shown that physical activity hasmany benefits for health. It can reduce the risk ofcertain chronic diseases, appears to relieve symptomof depression, helps to maintain independent living,and enhances overall quality of life (1). TheHealthyPeople 2000target is to increase levels of activityamong the elderly population so that no more than 2percent of persons 65 years of age and over engageno leisure-time physical activity.

In 1995, 71 percent of nondisabled persons 65years of age and over exercised at least once in arecent 2-week period. The proportion of persons whoexercised declined with age. Close to 75 percent ofpersons 65–74 years of age were active compared w60 percent of persons 85 years of age and over. Men65 years of age and over were more likely to exercisthan women.

Less strenuous forms of exercise were mostprevalent. Sixty-five percent of those who were notsedentary walked for exercise. Other common light amoderate activities included gardening (54 percentamong men and 38 percent among women) andstretching (26 percent among men and 32 percentamong women). Activities such as stair climbing,swimming, aerobics, and cycling were less frequentlyundertaken.

Regular exercise is important to obtain substantiahealth benefits. The recommended level for light tomoderate physical activity is 30 minutes each time onmost days of the week (1). When all forms of exercisare added together, two-thirds of persons 65 years oage and over who exercised (and who were able toestimate frequency) did not achieve recommendedlevels. For example, only about 18 percent of oldermen and women who walked for exercise did so for a

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

least 30 minutes 10 times in a 2-week period. Seven10 percent of those who gardened did so regularly.Even lower levels of activity were recorded for morechallenging forms of exercise such as jogging andweight lifting.

Reference1. U.S. Department of Health and Human Services. Physical activity and health: A report of the surgeon general. Atlanta, GA: U.S. Department of Health and Human Services,Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. 1996.

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

Figure 22. Percent who exercise and selected type of exercise among persons 65 years of age and over by sex:United States, 1995

0

0

10

10

20

20

30

30

40

40

50

50

60

60

70

70

80

80

Percent of nondisabled persons who exercise

Percent of nondisabled persons

Anyexercise

Walking

Gardening

Stretching

Swimming

Aerobics

Stairclimbing

*

Women

Men

................................................................................................................................... 57

............................................................................................... Health and Aging

* The number of men 65 years of age and over who participated in aerobic exercise was too small to calculate reliable rates.

NOTES: Figures are based on the noninstitutionalized population. Percents are age adjusted. Exercise is defined as doing at least 1 of 20 exercises,sports, or physically active hobbies at least once within a 2-week period. The percent engaging in a specific activity is calculated among persons whoengage in any exercise. See Technical Notes for list of activities.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Interview Survey.

Health

Status

er

dd

f

lnd

gver

nt

tmef

58 ........


ealth

Car

eAc

cess

and

Util

izat

ion

Caregivers

Older persons often receive help from others toperform routine activities. This help can allow an oldperson to remain in his or her home and avoidinstitutionalization. Caregivers who provide help maybe family members, friends, or paid employees.Caregivers provide assistance with a variety ofactivities including basic needs such as dressing anbathing, known as activities of daily living (ADL), another chores such as shopping, housework, andmanaging money, known as instrumental activities odaily living (IADL).

In 1995 among noninstitutionalized persons 70years of age and over, 34 percent received help orsupervision with at least one ADL or IADL. Over 12million caregivers were providing formal and informacare. Seventy percent of caregivers were women, a73 percent were unpaid or informal helpers. Thepercent of paid caregivers increased with age: amonthose receiving help, persons 85 years of age and owere 1.4 times as likely to have paid caregivers aspersons 70–74 years of age.

The majority of persons receiving help(56 percent) received it from a single caregiver. Thenumber of caregivers rose with age: among persons70–74 years of age who received help, 63 percentreceived it from one caregiver. Among persons 85years of age and over receiving help, only 44 percehad one caregiver. At each age, women had morecaregivers than men.

Among informal or unpaid caregivers, 91 percenwere family members and 51 percent lived in the sahousehold as the recipient of the help. One-fourth othe caregivers were spouses and slightly more thanone-half were children. Women were less likely thanmen to receive care from their spouses, due in large

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

part to the higher proportion of widowhood amongwomen than men. Informal caregivers provided helpon an average of 7 days in a 2-week period.

Older persons were more likely to receive helpfrom caregivers for IADL’s than for ADL’s. One-thirdof persons who received any assistance, had help withbathing or showering and nearly one-fourth had helpwith walking. However, 58 percent had another personhelp with shopping, and 56 percent received assistanceto get to places outside of walking distance. Eight outof ten older persons who received any help had acaregiver help with heavy housework.

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

Figure 23. Number of caregivers providing assistance with activities of daily living or instrumental activities ofdaily living to persons 70 years of age and over by age and sex: United States, 1995

Percent

MenWomen

75-79years

75-79years

85 yearsand over

85 yearsand over

70-74years

70-74years

80-84years

80-84years

0

10

20

30

40

50

60

70

1 caregiver

2 caregivers

3-4 caregivers

................................................................................................................................... 59

............................................................................................... Health and Aging

NOTES: Based on interviews conducted between October 1994 and March 1996 with noninstitutionalized persons. Caregivers provide help or supervisionwith at least one activity of daily living (ADL) or instrumental activity of daily living (IADL). See Technical Notes for definitions of ADL and IADL.


Health

Care

Accessand

Utilization

60 ...........


ealth

Car

eAc

cess

and

Util

izat

ion

Unmet Needs

Frequently, chronic diseases and impairmentsimpose limitations on what older persons can do forthemselves. Without personal assistance or the use ofdevices such as walkers, hand rails, or specialbreathing equipment, many older persons have unmetneeds that reduce quality of life and increase the riskof institutionalization (1).

In 1995, 14 percent of noninstitutionalized persons70 years of age and over had difficulty and neededhelp with one or more activities of daily living (ADL),such as bathing, dressing, and moving about the house.Twenty-six percent had difficulty and needed help forhousehold activities such as shopping, cleaning, andmeal preparation known as instrumental activities ofdaily living (IADL).

The majority of older persons in need receivedenough personal assistance to carry out these importanttasks. However, approximately 44 percent (1.4 million)of those who had difficulty and needed help had‘‘unmet needs,’’ that is they either had no assistance atall or required additional assistance. Roughly one-fifthhad unmet ADL needs and another one-fourth did notneed assistance with ADL’s, but had unmet need forassistance with IADL’s. In most cases those withunmet needs required direct hands-on help. Lackingthe necessary assistance with ADL’s, approximatelyone-half of those in need experienced a seriousnegative consequence such as burns from bath water,weight loss, or being chair- or bed-bound.

Men and women were equally likely to haveunmet ADL needs; however, women and youngerpersons were more likely than men and older personsto say they went without needed assistance withIADL’s.

Reference1. Allen SM, Mor V. The prevalence and consequences of unmet need: Contrastsbetween older and younger adults with disability. Med Care. 35(11):1132–48. 1997.

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

Figure 24. Percent with unmet needs among persons 70 years of age and over who need help with 1 or moreactivities of daily living or instrumental activities of daily living by age and sex: United States, 1995

70-84 years 70-84 years85 years and over 85 years and over

Women Men

0

5

10

15

20

25

30

35

Percent

Unmet need for ADL's Unmet need for IADL's only

*

................................................................................................................................... 61

............................................................................................... Health and Aging

* The number of men 85 years of age and over with unmet needs was too small to calculate reliable rates.

NOTES: Based on interviews conducted between October 1994 and March 1996 with noninstitutionalized persons. See Technical Notes for definitions ofactivities of daily living (ADL) and instrumental activities of daily living (IADL). Persons with unmet ADL needs may also have unmet IADL needs.


Health

Care

Accessand

Utilization

rs

62 .........


ealth

Car

eAc

cess

and

Util

izat

ion

Assistive Devices

Assistive devices are important tools in managinghealth and prolonging independent living. Devices suchas hearing aids, diabetic and respiratory equipment,and mobility aids such as canes and walkers can helpolder persons to remain in the community and toprevent further progression of a chronic disease orcondition.

Among noninstitutionalized persons 70 years ofage and over in 1995, 39 percent used an assistivedevice during the previous 12 months. The proportionof older persons using devices increased with age.Persons 85 years of age and over were twice as likelyto rely on assistive devices as persons 70–74 years ofage. In addition, the number of devices increased withage. Persons 85 years of age and over were twice aslikely to use three or more devices as persons 70–74years of age. The age pattern was similar for womenand men.

Mobility aids were the most common type ofassistive device. Seventeen percent of persons 70 yeaof age and over used a cane, and 10 percent used awalker. Other common devices include hearing aids(11 percent), respiratory equipment (8 percent), anddiabetic equipment (7 percent). The prevalence ofmobility aids and hearing aids increased sharply withage. Persons 85 years of age and over were five timesas likely to use a walker and over three times as likelyto use hearing aids as persons 70–74 years of age.

Disability increases reliance on assistive devices.Older persons who were limited in their activities ofdaily living (ADL) were 2.8 times as likely to useassistive devices as persons without limitations. In1995 nearly three-fourths of noninstitutionalizedpersons 70 years of age and over who had difficultywith these activities used at least one assistive devicein the previous 12 months. Four out of ten personswith at least one ADL limitation reported using a caneand three out of ten used a walker.

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

Figure 25. Assistive devices used among persons 70 years of age and over by age and sex: United States, 1995

0

10

20

30

40

50

60

70

Percent

Women Men

70-74years

70-74years

75-79years

75-79years

80-84years

80-84years

85 yearsand over

85 yearsand over

1 device

2 devices

3 or more devices

................................................................................................................................... 63

............................................................................................... Health and Aging

NOTES: Based on interviews conducted between October 1994 and March 1996 with noninstitutionalized persons. See Technical Notes for definitions ofassistive devices.


Health

Care

Accessand

Utilization

64 .........


ealth

Car

eAc

cess

and

Util

izat

ion

Physician Contacts

Older persons have more contacts with medicalproviders on average than do younger adults. In1994–96 persons 65 years of age and over had anaverage of 11.4 contacts per year with a physician orother personnel working under a physician’ssupervision for examination, diagnosis, treatment, oradvice. Adults 45–64 years of age averaged 7.2contacts per year. These contacts were by phone or atdoctors’ offices, hospital outpatient clinics andemergency rooms, clinics, home, or other places.

The number of contacts rose with age, from amean of 10 contacts per year among persons 65–74years of age to nearly 15 contacts per year amongpersons 85 years of age and over. Women on averagehad more contacts than men. Since 1990 the meannumber of physician contacts per year among olderpersons has increased by approximately two visits.

Older persons in fair or poor health had more thantwice the number of physician contacts per year aspersons in good to excellent health. Among persons 85years of age and over who reported their health as fairor poor, women had nearly 27 contacts with physiciansper year compared with 20 contacts among men.

One-half of physician contacts among persons 65years of age and over occurred in doctors’ offices;however, this percent declined with age. Theproportion of outpatient medical contacts among olderpersons that occurred in the home increased sharplywith age from 10 percent among persons 65–74 yearsof age to 38 percent among persons 85 years of ageand over. The percent of contacts occurring in thehome was higher among women than men. From 1990to 1996 the proportion of contacts in the homeincreased by 63 percent among persons 65 years of ageand over. This trend probably reflects the increased useof home health care services among the elderly.

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

Figure 26. Place of ambulatory physician contacts among persons 65 years of age and over by age and sex:United States, 1994–96

0 20 40 60 80 100

Percent of contacts

Doctor's office Hospital Home Phone Other

65-74years

65-74years

75-84years

75-84years

85 yearsand over

85 yearsand over

Men

Women

................................................................................................................................... 65

............................................................................................... Health and Aging

NOTES: Figures are based on the noninstitutionalized population. Physician contacts include contact with other medical personnel working under aphysician’s supervision and do not include contacts during overnight hospital stays. Persons with unknown place of contact are excluded fromcalculations. See Appendix II for definitions.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Interview Survey. See related Health,United States, 1999, tables 75–77.

Health

Care

Accessand

Utilization

e

jort

forr

5

ly5ly

ateen

66 .........


ealth

Car

eAc

cess

and

Util

izat

ion

Inpatient Health Care

Persons 65 years of age and over are majorconsumers of inpatient care. Although they representonly 13 percent of the total population in 1996, theyaccounted for 38 percent of the roughly 31 millionpatient discharges from non-Federal short-stayhospitals. Moreover, the average length of stay forolder persons exceeds that for younger adults.

Hospitalizations increased with age. Comparedwith those 65–74 years of age, persons 85 years of aand over had more than twice the rate of hospitaldischarge. The overall discharge rate for all diagnosecombined was higher for men than for women withineach of the three age groups.

Heart disease was the most common cause forhospitalization as determined by the first-listeddischarge diagnosis. The rate of hospitalization forheart disease increased substantially with age, andwithin each age group men had a higher rate thanwomen. Patient discharges from stroke, the other macirculatory disease, also increased with age. Except athe oldest ages where the rate was nearly the samemen and women, discharges from stroke were highefor men than women. Combined heart disease andstroke accounted for more than one-fourth of allhospital discharges among elderly men and women 8years of age and over.

Malignant neoplasms accounted for approximate6 percent of all hospital discharges among persons 6years of age and over, and the rate remained relativestable across the age groups.

Pneumonia and bronchitis combined wereresponsible for slightly less than 10 percent of allhospitalizations for elderly men and women. Theserates increased rapidly with age, and the combined rfor these two diseases was about twice as high for mas women among persons 85 years of age and over.

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

d

ge

s

Hospitalizations for fractures of all types weremore common among women than men within all agegroups. At the oldest ages, fractures accounted fornearly 10 percent of all discharges among women, thesecond most important cause of hospitalization amongthe listed diagnoses.

Hospital stays for persons 65 years of age andover were shorter in 1996 than a decade earlier. Theaverage length of stay of 6.5 days was 2 days less thanin 1986. Similar trends were evident for younger(65–74 years of age) and older (75 years of age andover) persons.

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

Figure 27. Hospital discharge rates in non-Federal short-stay hospitals for selected first-listed diagnoses amongpersons 65 years of age and over by age and sex: United States, 1996

Heartdisease

Heartdisease

Stroke

Stroke

Malignantneoplasms

Malignantneoplasms

Pneumonia

Pneumonia

Bronchitis

Bronchitis

Fractures

Fractures

0 25 50 75 100 125 150Discharges per 1,000 population

Women

Men

65-74 years

75-84 years

85 years and over

65-74 years

75-84 years

85 years and over

................................................................................................................................... 67

............................................................................................... Health and Aging

NOTES: For a description of the International Classification of Diseases code numbers for diagnoses, see Appendix II. Rates are based on the civilianpopulation as of July 1, 1996.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Hospital Discharge Survey. See related Health,United States, 1999, tables 90–93.

Health

Care

Accessand

Utilization

f

f

y

ar

ade.onsto

r.

r

t

68 ..........


ealth

Car

eAc

cess

and

Util

izat

ion

Influenza and PneumococcalVaccinations

An annual influenza vaccination is recommendedfor all persons 65 years of age and over; it offerssubstantial protection against complications from theinfluenza virus. During 1993–95 an average of55 percent of noninstitutionalized persons 65 years oage and over reported receiving a flu shot within theprevious 12 months. Older women and men hadsimilar levels of vaccination. Persons 75–84 years oage had a slightly higher level of coverage thanpersons 65–74 years of age.

Vaccination coverage varies by race and ethnicitApproximately 57 percent of non-Hispanic whitepersons 65 years of age and over received an annuvaccination for influenza. This was significantly highethan the level of coverage for non-Hispanic black(36 percent) or Hispanic (44 percent) older persons.

Vaccine coverage has increased in the past decBetween 1989 and 1995 influenza coverage for pers65 years of age and over nearly doubled. However,meet national immunization targets of 60 percentestablished inHealthy People 2000, coverage needs toexpand particularly among older black and Hispanicpersons (1).

A single-dose pneumococcal vaccine isrecommended for all adults 65 years of age and oveOverall, in 1993–95 about 29 percent of the oldernoninstitutionalized population reported ever havingreceived a pneumonia vaccination. Although vaccinecoverage was approximately the same for older menand women, coverage was highest among persons75–84 years of age. Roughly 30 percent of oldernon-Hispanic white persons were vaccinated forpneumonia, but only an estimated 16 percent of oldenon-Hispanic black and Hispanic persons werevaccinated.

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

.

l

Pneumococcal vaccination coverage more thandoubled between 1989 and 1995. However, coverageamong older persons, who are at greatest risk foradverse effects, needs to be greatly increased to meetthe Healthy People 2000target of 60 percent (1).

Older institutionalized persons are at especiallyhigh risk of contracting and suffering adverseconsequences of influenza or pneumonia. Results fromthe 1995 National Nursing Home Survey indicated thaamong residents with known influenza vaccinationstatus (79 percent), 79 percent received an influenzavaccination in the past 12 months, and 42 percent(among the 57 percent with known vaccination status)received a pneumococcal vaccination (2).

References1. National Center for Health Statistics. Healthy People 2000 Review, 1997. Hyattsville, Maryland: Public Health Service. 1997.

2. Greby SM, Singleton JA, Sneller VP, et al. Influenza and pneumococcal vaccination coverage in nursing homes, United States, 1995. Atlanta, Georgia: National ImmunizationProgram, Centers for Disease Control and Prevention. 1998.

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

Figure 28. Percent vaccinated against influenza and pneumococcal disease among persons 65 years of age andover by race and Hispanic origin: United States, 1993–95

White,non-Hispanic

White,non-Hispanic

Black,non-Hispanic

Black,non-Hispanic

Hispanic

Hispanic

0 10 20 30 40 50 60

Percent

Influenza

Pneumococcaldisease

................................................................................................................................... 69

............................................................................................... Health and Aging

NOTES: Figures are based on the noninstitutionalized population. For influenza, the percent vaccinated consists of persons who reported having a flushot during the past 12 months. For pneumococcal disease, the percent is persons who reported ever having a pneumonia vaccination.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Interview Survey.

Health

Care

Accessand

Utilization

lsn

ofes

s.d

erebutree

ttht of

nd

rnting

ece

Women Men

0

20

40

60

80

100

120

140

Current patients per 1,000 population

65-74years

75-84years

85 yearsand over

70 .........

Health and Aging ...............................................................................................

NOTES: Age is defined as age at interview. See Technical Notes for detailson calculations. Rates are based on the civilian population as of July 1,1996. See Appendix II for definition of home health care.

SOURCE: Centers for Disease Control and Prevention, National Center forHealth Statistics, National Home and Hospice Care Survey.

Figure 29. Home health care patients among persons 65years of age and over by age and sex: United States,1996

Hea

lthC

are

Acce

ssan

dU

tiliz

atio

n

Home Health Care

Home health care is an important alternative totraditional institutional care. Services such as medicatreatment, physical therapy, and homemaker serviceoften allow patients to be cared for at lower cost thaa nursing home or hospital and in the familiarsurroundings of their home. In 1996 nearly one-halfall home health care expenditures in the United Statwere paid by Medicare, the Federal health insuranceprogram for the elderly and disabled (1).

On an average day in 1996, approximately 1.7million persons 65 years of age and over, roughly 51per 1,000 population, were home health care patientUsage was higher for older women than for men, anthe rate increased with age. Women 85 years of ageand over had the highest level of current utilization,(130 current patients per 1,000) followed by men inthis same age group.

Home health care providers offered a variety ofservices. Nursing care was the most widely usedservice among older home health care patients;85 percent were receiving this service in 1996.Twenty-nine percent of the current home health carepatients used homemaker services. Older women wslightly more likely than men to use these services,the level of use was relatively constant across the thage groups. Other services frequently used includedphysical therapy, social services, and help withmedications.

Diseases of the circulatory system were the moscommon conditions leading to the use of home healcare services among older persons. Fourteen percencurrent home health care patients 65 years of age aover had a primary admission diagnosis of heartdisease, and another 9 percent listed cerebrovasculadiseases. Respiratory diseases and diabetes, accoufor about 9 percent each, were also commonconditions. Fractures were the primary admissiondiagnosis for approximately 4 percent of elderly homhealth care patients, with women patients having twithe rate of fractures as men.

.

Reference1. Levit KR, Lazenby HC, Braden BR, et al. National Health Expenditures, 1996. Health Care Financing Rev 19(1):161–200. 1997.

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

Figure 30. Home health care services received by current patients 65 years of age and over: United States, 1996

Nursing

Homemaker

Physicaltherapy

Socialservices

Medications

Continuoushome care

Occupationaltherapy

Physicianservices

Nutrition

All otherservices

Percent

0 20 40 60 80 100

................................................................................................................................... 71

............................................................................................... Health and Aging

NOTE: Home health care patients may receive one or more services per visit.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Home and Hospice Care Survey.

Health

Care

Accessand

Utilization

n

y

all

rends

re

72 .........


ealth

Car

eAc

cess

and

Util

izat

ion

Health Insurance

Medicare, the Federal health insurance programfor the elderly and disabled, provides health carecoverage for over 96 percent of the elderly populatioIn 1996 more than 33 million older persons in theUnited States were covered by Medicare.

In 1994–96, 16 percent of noninstitutionalizedpersons 65 years of age and over relied solely onMedicare to cover inpatient care in hospitals and tohelp pay the cost of doctors’ visits and other healthcare. Most elderly persons supplement Medicarecoverage with private or other publicly funded healthinsurance to pay a portion of the costs not covered bMedicare. A small proportion of persons 65 years ofage and over reported that they had only privateinsurance (3 percent) or no health care coverage at(less than 1 percent).

The distribution of types of health insurancecoverage was similar among women and men butvaried greatly by age, race, and ethnicity. Theproportion of older persons relying solely on Medicaor on Medicare combined with Medicaid (the State aFederal programs that pay for health care for personin need) increased with age, while the proportion ofpersons who have private insurance along with theirMedicare coverage declined with age.

Compared with non-Hispanic white persons,non-Hispanic black persons and Hispanics were molikely to have Medicare only or Medicare andMedicaid as their health care coverage. Less thanone-half of non-Hispanic black persons and Hispanicpersons 65 years of age and over reported that theyhad private insurance to supplement their Medicarecoverage. Nearly one-half of noninstitutionalizedHispanic persons 85 years of age and over had

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

.

Medicare combined with Medicaid as their health carecoverage.

The type of insurance affects access to health care.Older persons who had Medicare coverage only orwho had no health care coverage were less likely tohave a regular source of medical care than personswith Medicare supplemented by private or publicinsurance. In addition, elderly persons with Medicareonly were more likely to delay care or to go withoutmedical care than persons who had Medicare andprivate insurance (1).

Reference1. Cohen RA, Bloom B, Simpson G, Parsons PE. Access to health care part 3: Older adults. National Center for Health Statistics. Vital Health Stat 10(198). 1997.

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

Figure 31. Health insurance coverage among persons 65 years of age and over by age, race, Hispanic origin, andtype of insurance: United States, 1994–96

White,non-Hispanic

White,non-Hispanic

Black,non-Hispanic

Black,non-Hispanic

Hispanic

Hispanic

0 20 40 60 80 100Percent

85 years of ageand over

65-84 years

*

*

*

Medicare only Medicare/Medicaid Medicare/private Private only None

................................................................................................................................... 73

............................................................................................... Health and Aging

* The number of persons 85 years of age and over with private health insurance only or with no health insurance was too small to calculate reliablerates.

NOTES: Figures are based on the noninstitutionalized population. Figures exclude persons with unknown health insurance coverage. The categoryMedicare/Medicaid can include other public health insurance programs. The category Medicare/private includes a small number of persons who reportedthat they had Medicaid in addition to Medicare and private health insurance.

SOURCE: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Interview Survey. See related Health, UnitedStates, 1999, table 130.

Health

Care

Accessand

Utilization

sr

l

eth

re

ageer.

n;s

ert

lsof

g.

74 ..........


ealth

Car

eAc

cess

and

Util

izat

ion

Medicare Health MaintenanceOrganization Enrollment

A growing number of Medicare beneficiariesreceive their health care through health maintenanceorganizations (HMO’s). In 1997 over 4 million person65 years of age and over who received Medicare weenrolled in a managed care plan compared withapproximately one million in 1985. In 1997 the overarate of enrollment in Medicare managed care planswas 12 percent among persons 65 years of age andover. Levels of enrollment in managed care plans arlower among Medicare beneficiaries compared withoverall population, but the rate of growth has beenfaster in the 1990’s.

Participants in Medicare HMO’s usually havelower out-of-pocket costs for services covered byMedicare and often receive additional benefits notcovered by traditional fee-for-service Medicare, suchas prescription drugs. In turn, Medicare HMOparticipants are subject to many of the samerestrictions of other managed care plans and usuallyreceive their health care through a specific clinic ornetwork of providers. Research has shown thatMedicare HMO enrollees are generally healthier,younger, and less likely to be institutionalized or toreceive Medicaid than Medicare beneficiaries who anot enrolled in managed care plans (1).

Enrollment in Medicare HMO’s decreased withage from 13 percent among persons 65–74 years ofto 9 percent among persons 85 years of age and ovOverall levels of enrollment among persons 65 yearsof age and over were similar among women and mehowever, at the oldest ages, women were slightly leslikely than men to be enrolled in HMO’s. In addition,levels of enrollment were similar among white andblack persons, with black persons having slightly lowrates of participation in Medicare HMO’s at the oldesages.

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

e

l

e

Participation in Medicare HMO’s is unevenlydistributed throughout the country. The highest leveof participation are in the West. In 1997, 38 percentMedicare beneficiaries in California were enrolled inHMO’s compared with 34 percent in Arizona,22 percent in Florida, 13 percent in New York, and2 percent in Indiana. The following 10 States had noMedicare managed care plans in 1997: Alaska,Delaware, Idaho, Maine, Mississippi, Montana, NewHampshire, South Dakota, Tennessee, and Wyomin

Reference1. Zarabozo C, Taylor C, Hicks J. Medicare managed care: Numbers and trends. Health Care Financing Rev 17(3):243–61. 1996.

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

Figure 32. Percent of Medicare enrollees in health maintenance organizations by State: United States, 1997

................................................................................................................................... 75

............................................................................................... Health and Aging

NOTES: Data as of January 1997. Figure includes Medicare beneficiaries less than 65 years of age. Persons 65 years of age and over are 87 percentof all Medicare enrollees and 95 percent of Medicare HMO enrollees.

SOURCE: Health Care Financing Administration/OMC/BDMS. See related Health, United States, 1999, table 143.

Health

Care

Accessand

Utilization

e

s

l

as

ltf

men

ted–84

e,entgeing

ears

t

or

n

n

fnat

76 .........


ealth

Car

eAc

cess

and

Util

izat

ion

Cost of Heart Disease

Expenditures on personal health care include thgoods and services used to treat disease, such ashospital care, physician’s services, prescription drughome health care, and nursing home care. In 1995personal health care expenditures for heart diseaseamong persons 65 years of age and over wereestimated to be more than 58 billion dollars. Hospitacare and nursing home care accounted for overthree-fourths of the total expenditures for heart diseamong older persons.

Total expenditures for heart disease-related heacare decreased with age among men. Expendituresmen 65–74 years of age were 1.3 times as large asexpenditures for men 75–84 years of age and 3.2 tias large as expenditures among men 85 years of agand over. However, total expenditures among womewere higher among the older ages compared withexpenditures for persons 65–74 years of age, eventhough there were fewer women at older ages. Thelargest expenditures on personal health care attributo heart disease for women were among persons 75years of age.

For men, 69 percent of all personal healthexpenditures for heart disease were for hospital carbut the proportion decreased with age, from 75 percat 65–74 years of age to 51 percent at 85 years of aand over. In contrast to other health services, spendfor nursing home care increased with age from3 percent of total expenditures among men 65–74 yof age to 34 percent of health care for heart diseaseamong men 85 years of age and over.

Age-related patterns of spending were somewhadifferent for women than men. More was spent forhospital care than for any other service at ages lessthan 85 years, but the proportion of total spending f

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

,

e

hor

es

hospital care was less for women than men. Spendingfor nursing home care was much higher among womethan men and accounted for 58 percent of total healthcare expenditures attributed to heart disease for wome85 years of age and over.

In contrast to total spending, per capitaexpenditures for heart disease in 1995 increased withage among both women and men and for every type ocare. The amount spent per person was higher for methan women at younger ages, but women spent more85 years of age and over. Spending by men rose from$1,520 per person at 65–74 years of age to $2,290 at75–84 years of age and $3,850 at 85 years of age andover. Per capita spending among women increasedfrom $790 at the youngest ages to $1,770 at 75–84years of age and $4,220 for persons 85 years of ageand over.

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

Figure 33. Estimated amount of personal health care expenditures attributed to heart disease among persons 65years of age and over by age, sex, and type of health service: United States, 1995

65-74years

65-74years

75-84years

75-84years

85 yearsand over

85 yearsand over

Women

Men

0 2 4 6 8 10 12 14

Amount in billions of dollars

Hospitalcare

Physician and otherprofessional services

Home healthcare

Prescription drugsand medical durables

Nursing homecare

................................................................................................................................... 77

............................................................................................... Health and Aging

NOTE: Cost estimates are calculated from first-listed diagnoses of heart disease only.

SOURCE: Hodgson TA, Cohen AJ. Medical care expenditures for selected circulatory diseases: Opportunities for reducing national health expenditures.Med Care. Forthcoming.

Health

Care

Accessand

Utilization

rens

,re

alas

er

to

onhr.

r

overenn

ly

are

andofen

78 .........


ealth

Car

eAc

cess

and

Util

izat

ion

Cost of Diabetes

Diabetes is a common disease among the olderpopulation, affecting approximately 10 percent ofpersons 65 years of age and over; however, it is moprevalent among women and minority groups. Persowith diabetes are at higher risk for other chronicconditions such as heart disease, visual impairmentsand kidney disease. In 1995 total personal health caexpenditures for diabetes, including chroniccomplications and comorbidities associated withdiabetes, among persons 65 years of age and over,were estimated to be 26 billion dollars. Nearly one-hof this amount was for hospital care, and one-fifth wfor nursing home care.

Total spending on diabetes health care was highfor women than men. Spending decreased with age,although the declines were steeper among men thanwomen. Among men, expenditures declined from 5.6billion dollars among persons 65–74 years of age, to3.5 billion dollars among those 75–84 years of age,1.5 billion dollars among those 85 years of age andover. Among women, total spending was only6 percent lower among persons 75–84 years of agecompared with persons 65–74 years of age (6.1 billidollars compared with 6.5 billion dollars) even thougthe population was approximately 35 percent smalle

Hospital care was the largest expenditure forpersonal health care attributed to diabetes amongpersons 65 years of age and over, but it was a largeproportion of expenditures for men (53 percent) thanfor women (41 percent). The proportion ofexpenditures on nursing home care was more thantwice as high among women than men for all olderpersons; however, for persons 85 years of age andthe differences narrowed, and expenditures for womwere 1.5 times as large as those for men. Prescriptiodrugs and other medical durables were approximate

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

f

6 percent of total expenditures on diabetes health camong the older population.

In contrast to total spending, per capitaexpenditures for diabetes health care increased withage. The amount spent was similar among womenmen 65–84 years of age. Among persons 85 yearsage and over, per capita spending was higher for m($1,430) than for women ($1,170).

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

Figure 34. Estimated amount of personal health care expenditures attributed to diabetes among persons 65 yearsof age and over by age, sex, and type of health service: United States, 1995

65-74years

65-74years

75-84years

75-84years

85 yearsand over

85 yearsand over

0 1 2 3 4 5 6 7

Amount in billions of dollars

Hospitalcare

Physician and otherprofessional services

Home healthcare

Prescription drugsand medical durables

Nursing homecare

Women

Men

................................................................................................................................... 79

............................................................................................... Health and Aging

NOTE: Cost estimates are calculated from first-listed diagnoses of diabetes, chronic complications and other unrelated diagnoses attributed to diabetes,and certain comorbidities among persons with diabetes.

SOURCE: Hodgson TA, Cohen AJ. Medical care expenditures for diabetes, its chronic complications and comorbidities. Prev Med. Forthcoming.

Health

Care

Accessand

Utilization

en

n

r

I

w

ode

e

a-De

80 .......

Health and Aging ...............................................................................................Te

chni

calN

otes

Data SourcesAppendix I describes the data sources used in th

chartbook except for the Second Supplement on Agidescribed below.

Second Supplement on Aging (figures 11–13, 15and 16, 21, 23–25)

Nine of the figures in the chartbook are based odata from the Second Supplement on Aging to the1994 National Health Interview Survey (SOA II). TheSOA II, conducted by NCHS with the support of theNational Institute on Aging, is a survey of 9,447noninstitutionalized persons 70 years of age and ovewho were interviewed originally as part of the 1994National Health Interview Survey (NHIS). The SOA Iincludes measures of health and functioning, chronicconditions, use of assistive devices, housing andlong-term care, and social activities. The SOA II wasdesigned to replicate the Supplement on Aging (SOAto the 1984 NHIS to examine whether changes haveoccurred in the health and functioning of the olderpopulation between the mid-1980’s and themid-1990’s. The SOA served as the baseline for theLongitudinal Study on Aging (LSOA), which followedthe original 1984 cohort through subsequent interviein 1986, 1988, and 1990 and is continuing withpassive mortality followup. The SOA II serves as thebaseline for the Second Longitudinal Study on Aging(LSOA II).

The SOA II was implemented as part of theNational Health Interview Survey on Disability(NHIS-D), which was designed to help researchersunderstand disability, to estimate the prevalence ofcertain conditions, and to provide baseline statisticsthe effects of disabilities. The NHIS-D was conductein two phases. Phase 1 collected information from thhousehold respondent at the time of the 1994 NHIScore interview and was used as a screening instrumfor Phase 2. The screening criteria were broadlydefined, and more than 50 percent of persons 70 yeof age and over were included in the Phase 2 NHISinterviews. Persons 70 years of age and over who wnot included in Phase 2 NHIS-D received the SOA IIsurvey instrument, which was a subset of questionsfrom the NHIS-D.

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

g

)

s

n

nt

rs

re

While the 1994 NHIS core and NHIS-D Phase 1interviews took place in 1994, Phase 2 was conductedas a followup survey, 7–17 months after the coreinterviews. In the calculation of weights, therefore, thepost-stratification adjustment was based on thepopulation control counts from July 1, 1995, roughlythe midpoint of the Phase 2 survey period. As a result,the SOA II sample, based on all 1994 NHIS coreparticipants 70 years of age and over at the time of thePhase 2 NHIS-D interviews, is representative of the1995 noninstitutionalized population 70 years of ageand over. Refer to the documentation for the NHIS-Dand the SOA II for more details on the implementationof the surveys (1,2).

Institutionalized PopulationThe majority of figures in the chartbook are

calculated from data that represent thenoninstitutionalized older population in the UnitedStates.

However, figures on population (figure 1), nursinghome residence (figure 3), life expectancy andmortality (figures 5–9), rates of hospital discharges(figure 27), Medicare HMO enrollment (figure 32), andcost of health care expenditures for heart disease anddiabetes (figures 33and34) cover the total elderlypopulation, including persons living in institutions. Theestimates of home health care patients and servicesreceived by home health care patients (figures 29and30) are based on the total older population includingthe institutionalized population. These rates arecalculated from the National Home and Hospice CareSurvey, which is a sample survey of home healthagencies and hospices. A small percent of patients areliving in institutions when they receive services fromhome health care agencies such as physical oroccupational therapy.

Age AdjustmentThe age distribution of older women and men is

different: among persons 65 years of age and over, theaverage woman is older than the average man.Consequently, comparing rates for women and menamong the total older population may confound

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

afo

uic

d

4

o

fo

r

n

t,

an

d

a

s

on

t

............................................................................................... Health and AgingTechnicalNotes

differences in rates with differences in agecomposition.

In general, the chartbook presents rates by sexat least three age groups. When data are presentedthe total older population (65 years of age and over70 years of age and over) and sex differences arehighlighted, the rates are age adjusted.Appendix IIdescribes the age adjustment procedures and the soof the standard population. The prevalence of chronconditions (figure 11) is age adjusted using four agegroups (70–74 years, 75–79 years, 80–84 years, anyears and over). Rates of exercise (figure 22) are ageadjusted using three age groups (65–74 years, 75–8years, and 85 years and over).

Race and Ethnicity

The focus of the chartbook is age and sexdifferences in the health of the older population.Depending on the variable of interest, some datasources did not have sufficient numbers of observatito allow calculation of reliable estimates of the olderpopulation by race and ethnicity. When race andethnicity differences are presented, data are shownwhite and black persons or for non-Hispanic white,non-Hispanic black, and Hispanic persons, except fothe figure on deaths from all causes (figure 8).

Death rates for all causes (figure 8) are presentedfor five groups: white persons, black persons, Hispapersons, Asians or Pacific Islanders, and AmericanIndians or Alaska Natives.

Among persons 65–74 and 75–84 years of age,death rates for Hispanics, Asians or Pacific Islandersand American Indians or Alaska Natives were lowerthan the death rates for white persons; in addition, thdeath rates for black persons were higher than the rfor white persons. Among persons 85 years of age aover, white persons had higher death rates than othegroups.

There are various explanations for the race andethnicity patterns of mortality among the olderpopulation. Inconsistency in the reporting of race anethnicity in vital statistics (the source of numeratorsfor death rates) and in census data (the source ofdenominators for death rates) is one explanation. De

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

ndorr

rce

85

ns

r

ic

he

etesd

r

th

rates will be underestimated if persons who areidentified as Asian, American Indian, or Hispanic indata from the Census Bureau are reported as white ornon-Hispanic on death certificates.

Among Hispanics and Asians or Pacific Islanders,many of whom are foreign born, the ‘‘healthy migrant’’effect may be operating. Immigration is a selectiveprocess, and immigrants are usually healthier thanpeople who do not migrate. In addition, if foreign-bornpersons return to their homeland to die, then deathrates will be underestimated because their deaths willnot be counted in U.S. vital statistics (3). These broadracial and ethnic categories include native-born personand immigrants from many different countries anddiverse backgrounds. The overall death rate mayobscure differences in health and mortality betweensubgroups of these populations.

Death rates for American Indians and AlaskaNatives are regarded as understated because populatiestimates between 1980 and 1990 increased by45 percent, in part due to more people identifyingthemselves as American Indian and because there isevidence that American Indians are underreported ondeath certificates (4).

The racial ‘‘crossover’’ in mortality, with youngerblack persons having higher death rates than youngerwhite persons, but older black persons having lowerdeath rates, is a subject of debate. Some researchshows that age misreporting may have artificiallydepressed death rates; however, other research hassuggested that black persons who survive to the oldesages may be healthier than white persons and havelower mortality rates (5,6).

Other Measures and Methods

Living Arrangements (figure 2)The categories of living arrangements were

computed from data on family status and maritalstatus. Persons living ‘‘with spouse’’ may also beliving with other relatives or nonrelatives. Thecategory ‘‘with other relatives’’ does not includepersons living with a spouse. Persons living ‘‘withnonrelatives’’ does not include persons living withspouses or other relatives.

........................................................................ 81

82 .......


chni

calN

otes

Life Expectancy (figures 5 and 6)In figure5 the estimates of lif e expectancy at age

85 for the years 1950–90 are based on death ratescalculated from 3 years of data. The denominator ofthe rate is the population from the decennial census(1950, 1960, 1970, 1980, and 1990). The numerator isthe deaths occurring in the decennial census year andthe 2 surrounding years (1949–51, 1959–61, 1969–71,1979–81, and 1989–91).

Beginning in 1997 lif e table methodology wasrevised to construct complete lif e tables by singleyears of age that extend to age 100. Previously,abridged lif e tables were constructed for 5-year agegroups ending with the age group 85 years and over.In the revised methodology, Medicare data are used toadjust estimates of lif e expectancy at ages 85–100.Some of the increase in lif e expectancy from 1996 to1997 may be due to the change in methodology. Therace differences in lif e expectancy at the oldest agesare also affected by this change. See the forthcomingVital and Health Statistics report for furtherdiscussion (7).

Chronic Conditions (figure 11)Estimates of the prevalence of chronic conditions

are based on self-reports in the Second Supplement onAging. Respondents were asked whether they ‘‘everhad’’ various conditions common among older persons.For certain conditions, the respondents were asked afollowup question about whether they ‘‘stil l had’’ thecondition. The estimates of heart disease prevalenceare based on persons who said they ever had heartdisease, including coronary heart disease, angina, heartattack or myocardial infarction, or any other heartdisease. The estimates of the prevalence of respiratorydiseases are based on persons who said they ‘‘stillhad’’ chronic bronchitis, emphysema, or asthma.Estimates of cancer prevalence are calculated frompersons who reported that they ‘‘stil l had’’ cancer ofany kind. Arthritis prevalence was estimated frompersons who reported they ‘‘ever had’’ arthritis.Estimates of hypertension and diabetes were based onpersons who reported that they ‘‘stil l had’’ thecondition. The prevalence of hypertension in figure11,

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

based on self-reports in the Second Supplement onAging, differs from estimates based on NHANES III,which were measured in physical examinations. (SeeHealth, United States, 1999, table68.)

Visual and Hearing Impairments (figures 12 and 13)The prevalence of visual and hearing impairment

is based on self-reports in the Second Supplement onAging in response to several questions. Visualimpairment is defined as blindness in one or both eyesor any other trouble seeing with one or both eyes evenwhen wearing glasses. Hearing impairment is definedas deafness in one or both ears or any other troublehearing with one or both ears.

Osteoporosis (reduced hip bone density) (figure 14)The definitions of osteoporosis and osteopenia are

based on diagnostic criteria proposed by the WorldHealth Organization (8,9). Estimates are based on thetotal femur region.

There is no consensus at this time concerning thedefinition of low bone density in men. The estimatesof osteopenia and osteoporosis for men in figure14 aremade by comparing their levels of bone mineraldensity to the values for non-Hispanic white women20–29 years of age as measured in NHANES III.

Using the bone mineral density values of youngwhite women as the cutoff point for diagnosingosteoporosis in men gives aconservative estimate ofthe prevalence of this condition in men, since whitewomen have the lowest values of bone mineral density(10).

Physical Functioning and Disability (figure 15)Limitation in physical activities, activities of daily

living, and instrumental activities of daily living arebased on self-reports in the Second Supplement onAging.

Nine physical activities are measured in figure15:walking for a quarter of a mile; walking up 10 stepswithout resting; standing or being on one’s feet forabout 2 hours; sitting for about 2 hours; stooping,crouching, or kneeling; reaching up over one’s head;reaching out (as if to shake someone’s hand); using

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

o

s

n

c

th

f

le

l

,


one’s fingers to grasp or handle; and lifting or carryinsomething as heavy as 10 pounds.To determine severity of limitations in physicalactivities, respondents in SOA II were asked a seriesquestions. The first question was ‘‘By yourself and nusing aids, do you have any difficulty (name ofactivity)?’’ Persons who answered ‘‘yes’’ to thisquestion were then asked, ‘‘How much difficulty doyou have (name of activity), some, a lot, or are youunable to do it?’’ The category ‘‘perform withdifficulty’’ in figure 15 consists of persons whoreported that they had ‘‘some’’ or ‘‘a lot’’ of difficulty.

ADL and IADL

Researchers group the important tasks of dailyliving into two categories frequently referred to in thichartbook:

ADL - activities of daily living. ADL’s includeseven activities: bathing or showering; dressing; eatigetting in and out of bed or chairs; walking; gettingoutside; and using the toilet, including getting to thetoilet.

IADL - instrumental activities of daily living.IADL’s include six activities: preparing one’s ownmeals; shopping for groceries and personal items suas toilet items or medicines; managing one’s money,such as keeping track of expenses or paying bills;using the telephone; doing heavy housework, likescrubbing floors or washing windows; and doing lighhousework, like doing dishes, straightening up, or ligcleaning.

To determine severity of limitations in activities odaily living and instrumental activities of daily living,respondents in SOA II were asked a series ofquestions. The first question was: ‘‘Because of a heaor physical problem, do you have any difficulty (namof activity)? Persons who answered ‘‘yes’’ to thisquestion were then asked, ‘‘By yourself (and withoutusing special equipment), how much difficulty do youhave (name of activity), some, a lot, or are you unabto do it?’’ The category ‘‘perform with difficulty’’ in

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

g

oft

g;

h

t

th

e

figure 15consists of persons who reported that theyhad ‘‘some’’ or ‘‘a lot’’ of difficulty.

Conditions Associated with Disability (figure 16)In SOA II, persons who had any difficulty with

one or more activities of daily living were asked,‘‘What condition causes the trouble in (list of activitiespreviously mentioned)?’’ Respondents could name upto five conditions. Infigure 16the percent andrank-order of conditions are computed from allconditions mentioned.

The category of respiratory diseases includesasthma, bronchitis, emphysema, influenza, pneumoniaand other respiratory, lung, or breathing problems.

Overweight (figures 17 and 18)The categories of weight distribution reflect

current Federal guidelines for overweight and obesity(11). The categories are based on body mass index(BMI), a measure of weight for height (kilograms permeter squared). Low weight is defined as a BMI lessthan 19. Healthy weight is a BMI of 19–24.99.Overweight is a BMI of 25–29.99. Moderately obese isa BMI of 30–34.99, and severely obese is a BMIgreater than or equal to 35.

Oral Health (figures 19 and 20)In 1993 estimates of edentulism are based on data

from the1993 Healthy People 2000 Supplementto theNational Health Interview Survey. This supplementwas administered to one adult sample person perfamily in the second half of the year. Respondentswere asked two questions concerning loss of naturalteeth: if they had lost all of their upper natural teethand if they had lost all of their lower natural teeth.

Estimates of edentulism in 1983 are based on datafrom the 1983 NHIS core interview that obtainedinformation on all household members. One questionasked respondents if they had lost all of their naturalteeth.

........................................................................ 83

n

th

o

uto

e

in

e

g

e

a

r

tsed

r

84 .......


chni

calN

otes

Social Activity (figure 21)Social activity, estimated from the Second

Supplement on Aging, is defined as doing at least oof the seven following activities at least once in a2-week period: getting together with friends orneighbors; talking with friends or neighbors on thetelephone; getting together with any relatives notincluding those living with the respondent; talking wiany relatives on the telephone not including thoseliving with the respondent; going to church, temple,another place of worship for services or otheractivities; going to a show or movie, sports event, clmeeting, class, or other group event; and going outeat at a restaurant.

Exercise (figure 22)Estimates of exercise are based on data from th

1995 Healthy People 2000 Supplementto the NationalHealth Interview Survey. This supplement wasadministered to one adult sample person per familyone-half of the households in the 1995 NHIS core.Persons who were physically handicapped, asdetermined by the interviewer, were not askedquestions regarding specific types of exercise and wexcluded from the calculations.

Exercise is defined as doing at least 1 of thefollowing 20 exercises, sports, or physically activehobbies at least once within a 2-week period: walkinfor exercise; gardening or yard work; stretchingexercises; weightlifting or other exercises to increasmuscle strength; jogging or running; aerobics oraerobic dancing; riding a bicycle or exercise bike; stclimbing for exercise; swimming for exercise; playingtennis; playing golf; bowling; playing baseball orsoftball; playing handball, racquetball, or squash;skiing; playing basketball; playing volleyball; playingsoccer; playing football; or other exercises, sports, ophysically active hobbies not mentioned above.

Caregivers (figure 23)In the Second Supplement on Aging, responden

could name a maximum of four persons who providhelp with activities of daily living (ADL) orinstrumental activities of daily living (IADL). See

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

e

r

b

re

ir

notes forfigure 15above for definitions of ADL andIADL.

Assistive Devices (figure 25)The use of assistive devices in the Second

Supplement on Aging was defined as using any of thefollowing medical devices or supplies in the 12 monthsbefore the interview: tracheotomy tube; respirator;ostomy bag; catheterization equipment; glucosemonitor; diabetic equipment or supplies; inhaler;nebulizer; hearing aid; crutches; cane; walker;wheelchair; scooter; or feeding tube. Use of respiratoryequipment was defined as using a respirator, inhaler, onebulizer. Use of diabetic equipment was defined asusing a glucose monitor or other diabetic equipmentand supplies.

Influenza and Pneumonia (figure 28)Estimates of influenza and pneumonia vaccinations

are based on data from theHealthy People 2000Supplementto the 1993–95 National Health InterviewSurveys (NHIS). In 1994–95, this supplement wasadministered to one adult sample person per family inone-half of the households in the NHIS core. In 1993the supplement was administered to one adult sampleperson per family in the second half of the year.

Home Health Care (figures 29 and 30)Rates of home health care patients per 1,000

population are based on age at interview, calculatedfrom date of birth and date of interview. Persons withmissing date of birth information were excluded fromcalculations.

References

1. National Center for Health Statistics. Data file documentation, National HealthInterview Survey, Second Supplement on Aging, 1994 (machine readable data file anddocumentation). Hyattsville, Maryland. 1998.

2. National Center for Health Statistics. Data file documentation, National HealthInterview Survey of Disability, Phase 1, 1995 (machine readable data file anddocumentation). Hyattsville, Maryland. 1998.

3. Elo IT, Preston SH. Racial and ethnic differences in mortality at older ages. In:Martin LG, Soldo BJ, eds. Racial and ethnic differences in the health of olderAmericans. Washington: National Academy Press 10–42. 1997.

4. Sorlie PD, Rogot E, Johnson NJ. Validity of demographic characteristics on thedeath certificate. Epidemiology 3(2):181–4. 1992.

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


5. Preston SH, Elo IT, Rosenwaike I, Hill M. African-American mortality at older ages:Results of a matching study. Demography 33(2):193–209. 1996.

6. Manton KC, Stallard E, Wing S. Analyses of black and white differentials in the agetrajectory of mortality in two closed cohort studies. Stat Med 10:1043–59. 1991.

7. Anderson RN. A methodology for constructing complete life tables for the UnitedStates. National Center for Health Statistics. Vital and Health Stat (in preparation).

8. Kanis JL, Melton LJ, Christiansen C, et al. The diagnosis of osteoporosis. J BoneMiner Res 9:1137–41. 1994.

9. World Health Organization. Assessment of fracture risk and its application toscreening for postmenopausal osteoporosis. Technical Report Series no. 842. WHO,Geneva, Switzerland. 1994.

10. Looker AC, Orwoll ES, Johnston Jr. CC, et al. Prevalence of low femoral bonedensity in older U.S. adults from NHANES III. J Bone Miner Res 12(11):1761–8. 1997.

11. Report of the dietary guidelines advisory committee on the dietary guidelines forAmericans, 1995 to the Secretary of Health and Human Services and the Secretary ofAgriculture. U.S. Department of Agriculture, Agricultural Research Service. 1995.

................................................................................................................................... 85

Figure 1. Population 65 years of age and over

Year65 yearsand over

85 yearsand over

Number in millions

1950 . . . . . . . . . . . . . . . . . . 12.2 0.61960 . . . . . . . . . . . . . . . . . . 16.6 0.91970 . . . . . . . . . . . . . . . . . . 20.1 1.51980 . . . . . . . . . . . . . . . . . . 25.5 2.21990 . . . . . . . . . . . . . . . . . . 31.1 3.02000 . . . . . . . . . . . . . . . . . . 34.7 4.32010 . . . . . . . . . . . . . . . . . . 39.4 5.72020 . . . . . . . . . . . . . . . . . . 53.2 6.52030 . . . . . . . . . . . . . . . . . . 69.4 8.5

Figure 2. Living arrangements of persons 65 years of age and over

Age and type of arrangement Women Men

Percent

65–74 years

Living alone . . . . . . . . . . . . . . . . . . . . 32.1 14.6Living with spouse . . . . . . . . . . . . . . . 51.1 76.8Living with other relatives . . . . . . . . . . 14.8 5.7Living with nonrelatives only . . . . . . . . 2.0 2.9

75–84 years

Living alone . . . . . . . . . . . . . . . . . . . . 50.0 19.6Living with spouse . . . . . . . . . . . . . . . 31.5 68.2Living with other relatives . . . . . . . . . . 16.9 9.7Living with nonrelatives only . . . . . . . . 1.7 2.5

85 years and overLiving alone . . . . . . . . . . . . . . . . . . . . 58.6 29.2Living with spouse . . . . . . . . . . . . . . . 10.7 46.0Living with other relatives . . . . . . . . . . 28.4 17.5Living with nonrelatives only . . . . . . . . 2.3 7.3

Figure 3. Nursing home residents among persons 65 years of ageand over

Sex and age

Residents per 1,000 population

White Black

Rate SE Rate SE

Women65 years and over . . . 55.5 0.7 54.7 3.365–74 years. . . . . . . . 11.1 0.5 18.1 1.975–84 years. . . . . . . . 51.9 1.3 62.6 5.885 years and over . . . 222.5 4.3 203.3 15.9

Men

65 years and over . . . 25.3 0.7 41.1 3.165–74 years. . . . . . . . 8.6 0.5 20.7 2.475–84 years. . . . . . . . 32.6 1.3 57.3 6.585 years and over . . . 117.3 5.1 144.3 18.4

86 ...................................................................................................................................

Data Tables for Figures 1–34 .....................................................................

Figure 4. Percent in poverty among persons 65 years of age and over

Race, Hispanic origin, and sex Percent

White:Women . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6

Black:Women . . . . . . . . . . . . . . . . . . . . . . . . . . 28.8Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.8

Hispanic:Women . . . . . . . . . . . . . . . . . . . . . . . . . . 26.3Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.3

Figure 5. Life expectancy at birth, age 65, and age 85

Year

At birth At age 65 At age 85

Women Men Women Men Women Men

1950 . . . . . . . . . . 71.1 65.6 15.0 12.8 4.9 4.41960 . . . . . . . . . . 73.1 66.6 15.8 12.8 4.7 4.41970 . . . . . . . . . . 74.7 67.1 17.0 13.1 5.6 4.71980 . . . . . . . . . . 77.4 70.0 18.3 14.1 6.4 5.11990 . . . . . . . . . . 78.8 71.8 18.9 15.1 6.7 5.31992 . . . . . . . . . . 79.1 72.3 19.2 15.4 6.6 5.31993 . . . . . . . . . . 78.8 72.2 18.9 15.3 6.4 5.21994 . . . . . . . . . . 79.0 72.4 19.0 15.5 6.4 5.21995 . . . . . . . . . . 78.9 72.5 18.9 15.6 6.3 5.21996 . . . . . . . . . . 79.1 73.1 19.0 15.7 6.4 5.41997 . . . . . . . . . . 79.4 73.6 19.2 15.9 6.6 5.5

Figure 6. Life expectancy at birth, age 65, and age 85 by sex and race

Sex and age White Black

Women

At birth . . . . . . . . . . . . . . . . . . . . 79.9 74.7At age 65 . . . . . . . . . . . . . . . . . . 19.3 17.6At age 85 . . . . . . . . . . . . . . . . . . 6.6 6.7

Men

At birth . . . . . . . . . . . . . . . . . . . . 74.3 67.2At age 65 . . . . . . . . . . . . . . . . . . 16.0 14.2At age 85 . . . . . . . . . . . . . . . . . . 5.4 5.7

................................................................................................................................... 87

.................................................................... Data Tables for Figures 1–34

Figure 7. Death rates for all causes among persons 65 years of age and over

Age Women Men

65–69 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,530.1 2,556.770–74 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2,425.5 3,948.975–79 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3,763.5 5,831.380–84 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6,325.2 9,320.085–89 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11,202.6 15,261.790–94 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17,572.2 21,365.995 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25,556.3 26,078.3

Figure 8. Death rates for all causes among persons 65 years of age and over

Sex and age White BlackAsian or

Pacific Islander

AmericanIndian or

Alaska Native Hispanic

Women

65–74 years. . . . . . . . . . . . . . . . . . . . . . . . . 1,900.5 2,739.7 1,117.3 1,920.5 1,381.975–84 years. . . . . . . . . . . . . . . . . . . . . . . . . 4,786.3 5,669.3 3,052.1 3,531.6 3,220.585 years and over . . . . . . . . . . . . . . . . . . . . 14,681.4 13,701.7 8,414.1 5,773.6 8,708.6

Men

65–74 years. . . . . . . . . . . . . . . . . . . . . . . . . 3,122.7 4,298.3 1,892.6 2,847.2 2,251.775–84 years. . . . . . . . . . . . . . . . . . . . . . . . . 7,086.0 8,296.3 4,749.1 4,796.3 4,750.385 years and over . . . . . . . . . . . . . . . . . . . . 17,767.1 16,083.5 11,796.3 7,888.1 10,487.1

Figure 9. Death rates for selected leading causes among persons 65 years of age and over

Age and sexHeart

disease Cancer Stroke

Chronicobstructivepulmonary

diseasePneumonia/

influenza

65–74 years

Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529.4 676.8 120.1 136.1 42.9Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,031.1 1,058.4 153.1 201.3 74.3

75–84 years

Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,616.6 1,050.6 444.4 287.6 189.3Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2,443.6 1,770.2 488.7 469.6 301.6

85 years and over

Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6,013.7 1,439.2 1,618.4 424.5 933.7Men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6,658.5 2,712.5 1,500.7 902.8 1,250.5

88 ...................................................................................................................................


Figure 10. Fair or poor health among persons 65 years of age and over

Sex, race, and Hispanic origin

65 years and over 65–74 years 75–84 years 85 years and over

Percent SE Percent SE Percent SE Percent SE

Women

Non-Hispanic white. . . . . . . . . . . . . . . 25.7 0.4 22.5 0.5 28.3 0.7 33.6 1.3Non-Hispanic black. . . . . . . . . . . . . . . 42.2 1.3 40.7 1.6 44.7 2.1 44.0 3.8Hispanic . . . . . . . . . . . . . . . . . . . . . . 35.4 1.6 31.5 1.8 40.7 3.0 44.9 4.7All races . . . . . . . . . . . . . . . . . . . . . . 27.6 0.4 24.8 0.5 30.2 0.6 34.9 1.2

Men

Non-Hispanic white. . . . . . . . . . . . . . . 26.5 0.5 23.7 0.6 30.6 0.8 32.7 1.9Non-Hispanic black. . . . . . . . . . . . . . . 40.7 1.6 38.4 1.9 43.6 2.8 55.0 6.1Hispanic . . . . . . . . . . . . . . . . . . . . . . 34.6 1.8 31.3 2.0 40.3 4.0 49.1 7.6All races . . . . . . . . . . . . . . . . . . . . . . 28.0 0.4 25.4 0.5 31.7 0.8 35.0 1.8

SE Standard error.

Figure 11. Percent of persons 70 years of age and over who reportedselected chronic conditions

Chronic condition

Women Men

Percent SE Percent SE

Arthritis . . . . . . . . . . . . . . . . 63.3 0.8 49.5 0.9Hypertension . . . . . . . . . . . . 39.6 0.7 31.5 0.8Heart disease. . . . . . . . . . . . 24.1 0.6 30.0 0.8Diabetes . . . . . . . . . . . . . . . 10.4 0.4 11.6 0.5Respiratory diseases . . . . . . 10.3 0.4 11.0 0.5Stroke . . . . . . . . . . . . . . . . . 7.6 0.4 10.4 0.6Cancer . . . . . . . . . . . . . . . . 2.3 0.2 6.0 0.4

SE Standard error.

Figure 12. Prevalence of visual impairment among persons 70 years ofage and over

Race and age

Women Men


White:70 years and over . . . . . . . 19.0 0.7 15.6 0.770–74 years . . . . . . . . . . . 13.3 0.8 11.2 0.975–79 years . . . . . . . . . . . 16.6 1.1 15.9 1.380–84 years . . . . . . . . . . . 22.9 1.4 19.7 1.785 years and over . . . . . . . 32.9 1.9 27.4 2.4

Black:70 years and over . . . . . . . 18.8 1.8 19.7 2.770–74 years . . . . . . . . . . . 19.6 2.9 16.6 4.675–79 years . . . . . . . . . . . 17.2 2.9 17.7 3.480–84 years . . . . . . . . . . . 16.7 3.8 23.3 5.785 years and over . . . . . . . 23.3 4.4 31.6 9.0

SE Standard error.

................................................................................................................................... 89


Figure 15. Percent of persons 70 years of age and over who have difficulty performing 1 or more physical activities, activities of dailyliving, and instrumental activities of daily living

Sex

70 years and over 70–74 years 75–79 years 80–84 years 85 years and over

Percent SE Percent SE Percent SE Percent SE Percent SE

Women Perform with difficulty

Physical activity . . . . . . . 32.5 0.8 31.3 1.2 33.9 1.3 35.3 1.7 29.1 1.7ADL . . . . . . . . . . . . . . . 22.2 0.7 19.6 1.0 19.0 1.0 23.9 1.3 32.9 1.9IADL . . . . . . . . . . . . . . . 12.8 0.5 12.5 0.8 12.0 1.0 14.1 1.3 13.2 1.2

Men


Women Unable to perform


Men


SE Standard error.

Figure 13. Prevalence of hearing impairment among persons 70 yearsof age and over

Race and age

Women Men


White:70 years and over . . . . 29.2 0.8 41.6 1.070–74 years . . . . . . . . 21.3 1.1 35.4 1.475–79 years . . . . . . . . 25.8 1.1 42.7 1.880–84 years . . . . . . . . 34.5 1.6 47.1 2.385 years and over . . . . 48.4 2.0 55.9 2.9

Black:70 years and over . . . . 17.3 1.5 21.2 2.270–74 years . . . . . . . . 12.6 2.1 15.1 3.375–79 years . . . . . . . . 12.3 2.5 25.3 4.180–84 years . . . . . . . . 24.8 3.8 21.6 6.485 years and over . . . . 29.9 4.9 33.3 7.6

SE Standard error.

Figure 14. Prevalence of reduced hip bone density among persons 65 years of age and over

Age

Women Men

Osteoporosis Osteopenia Osteoporosis Osteopenia

Percent SE Percent SE Percent SE Percent SE

65 years and over . . . . . . . . . . 26.1 1.6 45.9 1.5 3.8 0.7 21.8 1.565–74 years. . . . . . . . . . . . . . . 19.0 2.0 46.9 2.1 2.0 0.6 17.9 1.675–84 years. . . . . . . . . . . . . . . 32.5 2.1 45.8 2.2 6.4 1.4 28.0 2.685 years and over . . . . . . . . . . 50.5 3.5 39.6 3.1 13.7 3.0 40.2 4.4

SE Standard error.

90 ...................................................................................................................................


Figure 16. Percent of persons 70 years of age and over who reportspecific conditions as a cause of limitation in activities of daily living

Type of condition Percent

Arthritis . . . . . . . . . . . . . . . . . . . . . . . . . 10.6Heart disease. . . . . . . . . . . . . . . . . . . . . 4.0Stroke . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6Respiratory . . . . . . . . . . . . . . . . . . . . . . 2.5Diabetes . . . . . . . . . . . . . . . . . . . . . . . . 1.5

Figure 17. Distribution of weight among persons 65–74 years of age

Body mass index Classification Women Men

Less than 19 . . . . . . . . Low 3.7 1.719–24.99 . . . . . . . . . . . Healthy 36.0 29.825–29.9 . . . . . . . . . . . . Overweight 33.5 44.430–34.9 . . . . . . . . . . . . Moderately obese 16.0 19.435 and over . . . . . . . . . Severely obese 10.9 4.6

Figure 18. Prevalence of obesity among persons 65–74 years of age

Sex1960–62

NHES1971–74

NHANES I1976–80

NHANES II1988–94

NHANES III

Women . . . . . . . . 23.2 22.0 21.5 26.9Men . . . . . . . . . . 10.4 10.9 13.2 24.1

Figure 19. Percent with untreated dental caries among dentate persons65 years of age and over

Age

Women Men


65 years and over . . . . . 27.5 1.8 35.2 1.965–74 years. . . . . . . . . . 24.7 2.0 33.9 2.375–84 years. . . . . . . . . . 33.5 2.5 37.0 2.885 years and over . . . . . 24.4 3.8 46.0 6.0

SE Standard error.

Figure 20. Prevalence of total tooth loss (edentulism) among persons 65 years of age and over

Age

1983 1993


65 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . 38.4 0.6 29.5 1.065–74 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34.1 0.7 25.5 1.175–84 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43.4 1.0 33.6 1.685 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . 54.2 1.8 43.5 4.2

SE Standard error.

................................................................................................................................... 91


Figure 21. Number of social activities in a 2-week period amongpersons 70 years of age and over

Age and numberof activities

Women Men


70 years and over

0 activities . . . . . . . . . . . . . . . . 1.6 0.2 2.4 0.21–2 activities . . . . . . . . . . . . . . 10.7 0.5 13.5 0.63–4 activities . . . . . . . . . . . . . . 29.8 0.7 30.3 0.95–7 activities . . . . . . . . . . . . . . 57.9 0.8 53.8 1.0

70–74 years


75–79 years


80–84 years


85 years and over


SE Standard error.

Figure 22. Percent who exercise and selected type of exercise amongpersons 65 years of age and over

Type of exercise

Women Men


Any exercise . . . . . . . . . 67.0 1.3 77.0 1.3

Walking . . . . . . . . . . . . . 67.3 1.5 62.1 1.8Gardening . . . . . . . . . . . 37.9 1.7 54.0 2.0Stretching . . . . . . . . . . . 31.9 1.5 25.7 1.6Swimming . . . . . . . . . . . 3.6 0.6 3.2 0.6Aerobics . . . . . . . . . . . . 4.9 0.7 * *Stair climbing . . . . . . . . . 9.2 1.0 4.6 0.8

SE Standard error.* Number in this category is too small to calculate reliable rates.

92 ...................................................................................................................................


Figure 23. Number of caregivers providing assistance with activities of daily living or instrumental activities of daily living to persons70 years of age and over

Sex and age

1 caregiver 2 caregivers 3–4 caregivers

Percent SE Percent SE Percent SE

Women70 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.8 0.6 10.8 0.5 7.9 0.570–74 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.4 0.9 7.0 0.6 4.6 0.575–79 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19.8 1.0 9.6 0.8 5.6 0.780–84 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.6 1.4 12.4 1.0 9.8 0.985 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . 27.9 1.7 20.5 1.6 18.7 1.6

Men70 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.2 0.6 5.4 0.4 3.8 0.370–74 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.9 0.9 3.3 0.5 2.1 0.475–79 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15.1 1.1 4.3 0.6 3.3 0.680–84 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.2 1.6 7.7 1.0 5.0 0.885 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . 26.9 2.5 14.1 1.8 10.7 1.7

SE Standard error.

Figure 24. Percent with unmet needs among persons 70 years of age and over who need help with 1 or more activities of daily living orinstrumental activities of daily living

Sex and age

Unmet need ADL’s Unmet need IADL’s only


Women70–84 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.4 1.7 30.5 1.885 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.5 2.3 15.2 2.0

Men70–84 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.9 2.3 19.9 2.485 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . * * * *

ADL’s Activities of daily living.IADL’s Instrumental activities of daily living.SE Standard error.* Number in this category is too small to calculate reliable rates.

................................................................................................................................... 93


Figure 25. Assistive devices used among persons 70 years of age and over

Age and number of devices

Women Men


70 years and over1 device . . . . . . . . . . . . . . . . . 21.2 0.6 21.5 0.72 devices. . . . . . . . . . . . . . . . . 10.1 0.5 10.1 0.53 or more devices. . . . . . . . . . . 7.8 0.3 8.3 0.5

70–74 years1 device . . . . . . . . . . . . . . . . . 17.0 0.8 19.5 1.12 devices. . . . . . . . . . . . . . . . . 7.6 0.6 7.1 0.73 or more devices. . . . . . . . . . . 6.0 0.5 6.5 0.6



85 years and over1 device . . . . . . . . . . . . . . . . . 31.7 1.8 27.5 2.52 devices. . . . . . . . . . . . . . . . . 18.5 1.6 19.2 2.13 or more devices. . . . . . . . . . . 13.5 1.2 13.3 1.8

SE Standard error.

Figure 26. Place of ambulatory physician contacts among persons 65 years of age and over

Sex and age

Place of contact

Doctor’s office Hospital Home Phone Other


Women

65 years and over . . . . . . . . . . 51.1 1.3 9.2 0.5 19.8 1.5 10.0 0.6 9.9 0.765–74 years. . . . . . . . . . . . . . . 55.5 1.6 10.8 0.7 10.9 1.6 10.7 0.8 12.1 1.075–84 years. . . . . . . . . . . . . . . 49.3 1.8 8.4 0.7 23.3 2.3 9.9 0.9 9.1 0.985 years and over . . . . . . . . . . 39.6 2.9 5.7 1.0 43.1 3.9 8.1 1.2 3.5 0.6

Men

65 years and over . . . . . . . . . . 54.3 1.6 11.7 0.8 13.5 1.9 8.6 0.6 11.9 0.865–74 years. . . . . . . . . . . . . . . 56.5 1.7 12.3 0.9 7.9 1.5 10.0 0.8 13.5 1.275 –84 years . . . . . . . . . . . . . . 51.1 3.1 11.8 1.6 19.7 4.0 7.1 0.8 10.3 1.385 years and over . . . . . . . . . . 53.0 4.4 7.6 2.2 25.5 4.8 5.6 1.3 8.4 1.9

SE Standard error.

94 ...................................................................................................................................


Figure 27. Hospital discharge rates in non-Federal short-stay hospitals for selected first-listed diagnoses among persons 65 years of ageand over

Diagnosis

Women Men

65–74years

75–84years

85 yearsand over

65–74years

75–84years

85 yearsand over

All diagnoses . . . . . . . . . . . . . . . . . . . . . 246.6 395.3 565.3 270.6 441.3 624.7

Heart Disease . . . . . . . . . . . . . . . . . . . . 53.4 89.1 112.7 73.5 110.3 131.0Stroke . . . . . . . . . . . . . . . . . . . . . . . . . . 12.0 26.4 39.5 15.3 31.9 39.8Malignant neoplasms . . . . . . . . . . . . . . . 19.8 19.8 16.3 23.5 27.2 31.3Pneumonia . . . . . . . . . . . . . . . . . . . . . . 10.8 21.1 46.0 13.0 29.9 73.3Bronchitis . . . . . . . . . . . . . . . . . . . . . . . 8.4 10.2 9.6 7.1 12.0 21.4Fractures . . . . . . . . . . . . . . . . . . . . . . . . 9.7 23.6 55.9 3.5 10.3 30.6

Figure 28. Percent vaccinated against influenza and pneumococcal disease among persons 65 years of age and over

Race and Hispanic origin

Influenza Pneumococal disease


White, non-Hispanic . . . . . . . . . . . . . . . . . . . 56.8 0.6 31.2 0.7Black, non-Hispanic . . . . . . . . . . . . . . . . . . . 36.4 1.5 16.2 1.4Hispanic . . . . . . . . . . . . . . . . . . . . . . . . . . . 44.0 3.2 15.9 1.7

SE Standard error.

Figure 29. Home health care patients among persons 65 years of ageand over

Sex65 yearsand over

65–74years

75–84years

85 yearsand over

Patients per 1,000 population

Women . . . . . . . . . . . . . 61.1 27.6 84.1 130.1Men . . . . . . . . . . . . . . . 36.5 18.2 55.4 99.5

Figure 30. Home health care services received by current patients65 years of age and over

Type of servicePercent

receiving service

Nursing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85.3Homemaker . . . . . . . . . . . . . . . . . . . . . . . . . 28.7Physical therapy. . . . . . . . . . . . . . . . . . . . . . 20.0Social services . . . . . . . . . . . . . . . . . . . . . . . 10.9Medications . . . . . . . . . . . . . . . . . . . . . . . . . 9.9Continuous home care . . . . . . . . . . . . . . . . . 5.8Occupational therapy . . . . . . . . . . . . . . . . . . 4.8Physician services . . . . . . . . . . . . . . . . . . . . 3.7Nutrition. . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4

All other services . . . . . . . . . . . . . . . . . . . . . 15.1

................................................................................................................................... 95


State Percent

New England

Maine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0New Hampshire . . . . . . . . . . . . . . . . . . . . . . . . 0Vermont . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Massachusetts . . . . . . . . . . . . . . . . . . . . . . . . . 16Rhode Island . . . . . . . . . . . . . . . . . . . . . . . . . . 15Connecticut . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Middle AtlanticNew York . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13New Jersey . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Pennsylvania . . . . . . . . . . . . . . . . . . . . . . . . . . 17

East North CentralOhio. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Indiana. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Illinois . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Michigan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Wisconsin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

West North CentralMinnesota. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18Iowa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Missouri . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10North Dakota . . . . . . . . . . . . . . . . . . . . . . . . . . 1South Dakota . . . . . . . . . . . . . . . . . . . . . . . . . . 0Nebraska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Kansas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

South AtlanticDelaware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0Maryland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8District of Columbia . . . . . . . . . . . . . . . . . . . . . . 21Virginia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1West Virginia . . . . . . . . . . . . . . . . . . . . . . . . . . 2North Carolina. . . . . . . . . . . . . . . . . . . . . . . . . . 1South Carolina . . . . . . . . . . . . . . . . . . . . . . . . . 1Georgia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Florida . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

State Percent

East South CentralKentucky . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Tennessee . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0Alabama. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Mississippi . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

West South CentralArkansas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Louisiana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Oklahoma. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Texas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

MountainMontana. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0Idaho . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0Wyoming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0Colorado . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26New Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . 17Arizona . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34Utah. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17Nevada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

PacificWashington. . . . . . . . . . . . . . . . . . . . . . . . . . . . 20Oregon. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37California . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38Alaska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0Hawaii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

96 ...................................................................................................................................


Figure 32. Percent of Medicare enrollees in health maintenance organizations

Figure 31. Health insurance coverage among persons 65 years of age and over

Age, race, and Hispanic origin

Medicare only Medicare/Medicaid Medicare/private Private only None


65–84 years

White, non-Hispanic . . . . . . . . . 13.9 0.9 5.1 0.3 77.9 1.0 2.7 0.2 0.5 0.1Black, non-Hispanic . . . . . . . . . 29.5 1.7 22.4 1.2 43.2 1.9 3.6 0.6 1.3 0.3Hispanic . . . . . . . . . . . . . . . . . 27.3 2.5 23.5 1.9 42.4 2.9 3.8 0.8 3.0 0.5

85 years of age and over

White, non-Hispanic . . . . . . . . . 20.4 1.4 7.5 0.8 70.2 1.6 * * * *Black, non-Hispanic . . . . . . . . . 38.6 4.5 28.1 3.3 32.5 4.8 * * * *Hispanic . . . . . . . . . . . . . . . . . 23.7 5.6 49.2 5.7 22.6 4.4 * * * *

SE Standard error.* Number in this category is too small to calculate reliable rates.

Figure 33. Estimated amount of personal health care expenditures attributed to heart disease among persons 65 years of age and over

Type of Health Service

Women Men

65 yearsand over

65–74years

75–84years

85 yearsand over

65 yearsand over

65–74years

75–84years

85 yearsand over

Amount in billions

All personal health care. . . . . . . . . . . . . . . . . 31.2 8.2 12.0 11.0 26.5 12.6 9.9 3.9Hospital care . . . . . . . . . . . . . . . . . . . . . . . . 14.2 5.2 5.9 3.0 18.2 9.4 6.8 2.0Physician and other professional services . . . . 3.0 1.2 1.2 0.6 3.2 1.8 1.1 0.3Home health care . . . . . . . . . . . . . . . . . . . . . 2.1 0.4 1.1 0.6 0.8 0.2 0.4 0.2Prescription drugs and medical durables. . . . . 1.8 0.7 0.8 0.4 1.6 0.9 0.6 0.2Nursing home care . . . . . . . . . . . . . . . . . . . . 10.2 0.7 3.1 6.4 2.7 0.4 1.0 1.3

Figures may not sum to totals due to rounding.

Figure 34. Estimated amount of personal health care expenditures attributed to diabetes among persons 65 years and over

Type of Health Service

Women Men

65 yearsand over

65–74years

75–84years

85 yearsand over

65 yearsand over

65–74years

75–84years

85 yearsand over

Amount in billions

All personal health care. . . . . . . . . . . . . . . . . 15.6 6.5 6.1 3.1 10.6 5.6 3.5 1.5Hospital care . . . . . . . . . . . . . . . . . . . . . . . . 6.4 3.1 2.2 1.1 5.6 3.4 1.7 0.6Physician and other professional services . . . . 2.1 1.1 0.7 0.3 1.5 0.9 0.5 0.1Home health care . . . . . . . . . . . . . . . . . . . . . 2.2 0.9 1.0 0.3 1.5 0.7 0.6 0.3Prescription drugs and medical durables. . . . . 0.9 0.5 0.3 0.1 0.7 0.5 0.2 0.1Nursing home care . . . . . . . . . . . . . . . . . . . . 4.0 0.8 1.9 1.3 1.2 0.2 0.6 0.4

Figures may not sum to totals due to rounding.

................................................................................................................................... 97


I. Sources and Limitations of Data 104

Introduction 104Department of Health and Human Services

Centers for Disease Control and PreventionNational Center for Health Statistics

National Vital Statistics System 105National Linked File of Live Births and Infant Deaths 108Compressed Mortality File 109National Survey of Family Growth 109National Health Interview Survey 110National Immunization Survey 111National Health and Nutrition Examination Survey 112National Health Provider Inventory (National Master Facility Inventory) 114National Home and Hospice Care Survey 114National Hospital Discharge Survey 115National Survey of Ambulatory Surgery 116National Nursing Home Survey 116National Ambulatory Medical Care Survey 118National Hospital Ambulatory Medical Care Survey 118

National Center for HIV, STD, and TB PreventionAIDS Surveillance 118

Epidemiology Program OfficeNational Notifiable Diseases Surveillance System 119

National Center for Chronic Disease Prevention and Health PromotionAbortion Surveillance 120

National Institute for Occupational Safety and HealthNational Traumatic Occupational Fatalities Surveillance System 120

Health Resources and Services AdministrationBureau of Health Professions

Physician Supply Projections 121Nurse Supply Estimates 121

Substance Abuse and Mental Health Services AdministrationOffice of Applied Studies

National Household Surveys on Drug Abuse 121Drug Abuse Warning Network 122Uniform Facility Data Set 123

Center for Mental Health ServicesSurveys of Mental Health Organizations 123

National Institutes of HealthNational Cancer Institute

Surveillance, Epidemiology, and End Results Program 124National Institute on Drug Abuse

Monitoring the Future Study (High School Senior Survey) 125Health Care Financing Administration

Office of the ActuaryEstimates of National Health Expenditures 125Estimates of State Health Expenditures 126

Medicare National Claims History Files 127Medicaid Data System 127Online Survey Certification and Reporting Database 128

.................................................................................................................................. 101

...........................................................................................Appendix Contents

Department of CommerceBureau of the Census

Census of Population 129Current Population Survey 129Population Estimates 129

Department of LaborBureau of Labor Statistics

Annual Survey of Occupational Injuries and Illnesses 130Consumer Price Index 130Employment and Earnings 131Employer Costs for Employee Compensation 131

Department of Veterans AffairsThe Patient Treatment File 132The Patient Census File 132The Outpatient Clinic File 132

Environmental Protection AgencyAerometric Information Retrieval System (AIRS) 132

United NationsDemographic Yearbook 133World Health Statistics Annual 133

Alan Guttmacher InstituteAbortion Survey 134

American Association of Colleges of Osteopathic Medicine 134American Association of Colleges of Pharmacy 134American Association of Colleges of Podiatric Medicine 134American Dental Association 134American Hospital Association

Annual Survey of Hospitals 135American Medical Association

Physician Masterfile 135Annual Census of Hospitals 135

Association of American Medical Colleges 136Association of Schools and Colleges of Optometry 136Association of Schools of Public Health 136InterStudy

National Health Maintenance Organization Census 136National League for Nursing 136

II. Glossary 138

Glossary Tables

I. Standard million age distribution used to adjust death rates to the U.S. population in 1940 138II. Numbers of live births and mother’s age groups used to adjust maternal mortality rates to live

births in the United States in 1970 139III. Populations and age groups used to age adjust NCHS survey data 139IV. Revision of theInternational Classification of Diseases, according to year of conference by which

adopted and years in use in the United States 141V. Cause-of-death codes, according to applicable revision ofInternational Classification of Diseases 142

VI. Codes for industries, according to theStandard Industrial Classification (SIC) Manual 150VII. Codes for diagnostic categories from theInternational Classification of Diseases, Ninth Revision,

Clinical Modification 151

102 ...............................................................................................................................

Appendix Contents ...........................................................................................

VIII. Codes for procedure categories from theInternational Classification of Diseases, Ninth Revision,Clinical Modification 152

IX. Mental health codes, according to applicable revision of theDiagnostic and Statistical Manual ofMental Disordersand International Classification of Diseases 154

III. Additional Data 165

.................................................................................................................................. 103

...........................................................................................Appendix Contents

n

a

o

s

hr

ivo

n

yre

o

d

f

104 .......

Appendix I ..............................................................................................................So

urce

sof

Dat

a

Introduction

This report consolidates the most current data othe health of the population of the United States, theavailability and use of health resources, and health cexpenditures. The information was obtained from thedata files and/or published reports of manygovernmental and nongovernmental agencies andorganizations. In each case, the sponsoring agencyorganization collected data using its own methods anprocedures. Therefore, the data in this report varyconsiderably with respect to source, method ofcollection, definitions, and reference period.

Much of the data presented in the detailed tableare from the ongoing data collection systems of theNational Center for Health Statistics. For an overviewof these systems, see: Kovar MG. Data systems of tNational Center for Health Statistics. National Centefor Health Statistics. Vital Health Stat 1(23). 1989.However, health care personnel data come primarilyfrom the Bureau of Health Professions, HealthResources and Services Administration, and theAmerican Medical Association. National healthexpenditures data were compiled by the office of theActuary, Health Care Financing Administration.

Although a detailed description and comprehensevaluation of each data source is beyond the scopethis appendix, users should be aware of the generalstrengths and weaknesses of the different datacollection systems. For example, population-basedsurveys obtain socioeconomic data, data on familycharacteristics, and information on the impact of anillness, such as days lost from work or limitation ofactivity. They are limited by the amount of informatioa respondent remembers or is willing to report.Detailed medical information, such as precisediagnoses or the types of operations performed, manot be known and so will not be reported. Health caproviders, such as physicians and hospitals, usuallyhave good diagnostic information but little or noinformation about the socioeconomic characteristicsindividuals or the impact of illnesses on individuals.

The populations covered by different datacollection systems may not be the same and

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

re

rd

e

ef

f

understanding the differences is critical to interpretingthe data. Data on vital statistics and nationalexpenditures cover the entire population. Most data onmorbidity and utilization of health resources coveronly the civilian noninstitutionalized population. Thus,statistics are not included for military personnel whoare usually young; for institutionalized people whomay be any age; or for nursing home residents whoare usually old.

All data collection systems are subject to error,and records may be incomplete or contain inaccurateinformation. People may not remember essentialinformation, a question may not mean the same thingto different respondents, and some institutions orindividuals may not respond at all. It is not alwayspossible to measure the magnitude of these errors ortheir impact on the data. Where possible, the tableshave notes describing the universe and the method ofdata collection to enable the user to place his or herown evaluation on the data. In many instances data donot add to totals because of rounding.

Some information is collected in more than onesurvey and estimates of the same statistic may varyamong surveys. For example, cigarette use is measureby the Health Interview Survey, the NationalHousehold Survey of Drug Abuse, and the Monitoringthe Future Survey. Estimates of cigarette use maydiffer among surveys because of different surveymethodologies, sampling frames, questionnaires,definitions, and tabulation categories.

Overall estimates generally have relatively smallsampling errors, but estimates for certain populationsubgroups may be based on small numbers and haverelatively large sampling errors. Numbers of births anddeaths from the vital statistics system representcomplete counts (except for births in those Stateswhere data are based on a 50-percent sample forcertain years). Therefore, they are not subject tosampling error. However, when the figures are used foranalytical purposes, such as the comparison of ratesover a period, the number of events that actuallyoccurred may be considered as one of a large series opossible results that could have arisen under the same

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

ga

s

s

ts

e

f

t

n

d

d

.............................................................................................................. Appendix ISourcesofD

ata

circumstances. When the number of events is smalland the probability of such an event is small,considerable caution must be observed in interpretinthe conditions described by the figures. Estimates thare unreliable because of large sampling errors orsmall numbers of events have been noted withasterisks in selected tables. The criteria used todesignate unreliable estimates are indicated as notethe applicable tables.

The descriptive summaries that follow provide ageneral overview of study design, methods of datacollection, and reliability and validity of the data.More complete and detailed discussions are found inthe publications referenced at the end of eachsummary. The data set or source is listed under theagency or organization that sponsored the datacollection.

Department of Health and Human Services

Centers for Disease Control and Prevention

National Center for Health Statistics

National Vital Statistics System

Through the National Vital Statistics System, theNational Center for Health Statistics (NCHS) collectsand publishes data on births, deaths, marriages, anddivorces in the United States. Fetal deaths areclassified and tabulated separately from other deathThe Division of Vital Statistics obtains information onbirths and deaths from the registration offices of allStates, New York City, the District of Columbia,Puerto Rico, the U.S. Virgin Islands, and Guam.Geographic coverage for births and deaths has beencomplete since 1933. U.S. data shown in detailedtables in this book are for the 50 States and theDistrict of Columbia, unless otherwise specified.

Until 1972 microfilm copies of all deathcertificates and a 50-percent sample of birth certificawere received from all registration areas and procesby NCHS. In 1972 some States began sending theirdata to NCHS through the Cooperative Health

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

t

to

.

esed

Statistics System (CHSS). States that participated inthe CHSS program processed 100 percent of theirdeath and birth records and sent the entire data file toNCHS on computer tapes. Currently, the data are sentto NCHS through the Vital Statistics CooperativeProgram (VSCP), following the same procedures as thCHSS. The number of participating States grew from 6in 1972 to 46 in 1984. Starting in 1985 all 50 Statesand the District of Columbia participated in VSCP.

In most areas practically all births and deaths areregistered. The most recent test of the completeness obirth registration, conducted on a sample of births from1964 to 1968, showed that 99.3 percent of all births inthe United States during that period were registered.No comparable information is available for deaths, butit is generally believed that death registration in theUnited States is at least as complete as birthregistration.

Demographic information on the birth certificatesuch as race and ethnicity is provided by the mother athe time of birth. Medical and health information isbased on hospital records. Demographic information othe death certificate is provided by the funeral directorbased on information supplied by an informant.Medical certification of cause of death is provided by aphysician, medical examiner, or coroner.

U.S. Standard Certificates—U.S. Standard LiveBirth and Death Certificates and Fetal Death Reportsare revised periodically, allowing careful evaluation ofeach item and addition, modification, and deletion ofitems. Beginning with 1989, revised standardcertificates replaced the 1978 versions. The 1989revision of the birth certificate includes items toidentify the Hispanic parentage of newborns and toexpand information about maternal and infant healthcharacteristics. The 1989 revision of the deathcertificate includes items on educational attainment anHispanic origin of decedents as well as changes toimprove the medical certification of cause of death.Standard certificates recommended by NCHS aremodified in each registration area to serve the area’sneeds. However, most certificates conform closely incontent and arrangement to the standard certificate, an

....................................................................... 105

bdn

rin

rt

e

a

a

.

e

sfs

nf

f

e

e

.

t

g

106 ......

Appendix I ..............................................................................................................So

urce

sof

Dat

a

all certificates contain a minimum data set specifiedNCHS. For selected items, reporting areas expandeduring the years spanned by this report. For items othe birth certificate, the number of reporting Statesincreased for mother’s education, prenatal care, mastatus, Hispanic parentage, and tobacco use; and odeath certificate, for educational attainment andHispanic origin of the decedent.

Maternal age—Mother’s age was reported on thebirth certificate by all States. Data are presented formothers age 10–49 years through 1996 and 10–54years starting in 1997, based on mother’s date of bior age as reported on the birth certificate. The age omother is edited for upper and lower limits. When thage of the mother is computed to be under 10 years55 years or over (50 years or over in 1964–96), it isconsidered not stated and imputed according to theof the mother from the previous birth record of thesame race and total birth order (total of fetal deathsand live births). Before 1963 not stated ages weredistributed in proportion to the known ages for eachracial group. Beginning in 1997, the birth rate for thematernal age group 45–49 years includes data formothers age 50–54 years in the numerator and is bon the population of women 45–49 years in thedenominator.

Maternal education—Mother’s education wasreported on the birth certificate by 38 States in 1970Data were not available from Alabama, Arkansas,California, Connecticut, Delaware, District ofColumbia, Georgia, Idaho, Maryland, New Mexico,Pennsylvania, Texas, and Washington. In 1975 thesdata were available from 4 additional States,Connecticut, Delaware, Georgia, Maryland, and theDistrict of Columbia, increasing the number of Statereporting mother’s education to 42 and the District oColumbia. Between 1980 and 1988 only three StateCalifornia, Texas, and Washington did not reportmother’s education. In 1988 mother’s education wasalso missing from New York State outside of NewYork City. In 1989–91 mother’s education was missionly from Washington and New York State outside oNew York City. Starting in 1992 mother’s education

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

y

talthe

hf

or

ge

sed

,

g

was reported by all 50 States and the District ofColumbia.

Prenatal care—Prenatal care was reported on thebirth certificate by 39 States and the District ofColumbia in 1970. Data were not available fromAlabama, Alaska, Arkansas, Connecticut, Delaware,Georgia, Idaho, Massachusetts, New Mexico,Pennsylvania, and Virginia. In 1975 these data wereavailable from 3 additional States, Connecticut,Delaware, and Georgia, increasing the number ofStates reporting prenatal care to 42 and the District oColumbia. Starting in 1980 prenatal care informationwas available for the entire United States.

Marital status—Mother’s marital status wasreported on the birth certificate by 39 States and theDistrict of Columbia in 1970, and by 38 states and thDistrict of Columbia in 1975. The incidence of birthsto unmarried women in States with no direct questionon marital status was assumed to be the same as theincidence in reporting States in the same geographicdivision. Starting in 1980 for States without a directquestion, marital status was inferred by comparing thparents’ and child’s surnames and other informationconcerning the father. In 1980 through 1996 maritalstatus was reported on the birth certificates of 41–45states. Beginning in 1997, all but four States(Connecticut, Michigan, Nevada, and New York)included a direct question on their birth certificates.

Hispanic births—In 1980 and 1981 information onbirths of Hispanic parentage was reported on the birthcertificate by the following 22 States: Arizona,Arkansas, California, Colorado, Florida, Georgia,Hawaii, Illinois, Indiana, Kansas, Maine, Mississippi,Nebraska, Nevada, New Jersey, New Mexico, NewYork, North Dakota, Ohio, Texas, Utah, and WyomingIn 1982 Tennessee, and in 1983 the District ofColumbia began reporting this information. Between1983 and 1987 information on births of Hispanicparentage was available for 23 States and the Districof Columbia. In 1988 this information becameavailable for Alabama, Connecticut, Kentucky,Massachusetts, Montana, North Carolina, andWashington, increasing the number of States reportin

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

n

.

iih

9t

.

rInfo

d

b

tnn

an


ata

information on births of Hispanic parentage to 30States and the District of Columbia. In 1989 thisinformation became available from an additional 17States, increasing the number of Hispanic-reportingStates to 47 and the District of Columbia. In 1989 oLouisiana, New Hampshire, and Oklahoma did notreport Hispanic parentage on the birth certificate. In1990 Louisiana began reporting Hispanic parentageHispanic origin of the mother was reported on thebirth certificates of 49 States and the District ofColumbia in 1991 and 1992; only New Hampshire dnot provide this information. Starting in 1993 Hispanorigin of mother was reported by all 50 States and tDistrict of Columbia. In 1990, 99 percent of birthrecords included information on mother’s origin.

Tobacco use—Information on tobacco use duringpregnancy became available for the first time in 198with the revision of the U.S. Standard Birth CertificaIn 1989 data on tobacco use were collected by 43States and the District of Columbia. The followingStates did not require the reporting of tobacco use othe birth certificate: California, Indiana, Louisiana,Nebraska, New York, Oklahoma, and South Dakota1990 information on tobacco use became availablefrom Louisiana and Nebraska increasing the numbereporting States to 45 and the District of Columbia.1991–93 information on tobacco use was available46 States and the District of Columbia with theaddition of Oklahoma to the reporting area; and in1994–97, for 46 States, the District of Columbia, anNew York City.

Education of decedent—Information oneducational attainment of decedents became availafor the first time in 1989 due to the revision of theU.S. Standard Certificate of Death. Mortality data byeducational attainment for 1989 was based on datafrom 20 States and by 1994–96 increased to 45 Staand the District of Columbia. In 1994–96 the followiStates either did not report educational attainment othe death certificate or the information was more tha20 percent incomplete: Georgia, Kentucky, OklahomRhode Island, and South Dakota. In 1997 informatioon decedent’s education became available from

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

ly

dce

e.

n

In

of

r

le

esg

n,

Oklahoma, increasing the reporting area to 46 Statesand the District of Columbia. Information on the deathcertificate about the decedent’s educational attainmentis reported by the funeral director based oninformation provided by an informant such as next ofkin.

Calculation of unbiased death rates by educationalattainment based on the National Vital StatisticsSystem requires that the reporting of education on thedeath certificate be complete and consistent with thereporting of education on the Current PopulationSurvey, the source of population estimates that formthe denominators for death rates. Death records witheducation not stated have not been included in thecalculation of rates. Therefore the levels of the ratesshown in this report are underestimated byapproximately the percent not stated, which rangedfrom 3 to 5 percent.

The validity of information about the decedent’seducation was evaluated by comparing self-reportededucation obtained in the Current Population Surveywith education on the death certificate for decedents inthe National Longitudinal Mortality Survey (NLMS).(Sorlie PD, Johnson NJ: Validity of educationinformation on the death certificate,Epidemiology7(4):437–439, 1996.) Another analysis comparedself-reported education collected in the first NationalHealth and Nutrition Examination Survey (NHANES I)with education on the death certificate for decedents inthe NHANES I Epidemiologic Followup Study.(Makuc DM, Feldman JJ, Mussolino ME: Validity ofeducation and age as reported on death certificates,American Statistical Association 1996 Proceedings ofthe Social Statistics Section, 102–6, 1997.) Results ofboth studies indicated that there is a tendency for somepeople who did not graduate from high school to bereported as high school graduates on the deathcertificate. This tendency results in overstating thedeath rate for high school graduates and understatingthe death rate for the group with less than 12 years ofeducation. The bias was greater among older thanyounger decedents and somewhat greater among blackthan white decedents.

....................................................................... 107

ion

nsof

oofat

to

ereofn

o

a

s

t

g

s,)

,

Toith

s

108 .......

Appendix I ..............................................................................................................So

urce

sof

Dat

a

In addition, educational gradients in death ratesbased on the National Vital Statistics System werecompared with those based on the NLMS, aprospective study of persons in the Current PopulatSurvey. Results of these comparisons indicate thateducational gradients in death rates based on theNational Vital Statistics System were reasonablysimilar to those based on the NLMS for white perso25–64 years of age and black persons 25–44 yearsage. The number of deaths for persons of Hispanicorigin in the NLMS was too small to permitcomparison for this ethnic group.

Hispanic deaths—In 1985 mortality data byHispanic origin of decedent were based on deaths tresidents of the following 17 States and the DistrictColumbia whose data on the death certificate wereleast 90 percent complete on a place-of-occurrencebasis and of comparable format: Arizona, Arkansas,California, Colorado, Georgia, Hawaii, Illinois,Indiana, Kansas, Mississippi, Nebraska, New York,North Dakota, Ohio, Texas, Utah, and Wyoming. In1986 New Jersey began reporting Hispanic origin ofdecedent, increasing the number of reporting States18 and the District of Columbia in 1986 and 1987. In1988 Alabama, Kentucky, Maine, Montana, NorthCarolina, Oregon, Rhode Island, and Washington wadded to the reporting area, increasing the numberStates to 26 and the District of Columbia. In 1989 aadditional 18 States were added, increasing theHispanic reporting area to 44 States and the DistrictColumbia. In 1989 only Connecticut, Louisiana,Maryland, New Hampshire, Oklahoma, and Virginiawere not included in the reporting area. Starting with1990 data in this book, the criterion was changed toinclude States whose data were at least 80 percentcomplete. In 1990 Maryland, Virginia, andConnecticut, in 1991 Louisiana, and in 1993 NewHampshire were added, increasing the reporting arefor Hispanic origin of decedent to 47 States and theDistrict of Columbia in 1990, 48 States and theDistrict of Columbia in 1991 and 1992, and 49 Stateand the District of Columbia in 1993–96. OnlyOklahoma did not provide this information in

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

f

1993–96. Starting in 1997 Hispanic origin of decedenwas reported by all 50 States and the District ofColumbia. Based on data from the U.S. Bureau of theCensus, the 1990 reporting area encompassed99.6 percent of the U.S. Hispanic population. In 1990more than 96 percent of death records includedinformation on origin of decedent.

Alaska data—For 1995 the number of deathsoccurring in Alaska is in error for selected causesbecause NCHS did not receive changes resulting fromamended records and because of errors in processinthe cause of death data. Differences are concentratedamong selected causes of death, principally Symptomsigns, and ill-defined conditions (ICD-9 Nos. 780–799and external causes.

For more information, see: National Center forHealth Statistics,Technical Appendix, Vital Statistics ofthe United States, 1992, Vol. I, Natality, DHHS Pub.No. (PHS)96–1100 and Vol. II, Mortality, Part A,DHHS Pub. No. (PHS) 96–1101, Public HealthService. Washington. U.S. Government Printing Office1996.

National Linked File of Live Births and Infant Deaths

National linked files of live births and infantdeaths are data sets for research on infant mortality.create these data sets, death certificates are linked wcorresponding birth certificates for infants who die inthe United States before their first birthday. Linkeddata files include all of the variables on the nationalnatality file, including the more accurate racial andethnic information, as well as the variables on thenational mortality file, including cause of death andage at death. The linkage makes available for theanalysis of infant mortality extensive information fromthe birth certificate about the pregnancy, maternal riskfactors, and infant characteristics and health items atbirth. Each year, 97–98 percent of infant death recordare linked to their corresponding birth records.

National linked files of live births and infantdeaths were first produced for the 1983 birth cohort.Birth cohort linked file data are available for 1983–91and period linked file data for 1995 and 1996. While

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

e

ex

a

2

4

ors

d

alire.

d.

d

’s,

1


ata

birth cohort linked files have methodologicaladvantages, their production incurs substantial delayin data availability, since it is necessary to wait untilthe close of a second data year to include all infantdeaths to the birth cohort. Starting with data year1995, more timely linked file data are produced in aperiod data format, preceding the release of thecorresponding birth cohort format. Other changes tothe data set starting with 1995 data include theaddition of record weights to correct for the2.2–2.5 percent of records that could not be linked athe addition of an imputation for not statedbirthweight. For more information, see: Prager K.Infant mortality by birthweight and othercharacteristics: United States, 1985 birth cohort.National Center for Health Statistics. Vital Health Sta20(24). 1994; MacDorman MF, Atkinson JO. Infantmortality statistics from the 1996 period linkedbirth/death data set. Monthly vital statistics report; vo46 no 12, supp. Hyattsville, Maryland: National Centfor Health Statistics. 1998.

Compressed Mortality File

The Compressed Mortality File (CMF) used tocompute death rates by urbanization level is a countlevel national mortality and population database. Themortality data base of CMF is derived from thedetailed mortality files of the National Vital StatisticsSystem starting with 1968. The population data baseCMF is derived from intercensal and postcensalpopulation estimates and census counts of the residpopulation of each U.S. county by age, race, and seCounties are categorized according to level ofurbanization based on an NCHS-modified version ofthe 1993 rural-urban continuum codes for metropolitand nonmetropolitan counties developed by theEconomic Research Service, U.S. Department ofAgriculture. SeeAppendix II, Urbanization. For moreinformation about the CMF, contact: D. Ingram,Analytic Studies Branch, Division of Health andUtilization Analysis, National Center for HealthStatistics, 6525 Belcrest Road, Hyattsville, MD 2078

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

s

nd

t

lr

y

of

nt.

n

.

National Survey of Family Growth

Data from the National Survey of Family Growth(NSFG) are based on samples of women ages 15–4years in the civilian noninstitutionalized population ofthe United States. The first and second cycles,conducted in 1973 and 1976, excluded most womenwho had never been married. The third, fourth, andfifth cycles, conducted in 1982, 1988, and 1995,included all women ages 15–44 years.

The purpose of the survey is to provide nationaldata on factors affecting birth and pregnancy rates,adoption, and maternal and infant health. These factinclude sexual activity, marriage, divorce andremarriage, unmarried cohabitation, contraception ansterilization, infertility, breastfeeding, pregnancy loss,low birthweight, and use of medical care for familyplanning and infertility.

Interviews are conducted in person by professionfemale interviewers using a standardized questionnaIn 1973–88 the average interview length was about 1hour. In 1995 the average interview lasted about 1hour and 45 minutes. In all cycles black women weresampled at higher rates than white women, so thatdetailed statistics for black women could be produce

Interviewing for Cycle 1 of NSFG was conductedfrom June 1973 to February 1974. Counties andindependent cities of the United States were sampleto form a frame of primary sampling units (PSU’s),and 101 PSU’s were selected. From these 101 PSU10,879 women 15–44 years of age were selected;9,797 of these were interviewed. Most never-marriedwomen were excluded from the 1973 NSFG.

Interviewing for Cycle 2 of NSFG was conductedfrom January to September 1976. From 79 PSU’s,10,202 eligible women were identified; of these, 8,61were interviewed. Again, most never-married womenwere excluded from the sample for the 1976 NSFG.

Interviewing for Cycle 3 of NSFG was conductedfrom August 1982 to February 1983. The sampledesign was similar to that in Cycle 2: 31,027households were selected in 79 PSU’S. Householdscreener interviews were completed in 29,511households (95.1 percent). Of the 9,964 eligible

....................................................................... 109

sl

t

5

f

w

t

i

r

lindsfd

l

e

r

e

th

110 ........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

women identified, 7,969 were interviewed. For the firtime in NSFG, Cycle 3 included women of all maritastatuses.

Interviewing for Cycle 4 was conducted betweenJanuary and August 1988. The sample was obtainedfrom households that had been interviewed in theNational Health Interview Survey in the 18 monthsbetween October 1, 1985, and March 31, 1987. Forfirst time, women living in Alaska and Hawaii wereincluded so that the survey covered women from thenoninstitutionalized population of the entire UnitedStates. The sample was drawn from 156 PSU’s; 10,eligible women ages 15–44 years were sampled.Interviews were completed with 8,450 women.

Between July and November of 1990, 5,686women were interviewed by telephone in the firstNSFG telephone reinterview. The average length ofinterview in 1990 was 20 minutes. The response ratefor the 1990 telephone reinterview was 68 percent othose responding to the 1988 survey and still eligiblefor the 1990 survey.

Interviewing for Cycle 5 of NSFG was conductedbetween January and October of 1995. The sampleobtained from households that had been interviewed198 PSU’s in the National Health Interview Survey in1993. Of the 13,795 eligible women in the sample,10,847 were interviewed. For the first time, Hispanicas well as black women were sampled at a higher rathan other women.

In order to make national estimates from thesample for the millions of women ages 15–44 yearsthe United States, data for the interviewed samplewomen were (a) inflated by the reciprocal of theprobability of selection at each stage of sampling (foexample, if there was a 1 in5,000 chance that awoman would be selected for the sample, her sampweight was 5,000), (b) adjusted for nonresponse, an(c) forced to agree with benchmark population valuebased on data from the Current Population Survey othe U.S. Bureau of the Census (this last step is calle‘‘poststratification’’).

Quality control procedures for selecting andtraining interviewers, coding, editing, and processing

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

t

he

66

asin

e

n

g

the data, were built into NSFG to minimizenonsampling error.

More information on the methodology of NSFG isavailable in the following reports: French DK. NationaSurvey of Family Growth, Cycle I: Sample design,estimation procedures, and variance estimation.National Center for Health Statistics. Vital Health Stat2(76). 1978; Grady WR. National Survey of FamilyGrowth, Cycle II: Sample design, estimationprocedures, and variance estimation. National Centerfor Health Statistics. Vital Health Stat 2(87). 198l;Bachrach CA, Horn MC, Mosher WD, Shimizu I.National Survey of Family Growth, Cycle III: Sampledesign, weighting, and variance estimation. NationalCenter for Health Statistics. Vital Health Stat 2(98).1985; Judkins DR, Mosher WD, Botman SL. NationalSurvey of Family Growth: Design, estimation, andinference. National Center for Health Statistics. VitalHealth Stat 2(109). 1991; Goksel H, Judkins DR,Mosher WD. Nonresponse adjustments for a telephonfollowup to a National In-Person Survey. Journal ofOfficial Statistics 8(4):417–32. 1992; Kelly JE, MosherWD, Duffer AP, Kinsey SH. Plan and operation of the1995 National Survey of Family Growth. Vital HealthStat 1(36). 1997; Potter FJ, Iannacchione VG, MosheWD, Mason RE, Kavee JD. Sampling weights,imputation, and variance estimation in the 1995National Survey of Family Growth. Vital Health Stat2(124). 1998.

National Health Interview Survey

The National Health Interview Survey (NHIS) is acontinuing nationwide sample survey in which data arcollected through personal household interviews.Information is obtained on personal and demographiccharacteristics including race and ethnicity byself-reporting or as reported by an informant.Information is also obtained on illnesses, injuries,impairments, chronic conditions, utilization of healthresources, and other health topics. The householdquestionnaire is reviewed each year with special healtopics being added or deleted. For most health topicsdata are collected over an entire calendar year.

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

un

kly

e

d

g

jo

a

0t

4

.e

a

e

5r

ta

s

f7

eey

,

or


ata

The sample design plan of NHIS follows amultistage probability design that permits a continuosampling of the civilian noninstitutionalized populatioresiding in the United States. The survey is designedsuch a way that the sample scheduled for each weerepresentative of the target population and the weeksamples are additive over time. The response rate fothe ongoing portion of the survey (core) has beenbetween 94 and 98 percent over the years. Responsrates for special health topics (supplements) havegenerally been lower. For example the response ratewas 80 percent for the 1994 Year 2000 Supplement,which included questions about cigarette smoking anuse of such preventive services as mammography.

In 1985 NHIS adopted several new sample desifeatures although, conceptually, the sampling planremained the same as the previous design. Two machanges included reducing the number of primarysampling locations from 376 to 198 for samplingefficiency and oversampling the black population toimprove the precision of the statistics. The sample wdesigned so that a typical NHIS sample for the datacollection years 1985–94 consisted of approximately7,500 segments containing about 59,000 assignedhouseholds. Of these households, an expected 10,0were vacant, demolished, or occupied by persons noin the target population of the survey. The expectedsample of 49,000 occupied households yielded aprobability sample of about 127,000 persons. In 199there was a sample of 116,179 persons.

In 1995 the NHIS sample was redesigned againMajor design changes included increasing the numbof primary sampling units from 198 to 358 andoversampling the black and Hispanic populations toimprove the precision of the statistics. The sample wdesigned so that a typical NHIS sample for the datacollection years 1995–2004 will consist ofapproximately 7,000 segments. The expected sampl44,000 occupied respondent households will yield aprobability sample of about 106,000 persons. In 199there was a sample of 102,467 persons. In 1996 thewas a smaller sample of 63,402 persons because pa

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

s

inis

r

n

r

s

0

r

s

of

ert

of the sample was reserved for use in testing the newquestionnaire instrument (1997).

In 1997 the questionnaire was redesigned and dawere collected using a computer assisted personalinterview (CAPI). The CAPI instrument wasadministered using a laptop computer with interviewerentering responses directly in the computer during theinterview. In 1997 the interviewed sample consisted o39,832 households yielding 40,623 families or 103,47persons. Because of the extensive redesign of thequestionnaire and the introduction of the CAPI methodof data collection, 1997 data may differ from earlieryears.

A description of the survey design, the methodsused in estimation, and the general qualifications of thdata obtained from the survey are presented in: MassJT, Moore TF, Parsons VL, Tadros W. Design andestimation for the National Health Interview Survey,1985–94. National Center for Health Statistics. VitalHealth Stat 2(110). 1989; Kovar MG, Poe GS. TheNational Health Interview Survey design, 1973–84,and procedures, 1975–83. National Center for HealthStatistics. Vital Health Stat 1(18). 1985; Hendershot GAdams P, Marano M, Benaissa S. Current estimatesfrom the National Health Interview Survey, 1996.National Center for Health Statistics. Vital Health Stat10(200). 1999.

National Immunization Survey

The National Immunization Survey (NIS) is acontinuing nationwide telephone sample surveytogather data on children 19–35 months of age.Estimates of vaccine-specific coverage are available fnational, State, and 28 urban areas considered to behigh risk for undervaccination.

NIS uses a two-phase sample design. First, arandom-digit-dialing (RDD) sample of telephonenumbers is drawn. When households with age-eligiblechildren are contacted, the interviewer collectsinformation on the vaccinations received by allage-eligible children. In 1997 the overall response ratewas 69 percent, yielding data for 32,742 children aged19–35 months. The interviewer also collects

....................................................................... 111

ni

d

y

u

e

n

n

,:

al

l

d

,

112 ........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

information on the vaccination providers. In the secophase, all vaccination providers are contacted by maVaccination information from providers was obtainedfor 68 percent of all children who were eligible forprovider followup in 1997. Providers’ responses arecombined with information obtained from thehouseholds to provide a more accurate estimate ofvaccination coverage levels. Final estimates areadjusted for noncoverage of nontelephone househol

A description of the survey design and themethods used in estimation are presented in: MasseJT. Estimating the response rate in a two stagetelephone survey. Proceedings of the Section onSurvey Research Methods. Alexandria, Virginia:American Statistical Association. 1995.

National Health and Nutrition Examination Survey

For the first program or cycle of the NationalHealth Examination Survey (NHES I), 1960–62, datawere collected on the total prevalence of certainchronic diseases as well as the distributions of variophysical and physiological measures, including bloodpressure and serum cholesterol levels. For thatprogram, a highly stratified, multistage probabilitysample of 7,710 adults, of whom 86.5 percent wereexamined, was selected to represent the 111 millioncivilian noninstitutionalized adults 18–79 years of agin the United States at that time. The sample areasconsisted of 42 primary sampling units (PSU’s) fromthe 1,900 geographic units.

NHES II (1963–65) and NHES III (1966–70)examined probability samples of the nation’snoninstitutionalized children between the ages of 6 a11 years (NHES II) and 12 and 17 years (NHES III)focusing on factors related to growth and developmeBoth cycles were multistage, stratified probabilitysamples of clusters of households in land-basedsegments and used the same 40 PSU’s. NHES IIsampled 7,417 children with a response rate of96 percent. NHES III sampled 7,514 youth with aresponse rate of 90 percent.

For more information on NHES I, see: Gordon TMiller HW. Cycle I of the Health Examination Survey

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

dl.

s.

s

d

t.

Sample and response, United States, 1960–62. NationCenter for Health Statistics. Vital Health Stat 11(1).1974. For more information on NHES II, see: Plan,operation, and response results of a program ofchildren’s examinations. National Center for HealthStatistics. Vital Health Stat 1(5). 1967. For moreinformation on NHES III, see: Schaible, WL. Qualitycontrol in a National Health Examination Survey.National Center for Health Statistics. Vital Health Stat2(44). 1972.

In 1971 a nutrition surveillance component wasadded and the survey name was changed to theNational Health and Nutrition Examination Survey(NHANES). In NHANES I, conducted from 1971 to1974, a major purpose was to measure and monitorindicators of the nutrition and health status of theAmerican people through dietary intake data,biochemical tests, physical measurements, and clinicaassessments for evidence of nutritional deficiency.Detailed examinations were given by dentists,ophthalmologists, and dermatologists with anassessment of need for treatment. In addition, datawere obtained for a subsample of adults on overallhealth care needs and behavior, and more detailedexamination data were collected on cardiovascular,respiratory, arthritic, and hearing conditions.

The NHANES I target population was the civiliannoninstitutionalized population 1–74 years of ageresiding in the coterminous United States, except forpeople residing on any of the reservation lands setaside for the use of American Indians. The sampledesign was a multistage, stratified probability sampleof clusters of persons in land-based segments. Thesample areas consisted of 65 PSU’s selected from the1,900 PSU’s in the coterminous United States. Asubsample of persons 25–74 years of age was selecteto receive the more detailed health examination.Groups at high risk of malnutrition were oversampledat known rates throughout the process. Householdinterviews were completed for more than 96 percent ofthe 28,043 persons selected for the NHANES I sampleand about 75 percent (20,749) were examined.

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

fn

or

s

n

e

n

S

lral

an

l

ent

h

rs


ata

For NHANES II, conducted from 1976 to 1980,the nutrition component was expanded from the onefielded for NHANES I. In the medical area primaryemphasis was placed on diabetes, kidney and liverfunctions, allergy, and speech pathology. TheNHANES II target population was the civiliannoninstitutionalized population 6 months–74 years oage residing in the United States, including Alaska aHawaii.

NHANES II utilized a multistage probabilitydesign that involved selection of PSU’s, segments(clusters of households) within PSU’s, households,eligible persons, and finally, sample persons. Thesample design provided for oversampling among thpersons 6 months–5 years of age, those 60–74 yeaage, and those living in poverty areas. A sample of27,801 persons was selected for NHANES II. Of thisample 20,322 (73.1 percent) were examined. Raceinformation for NHANES I and NHANES II wasdetermined primarily by interviewer observation.

The estimation procedure used to produce natiostatistics for NHANES I and NHANES II involvedinflation by the reciprocal of the probability ofselection, adjustment for nonresponse, andpoststratified ratio adjustment to population totals.Sampling errors also were estimated to measure threliability of the statistics.

For more information on NHANES I, see: MillerHW. Plan and operation of the Health and NutritionExamination Survey, United States, 1971–73. NatioCenter for Health Statistics. Vital Health Stat 1(10a)and 1(10b). 1977 and 1978; and Engel A, Murphy RMaurer K, Collins E. Plan and operation of theNHANES I Augmentation Survey of Adults 25–74years, United States 1974–75. National Center forHealth Statistics. Vital Health Stat 1(14). 1978.

For more information on NHANES II, see:McDowell A, Engel A, Massey JT, Maurer K. Planand operation of the second National Health andNutrition Examination Survey, 1976–80. NationalCenter for Health Statistics. Vital Health Stat 1(15).1981. For information on nutritional applications ofthese surveys, see: Yetley E, Johnson C. 1987.

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

d

ses of

al

al

,

Nutritional applications of the Health and NutritionExamination Surveys (HANES). Ann Rev Nutr7:441–63.

The Hispanic Health and Nutrition ExaminationSurvey (HHANES), conducted during 1982–84, wassimilar in content and design to the previous NationaHealth and Nutrition Examination Surveys. The majodifference between HHANES and the previous nationsurveys is that HHANES employed a probabilitysample of three special subgroups of the populationliving in selected areas of the United States rather tha national probability sample. The three HHANESuniverses included approximately 84, 57, and59 percent of the respective 1980 Mexican-, Cuban-,and Puerto Rican-origin populations in the continentaUnited States. The Hispanic ethnicity of thesepopulations was determined by self-report.

In the HHANES three geographically andethnically distinct populations were studied: MexicanAmericans living in Texas, New Mexico, Arizona,Colorado, and California; Cuban Americans living inDade County, Florida; and Puerto Ricans living inparts of New York, New Jersey, and Connecticut. Inthe Southwest 9,894 persons were selected (75 percor 7,462 were examined), in Dade County 2,244persons were selected (60 percent or 1,357 wereexamined), and in the Northeast 3,786 persons wereselected (75 percent or 2,834 were examined).

For more information on HHANES, see: MaurerKR. Plan and operation of the Hispanic Health andNutrition Examination Survey, 1982–84. NationalCenter for Health Statistics. Vital Health Stat 1(19).1985.

The third National Health and NutritionExamination Survey (NHANES III) is a 6-year surveycovering the years 1988–94. Over the 6-year period,39,695 persons were selected for the survey of whic30,818 (77.6 percent) were examined in the mobileexamination center.

The NHANES III target population is the civiliannoninstitutionalized population 2 months of age andover. The sample design provides for oversamplingamong children 2–35 months of age, persons 70 yea

........................................................................ 113

e

g

r

t

c

)i

as

o

).

).y

n

a

iests

ren

te

d

114 ........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

of age and over, black Americans, and MexicanAmericans. Race is reported for the household by threspondent.

Although some of the specific health areas havechanged from earlier NHANES surveys, the followingoals of the NHANES III are similar to those ofearlier NHANES surveys:

to estimate the national prevalence of selecteddiseases and risk factors

to estimate national population referencedistributions of selected health parameters

to document and investigate reasons for seculatrends in selected diseases and risk factors

Two new additional goals for the NHANES III surveyare:

to contribute to an understanding of diseaseetiology

to investigate the natural history of selecteddiseases

For more information on NHANES III, see: EzzaTM, Massey JT, Waksberg J, et al. Sample design:Third National Health and Nutrition ExaminationSurvey. National Center for Health Statistics. VitalHealth Stat 2(113). 1992; Plan and operation of theThird National Health and Nutrition ExaminationSurvey, 1988–94. National Center for Health StatistiVital Health Stat 1(32). 1994.

National Health Provider Inventory (National MasterFacility Inventory)

The National Master Facility Inventories (NMFI’swere a series of surveys of inpatient health facilitiesthe United States. They included hospitals, nursingrelated care homes, and other custodial care facilitieThe last NMFI was conducted in 1982. In 1986 adifferent inventory was conducted, the Inventory ofLong-Term Care Places (ILTCP). This was a surveynursing and related care homes and facilities for thementally retarded. In 1991 the National HealthProvider Inventory (NHPI), which was a survey of

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

i

s.

nnd.

f

nursing homes, board and care homes, home healthagencies, and hospices, was conducted. The NMFI,ILTCP, and NHPI were used as a basis for samplingframes for other surveys conducted by the NationalCenter for Health Statistics (National Nursing HomeSurvey and National Home and Hospice Care Survey

National Home and Hospice Care Survey

The National Home and Hospice Care Survey(NHHCS) is a sample survey of health agencies andhospices. Initiated in 1992, it was also conducted in1993, 1994, and 1996. The original sampling frameconsisted of all home health care agencies andhospices identified in the 1991 National HealthProvider Inventory (NHPI). The 1992 samplecontained 1,500 agencies. These agencies wererevisited during the 1993 survey (excluding agenciesthat had been found to be out of scope for the surveyIn 1994 in-scope agencies identified in the 1993 survewere revisited, with 100 newly identified agenciesadded to the sample. For 1996 the universe was agaiupdated and a new sample of 1,200 agencies wasdrawn.

The sample design for the 1992–94 NHHCS wasstratified three-stage probability design. Primarysampling units were selected at the first stage, agencwere selected at the second stage, and current patienand discharges were selected at the third stage. Thesample design for the 1996 NHHCS has a two-stageprobability design in which agencies were selected atthe first stage and current patients and discharges weselected at the second stage. Current patients were othe rolls of the agency as of midnight on the daybefore the survey. Discharges were selected to estimathe number of discharges from the agency during theyear before the survey.

After the samples were selected, a patientquestionnaire was completed for each current patientand discharge by interviewing the staff member mostfamiliar with the care provided to the patients. Therespondent was requested to refer to the medicalrecords for each patient. For additional informationsee: Haupt BJ. Development of the National Home an

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

a

n

h

o

g

l

em

e

s

g

ha

h

n

d

a

r

t


ata

Hospice Care Survey. National Center for HealthStatistics. Vital Health Stat 1(33). 1994.

National Hospital Discharge Survey

The National Hospital Discharge Survey (NHDS)is a continuing nationwide sample survey of short-sthospitals in the United States. The scope of NHDSencompasses patients discharged from noninstitutionhospitals, exclusive of military and Department ofVeterans Affairs hospitals, located in the 50 States athe District of Columbia. Only hospitals having six ormore beds for patient use are included in the surveyand before 1988 those in which the average length ostay for all patients was less than 30 days. In 1988 tscope was altered slightly to include all general andchildren’s general hospitals regardless of the lengthstay. Although all discharges of patients from thesehospitals are within the scope of the survey, discharof newborn infants from all hospitals are excludedfrom this report.

The original sample was selected in 1964 from aframe of short-stay hospitals listed in the NationalMaster Facility Inventory. A two-stage stratified sampdesign was used, and hospitals were stratifiedaccording to bed size and geographic region. Samplhospitals were selected with probabilities ranging frocertainty for the largest hospitals to 1 in 40 for thesmallest hospitals. Within each sample hospital, asystematic random sample of discharges was selectfrom the daily listing sheet. Initially, thewithin-hospital sampling rates for selecting dischargevaried inversely with the probability of hospitalselection so that the overall probability of selecting adischarge was approximately the same across thesample. Those rates were adjusted for individualhospitals in subsequent years to control the reportinburden of those hospitals.

In 1985, for the first time, two data collectionprocedures were used for the survey. The first was ttraditional manual system of sample selection and dabstraction. In the manual system, sample selectionand transcription of information from the hospitalrecords to abstract forms were performed by either t

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

y

al

d

fe

f

es

e

d

eta

e

hospital staff or representatives of NCHS or both. Thesecond was an automated method, used inapproximately 17 percent of the sample hospitals in1985, involving the purchase of data tapes fromcommercial abstracting services. These tapes were thesubjected to NCHS sampling, editing, and weightingprocedures.

In 1988 NHDS was redesigned. The hospitals withthe most beds and/or discharges annually were selectewith certainty, but the remaining sample was selectedusing a three-stage stratified design. The first stage issample of PSU’s used by the National HealthInterview Survey. Within PSU’s, hospitals werestratified or arrayed by abstracting status (whethersubscribing to a commercial abstracting service) andwithin abstracting status arrayed by type of service andbed size. Within these strata and arrays, a systematicsampling scheme with probability proportional to theannual number of discharges was used to selecthospitals. The rates for systematic sampling ofdischarges within hospitals vary inversely withprobability of hospital selection within PSU. Dischargerecords from hospitals submitting data via commercialabstracting services and selected State data systems(approximately 38 percent of sample hospitals in 1996)were arrayed by primary diagnoses, patient sex andage group, and date of discharge before sampling.Otherwise, the procedures for sampling dischargeswithin hospitals are the same as those used in the priodesign.

In 1994 the hospital sample was updated bycontinuing the sampling process among hospitals thatwere NHDS-eligible for the sampling frame in 1994but not in 1991. The additional hospitals were added athe end of the list for the strata where they belonged,and the systematic sampling was continued as if theadditional hospitals had been present during the initialsample selection. Hospitals that were no longerNHDS-eligible were deleted. A similar updatingprocess occurred in 1991.

The basic unit of estimation for NHDS is thesample patient abstract. The estimation procedureinvolves inflation by the reciprocal of the probability

....................................................................... 115

re

J,

h

raas

9

r

thrdas

k,a

to

,l

e

116 .......

Appendix I ..............................................................................................................So

urce

sof

Dat

a

of selection, adjustment for nonresponding hospitalsand missing abstracts, and ratio adjustments to fixedtotals. In 1996, 525 hospitals were selected, 507 wewithin scope, 480 participated, and 282,000 medicalrecords were abstracted.

For more detailed information on the design ofNHDS and the magnitude of sampling errorsassociated with the NHDS estimates, see: Graves EOwings MF. 1996 Summary: National HospitalDischarge Survey. Advance data from vital and healtstatistics; no 301. Hyattsville, Maryland: NationalCenter for Health Statistics. 1998; and Haupt BJ,Kozak LJ. Estimates from two survey designs:National Hospital Discharge Survey. National Centerfor Health Statistics. Vital Health Stat 13(111). 1992.

National Survey of Ambulatory Surgery

The National Survey of Ambulatory Surgery(NSAS) is a nationwide sample survey of ambulatorysurgery patient discharges from short-stay non-Fedehospitals and freestanding surgery centers. NSAS wconducted annually between 1994 and 1996. Thesample consisted of eligible hospitals listed in the 19SMG Hospital Market Database and the 1993 SMGFreestanding Outpatient Surgery Center Database oMedicare Provider-of-Service files. Facilitiesspecializing in dentistry, podiatry, abortion, familyplanning, or birthing were excluded.

A three-state stratified cluster design was used,and facilities were stratified according to primarysampling unit (PSU). The second stage consisted ofselection of facilities from sample PSU’s, and the thistage consisted of a systematic random sample of cfrom all locations within a facility where ambulatorysurgery was performed. Locations within hospitalsdedicated exclusively to dentistry, podiatry, pain blocabortion, or small procedures (sometimes referred to‘‘lump and bump’’ rooms) were not included. In 1996of the 751 hospitals and freestanding ambulatorysurgery centers selected for the survey, 601 werein-scope, and 488 responded for an overall responserate of 81 percent. These facilities providedinformation for approximately 125,000 ambulatory

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

l

3

e

es

s

surgery discharges. Up to six procedures were codedthe International Classification of Diseases, 9thRevision, Clinical Modification. Estimates were derivedusing a multistage estimation procedure: inflation byreciprocals of the probabilities of selection; adjustmentfor nonresponse; and population weighting ratioadjustments.

For more detailed information on the design ofNSAS, see: McLemore T, Lawrence L. Plan andOperation of the National Survey of AmbulatorySurgery. National Center for Health Statistics. VitalHealth Stat 1(37). 1997.

National Nursing Home Survey

NCHS has conducted five National Nursing HomeSurveys. The first survey was conducted from August1973 to April 1974; the second survey from May 1977to December 1977; the third from August 1985 toJanuary 1986; the fourth from July 1995 to December1995; and the fifth from July 1997 to December 1997.

Much of the background information andexperience used to develop the first National NursingHome Survey was obtained from a series of three adhoc sample surveys of nursing and personal carehomes called the Resident Places Surveys (RPS-1, -2-3). The three surveys were conducted by the NationaCenter for Health Statistics during April–June 1963,May–June 1964, and June–August 1969. During thefirst survey, RPS-1, data were collected on nursinghomes, chronic disease and geriatric hospitals, nursinghome units, and chronic disease wards of general andmental hospitals. RPS-2 concentrated mainly onnursing homes and geriatric hospitals. During the thirdsurvey, RPS-3, nursing and personal care homes in thcoterminous United States were sampled.

For the initial National Nursing Home Survey(NNHS) conducted in 1973–74, the universe includedonly those nursing homes that provided some level ofnursing care. Homes providing only personal ordomiciliary care were excluded. The sample of 2,118homes was selected from the 17,685 homes thatprovided some level of nursing care and were listed inthe 1971 National Master Facility Inventory (NMFI) orthose that opened for business in 1972. Data were

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

n

wn

si

al

e

,

f


ata

obtained from about 20,600 staff and 19,000 resideResponse rates were 97 percent for facilities,88 percent for expenditures, 98 percent for residentsand 82 percent for staff.

The scope of the 1977 NNHS encompassed alltypes of nursing homes, including personal care anddomiciliary care homes. The sample of about 1,700facilities was selected from 23,105 nursing homes inthe sampling frame, which consisted of all homeslisted in the 1973 NMFI and those opening forbusiness between 1973 and December 1976. Dataobtained from about 13,600 staff, 7,000 residents, a5,100 discharged residents. Response rates were95 percent for facilities, 85 percent for expenses,81 percent for staff, 99 percent for residents, and97 percent for discharges.

The scope of the 1985 NNHS was similar to the1973–74 survey in that it excluded personal ordomiciliary care homes. The sample of 1,220 homewas selected from a sampling frame of 20,479 nursand related care homes. The frame consisted of allhomes in the 1982 NMFI; homes identified in the 19Complement Survey of NMFI as ’’missing‘‘ from the1982 NMFI; facilities that opened for businessbetween 1982 and June 1984; and hospital-basednursing homes obtained from the Health CareFinancing Administration. Information on the facilitywas collected through a personal interview with theadministrator. Accountants were asked to completequestionnaire on expenditures or provide a financiastatement. Resident data were provided by a nursefamiliar with the care provided to the resident. Thenurse relied on the medical record and personalknowledge of the resident. In addition to employeedata that were collected during the interview with thadministrator, a sample of registered nurses complea self-administered questionnaire. Discharge data wbased on information recorded in the medical recordAdditional data about the current and dischargedresidents were obtained in telephone interviews withnext of kin. Data were obtained from 1,079 facilities2,763 registered nurses, 5,243 current residents, an6,023 discharges. Response rates were 93 percent

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

ts.

,

ered

ng

82

tedere.

dor

facilities, 68 percent for expenses, 80 percent forregistered nurses, 97 percent for residents, 95 percentfor discharges, and 90 percent for next of kin.

The scope of the 1995 and 1997 NNHS wassimilar to the 1985 and the 1973–74 NNHS in thatthey included only nursing homes that provided somelevel of nursing care. Homes providing only personalor domiciliary care were excluded. The 1995 sample of1,500 homes was selected from a sampling frame of17,500 nursing homes. The frame consisted of anupdated version of the 1991 National Health ProviderInventory (NHPI). Data were obtained from about1,400 nursing homes and 8,000 current residents. Dataon current residents were provided by a staff memberfamiliar with the care received by residents and frominformation contained in resident’s medical records.

The 1997 sample of 1,488 nursing homes was thesame basic sample used in 1995. Excluded wereout-of-scope and out-of-business places identified inthe 1995 survey and included were a small number ofadditions to the sample from a supplemental frame ofplaces not in the 1995 frame. The 1997 NNHSincluded the discharge component not available in the1995 survey.

Statistics for all five surveys were derived by aratio-estimation procedure. Statistics were adjusted forfailure of a home to respond, failure to fill out one ofthe questionnaires, and failure to complete an item ona questionnaire.

For more information on the 1973–74 NNHS, see:Meiners MR. Selected operating and financialcharacteristics of nursing homes, United States,1973–74 National Nursing Home Survey. NationalCenter for Health Statistics. Vital Health Stat 13(22).1975. For more information on the 1977 NNHS, see:Van Nostrand JF, Zappolo A, Hing E, et al. TheNational Nursing Home Survey, 1977 summary for theUnited States. National Center for Health Statistics.Vital Health Stat 13(43). 1979. For more informationon the 1985 NNHS, see: Hing E, Sekscenski E,Strahan G. The National Nursing Home Survey: 1985summary for the United States. National Center forHealth Statistics. Vital Health Stat 13(97). 1985. For

........................................................................ 117

.

e;

in

ed

e‘gens

.its

,

d

m

an

sw

.

ts

.n

to

:

at

o

118 ........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

more information on the 1995 NNHS, see: Strahan GAn overview of nursing homes and their currentresidents: Data from the 1995 National Nursing HomSurvey. Advance data from vital and health statisticsno 280. Hyattsville, Maryland: National Center forHealth Statistics. 1997. For more information on the1997 NNHS, see the Advance Data report availablethe summer of 1999.

National Ambulatory Medical Care Survey

The National Ambulatory Medical Care Survey(NAMCS) is a continuing national probability sampleof ambulatory medical encounters. The scope of thesurvey covers physician-patient encounters in theoffices of non-Federally employed physicians classifiby the American Medical Association or AmericanOsteopathic Association as ’’office-based, patient carphysicians. Patient encounters with physicians engain prepaid practices (health maintenance organizatio(HMO’s), independent practice organizations (IPA’s),and other prepaid practices) are included in NAMCSExcluded are visits to hospital-based physicians, visto specialists in anesthesiology, pathology, andradiology, and visits to physicians who are principallyengaged in teaching, research, or administration.Telephone contacts and nonoffice visits are excludedalso.

A multistage probability design is employed. Thefirst-stage sample consists of 84 primary samplingunits (PSU’s) in 1985 and 112 PSU’s in 1992 selectefrom about 1,900 such units into which the UnitedStates has been divided. In each sample PSU, a saof practicing non-Federal office-based physicians isselected from master files maintained by the AmericMedical Association and the American OsteopathicAssociation. The final stage involves systematicrandom samples of office visits during randomlyassigned 7-day reporting periods. In 1985 the surveyexcluded Alaska and Hawaii. Starting in 1989 thesurvey included all 50 States.

For the 1997 survey a sample of 2,498 physicianwas selected. The physician response rate for 199769 percent, providing data on 24,715 records.

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

‘d

ple

as

The estimation procedure used in NAMCSbasically has three components: inflation by thereciprocal of the probability of selection, adjustmentfor nonresponse, and ratio adjustment to fixed totals

For more detailed information on NAMCS, see:Woodwell, DA. National Ambulatory Medical CareSurvey: 1997 summary. Advance data from vital andhealth statistics; no 305. Hyattsville, Maryland:National Center for Health Statistics. 1999.

National Hospital Ambulatory Medical Care Survey

The National Hospital Ambulatory Medical CareSurvey (NHAMCS), initiated in 1992, is a continuingannual national probability sample of visits by patiento emergency departments (ED’s) and outpatientdepartments (OPD’s) of non-Federal, short-stay, orgeneral hospitals. Telephone contacts are excluded

A four-stage probability sample design is used iNHAMCS, involving samples of primary samplingunits (PSU’s), hospitals with ED’s and/or OPD’swithin PSU’s, ED’s within hospitals and/or clinicswithin OPD’s, and patient visits within ED’s and/orclinics. In 1997 the hospital response rate forNHAMCS was 95 percent. Hospital staff were askedcomplete Patient Record forms for a systematicrandom sample of patient visits occurring during arandomly assigned 4-week reporting period. In 1997the number of Patient Record forms completed forED’s was 22,209 and for OPD’s was 30,107.

For more detailed information on NHAMCS, seeMcCaig LF, McLemore T. Plan and operation of theNational Hospital Ambulatory Medical Care Survey.National Center for Health Statistics. Vital Health St1(34). 1994.

National Center for HIV, STD, and TB Prevention

AIDS Surveillance

Acquired immunodeficiency syndrome (AIDS)surveillance is conducted by health departments ineach State, territory, and the District of Columbia.Although surveillance activities range from passive tactive, most areas employ multifaceted active

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

rdl

ro

se

,l

se

smte

n

y


ata

surveillance programs, which include four majorreporting sources of AIDS information: hospitals andhospital-based physicians, physicians in nonhospitalpractice, public and private clinics, and medical recosystems (death certificates, tumor registries, hospitadischarge abstracts, and communicable diseasereports). Using a standard confidential case reportform, the health departments collect informationwithout personal identifiers, which is coded andcomputerized either at the Centers for Disease Contand Prevention (CDC) or at health departments fromwhich it is then transmitted electronically to CDC.

AIDS surveillance data are used to detectepidemiologic trends, to identify unusual casesrequiring followup, and for semiannual publication inthe HIV/AIDS Surveillance Report. Studies todetermine the completeness of reporting of AIDS cameeting the national surveillance definition suggestreporting at greater than or equal to 90 percent.

For more information on AIDS surveillance, see:Centers for Disease Control and Prevention.HIV/AIDSSurveillance Report, published semiannually; orcontact: Chief, Surveillance Branch, Division ofHIV/AIDS, National Center for HIV, STD, and TBPrevention (NCHSTP), Centers for Disease Controland Prevention, Atlanta, GA 30333; orvisit the NCHSTP home page athttp://www.cdc.gov/nchstp/od/nchstp.html.

Epidemiology Program Office

National Notifiable Diseases Surveillance System

The Epidemiology Program Office (EPO) of CDCin partnership with the Council of State and TerritoriaEpidemiologists (CSTE), operates the NationalNotifiable Diseases Surveillance System. The purpoof this system is primarily to provide weeklyprovisional information on the occurrence of diseasedefined as notifiable by CSTE. In addition, the systealso provides summary data on an annual basis. Staepidemiologists report cases of notifiable diseases toEPO, and EPO tabulates and publishes these data ithe Morbidity and Mortality Weekly Report(MMWR)

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

l

s

and theSummary of Notifiable Diseases, United States(entitledAnnual Summarybefore 1985). Notifiabledisease surveillance is conducted by public healthpractitioners at local, State, and national levels tosupport disease prevention and control activities.

Notifiable disease reports are received from 52areas in the United States and 5 territories. Tocalculate U.S. rates, data reported by 50 States, NewYork City, and the District of Columbia, are used.(New York State is reported as Upstate New York,which excludes New York City.)

CSTE and CDC annually review the status ofnational infectious disease surveillance and recommendadditions or deletions to the list of nationally notifiablediseases based on the need to respond to emergingpriorities. For example, genital chlamydial infectionsbecame nationally notifiable in 1995. However,reporting nationally notifiable diseases to CDC byStates is voluntary. Reporting is currently mandated bylaw or regulation only at the State level. Therefore, thelist of diseases that are considered notifiable variesslightly by State. For example, reporting of mumps toCDC is not done by some States in which this diseaseis not notifiable to local or State authorities.

Completeness of reporting varies because not allcases receive medical care and not all treatedconditions are reported. Estimates of underreporting ofsome diseases have been made. For example, it isestimated that only 22 percent of cases of congenitalrubella syndrome are reported. Only 10–15 percent ofall measles cases were reported before the institutionof the Measles Elimination Program in 1978. Recentinvestigations suggest that fewer than 50 percent ofmeasles cases were reported following an outbreak inan inner city and that 40 percent of hospitalizedmeasles cases are currently reported. Data from a studof pertussis suggest that only one-third of severe casescausing hospitalization or death are reported. Datafrom a study of tetanus deaths suggest that only40 percent of tetanus cases are reported to CDC.

For more information, see: Centers for DiseaseControl and Prevention, Summary of NotifiableDiseases, United States, 1997.Morbidity and Mortality

...................................................................... 119

http://www.cdc.gov/nchstp/od/nchstp.html

d

sls

he

ga

tngnde

ttees

,

e

r

120 ........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

Weekly Report, 46(53), Public Health Service, DHHS,Atlanta, GA, 1998; or write: Chief, SurveillanceSystems Branch, Division of Public HealthSurveillance and Informatics. Epidemiology ProgramOffice, Centers for Disease Control and Prevention,1600 Clifton Road, MS C08, Atlanta, GA 30333; orvisit the EPO home page athttp://www.cdc.gov/epo.

National Center for Chronic Disease Preventionand Health Promotion

Abortion Surveillance

In 1969 CDC began abortion surveillance todocument the number and characteristics of womenobtaining legal induced abortions, monitor unintendepregnancy, and assist efforts to identify and reducepreventable causes of morbidity and mortalityassociated with abortions. For each year since 1969abortion data have been available from 52 reportingareas: 50 States, the District of Columbia, and NewYork City. The total number of legal induced abortionis available from all reporting areas; however, not alareas collect information regarding the characteristicof women who obtain abortions. Furthermore thenumber of States reporting each characteristic and tnumber of States with complete data for eachcharacteristic vary from year to year. State data withmore than 15 percent unknown for a givencharacteristic are excluded from the analysis of thatcharacteristic.

For 47 reporting areas, data concerning thenumber and characteristics of women who obtain leinduced abortions are provided by central healthagencies such as State health departments and thehealth departments of New York City and the Districof Columbia. For the other five areas, data concernithe number of abortions are provided by hospitals aother medical facilities. In general the procedures arreported by the State in which the procedure isperformed. However, two reporting areas (the Districof Columbia and Wisconsin) report abortions by Staof residence; occurrence data are unavailable for thareas.

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

l

e

The total number of abortions reported to CDC isabout 10 percent less than the total estimatedindependently by the Alan Guttmacher Institute, anot-for-profit organization for reproductive healthresearch, policy analysis, and public education.

For more information, see: Centers for DiseaseControl and Prevention, CDC Surveillance SummariesJuly 3, 1998.Morbidity and Mortality Weekly Report1998;47 (NoSS-2), Abortion Surveillance - UnitedStates, 1995; or contact: Director, Division ofReproductive Health, National Center for ChronicDisease Prevention and Health Promotion(NCCDPHP), Centers for Disease Control andPrevention Atlanta, GA 30333; or visit the NCCDPHPhome page athttp://www.cdc.gov/nccdphp.

National Institute for Occupational Safety andHealth

National Traumatic Occupational Fatalities SurveillanceSystem

The National Traumatic Occupational Fatalities(NTOF) surveillance system is compiled by theNational Institute for Occupational Safety and Health(NIOSH) based on information taken from deathcertificates. Certificates are collected from 52 vitalstatistics reporting units (the 50 States, New York City,and the District of Columbia) based on the followingcriteria: age 16 years or over, an external cause ofdeath (ICD-9, E800-E999), and a positive response tothe ‘‘Injury at work?’’ item.

For the period of this analysis there were nostandardized guidelines regarding the completion of th‘‘Injury at work?’’ item on the death certificate, thus,numbers and rates of occupational injury deaths fromNTOF should be regarded as the lower bound for thetrue number of these events. Operational guidelines fothe completion of the ‘‘Injury at work?’’ item havebeen developed by NIOSH in conjunction with theNational Center for Health Statistics, the NationalAssociation for Public Health Statistics andInformation Systems, and the National Center forEnvironmental Health and were disseminated in 1992

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

http://www.cdc.gov/epo

http://www.cdc.gov/nccdphp

t.

h

,

th

e

m

fl

e.

f

re


ata

for implementation in 1993. This should improve deacertificate-based surveillance of work-related injuries

The denominator data for the calculation of ratesby industry division were obtained from the U.S.Bureau of Labor Statistics’ annual averageemployment data. All of the rates presented are for tU.S. civilian labor force.

For further information on NTOF, see DHHS(NIOSH). Publication No. 93–108,Fatal Injuries toWorkers in the United States, 1980–1989: A DecadeSurveillance; or contact: Director, Division of SafetyResearch, National Institute for Occupational Safetyand Health, 1095 Willowdale Road, Mailstop P-1172Morgantown, WV 26505; or visit the NIOSH homepage athttp://www.cdc.gov/niosh.

Health Resources and Services Administration

Bureau of Health Professions

Physician Supply Projections

Physician supply projections in this report arebased on a model developed by the Bureau of HealProfessions to forecast the supply of physicians byspecialty, activity, and State of practice. The 1995supply of active physicians (M.D.’s) was used as thestarting point for the most recent projections of activphysicians. The major source of data used to obtain1995 figures was the American Medical Association(AMA) Physician Masterfile.

In the first stage of the projections, graduates froU.S. schools of allopathic (M.D.) and osteopathic(D.O.) medicine and internationally trained additionswere estimated on a year-by-year basis. Estimates ofirst-year enrollments, student attrition, other medicaschool-related trends, and a model of netinternationally trained medical graduate immigrationwere used in deriving these annual additions. Theseyear-by-year additions were then combined with thealready existing active supply in a given year toproduce a preliminary estimate of the active workforce in each succeeding year. These estimates werthen reduced to account for mortality and retirement

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

h

e

of

Gender-specific mortality and retirement losses werecomputed by 5-year age cohorts on an annual basis,using age distributions and mortality and retirementrates based on the AMA data.

For more information, see: Bureau of HealthProfessions,Health Personnel in the United StatesNinth Report to Congress, 1993, DHHS Pub. No.HRS-P-OD-94–1, Health Resources and ServicesAdministration, Rockville, MD.

Nurse Supply Estimates

Nursing estimates in this report are based on amodel developed by the Bureau of Health Professionsto meet the requirements of Section 951, P.L. 94–63.The model estimates the following for each State: (a)population of nurses currently licensed to practice; (b)supply of full- and part-time practicing nurses (oravailable to practice); and (c) full-time equivalentsupply of nurses practicing full time plus one-half ofthose practicing part time (or available on that basis).

The three estimates are divided into three levels ohighest educational preparation: associate degree ordiploma, baccalaureate, and master’s and doctorate.

Among the factors considered are new graduates,changes in educational status, nursing employmentrates, age, migration patterns, death rates, and licensuphenomena. The base data for the model are derivedfrom the National Sample Surveys of RegisteredNurses, conducted by the Division of Nursing, Bureauof Health Professions, HRSA. Other data sourcesinclude National League for Nursing for data onnursing education and National Council of StateBoards of Nursing for data on licensure.

Substance Abuse and Mental Health ServicesAdministration

Office of Applied Studies

National Household Surveys on Drug Abuse

Data on trends in use of marijuana, cigarettes,alcohol, and cocaine among persons 12 years of age

....................................................................... 121

http://www.cdc.gov/niosh

ine

.

b

g

nd

at

s

anhe

ra

dse

-A

e

asd

is

.

f

122 .........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

and over are from the National Household Survey onDrug Abuse (NHSDA). The 1997 survey is the 17tha series that began in 1971 under the auspices of thNational Commission on Marijuana and Drug AbuseFrom 1974 to September 1992, the survey wassponsored by the National Institute on Drug Abuse.Since October 1992, the survey has been sponsoredthe Substance Abuse and Mental Health ServicesAdministration (SAMHSA).

Since 1991 the National Household Survey onDrug Abuse has covered the civiliannoninstitutionalized population 12 years of age andover in the United States. This includes civilians livinon military bases and persons living innoninstitutionalized group quarters, such as collegedormitories, rooming houses, and shelters. Hawaii aAlaska were included for the first time in 1991.

In 1994 the survey underwent major changes thaffect the reporting of substance abuse prevalencerates. New questionnaire and data editing procedurewere implemented to improve the measurement oftrends in prevalence and to enhance the timelinessquality of the data. Because it was anticipated that tnew methodology would affect the estimates ofprevalence, the 1994 NHSDA was designed to genetwo sets of estimates. The first set, called the 1994-Aestimates, was based on the same questionnaire anediting method that were used in 1993. The secondcalled the 1994-B estimates, was based on the newquestionnaire and editing methodology. A descriptionof this new methodology can be found in AdvanceReport 10, available from SAMHSA. Because of the1994 changes, many of the estimates from the 1994and earlier NHSDA’s are not comparable withestimates from the 1994-B and later NHSDA’s. To bable to describe long-term trends in drug useaccurately, an adjustment procedure was developedapplied to the pre-1994 estimates. This adjustment uthe 1994 split sample design to estimate the magnituof the impact of the new methodology for each drugcategory. The adjusted estimates are presented in thvolume ofHealth, United States. A description of theadjustment method can be found in Advance ReportNumber 18,Appendix A, available from SAMHSA.

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

y

d

te

t,

ndese

The 1997 survey employed a multistageprobability sample design. Young people (age 12–34years), black Americans, and Hispanics wereoversampled. The sample included 24,505 respondentsThe screening and interview response rates were92.7 percent and 78.3 percent, respectively.

For more information on the National HouseholdSurvey on Drug Abuse (NHSDA), see: NHSDA Series:H-5 National Household Survey on Drug Abuse MainFindings 1996, H-6 Preliminary Results from the 1997National Household Survey on Drug Abuse, H-7National Household Survey on Drug Abuse: PopulationEstimates 1997; or write: Office of Applied Studies,Substance Abuse and Mental Health ServicesAdministration, Room 16C-06, 5600 Fishers Lane,Rockville, MD 20857; or visit the SAMHSA homepage athttp://www.samhsa.gov.

Drug Abuse Warning Network

The Drug Abuse Warning Network (DAWN) is alarge-scale, ongoing drug abuse data collection systembased on information from emergency room andmedical examiner facilities. DAWN collectsinformation about those drug abuse occurrences thathave resulted in a medical crisis or death. The majorobjectives of the DAWN data system include themonitoring of drug abuse patterns and trends, theidentification of substances associated with drug abuseepisodes, and the assessment of drug-relatedconsequences and other health hazards.

Hospitals eligible for DAWN are non-Federal,short-stay general hospitals that have a 24-houremergency room. Since 1988 the DAWN emergencyroom data have been collected from a representativesample of these hospitals located throughout thecoterminous United States, including 21 oversampledmetropolitan areas. Within each facility, a designatedDAWN reporter is responsible for identifying drugabuse episodes by reviewing official records andtranscribing and submitting data on each case. Thedata from this sample are used to generate estimates othe total number of emergency room drug abuse

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

http://www.samhsa.gov

6

in

aw

dt

tl

a

b

d6

t

s

ly

.

ut

h),

al

r

ic


ata

episodes and drug mentions in all such hospitals. Aresponse rate of 74 percent was obtained in the 199survey.

A methodology for generating comparableestimates for years before 1988 was developed, takadvantage of historical data on the characteristics ofthe universe of eligible hospitals and the extensive dfiles compiled over the years by DAWN. After the neprobability sample for DAWN was implemented in1988, old and new DAWN sample data were collectefor a period of 1 year. This overlap period was usedevaluate various procedures for weighting the oldsample data (from 1978 to 1987). The procedure thaconsistently produced reliable estimates for a particumetropolitan area was selected as the weightingscheme for that area and used to generate all estimfor that area for years before 1988. These historicalestimates are available in Advance Report 16, availafrom SAMHSA.

For further information, see: Series I, Number14-A The Drug Abuse Warning Network (DAWN)Annual Data, 1994; Advance Report 14: HistoricalEstimates from the Drug Abuse Warning Network;DAWN Series: D-5 Mid-Year 1997 PreliminaryEmergency Department Data from the Drug AbuseWarning Network and D-6 Drug Abuse WarningNetwork - Annual Medical Examiner Data 1996 orwrite: Office of Applied Studies, Substance Abuse anMental Health Services Administration, Room 16C-05600 Fishers Lane, Rockville, MD 20857; or visit theSAMHSA home page athttp://www.samhsa.gov.

Uniform Facility Data Set

The Uniform Facility Data Set (UFDS), is part ofthe Drug and Alcohol Services Information System(DASIS) maintained by the Substance Abuse andMental Health Services Administration. UFDS is acensus of all substance abuse treatment and prevenfacilities that are licensed, certified, or otherwiserecognized by the individual State substance abuseagencies, and an additional group of substance abutreatment facilities identified from other sources. Itseeks information from all specialized facilities that

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

g

ta

o

ar

tes

le

,

ion

e

treat substance abuse. These include facilities that ontreat substance abuse, as well as specialty substanceabuse units operating within larger mental health (forexample, community mental health centers), generalhealth (for example, hospitals), social service (forexample, family assistance centers), and criminaljustice (for example, probation departments) agenciesUFDS solicits data concerning facility and clientcharacteristics for a specific reference day (on or aboOctober 1) including number of individuals intreatment, substance of abuse (alcohol, drugs, or bottypes of services, and source of revenue. Public andprivate facilities are included.

Treatment facilities contacted through UFDS areidentified from the National Master Facility Inventory(NMFI), which lists all State-sanctioned substanceabuse treatment and prevention facilities and additiontreatment facilities identified through businessdirectories and other sources. In 1996, onlyState-sanctioned facilities were included in thepublished tables. The 1997 data include, for the firsttime, the facilities identified through businessdirectories and other sources. Response rates to thesurvey were 86 and 88 percent in 1996 and 1997respectively.

For further information on UFDS, contact: Officeof Applied Studies, Substance Abuse and MentalHealth Services Administration, Room 16–105, 5600Fishers Lane, Rockville, MD 20857; or visit the OASstatistical information section of the SAMHSA homepage:http://www.samhsa.gov.

Center for Mental Health Services

Surveys of Mental Health Organizations

The Survey and Analysis Branch of the Divisionof State and Community Systems Developmentconducts a biennial inventory of mental healthorganizations (IMHO) and general hospital mentalhealth services (GHMHS). One version is designed fospecialty mental health organizations and another fornon-Federal general hospitals with separate psychiatrservices. The response rate to most of the items on

....................................................................... 123



t.te

h

n

e

d

er

n

’s

tr

s

fo

te

als

f

ed

:

te

to

w

us.

a

124 .......

Appendix I ..............................................................................................................So

urce

sof

Dat

a

these inventories is relatively high (90 percent orbetter) as is the rate for data presented in this reporHowever, for some inventory items, the response ramay be somewhat lower.

IMHO and GHMHS are the primary sources forCenter for Mental Health Services data included in treport. This data system is based on questionnairesmailed every other year to mental health organizatioin the United States, including psychiatric hospitals,non-Federal general hospitals with psychiatric servicDepartment of Veterans Affairs psychiatric services,residential treatment centers for emotionally disturbechildren, freestanding outpatient psychiatric clinics,partial care organizations, freestanding day-nightorganizations, and multiservice mental healthorganizations, not elsewhere classified.

Federally funded community mental health cent(CMHC’s) were included separately through 1980. In1981, with the advent of block grants, the changes idefinition of CMHC’s and the discontinuation ofCMHC monitoring by the Center for Mental HealthServices, organizations formerly classified as CMHChave been reclassified as other organization types,primarily ‘‘multiservice mental health organizations,not elsewhere classified,’’ and ‘‘freestanding psychiaoutpatient clinics.’’

Beginning in 1983 any organization that provideservices in any combination of two or more services(for example, outpatient plus partial care, residentialtreatment plus outpatient plus partial care) and isneither a hospital nor a residential treatment centeremotionally disturbed children is classified as amultiservice mental health organization.

Other surveys conducted by the Survey andAnalysis Branch encompass samples of patientsadmitted to State and county mental hospitals, privamental hospitals, multiservice mental healthorganizations, the psychiatric services of non-Federgeneral hospitals and Department of Veterans Affairmedical centers, residential treatment centers foremotionally disturbed children, and freestandingoutpatient and partial care programs. The purpose o

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

is

s

s,

s

ic

r

these surveys is to determine the sociodemographic,clinical, and treatment characteristics of patients servby these facilities.

For more information, write: Survey and AnalysisBranch, Division of State and Community SystemsDevelopment, Center for Mental Health Services,Room 15C-O4, 5600 Fishers Lane, Rockville, MD20857. For further information on mental health, seeCenter for Mental Health Services,Mental Health,United States, 1998.Manderscheid R, Henderson MJ,eds. DHHS Pub. No. (SMA) 99–3285. Washington:Superintendent of Documents, U.S. GovernmentPrinting Office. 1998; or visit the Center for MentalHealth Services home page athttp://www.samhsa.gov/cmhs/cmhs.htm.

National Institutes of Health

National Cancer Institute

Surveillance, Epidemiology, and End Results Program

In the Surveillance, Epidemiology, and EndResults (SEER) Program the National Cancer Institu(NCI) contracts with 11 population-based registriesthroughout the United States to provide data on allresidents diagnosed with cancer during the year andprovide current followup information on all previouslydiagnosed patients.

This report covers residents of one of thefollowing geographic areas at the time of their initialdiagnosis of cancer: Atlanta, Georgia; Detroit,Michigan; Seattle-Puget Sound, Washington; SanFrancisco-Oakland, California; Connecticut; Iowa; NeMexico; Utah; and Hawaii.

Population estimates used to calculate incidencerates are obtained from the U.S. Bureau of the CensNCI uses estimation procedures as needed to obtainestimates for years and races not included in the datprovided by the U.S. Bureau of the Census. Ratespresented in this report may differ somewhat fromprevious reports due to revised population estimatesand the addition and deletion of small numbers ofincidence cases.

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

http://www.samhsa.gov/cmhs/cmhs.htm

s.ci

to

)5

ednd

le

t

e

de

of

d

dsd


ata

Life tables used to determine normal lifeexpectancy when calculating relative survival rateswere obtained from NCHS and in-house calculationSeparate life tables are used for each race-sex-spegroup included in the SEER Program.

For further information, see: National CancerInstitute,Cancer Statistics Review, 1973–95by L.A.G.Ries, et al. Public Health Service. Bethesda, MD,1998; or visit the SEER home page:http://www-seer.ims.nci.nih.gov.

National Institute on Drug Abuse

Monitoring the Future Study (High School SeniorSurvey)

Monitoring the Future Study (MTF) is alarge-scale epidemiological survey of drug use andrelated attitudes. It was initiated by the NationalInstitute on Drug Abuse (NIDA) in 1975 and isconducted annually through a NIDA grant awardedthe University of Michigan’s Institute for SocialResearch. MTF is composed of three substudies: (aannual survey of high school seniors initiated in 197(b) ongoing panel studies of representative samplesfrom each graduating class that have been conductby mail since 1976; and (c) annual surveys of 8th a10th graders initiated in 1991.

The survey design is a multistage random sampwith stage one being the selection of particulargeographic areas, stage two the selection of one ormore schools in each area, and stage three theselection of students within each school. Data arecollected using self-administered questionnairesadministered in the classroom by representatives ofInstitute for Social Research. Dropouts and studentswho are absent on the day of the survey are excludRecognizing that the dropout population is at higherrisk for drug use, this survey was expanded to inclusimilar nationally representative samples of 8th and10th graders in 1991. Statistics that are published inthe Dropout Rates in the United States: 1996(published by the National Center for EducationalStatistics, Pub. No. 98–250) stated that among pers15–16 years of age, 3.5 percent have dropped out o

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

fic

;

he

d.

ns

school. Among persons 17 years of age, 3.4 percenthave dropped out of school, while the dropout percentincreases to 5.9 percent of persons 18 years of age, anto 8.9 percent for persons 19 years of age. Therefore,surveying eighth graders (where drop out rates aremuch lower than for high school seniors) should beeffective for picking up students at higher risk for druguse.

Approximately 50,000 8th, 10th, and 12th gradersare surveyed annually. In 1998 the annual seniorsamples are comprised of 15,780 seniors in 144 publicand private high schools nationwide, selected to berepresentative of all seniors in the continental UnitedStates. The 10th grade samples involve about 15,419students in 129 schools in 1998, and the 1998 eighthgrade samples have 18,667 students in 149 schools.

For further information on Monitoring the FutureStudy, see: National Institute on Drug Abuse, NationalSurvey Results on Drug Use from Monitoring theFuture Study, 1975–1997, vol I, secondary schoolstudents. NIH Pub. No. 98–4345. Washington: PublicHealth Service. 1998; or visit the NIDA home page athttp://www.nida.nih.govor University of Michigan’swebsite,http://www.isr.umich.edu/src/mtf/.

Health Care Financing Administration

Office of the Actuary

Estimates of National Health Expenditures

Estimates of expenditures for health (NationalHealth Accounts) are compiled annually by type ofexpenditure and source of funds.

Estimates of expenditures for health services comefrom an array of sources. The American HospitalAssociation (AHA) data on hospital finances are theprimary source for estimates relating to hospital care.The salaries of physicians and dentists on the staffs ofhospitals, hospital outpatient clinics, hospital-basedhome health agencies, and nursing home care providein the hospital setting are considered to be componentof hospital care. Expenditures for home health care anfor services of health professionals (for example,doctors, chiropractors, private duty nurses, therapists,

....................................................................... 125

http://www-seer.ims.nci.nih.gov

http://www.nida.nih.gov

http://www.isr.umich.edu/src/mtf/

ia

ntn

srp

lin

b

d

n.

p

ti

b

a

tus’

h

are

neor

hem

al

l

126 .........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

and podiatrists) are estimated primarily using acombination of data from the U.S. Bureau of theCensus’ Service Annual Survey and the quinquennCensus of Service Industries.

The estimates of retail spending for prescriptiodrugs are based on results of a HCFA-sponsored sconducted by the Actuarial Research Corporation aon industry data on prescription drug transactions.Expenditures for other medical nondurables and viproducts and other medical durables purchased inoutlets are based on estimates of personal consumexpenditures prepared by the U.S. Department ofCommerce’s Bureau of Economic Analysis, U.S.Bureau of Labor Statistics/Consumer ExpenditureSurvey, and the 1987 National Medical ExpenditureSurvey conducted by the Agency for Health CarePolicy and Research. Those durable and nondurabproducts provided to inpatients in hospitals or nurshomes, and those provided by licensed professionathrough home health agencies are excluded here,are included with the expenditure estimates of theprovider service category.

Nursing home expenditures cover care rendereestablishments providing inpatient nursing andhealth-related personal care through active treatmeprograms for medical and health-related conditionsThese establishments cover skilled nursing andintermediate care facilities, including those for thementally retarded. Spending estimates are based udata from the U.S. Bureau of the Census ServicesAnnual Survey, and the quinequennial Census ofService Industries.

Expenditures for construction include those speon the erection or renovation of hospitals, nursinghomes, medical clinics, and medical research facilibut not for private office buildings providing officespace for private practitioners. Expenditures fornoncommercial research (the cost of commercialresearch by drug companies is assumed to beimbedded in the price charged for the product; toinclude this item again would result in doublecounting) are developed from information gatheredthe National Institutes of Health.

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

l

udyd

ionetailtion

eg

ls orut

in

t

on

nt

es,

y

Source of funding estimates likewise come frommultiplicity of sources. Data on the Federal healthprograms are taken from administrative recordsmaintained by the servicing agencies. Among thesources used to estimate State and local governmenspending for health are the U.S. Bureau of the CensGovernment Financesand Social SecurityAdministration reports on State-operated Workers’Compensation programs. Federal and State-localexpenditures for education and training of medicalpersonnel are excluded from these measures wherethey are separable. For the private financing of healtcare, data on the financial experience of healthinsurance organizations come from special Health CFinancing Administration analyses of private healthinsurers, and from the Bureau of Labor Statistics’survey on the cost of employer-sponsored healthinsurance and on consumer expenditures. Informatioon out-of-pocket spending from the U.S. Bureau of thCensus’ Services Annual Survey, U.S. Bureau of LabStatistics’ Consumer Expenditure Survey, the 1987National Medical Expenditure Survey conducted by tAgency for Health Care Policy and Research, and froprivate surveys conducted by the American HospitalAssociation, American Medical Association, and theAmerican Dental Association are used to developestimates of direct spending by customers.

For more specific information on definitions,sources and methods used in the National HealthAccounts, see: National Health Accounts: Lessonsfrom the U.S. Experience, by Lazenby HC, Levit KR,Waldo DR, et al. Health Care Financing Review, vol14 no 4. Health Care Financing Administration.Washington: Public Health Service. 1992 and NationHealth Expenditures, 1994, Levit KR, Lazenby HC,Sivarajan L, et al. Health Care Financing Review, vo17 no 3. Health Care Financing Administration.Washington: Public Health Service. 1996.

Estimates of State Health Expenditures

Estimates of spending by State are created usingthe same definitions of health care sectors used inproducing the National Health Expenditures (NHE).

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

ls

e

ve

.ic

y

d

vupsi

of

.

de.


ata

The same data sources used in creating NHE are aused to create State estimates whenever possible.Frequently, however, surveys that are used to creatvalid national estimates lack sufficient size to createvalid State level estimates. In these cases, alternatidata sources that best represent the State-by-Statedistribution of spending are substituted and the U.Saggregate expenditures for the specific type of servor source of funds are used to control the level ofState-by-State distributions. This procedure implicitlassumes that national spending estimates can becreated more accurately than State specificexpenditures.

Despite definitional correspondence, NHE differfrom the sum of State estimates. NHE includeexpenditures for persons living in U.S. territories anfor military and Federal civilian employees and theirfamilies stationed overseas. The sum of the State leexpenditures exclude health spending for those groFor hospital care, exclusion of purchases of servicenon-U.S. areas accounts for a 0.9 percent reductionhospital expenditures from those measured as partNHE.

For more information, contact: Office of theActuary, Health Care Financing Administration, 7500Security Blvd., Baltimore, MD 21244–1850.

Medicare National Claims History FilesThe Medicare Common Working File (CWF) is a

Medicare Part A and Part B benefit coordination andclaims validation system. There are two NationalClaims History (NCH) files, the NCH100 percent-Nearline File, and the NCH BeneficiaryProgram Liability (BPL) File. The NCH files containclaims records and Medicare beneficiary informationThe NCH 100 percent Nearline File contains allinstitutional and physician/supplier claims from theCWF. It provides records of every claim submitted,including all adjustment claims. The NCH BPL filecontains Medicare Part A and Part B beneficiaryliability information (such as deductible andcoinsurance amounts remaining). The records incluall Part A and Part B utilization and entitlement data

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

o

e

els.inn

Records for 1997 were maintained on more than 38million enrollees and 48,826 institutional providersincluding 6,246 hospitals, 14,619 skilled nursingfacilities, 10,487 home health agencies, 2,239 hospices,2,689 outpatient physical therapy, 472 comprehensiveoutpatient rehabilitation facilities, 3,274 end state renaldialysis facilities, 3,447 rural health clinics, 1,175community mental health centers, 2,406 ambulatorysurgical centers, and 1,772 federally qualified healthcenters. About 708 million claims were processed infiscal year 1996.

Data from the NCH files provide informationabout enrollee use of benefits for a point in time orover an extended period. Statistical reports areproduced on enrollment, characteristics of participatingproviders, reimbursement, and services used.

For further information on the NCH files see:Health Care Financing Administration, Office ofInformation Services, Enterprise Data Base Group,Division of Information Distribution, Data UsersReference Guide or call the Medicare Hotline at410–786–3689.

For further information on Medicare visit theHCFA home page athttp://www.hcfa.gov.

Medicaid Data SystemThe majority of Medicaid data are compiled from

forms submitted annually by State Medicaid agenciesto the Health Care Financing Administration (HCFA)for Federal fiscal years ending September 30 on theForm HCFA-2082,Statistical Report on Medical Care:Eligibles, Recipients, Payments, and Services.

When using the data keep the following caveats inmind:

Counts of recipients and eligibles categorized bybasis of eligibility generally count each person onlyonce based on the person’s basis of eligibility as offirst appearance on the Medicaid rolls during theFederal fiscal year covered by the report. Note,however, that some States report duplicated counts ofrecipients; that is, they report an individual in as manycategories as the individual had different eligibility

....................................................................... 127

http://www.hcfa.gov

a

g

s

s

d

m1

lth

n

f

t

128 .......

Appendix I ..............................................................................................................So

urce

sof

Dat

a

statuses during the year. In such cases, the sum ofbasis-of-eligibility cells will be greater than the ‘‘totalrecipients’’ number.

Expenditure data include payments for all claimsadjudicated or paid during the fiscal year covered bythe report. Note that this is not the same as summinpayments for services that were rendered during thereporting period.

Some States fail to submit the HCFA-2082 for aparticular year. When this happens, HCFA estimatesthe current year’s HCFA-2082 data for missing Statebased upon prior year’s submissions and informationthe State entered on Form HCFA-64 (the form Stateuse to claim reimbursement for Federal matching funfor Medicaid).

HCFA-2082’s submitted by States frequentlycontain obvious errors in one or more cells in theform. For cells obviously in error, HCFA estimatesvalues that appear to be more reasonable.

The Medicaid data presented inHealth, UnitedStatesare from the Medicaid statistical system (usingform HCFA-2082) and may differ from data presenteelsewhere using the quarterly financial reports (formHCFA-64) submitted by States for reimbursement.Vendor payments from the Medicaid statistical systeexclude disproportionate share hospital payments ($billion in 1993) and payments to health maintenanceorganizations and Medicare ($6 billion in 1993).

For further information on Medicaid data, see:Health Care Financing Review: Medicare andMedicaid Statistical Supplement, 1995, HCFA Pub. No.0374, Health Care Financing Administration,Baltimore, MD. U.S. Government Printing Office,Sept. 1995; or visit the HCFA home page athttp://www.hcfa.gov.

Online Survey Certification and Reporting DatabaseThe Online Survey Certification and Reporting

(OSCAR) database has been maintained by the HeaCare Financing Administration (HCFA) since 1992.OSCAR is an updated version of the Medicare and

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

ll

ds

7

Medicaid Automated Certification System that hasbeen in existence since 1972. OSCAR is anadministrative database containing detailed informatioon all Medicare and Medicaid health care providers inaddition to all currently certified Medicare andMedicaid nursing home facilities in the United Statesand Territories. (Data for the territories are not shownin this report.) The purpose of the nursing homefacility survey certification process is to ensure thatnursing facilities meet the current HCFA long-termcare requirements and thus can participate in servingMedicare and Medicaid beneficiaries. Included in theOSCAR database are all certified nursing facilities,certified hospital-based nursing homes, and certifiedunits for other types of nursing home facilities (forexample, life care communities or board and carehomes). Facilities not included in OSCAR are allnoncertified facilities (that is, facilities that are onlylicensed by the State and are limited to privatepayment sources), and nursing homes that are part othe Department of Veterans Affairs. Also excluded arenursing homes that are intermediate care facilities forthe mentally retarded.

Information on the number of beds, residents, andresident characteristics are collected during aninspection of all certified facilities. All certified nursinghomes are inspected by representatives of the Statesurvey agency (generally the Department of Health) aleast once every 15 months. The information presenton OSCAR is based on each facility’s ownadministrative record system in addition to interviewswith key administrative staff members.

For more information, see: HCFA: OSCAR datausers reference guide, 1995, available from HCFA,Health Standards and Quality Bureau, HCFA/HSQBS2–11–07, 7500 Security Boulevard, Baltimore, MD21244; or visit the HCFA home page athttp://www.hcfa.gov.

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

http://www.hcfa.gov

http://www.hcfa.gov

0

e

te

at

at

o

of

rtInd

t

nd

alf


ata

Department of Commerce

Bureau of the Census

Census of Population

The census of population has been taken in theUnited States every 10 years since 1790. In the 199census, data were collected on sex, race, age, andmarital status from 100 percent of the enumeratedpopulation. More detailed information such as incomeducation, housing, occupation, and industry werecollected from a representative sample of thepopulation. For most of the country, one out of sixhouseholds (about 17 percent) received the moredetailed questionnaire. In places of residence estimato have less than 2,500 population, 50 percent ofhouseholds received the long form.

For more information on the 1990 census, see:U.S. Bureau of the Census,1990 Census ofPopulation, General Population Characteristics, Series1990, CP-1; or visit the Census Bureau home pagehttp://www.census.gov.

Current Population Survey

The Current Population Survey (CPS) is ahousehold sample survey of the civiliannoninstitutionalized population conducted monthly bythe U.S. Bureau of the Census. CPS provides estimof employment, unemployment, and othercharacteristics of the general labor force, thepopulation as a whole, and various other subgroupsthe population.

The 1998 CPS sample is located in 754 sampleareas, with coverage in every State and the DistrictColumbia. In an average month during 1998, thenumber of housing units or living quarters eligible fointerview was about 50,000; of these about 7 percenwere, for various reasons, unavailable for interview.1994 major changes were introduced, which includecomplete redesign of the questionnaire and theintroduction of computer-assisted interviewing for theentire survey. In addition, there were revisions to somof the labor force concepts and definitions.

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

,

d

es

f

a

e

The estimation procedure used involves inflationby the reciprocal of the probability of selection,adjustment for nonresponse, and ratio adjustment.Beginning in 1994 new population controls based onthe 1990 census adjusted for the estimated populationundercount were utilized.

For more information, see: U.S. Bureau of theCensus,The Current Population Survey, Design andMethodology, Technical Paper 40, Washington, U.S.Government Printing Office, Jan. 1978; U.S.Department of Labor, Bureau of Labor Statistics,Employment and Earnings, Feb. 1994, vol 41 no 2 andFeb. 1995, vol 42 no 2, Washington: U.S. GovernmentPrinting Office, Feb. 1994 and Feb. 1995; or visit theCPS home page athttp://www.bls.census.gov.

Population Estimates

National population estimates are derived by usingdecennial census data as benchmarks and dataavailable from various agencies as follows: births anddeaths (National Center for Health Statistics);immigrants (Immigration and Naturalization Service);Armed Forces (Department of Defense); net movemenbetween Puerto Rico and the U.S. mainland (PuertoRico Planning Board); and Federal employees abroad(Office of Personnel Management and Department ofDefense). State estimates are based on similar data aalso on a variety of data series, including schoolstatistics from State departments of education andparochial school systems. Current estimates areconsistent with official decennial census figures and donot reflect estimated decennial censusunderenumeration.

After decennial population censuses, intercensalpopulation estimates for the preceding decade areprepared to replace postcensal estimates. Intercensalpopulation estimates are more accurate than postcensestimates because they take into account the census opopulation at the beginning and end of the decade.Intercensal estimates have been prepared for the1960’s, 1970’s, and 1980’s to correct the ’’error ofclosure‘‘ or difference between the estimatedpopulation at the end of the decade and the census

...................................................................... 129

http://www.bls.census.gov

http://www.bls.census.gov

n

eh

)inl

n

1

l

e

thte

as

d

.

130 ......

Appendix I ..............................................................................................................So

urce

sof

Dat

a

count for that date. The error of closure at the natiolevel was quite small during the 1960’s (379,000).However, for the 1970’s it amounted to almost 5million and for the 1980’s, 1.5 million.

For more information, see: U.S. Bureau of theCensus, U.S. population estimated by age, sex, racand Hispanic origin: 1990–96, release PPL-57, Marc1997; or visit the Census Bureau home page:http://www.census.gov.

Department of Labor

Bureau of Labor Statistics

Annual Survey of Occupational Injuries and Illnesses

Since 1971 the Bureau of Labor Statistics (BLShas conducted an annual survey of establishmentsthe private sector to collect statistics on occupationainjuries and illnesses. The Survey of OccupationalInjuries and Illnesses is based on records thatemployers maintain under the Occupational Safety aHealth Act. Excluded from the survey areself-employed individuals; farmers with fewer than 1employees; employers regulated by other Federalsafety and health laws; and Federal, State, and locagovernment agencies.

Data are obtained from a sample of approximat250,000 establishments, that is, single physicallocations where business is conducted or whereservices of industrial operations are performed. Anindependent sample is selected for each State andDistrict of Columbia that represents industries in thajurisdiction. BLS includes all the State samples in thnational sample.

Establishments included in the survey areinstructed in a mailed questionnaire to providesummary totals of all entries for the previous calendyear to its Log and Summary of Occupational Injurieand Illnesses (OSHA No. 200 form). Additionally,from the selected establishments, approximately550,000 injuries and illnesses with days away fromwork are sampled in order to obtain demographic andetailed case characteristic information. An

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

al

,

d

ly

e

r

occupational injury is any injury, such as a cut,fracture, sprain, or amputation, that results from awork-related event or from a single instantaneousexposure in the work environment. An occupationalillness is any abnormal condition or disorder, otherthan one resulting from an occupational injury, causedby exposure to factors associated with employment. Itincludes acute and chronic illnesses or disease thatmay be caused by inhalation, absorption, ingestion, ordirect contact. Lost workday cases are cases thatinvolve days away from work, or days of restrictedwork activity, or both. The response rate is about92 percent.

For more information, see: Bureau of LaborStatistics, Occupational Injuries and Illnesses: Counts,Rates, and Characteristics, 1993. BLS Bulletin 2478,U.S. Department of Labor, Washington, D.C., August1996; or visit the BLS home page athttp://www.bls.gov.

Consumer Price Index

The Consumer Price Index (CPI) is a monthlymeasure of the average change in the prices paid byurban consumers for a fixed market basket of goodsand services. The all-urban index (CPI-U) introducedin 1978 covers residents of metropolitan areas as wellas residents of urban parts of non-metropolitan areas(about 87 percent of the United States population in1990).

In calculating the index, price changes for thevarious items in each location were averaged togetherwith weights that represent their importance in thespending of all urban consumers. Local data were thencombined to obtain a U.S. city average.

The index measures price changes from adesignated reference date, 1982–84, which equals 100An increase of 22 percent, for example, is shown as122. This change can also be expressed in dollars asfollows: the price of a base period ‘‘market basket’’ ofgoods and services bought by all urban consumers hasrisen from $10 in 1982–84 to $16.30 in 1998.

The current revision of CPI, projected to becompleted in 2000, reflects spending patterns based on

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

http://www.census.gov

http://www.bls.gov

inleo

r

s

t o

ee

ie

,

reesis,

rts

tix

K


ata

the Survey of Consumer Expenditures from 1993 to1995, the 1990 Census of Population, and the ongoPoint-of-Purchase Survey. Using an improved sampdesign, prices for the goods and services required tcalculate the index are collected in urban areasthroughout the country and from retail and serviceestablishments. Data on rents are collected fromtenants of rented housing and residents ofowner-occupied housing units. Food, fuels, and othegoods and services are priced monthly in urbanlocations. Price information is obtained through visitor calls by trained BLS field representatives usingcomputer-assisted telephone interviews.

The earlier 1987 revision changed the treatmenhealth insurance in the cost-weight definitions formedical care items. This change has no effect on thfinal index result but provides a clearer picture of throle of health insurance in the CPI. As part of therevision, three new indexes have been created byseparating previously combined items, for example,eye care from other professional services and inpatand outpatient treatment from other hospital andmedical care services.

Effective January 1997 the hospital index wasrestructured by combining the three categories roominpatient services and outpatient services into onecategory, hospital services. Differentiation betweeninpatient and outpatient and among service types aunder this broad category. In addition new procedurfor hospital data collection identify a payor, diagnosand the payor’s reimbursement arrangement fromselected hospital bills.

A new geographic sample and item structure weintroducted in January 1998 and expenditure weighwere updated to 1993 to 1995. Pricing of a newhousing sample using computer-assisted data collecwas started in June 1998. In January 1999 the indewill be rebased from the 1982–84 time period to1993–95.

For more information, see: Bureau of LaborStatistics,Handbook of Methods, BLS Bulletin 2490,U.S. Department of Labor, Washington, Apr. 1997; IFord and P Sturm. CPI revision provides more

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

g

f

nt

e

on

accuracy in the medical care services component,Monthly Labor Review, U.S. Department of Labor,Bureau of Labor Statistics, Washington, Apr. 1988; orvisit the BLS home page athttp://www.bls.gov.

Employment and Earnings

The Division of Monthly Industry EmploymentStatistics and the Division of Employment andUnemployment Analysis of the Bureau of LaborStatistics publish data on employment and earnings.The data are collected by the U.S. Bureau of theCensus, State Employment Security Agencies, andState Departments of Labor in cooperation with BLS.

The major data source is the Current PopulationSurvey (CPS), a household interview survey conductedmonthly by the U.S. Bureau of the Census to collectlabor force data for BLS. CPS is described separatelyin this appendix. Data based on establishment recordsare also compiled each month from mail questionnairesby BLS, in cooperation with State agencies.

For more information, see: U.S. Department ofLabor, Bureau of Labor Statistics,Employment andEarnings, Jan. 1999, vol 46 no 1, Washington: U.S.Government Printing Office. Jan. 1999.

Employer Costs for Employee Compensation

Employer costs for employee compensation coverall occupations in private industry, excluding farms andhouseholds and State and local governments. Thesecost levels are published once a year with the payrollperiod including March 12th as the reference period.

The cost levels are based on compensation costdata collected for the Bureau of Labor StatisticsEmployment Cost Index (ECI), released quarterly.Employee Benefits Survey (EBS) data are jointlycollected with ECI data. Cost data were collected fromthe ECI’s March 1993 sample that consisted of about23,000 occupations within 4,500 sample establishmentsin private industry and 7,000 occupations within 1,000establishments in State and local governments. Thesample establishments are classified industry categoriesbased on the 1987 Standard Industrial Classification(SIC) system, as defined by the U.S. Office of

....................................................................... 131

http://www.bls.gov

a

aa

sh

A

sf

ft

r

trl

n

132 ......

Appendix I ..............................................................................................................So

urce

sof

Dat

a

Management and Budget. Within an establishment,specific job categories are selected to represent bromajor occupational groups such as professionalspecialty and technical occupations. The cost levelscalculated with current employment weights each ye

For more information, see: U.S. Department ofLabor, Bureau of Labor Statistics,Employment CostIndexes and Levels, 1975–95, Bulletin 2466, Oct.1995.

Department of Veterans Affairs

Data are obtained from the Department ofVeterans Affairs (VA) administrative data systems.These include budget, patient treatment, patient cenand patient outpatient clinic information. Data from tthree patient files are collected locally at each VAmedical center and are transmitted to the nationaldatabank at the VA Austin Automated Center wherethey are stored and used to provide nationwidestatistics, reports, and comparisons.

The Patient Treatment File

The patient treatment file (PTF) collects data, atthe time of the patient’s discharge, on each episodeinpatient care provided to patients at VA hospitals, Vnursing homes, VA domiciliaries, community nursinghomes, and other non-VA facilities. The PTF recordcontains the scrambled social security number, dateinpatient treatment, date of birth, State and county oresidence, type of disposition, place of disposition adischarge, as well as the ICD–9–CM diagnostic andprocedure or operative codes for each episode of ca

The Patient Census File

The patient census file collects data on eachpatient remaining in a VA medical facility at midnighon a selected date of each year, normally Septembe30. This file includes patients admitted to VA hospitaVA nursing homes, and VA domiciliaries. The censusrecord includes information similar to that reported ithe patient treatment file record.

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

der

rer.

us,e

of

of

er

e.

s,

The Outpatient Clinic File

The outpatient clinic file (OPC) collects data oneach instance of medical treatment provided to aveteran in an outpatient setting. The OPC recordincludes the age, scrambled social security number,State and county of residence, VA eligibility code,clinic(s) visited, purpose of visit, and the date of visitfor each episode of care.

For more information, write: Department ofVeterans Affairs, National Center for Veteran Analysisand Statistics, Biometrics Division 008Cl2, 810Vermont Ave., NW, Washington, DC 20420; or visitthe VA home page athttp://www.va.gov.

Environmental Protection Agency

Aerometric Information Retrieval System (AIRS)

The Environmental Protection Agency’sAerometric Information Retrieval System (AIRS)compiles data on ambient air levels of particulatematter smaller than 10 microns (PM-10), lead, carbonmonoxide, sulphur dioxide, nitrogen dioxide, andtropospheric ozone. These pollutants were identified inthe Clean Air Act of 1970 and in its 1977 and 1990amendments because they pose significant threats topublic health. The National Ambient Air QualityStandards (NAAQS) define for each pollutant themaximum concentration level (micrograms per cubicmeter) that cannot be exceeded during specific timeintervals. Data shown in this publication reflectattainment of NAAQS during a 12-month period basedon analysis using county level air monitoring datafrom AIRS and population data from the Bureau of theCensus.

Data are collected at State and local air pollutionmonitoring sites. Each site provides data for one ormore of the six pollutants. The number of sites hasvaried, but generally increased over the years. In 1993there were 4,469 sites, 4,668 sites in 1994, and 4,800sites in 1995. The monitoring sites are locatedprimarily in heavily populated urban areas. Air quality

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

http://www.va.gov

d

s

1la

a

or

n

a

ra

th

c

e

tofdnd

re

le

e

the

dry


ata

for less populated areas is assessed through acombination of data from supplemental monitors anair pollution models.

For more information, see: EnvironmentalProtection Agency,National Air Quality and EmissionTrend Report, 1994, EPA-454/R-95–014, ResearchTriangle Park, NC, Oct. 1995, or write: Office of AirQuality Planning and Standards, EnvironmentalProtection Agency, Research Triangle Park, NC 277For additional information on this measure and simimeasures used to track the Healthy People 2000Objectives and Health Status Indicators, see: NationCenter for Health Statistics,Monitoring Air Quality inHealthy People 2000, Statistical Notes, No. 9.Hyattsville, Maryland: 1995; or visit the EPA AIRShome page athttp://www.epa.gov/airs/airs.html.

United Nations

Demographic Yearbook

The Statistical Office of the United Nationsprepares theDemographic Yearbook, a comprehensivecollection of international demographic statistics.

Questionnaires are sent annually and monthly tmore than 220 national statistical services and otheappropriate government offices. Data forwarded onthese questionnaires are supplemented, to the extepossible, by data taken from official nationalpublications and by correspondence with the nationstatistical services. To ensure comparability, rates,ratios, and percents have been calculated in thestatistical office of the United Nations.

Lack of international comparability betweenestimates arises from differences in concepts,definitions, and time of data collection. Thecomparability of population data is affected by sevefactors, including (a) the definitions of the totalpopulation, (b) the definitions used to classify thepopulation into its urban and rural components, (c)difficulties relating to age reporting, (d) the extent ofover- or underenumeration, and (e) the quality ofpopulation estimates. The completeness and accuraof vital statistics data also vary from one country to

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

1.r

l

t

l

l

e

y

another. Differences in statistical definitions of vitalevents may also influence comparability.

For more information, see: United Nations,Demographic Yearbook 1996, United Nations, NewYork, NY. 1998; or visit the United Nations home pagat http://www.un.orgor their website locator athttp://www.unsystem.org.

World Health Statistics Annual

The World Health Organization (WHO) preparesthe World Health Statistics Annual, an annual volumeof information on vital statistics and causes of deathdesigned for use by the medical and public healthprofessions. Each volume is the result of a joint efforby the national health and statistical administrationsmany countries, the United Nations, and WHO. UniteNations estimates of vital rates and population size acomposition, where available, are reprinted directly inthe Statistics Annual. For those countries for which theUnited Nations does not prepare demographicestimates, primarily smaller populations, the latestavailable data reported to the United Nations andbased on reasonably complete coverage of events aused.

Information published on late fetal and infantmortality is based entirely on official national dataeither reported directly or made available to WHO.

Selected life table functions are calculated fromthe application of a uniform methodology to nationalmortality data provided to WHO, in order to enhancetheir value for international comparisons. The life tabprocedure used by WHO may often lead todiscrepancies with national figures published bycountries, due to differences in methodology or degreof age detail maintained in calculations.

The international comparability of estimatespublished in theWorld Health Statistics Annualisaffected by the same problems discussed above forDemographic Yearbook. Cross-national differences instatistical definitions of vital events, in thecompleteness and accuracy of vital statistics data, anin the comparability of population data are the primafactors affecting comparability.

....................................................................... 133

http://www.un.org

http://www.unsystem.org

http://www.epa.gov/airs/airs.html

nted

sa

of

t

tha

er

rhels

s.

t

ts

134 ........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

For more information, see: World HealthOrganization,World Health Statistics Annual 1996,World Health Organization, Geneva, Switzerland,1998; or visit the WHO home page athttp://www.who.org.

Alan Guttmacher Institute

Abortion Survey

The Alan Guttmacher Institute (AGI) conducts aannual survey of abortion providers. Data are collecfrom hospitals, nonhospital clinics, and physiciansidentified as providers of abortion services. A universurvey of 3,092 hospitals, nonhospital clinics, andindividual physicians was compiled. To assess thecompleteness of the provider and abortion counts,supplemental surveys were conducted of a sampleobstetrician-gynecologists and a sample of hospitals(not in original universe) that were identified asproviding abortion services through the AmericanHospital Association Survey.

The number of abortions estimated by AGIthrough the mid to late 1980’s was about 20 percenmore than the number reported to the Centers forDisease Control and Prevention (CDC). Since 1989AGI estimates have been about 12 percent higher ththose reported by CDC.

For more information, write: The Alan GuttmachInstitute, 120 Wall Street, New York, NY 10005; orvisit AGI’s home page athttp://www.agi-usa.org.

American Association of Colleges ofOsteopathic Medicine

The American Association of Colleges ofOsteopathic Medicine (AACOM) compiles data onvarious aspects of osteopathic medical education fodistribution to the profession, the government, and tpublic. Questionnaires are sent annually to all schooof osteopathic medicine requesting information oncharacteristics of applicants and students, curricula,faculty, grants, contracts, revenues, and expenditureThe response rate is 100 percent.

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

l

en

For more information, see:Annual StatisticalReport, 1997, American Association of Colleges ofOsteopathic Medicine: Rockville, Maryland. 1997; orvisit the AACOM home page athttp://www.aacom.org.

American Association of Colleges of Pharmacy

The American Association of Colleges ofPharmacy (AACP) compiles data on the Colleges ofPharmacy, including information on student enrollmenand types of degrees conferred. Data are collectedthrough an annual survey; the response rate is100 percent.

For further information, see: Profile of PharmacyStudents. The American Association of Colleges ofPharmacy, 1426 Prince Street, Alexandria, VA 22314;or visit the AACP home page athttp://www.aacp.org.

American Association of Colleges of PodiatricMedicine

The American Association of Colleges of PodiatricMedicine (AACPM) compiles data on the Colleges ofPodiatric Medicine, including information on theschools and enrollment. Data are collected annuallythrough written questionnaires. The response rate is100 percent.

For further information, write: The AmericanAssociation of Colleges of Podiatric Medicine, 1350Piccard Drive, Suite 322, Rockville, MD 20850–4307;or visit the AACPM home page athttp://www.aacpm.org.

American Dental Association

The Division of Educational Measurement of theAmerican Dental Association (ADA) conducts annualsurveys of predoctoral dental educational institutions.The questionnaire, mailed to all dental schools, collecinformation on student characteristics, financialmanagement, and curricula.

For more information, see: American DentalAssociation,1996/97 Survey of predoctoral dental

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

http://www.who.org

http://www.aacom.org.

http://www.aacp.org

http://www.agi-usa.org

http://www.aacpm.org

en

te

ndoneang

nsg

letarohi

y

b.e

end

e

ual

nt to

s,on

50

e

ls,

s


ata

educational institutions. Chicago, Illinois, 1997; orvisit the ADA home page athttp://www.ada.org.

American Hospital Association

Annual Survey of Hospitals

Data from the American Hospital Association(AHA) annual survey are based on questionnaires sto all hospitals, AHA-registered and nonregistered, ithe United States and its associated areas. U.S.government hospitals located outside the United Stawere excluded. Questionnaires were mailed to allhospitals on AHA files. For nonreporting hospitals afor the survey questionnaires of reporting hospitalswhich some information was missing, estimates wermade for all data except those on beds, bassinets,facilities. Data for beds and bassinets of nonreportinhospitals were based on the most recent informatioavailable from those hospitals. Facilities and serviceand inpatient service area data include only reportinhospitals and, therefore, do not include estimates.

Estimates of other types of missing data werebased on data reported the previous year, if availabWhen unavailable, the estimates were based on dafurnished by reporting hospitals similar in size, contmajor service provided, length of stay, and geograpand demographic characteristics.

For more information on the AHA Annual Surveof Hospitals, see: American Hospital Association,(Health Forum),Hospital Statistics, 1999 ed. Chicago.1999; or visit an AHA page athttp://www.aha.org.

American Medical Association

Physician Masterfile

A masterfile of physicians has been maintainedthe American Medical Association (AMA) since 1906The Physician Masterfile contains data on almost evphysician in the United States, members andnonmembers of AMA, and on those graduates ofAmerican medical schools temporarily practicingoverseas. The file also includes graduates of

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

nt

s

d

.

l,c

y

ry

international medical schools who are in the UnitedStates and meet education standards for primaryrecognition as physicians.

A file is initiated on each individual upon entryinto medical school or, in the case of internationalgraduates, upon entry into the United States. Betwe1969–85 a mail questionnaire survey was conducteevery 4 years to update the file information onprofessional activities, self-designated area ofspecialization, and present employment status. Sinc1985 approximately one-third of all physicians aresurveyed each year.

For more information on the AMA PhysicianMasterfile, see: Division of Survey and DataResources, American Medical Association,PhysicianCharacteristics and Distribution in the U.S., 1999 ed.Chicago. 1999; or visit the AMA home page athttp://www.ama-assn.org.

Annual Census of Hospitals

From 1920 to 1953 the Council on MedicalEducation and Hospitals of the AMA conducted anncensuses of all hospitals registered by AMA.

In each annual census, questionnaires were sehospitals asking for the number of beds, bassinets,births, patients admitted, average census of patientlists of staff doctors and interns, and other informatiof importance at the particular time. Response rateswere always nearly 100 percent.

The community hospital data from 1940 and 19presented in this report were calculated usingpublished figures from the AMA Annual Census ofHospitals. Although the hospital classification schemused by AMA in published reports is not strictlycomparable with the definition of community hospitamethods were employed to achieve the greatestcomparability possible.

For more information on the AMA Annual Censuof Hospitals, see: American Medical Association,Hospital service in the United States,Journal of theAmerican Medical Association,16(11):1055–1144.1941; or visit the AMA home page athttp://www.ama-assn.org.

....................................................................... 135

http://www.ada.org

http://www.ama-assn.org

http://www.aha.org

http://www.ama-assn.org

nte

m

r

y.

ze

h,

1is

d.

f

136 ........

Appendix I ..............................................................................................................So

urce

sof

Dat

a

Association of American Medical Colleges

The Association of American Medical Colleges(AAMC) collects information on student enrollment imedical schools through the annual Liaison Commiton Medical Education questionnaire, the fallenrollment questionnaire, and the American MedicalCollege Application Service (AMCAS) data system.Other data sources are the institutional profile systethe premedical students questionnaire, the minoritystudent opportunities in medicine questionnaire, thefaculty roster system, data from the Medical CollegeAdmission Test, and one-time surveys developed fospecial projects.

For more information, see: Association ofAmerican Medical Colleges:Statistical InformationRelated to Medical Education. Washington. 1997; orvisit the AAMC home page athttp://www.aamc.org.

Association of Schools and Colleges ofOptometry

The Association of Schools and Colleges ofOptometry (ASCO) compiles data on the variousaspects of optometric education including data onschools and enrollment. Questionnaires are sentannually to all the schools and colleges of optometrThe response rate is 100 percent.

For further information, write: Annual Survey ofOptometric Educational Institutions, Association ofSchools and Colleges of Optometry, 6110 ExecutiveBlvd., Suite 690, Rockville, MD 20852; or visit theASCO home page athttp://www.opted.org.

Association of Schools of Public HealthThe Association of Schools of Public Health

(ASPH) compiles data on the 28 schools of publichealth in the United States and Puerto Rico.Questionnaires are sent annually to all memberschools, and the response rate is 100 percent.

Unlike health professional schools that emphasispecific clinical occupations, schools of public healthoffer study in specialty areas such as biostatistics,epidemiology, environmental and occupational healt

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

e

,

health administration, health planning, nutrition,maternal and child health, social and behavioralsciences, and other population-based sciences.

For further information, write: Association ofSchools of Public Health, 1660 L Street, NW, Suite204, Washington, D.C. 20036–5603; or visit the ASPHhome page athttp://www.asph.org.

InterStudy

National Health Maintenance Organization Census

From 1976 to 1980 the Office of HealthMaintenance Organizations conducted a census ofhealth maintenance organizations (HMO’s). Since 198InterStudy has conducted the census. A questionnairesent to all HMO’s in the United States asking forupdated enrollment, profit status, and Federalqualification status. New HMO’s are also asked toprovide information on model type. When necessary,information is obtained, supplemented, or clarified bytelephone. For nonresponding HMO’s State-suppliedinformation or the most current available data are use

In 1985 a large increase in the number of HMO’sand enrollment was partly attributable to a change inthe categories of HMO’s included in the census:Medicaid-only and Medicare-only HMO’s have beenadded. Also component HMO’s, which have their owndiscrete management, can be listed separately;whereas, previously the oldest HMO reported for all oits component or expansion sites, even when thecomponents had different operational dates or weredifferent model types.

For further information, see:The InterStudyCompetitive Edge,1995. InterStudy Publications, St.Paul, MN 55104; or visit the InterStudy home page athttp://www.hmodata.com.

National League for Nursing

The division of research of the National Leaguefor Nursing (NLN) conducts The Annual Survey ofSchools of Nursing in October of each year.Questionnaires are sent to all graduate nursing

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

http://www.asph.org

http://www.aamc.org

http://www.opted.org

http://www.hmodata.com


ata

programs (master’s and doctoral), baccalaureateprograms designed exclusively for registered nurses,basic registered nursing programs (baccalaureate,associate degree, and diploma), and licensed practicalnursing programs. Data on enrollments, first-timeadmissions, and graduates are completed for allnursing education programs. Response rates ofapproximately 80 percent are achieved for other areasof inquiry.

For more information, see: National League forNursing,Nursing Data Review, 1997, New York, NY;or visit the NLN home page athttp://www.nln.org.

.................................................................................................................................. 137

http://www.nln.org

s

l

y

D

,e

,

at

h

Table I. Standard million age distribution used to adjust death ratesto the U.S. population in 1940

AgeStandard

million

All ages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,000,000

Under 1 year . . . . . . . . . . . . . . . . . . . . . . . . . 15,3431–4 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 64,7185–14 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 170,35515–24 years . . . . . . . . . . . . . . . . . . . . . . . . . . 181,67725–34 years . . . . . . . . . . . . . . . . . . . . . . . . . . 162,06635–44 years . . . . . . . . . . . . . . . . . . . . . . . . . . 139,23745–54 years . . . . . . . . . . . . . . . . . . . . . . . . . . 117,81155–64 years . . . . . . . . . . . . . . . . . . . . . . . . . . 80,29465–74 years . . . . . . . . . . . . . . . . . . . . . . . . . . 48,42675–84 years . . . . . . . . . . . . . . . . . . . . . . . . . . 17,30385 years and over . . . . . . . . . . . . . . . . . . . . . . 2,770

138 .........

Appendix II ............................................................................................................G

loss

ary

The glossary is an alphabetical listing of termsused inHealth, United States. It includes crossreferences to related terms and synonyms. It alsocontains the standard populations used for ageadjustment andInternational Classification of Disease(ICD) codes for cause of death and diagnostic andprocedure categories.

Abortion —The Centers for Disease Control andPrevention’s (CDC) surveillance system counts legainduced abortions only. For surveillance purposes,legal abortion is defined as a procedure performed blicensed physician or someone acting under thesupervision of a licensed physician to induce thetermination of a pregnancy.

Acquired immunodeficiency syndrome(AIDS)—All 50 States and the District of Columbiareport AIDS cases to CDC using a uniform casedefinition and case report form. The case reportingdefinitions were expanded in 1985(MMWR1985;34:373–5); 1987(MMWR1987; 36 (supp. no. 1S):1S–15S); and 1993(MMWR1992; 41 (no. RR-17):1–19). These data are published semiannually by Cin HIV/AIDS Surveillance Report. See relatedHumanimmunodeficiency virus (HIV) infection.

Active physician—SeePhysician.

Addition —An addition to a psychiatricorganization is defined by the Center for MentalHealth Services as a new admission, a readmissionreturn from long-term leave, or a transfer from anothservice of the same organization or anotherorganization. See relatedMental health disorder;Mental health organization; Mental health servicetype.

Admission—The American Hospital Associationdefines admissions as patients, excluding newbornsaccepted for inpatient services during the surveyreporting period. See relatedDays of care; Discharge;Patient.

Age—Age is reported as age at last birthday, this, age in completed years, often calculated bysubtracting date of birth from the reference date, wit

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

a

C

ar

the reference date being the date of the examination,interview, or other contact with an individual.

Age adjustment—Age adjustment, using thedirect method, is the application of age-specific rates ina population of interest to a standardized agedistribution in order to eliminate differences inobserved rates that result from age differences inpopulation composition. This adjustment is usuallydone when comparing two or more populations at onepoint in time or one population at two or more pointsin time.

Age-adjusted death rates are calculated by thedirect method as follows:

∑i= 1

n

ri × (pi /P)

whereri = age-specific death rates for the populationof interest

pi = standard population in age groupi

P = ∑i= 1

n

pi for the age groups thatcomprise the age range ofthe rate being age adjusted

n = total number of age groups over theage range of the age-adjusted rate

Mortality data—Death rates are age adjusted tothe U.S. standard million population (relative agedistribution of 1940 enumerated population of theUnited States totaling 1,000,000) (table I).

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

1ea

–4arsup.

on,

).

eandand

Table III. Populations and age groups used to age adjust NCHSsurvey data

Population, survey, and ageNumber inthousands

U.S. civilian noninstitutionalized population in 1970NHIS, NHDS, NSAS, NAMCS, and NHAMCS

All ages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199,584

Under 15 years . . . . . . . . . . . . . . . . . . . . . . . . 57,74515–44 years . . . . . . . . . . . . . . . . . . . . . . . . . . 81,18945–64 years . . . . . . . . . . . . . . . . . . . . . . . . . . 41,53765 years and over . . . . . . . . . . . . . . . . . . . . . . 19,113

65–74 years. . . . . . . . . . . . . . . . . . . . . . . . . 12,22475 years and over . . . . . . . . . . . . . . . . . . . . . 6,889

NHIS smoking data

18 years and over . . . . . . . . . . . . . . . . . . . . . . 130,15825 years and over . . . . . . . . . . . . . . . . . . . . . . 107,694

18–24 years . . . . . . . . . . . . . . . . . . . . . . . . . . 22,46425–34 years . . . . . . . . . . . . . . . . . . . . . . . . . . 24,43035–44 years . . . . . . . . . . . . . . . . . . . . . . . . . . 22,61445–64 years . . . . . . . . . . . . . . . . . . . . . . . . . . 41,53765 years and over . . . . . . . . . . . . . . . . . . . . . . 19,113

NHIS health status and health care coverage data

All ages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199,584

Under 18 years . . . . . . . . . . . . . . . . . . . . . . . . 69,42618–44 years . . . . . . . . . . . . . . . . . . . . . . . . . . 69,50845–64 years . . . . . . . . . . . . . . . . . . . . . . . . . . 41,53765–74 years . . . . . . . . . . . . . . . . . . . . . . . . . . 12,22475 years and over . . . . . . . . . . . . . . . . . . . . . . 6,889

U.S. resident population in 1980NHES and NHANES

6–11 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 20,834

............................................................................................... Appendix IIGlossary

Age-adjusted death rates are calculated usingage-specific death rates per 100,000 populationrounded to 1 decimal place. Adjustment is based onage groups with 2 exceptions. First, age-adjusted drates for black males and black females in 1950 arebased on nine age groups, with under 1 year and 1years of age combined as one group and 75–84 yeand 85 years of age and over combined as one groSecond, age-adjusted death rates by educationalattainment for the age group 25–64 years are basedfour 10-year age groups (25–34 years, 35–44 years45–54 years, and 55–64 years).

The rate for years of potential life lost (YPLL)before age 75 years is age adjusted to the U.S.standard million population (table I) and is based oneight age groups (under 1 year, 1–14 years, 15–24years, and 10-year age groups through 65–74 years

Maternal mortality rates for Complications ofpregnancy, childbirth, and the puerperium arecalculated as the number of deaths per 100,000 livebirths. These rates are age adjusted to the 1970distribution of live births by mother’s age in theUnited States as shown intable II. See relatedRate:Death and related rates; Years of potential life lost.

National Health Interview Survey—Data from theNational Health Interview Survey (NHIS) are ageadjusted to the 1970 civilian noninstitutionalizedpopulation shown intable III. The 1970 civiliannoninstitutionalized population is derived as follows:Civilian noninstitutionalized population = civilianpopulation on July 1, 1970 – institutionalized

Table II. Numbers of live births and mother’s age groups used toadjust maternal mortality rates to live births in the United States in1970

Mother’s age Number

All ages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3,731,386

Under 20 years . . . . . . . . . . . . . . . . . . . . . . . . 656,46020–24 years . . . . . . . . . . . . . . . . . . . . . . . . . . 1,418,87425–29 years . . . . . . . . . . . . . . . . . . . . . . . . . . 994,90430–34 years . . . . . . . . . . . . . . . . . . . . . . . . . . 427,80635 years and over . . . . . . . . . . . . . . . . . . . . . . 233,342

SOURCE: U.S. Bureau of the Census: Population estimates andprojections. Current Population Reports. Series P-25, No. 499.Washington. U.S. Government Printing Office, May 1973.

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

1th

population. Institutionalized population = (1 –proportion of total population not institutionalized onApril 1, 1970) × total population on July 1, 1970.

Most of the data from NHIS (except as notedbelow and intable III) are age adjusted using four aggroups: under 15 years, 15–44 years, 45–64 years,65 years and over. The NHIS data on health status

6–8 years . . . . . . . . . . . . . . . . . . . . . . . . . . 9,7779–11 years . . . . . . . . . . . . . . . . . . . . . . . . . 11,057

12–17 years . . . . . . . . . . . . . . . . . . . . . . . . . . 23,41012–14 years. . . . . . . . . . . . . . . . . . . . . . . . . 10,94515–17 years. . . . . . . . . . . . . . . . . . . . . . . . . 12,465

20–74 years . . . . . . . . . . . . . . . . . . . . . . . . . . 144,12020–34 years. . . . . . . . . . . . . . . . . . . . . . . . . 58,40135–44 years. . . . . . . . . . . . . . . . . . . . . . . . . 25,63545–54 years. . . . . . . . . . . . . . . . . . . . . . . . . 22,80055–64 years. . . . . . . . . . . . . . . . . . . . . . . . . 21,70365–74 years. . . . . . . . . . . . . . . . . . . . . . . . . 15,581

SOURCE: Calculated from U.S. Bureau of Census: Estimates of thePopulation of the United States by Age, Sex, and Race: 1970 to 1977.Population Estimates and Projections. Current Population Reports.Series P–25, No. 721, Washington. U.S. Government Printing Office,April 1978.

...................................................................... 139

ars

ed

65

g

ars

l70

n

ge8

e

p

Ssnbcoc .Peo

nr

co

w

ve

IdnttlnbpAdoP

140 .......

Appendix II ............................................................................................................G

loss

ary

health care coverage are age adjusted for thepopulation under 65 years of age using three agegroups: under 18 years, 18–44 years, and 45–64 yeand for the population 65 years and over using twoage groups: 65–74 years and 75 years and over. ThNHIS data on smoking in the population 18 years anover are age adjusted using five age groups: 18–24years, 25–34 years, 35–44 years, 45–64 years, andyears and over. The NHIS data on smoking in thepopulation 25 years and over are age adjusted usinfour age groups: 25–34 years, 35–44 years, 45–64years, and 65 years and over. The NHIS data on nousual source of health care among adults are ageadjusted using three groups: 18–24 years, 25–44 yeand 45–64 years.

Health Care Surveys—Data from the four healthcare surveys, National Hospital Discharge Survey(NHDS), National Survey of Ambulatory Surgery(NSAS), National Ambulatory Medical Care Survey(NAMCS), and National Hospital Ambulatory MedicaCare Survey (NHAMCS) are age adjusted to the 19civilian noninstitutionalized population using five agegroups: under 15 years, 15–44 years, 45–64 years,65–74 years, and 75 years and over (table III).

National Health and Nutrition ExaminationSurvey— Data from the National Health ExaminationSurvey (NHES) and the National Health and NutritioExamination Survey (NHANES) are age adjusted tothe 1980 U.S. resident population using five agegroups for adults: 20–34 years, 35–44 years, 45–54years, 55–64 years, and 65–74 years (table III). Datafor children aged 6–11 years and 12–17 years are aadjusted within each group using two subgroups: 6–years and 9–11 years; and 12–14 years and 15–17years (table III).

AIDS—SeeAcquired immunodeficiency syndrom.

Air quality standards —SeeNational ambient airquality standards.

Air pollution —SeePollutant.

Alcohol abuse treatment clients—SeeSubstanceabuse treatment clients.

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

;

,

Ambulatory care—Health care provided toersons without their admission to a health facility.

Ambulatory surgery—According to the Nationalurvey of Ambulatory Surgery (NSAS), ambulatoryurgery refers to previously scheduled surgical andonsurgical procedures performed on an outpatientasis in a hospital or freestanding ambulatory surgeryenter’s general or main operating rooms, satelliteperating rooms, cystoscopy rooms, endoscopy rooms,ardiac catheterization labs, and laser procedure roomsrocedures performed in locations dedicatedxclusively to dentistry, podiatry, abortion, pain block,r small procedures were not included.

In NSAS, data on up to six surgical andon-surgical procedures are collected and coded. SeeelatedOutpatient surgery.

Average annual rate of change(percenthange)—In this report average annual rates of changer growth rates are calculated as follows:

[(Pn /Po )1/N – 1] × 100

herePn = later time periodPo = earlier time periodN = number of years in interval.

This geometric rate of change assumes that aariable increases or decreases at the same rate duringach year between the two time periods.

Average length of stay—In the National Healthnterview Survey, the average length of stay perischarged patient is computed by dividing the totalumber of hospital days for a specified group by the

otal number of discharges for that group. Similarly, inhe National Hospital Discharge Survey, the averageength of stay is computed by dividing the totalumber of days of care, counting the date of admissionut not the date of discharge, by the number ofatients discharged. The American Hospitalssociation computes the average length of stay byividing the number of inpatient days by the numberf admissions. See relatedDays of care; Discharge;atient.

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

geg

n

e

fehr

a

s

sa

Table IV. Revision of the International Classification of Diseases,according to year of conference by which adopted and years in usein the United States

Revision of theInternational Classification

of Diseases

Year ofconference bywhich adopted

Years inuse in

United States

First. . . . . . . . . . . . . . . . . 1900 1900–1909Second. . . . . . . . . . . . . . . 1909 1910–1920Third . . . . . . . . . . . . . . . . 1920 1921–1929Fourth . . . . . . . . . . . . . . . 1929 1930–1938Fifth. . . . . . . . . . . . . . . . . 1938 1939–1948Sixth . . . . . . . . . . . . . . . . 1948 1949–1957Seventh . . . . . . . . . . . . . . 1955 1958–1967Eighth . . . . . . . . . . . . . . . 1965 1968–1978Ninth . . . . . . . . . . . . . . . . 1975 1979–present


Bed—Any bed that is set up and staffed for useby inpatients is counted as a bed in a facility. For theAmerican Hospital Association the count is the averanumber of beds, cribs, and pediatric bassinets durinthe entire reporting period. In the Health CareFinancing Administration’s Online Survey Certificatioand Reporting database, all beds in certified facilitiesare counted on the day of certification inspection. ThWorld Health Organization defines a hospital bed asone regularly maintained and staffed for theaccommodation and full-time care of a succession oinpatients and situated in a part of the hospital whercontinuous medical care for inpatients is provided. TCenter for Mental Health Services counts the numbeof beds set up and staffed for use in inpatient andresidential treatment services on the last day of thesurvey reporting period. See relatedHospital; Mentalhealth organization; Mental health service type;Occupancy rate.

Birth cohort —A birth cohort consists of allpersons born within a given period of time, such ascalendar year.

Birth rate —SeeRate: Birth and related rates.

Birthweight —The first weight of the newbornobtained after birth. Low birthweight is defined as lesthan 2,500 grams or 5 pounds 8 ounces. Very lowbirthweight is defined as less than 1,500 grams or 3pounds 4 ounces. Before 1979 low birthweight wasdefined as 2,500 grams or less and very lowbirthweight as 1,500 grams or less.

Body mass index (BMI)— BMI is a measure thatadjusts body weight for height. It is calculated asweight in kilograms divided by height in meterssquared. Sex- and age-specific cut points of BMI areused in this book in the definition of overweight forchildren and adolescents. Healthy weight for adults idefined as a BMI of 19 to less than 25; overweight,greater than or equal to a BMI of 25; and obesity, asgreater than or equal to a BMI of 30. BMI cut pointsare defined in the Report of the Dietary GuidelinesAdvisory Committee on the Dietary Guidelines forAmericans, 1995. U.S. Department of Agriculture,Agricultural Research Service, Dietary Guidelines

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

e

s

Advisory Committee. 1995. pp.23–4; ClinicalGuidelines on the Identification, Evaluation, andTreatment of Overweight and Obesity in Adults: TheEvidence Report. National Institutes of Health.National Heart, Lung, and Blood Institute. in press;and in the Healthy People 2010 Objectives: Draft forPublic Comment. September 15, 1998. Objectives 2.1,2.2, and 2.3.

Cause of death—For the purpose of nationalmortality statistics, every death is attributed to oneunderlying condition, based on information reported onthe death certificate and utilizing the international rulesfor selecting the underlying cause of death from thereported conditions. Beginning with 1979 theInternational Classification of Diseases, Ninth Revision(ICD-9) has been used for coding cause of death. Datafrom earlier time periods were coded using theappropriate revision of the ICD for that time period.(Seetables IVandV.) Changes in classification ofcauses of death in successive revisions of the ICD mayintroduce discontinuities in cause-of-death statisticsover time. For further discussion, see TechnicalAppendix in National Center for Health Statistics:VitalStatistics of the United States, 1990, Volume II,Mortality, Part A. DHHS Pub. No. (PHS) 95–1101,Public Health Service, Washington, U.S. GovernmentPrinting Office, 1994. See relatedHumanimmunodeficiency virus infection; InternationalClassification of Diseases, Ninth Revision.

Cause-of-death ranking—Cause-of-death rankingfor infants is based on the List of 61 Selected Causesof Infant Death and HIV infection (ICD-9 Nos.

....................................................................... 141

Table V. Cause-of-death codes, according to applicable revision of International Classification of Diseases

Cause of death

Code numbers

Sixth Revision Seventh Revision Eighth Revision Ninth Revision

Communicable diseases. . . . . . . . . . . . . . . . . . . . . . . 001–139, 460–466, 480–487Chronic and other non-communicablediseases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140–459, 467–479, 488–799

Injury and adverse effects . . . . . . . . . . . . . . . . . . . . . . E800–E999

Meningococcal infection . . . . . . . . . . . . . . . . . . . . . . . 036Septicemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 038Human immunodeficiency virus infection1 . . . . . . . . . . . . *042–*044Malignant neoplasms. . . . . . . . . . . . . . . . 140–205 140–205 140–209 140–208

Colorectal . . . . . . . . . . . . . . . . . . . . . 153–154 153–154 153–154 153, 154Malignant neoplasm of peritoneum andpleura . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158, 163.0 158, 163

Respiratory system . . . . . . . . . . . . . . . 160–164 160–164 160–163 160–165Malignant neoplasm of trachea, bronchusand lung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

Breast . . . . . . . . . . . . . . . . . . . . . . . 170 170 174 174–175Prostate . . . . . . . . . . . . . . . . . . . . . . 177 177 185 185

Benign neoplasms . . . . . . . . . . . . . . . . . . . . . . . . . . 210–239Diabetes mellitus . . . . . . . . . . . . . . . . . . 260 260 250 250Anemias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280–285Meningitis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320–322Alzheimer’s disease . . . . . . . . . . . . . . . . . . . . . . . . . 331.0Diseases of heart . . . . . . . . . . . . . . . . . . 410–443 400–402, 410–443 390–398, 402, 404, 410–429 390–398, 402, 404–429

Ischemic heart disease . . . . . . . . . . . . . . . . . . . . . . 410–414Cerebrovascular diseases. . . . . . . . . . . . . 330–334 330–334 430–438 430–438Atherosclerosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440Pneumonia and influenza . . . . . . . . . . . . . 480–483, 490–493 480–483, 490–493 470–474, 480–486 480–487Chronic obstructive pulmonary diseases . . . . 241, 501, 502, 527.1 241, 501, 502, 527.1 490–493, 519.3 490–496Coalworkers’ pneumoconiosis . . . . . . . . . . . . . . . . 515.1 500Asbestosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 515.2 501Silicosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 515.0 502Chronic liver disease and cirrhosis . . . . . . . 581 581 571 571Nephritis, nephrotic syndrome, andnephrosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 580–589

Complications of pregnancy, childbirth, andthe puerperium. . . . . . . . . . . . . . . . . . . 640–689 640–689 630–678 630–676

Congenital anomalies . . . . . . . . . . . . . . . . . . . . . . . . 740–759Certain conditions originating in the perinatalperiod . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 760–779Newborn affected by maternal complica-tions of pregnancy . . . . . . . . . . . . . . . . . . . . . . . . 761

Newborn affected by complications ofplacenta, cord, and membranes . . . . . . . . . . . . . . . . 762

Disorders relating to short gestation andunspecified low birthweight . . . . . . . . . . . . . . . . . . . 765

Birth trauma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 767Intrauterine hypoxia and birth asphyxia . . . . . . . . . . . . 768Respiratory distress syndrome . . . . . . . . . . . . . . . . . 769Infections specific to the perinatal period . . . . . . . . . . . 771

Sudden infant death syndrome. . . . . . . . . . . . . . . . . . . 798.0Unintentional injuries2 . . . . . . . . . . . . . . . E800–E962 E800–E962 E800–E949 E800–E949

Motor vehicle-related injuries2 . . . . . . . . . E810–E835 E810–E835 E810–E823 E810–E825Suicide . . . . . . . . . . . . . . . . . . . . . . . . E963, E970–E979 E963, E970–E979 E950–E959 E950–E959Homicide and legal intervention . . . . . . . . . E964, E980–E985 E964, E980–E985 E960–E978 E960–E978Firearm-related injuries . . . . . . . . . . . . . . . . . . . . E922, E955, E965,

E970, E985E922, E955.0–E955.4,E965.0–E965.4, E970,E985.0–E985.4

. . . Category not applicable.1Categories for coding human immunodeficiency virus infection were introduced in 1987. The * indicates codes are not part of the Ninth Revision.2In the public health community, the term ‘‘unintentional injuries’’ is preferred to ‘‘accidents and adverse effects’’ and ‘‘motor vehicle-related injuries’’ to ‘‘motor vehicleaccidents.’’

142 ..................................................................................................................................

Appendix II ............................................................................................................

s

ty

eed

,

inth

r

g,

rd

fy

ed

to

r a

e


*042–*044). Cause-of-death ranking for other ages ibased on the List of 72 Selected Causes of Death, Hinfection, and Alzheimer’s disease. The List of 72Selected Causes of Death was adapted from one ofspecial lists for mortality tabulations recommended bthe World Health Organization for use with theNinthRevision of the International Classification of Diseas.Two group titles—Certain conditions originating in thperinatal period and Symptoms, signs, and ill-defineconditions—are not ranked from the List of 61Selected Causes of Infant Death; and two grouptitles—Major cardiovascular diseases and Symptomssigns, and ill-defined conditions—are not ranked fromthe List of 72 Selected Causes. In addition, categorytitles that begin with the words ‘‘Other’’ and ‘‘Allother’’ are not ranked. The remaining category titlesare ranked according to number of deaths to determthe leading causes of death. When one of the titlesrepresent a subtotal is ranked (for example,unintentional injuries), its component parts are notranked (in this case, motor vehicle crashes and allother unintentional injuries). See relatedInternationalClassification of Diseases, Ninth Revision.

Civilian noninstitutionalized population;Civilian population —SeePopulation.

Cocaine-related emergency room episodes—TheDrug Abuse Warning Network monitors selectedadverse medical consequences of cocaine and othedrug abuse episodes by measuring contacts withhospital emergency rooms. Contacts may be for druoverdose, unexpected drug reactions, chronic abusedetoxification, or other reasons in which drug use isknown to have occurred.

Cohort fertility —Cohort fertility refers to thefertility of the same women at successive ages. Womborn during a 12-month period comprise a birth cohoCohort fertility for birth cohorts of women is measureby central birth rates, which represent the number obirths occurring to women of an exact age divided bthe number of women of that exact age. Cumulativebirth rates by a given exact age represent the totalchildbearing experience of women in a cohort up to

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

IV

he

s

eat

ent.

that age. Cumulative birth rates are sums of centralbirth rates for specified cohorts and show the numberof children ever born up to the indicated age. Forexample, the cumulative birth rate for women exactly30 years of age as of January 1, 1960, is the sum ofthe central birth rates for the 1930 birth cohort for theyears 1944 (when its members were age 14) through1959 (when they were age 29). Cumulative birth ratesare also calculated for specific birth orders at eachexact age of woman. The percent of women who havnot had at least one live birth by a certain age is founby subtracting the cumulative first birth rate forwomen of that age from 1,000 and dividing by 10. Formethod of calculation, see Heuser RL.Fertility tablesfor birth cohorts by color: United States, 1917–73.Rockville, Maryland. NCHS. 1976. See relatedRate:Birth and related rates.

Community hospitals—SeeHospital.

Compensation—SeeEmployer costs for employeecompensation.

Completed fertility rate —SeeRate: Birth andrelated rates.

Condition—A health condition is a departure froma state of physical or mental well-being. Animpairment is a health condition that includes chronicor permanent health defects resulting from disease,injury, or congenital malformations. All healthconditions, except impairments, are coded accordingthe International Classification of Diseases, NinthRevision, Clinical Modification (ICD–9–CM).

Based on duration, there are two categories ofconditions, acute and chronic. In the National HealthInterview Survey, anacute conditionis a condition thathas lasted less than 3 months and has involved eithephysician visit (medical attention) or restricted activity.A chronic conditionrefers to any condition lasting 3months or more or is a condition classified as chronicregardless of its time of onset (for example, diabetes,heart conditions, emphysema, and arthritis). TheNational Nursing Home Survey uses a specific list ofchronic conditions, also disregarding time of onset. Se

....................................................................... 143

yy

fis

s

an

a

te

l

n

,

n

io

144 ........

Appendix II ............................................................................................................G

loss

ary

relatedInternational Classification of Diseases, NinthRevision, Clinical Modification.

Consumer Price Index (CPI)—CPI is preparedby the U.S. Bureau of Labor Statistics. It is a monthlmeasure of the average change in the prices paid burban consumers for a fixed market basket of goodsand services. The medical care component of CPIshows trends in medical care prices based on speciindicators of hospital, medical, dental, and drug priceA revision of the definition of CPI has been in usesince January 1988. See relatedGross domesticproduct; Health expenditures, national.

Crude birth rate; Crude death rate —SeeRate:Birth and related rates; Rate: Death and related rate.

Current smoker—In 1992 the definition ofcurrent smoker in the National Health InterviewSurvey (NHIS) was modified to specifically includepersons who smoked on ‘‘some days.’’ Before 1992current smoker was defined by the following questiofrom the NHIS survey ‘‘Have you ever smoked 100cigarettes in your lifetime?’’ and ‘‘Do you smokenow?’’ (traditional definition). In 1992 data werecollected for half the respondents using the traditionsmoking questions and for the other half ofrespondents using a revised smoking question (‘‘Doyou smoke every day, some days, or not at all?’’). Anunpublished analysis of the 1992 traditional smokingmeasure revealed that the crude percent of currentsmokers 18 years of age and over remained the samas 1991. The statistics for 1992 combine data collecusing the traditional and the revised questions. Forfurther information on survey methodology and sampsizes pertaining to the NHIS cigarette data for datayears 1965–92 and other sources of cigarette smokidata available from the National Center for HealthStatistics, see: National Center for Health Statistics,Biographies and Data Sources, Smoking Data GuideNo. 1, DHHS Pub. No. (PHS) 91–1308-1, PublicHealth Service. Washington. U.S. Government PrintiOffice. 1991.

Starting with 1993 data estimates of cigarettesmoking prevalence are based on the revised definit

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

c.

s

l

ed

e

g

g

n

that is considered a more complete estimate ofsmoking prevalence. In 1993–95 estimates of cigarettesmoking prevalence were based on a half-sample.Smoking data were not collected in 1996.

Days of care—According to the AmericanHospital Association, days, hospital days, or inpatientdays are the number of adult and pediatric days of carerendered during the entire reporting period. Days ofcare for newborns are excluded.

In the National Health Interview Survey, hospitaldays during the year refer to the total number ofhospital days occurring in the 12-month period beforethe interview week. A hospital day is a night spent inthe hospital for persons admitted as inpatients.

In the National Hospital Discharge Survey, days ofcare refers to the total number of patient daysaccumulated by patients at the time of discharge fromnon-Federal short-stay hospitals during a reportingperiod. All days from and including the date ofadmission but not including the date of discharge arecounted. See relatedAdmission; Average length ofstay; Discharge; Hospital; Patient.

Death rate—SeeRate: Death and related rates.

Dental visit—The National Health InterviewSurvey considers dental visits to be visits to a dentist’soffice for treatment or advice, including services by atechnician or hygienist acting under the dentist’ssupervision. Services provided to hospital inpatientsare not included. Dental visits are based on a 12-monthrecall period.

Diagnosis—SeeFirst-listed diagnosis.

Diagnostic and other nonsurgicalprocedures—SeeProcedure.

Discharge—The National Health Interview Surveydefines a hospital discharge as the completion of anycontinuous period of stay of one night or more in ahospital as an inpatient, not including the period ofstay of a well newborn infant. According to theNational Hospital Discharge Survey and the AmericanHospital Association, discharge is the formal release ofan inpatient by a hospital (excluding newborn infants),

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

n

e

cn

t

k

ae

r

ls

s;

ryy

r

s


that is, the termination of a period of hospitalization(including stays of 0 nights) by death or by dispositioto a place of residence, nursing home, or anotherhospital. See relatedAdmission; Average length ofstay; Days of care; Patient.

Domiciliary care homes—SeeNursing home.

Drug abuse treatment clients—SeeSubstanceabuse treatment clients.

Emergency department—According to theNational Hospital Ambulatory Medical Care Survey(NHAMCS), an emergency department is a hospitalfacility for the provision of unscheduled outpatientservices to patients whose conditions requireimmediate care and is staffed 24 hours a day. Off-sitemergency departments open less than 24 hours areincluded if staffed by the hospital’s emergencydepartment. An emergency department visit is a direpersonal exchange between a patient and a physiciaor other health care providers working under thephysician’s supervision, for the purpose of seekingcare and receiving personal health services. See relaHospital; Outpatient department.

Employer costs for employee compensation—Ameasure of the average cost per employee hour worto employers for wages and salaries and benefits.Wages and salaries are defined as the hourlystraight-time wage rate, or for workers not paid on anhourly basis, straight-time earnings divided by thecorresponding hours. Straight-time wage and salaryrates are total earnings before payroll deductions,excluding premium pay for overtime and for work onweekends and holidays, shift differentials,nonproduction bonuses, and lump-sum paymentsprovided in lieu of wage increases. Productionbonuses, incentive earnings, commission payments,cost-of-living adjustments are included in straight-timwage and salary rates. Benefits covered are paidleave—paid vacations, holidays, sick leave, and otheleave; supplemental pay—premium pay for overtimeand work on weekends and holidays, shift differentianonproduction bonuses, and lump-sum payments

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

t

ed

ed

nd

,

provided in lieu of wage increases; insurancebenefits—life, health, and sickness and accidentinsurance; retirement and savings benefits—pensionand other retirement plans and savings and thrift planlegally required benefits—social security, railroadretirement and supplemental retirement, railroadunemployment insurance, Federal and Stateunemployment insurance, workers’ compensation, andother benefits required by law, such as State temporadisability insurance; and other benefits—severance paand supplemental unemployment plans.

Expenditures—SeeHealth expenditures, national.

Family income—For purposes of the NationalHealth Interview Survey and National Health andNutrition Examination Survey, all people within ahousehold related to each other by blood, marriage, oadoption constitute a family. Each member of a familyis classified according to the total income of thefamily. Unrelated individuals are classified accordingto their own income. Family income is the totalincome received by the members of a family (or by anunrelated individual) in the 12 months before theinterview. Family income includes wages, salaries,rents from property, interest, dividends, profits and feefrom their own businesses, pensions, and help fromrelatives. Family income has generally beencategorized into approximate quintiles in the tables.

Federal hospitals—SeeHospital.

Federal physicians—SeePhysician.

Fee-for-service health insurance—This is private(commercial) health insurance that reimburses healthcare providers on the basis of a fee for each healthservice provided to the insured person. Also known asindemnity health insurance. See relatedHealthinsurance coverage.

Fertility rate —SeeRate: Birth and related rates.

Fetal death—In the World Health Organization’sdefinition, also adopted by the United Nations and theNational Center for Health Statistics, a fetal death isdeath before the complete expulsion or extraction from

...................................................................... 145

ft

oa

s

r

a

r

s

ly

c

146 ........

Appendix II ............................................................................................................G

loss

ary

its mother of a product of conception, irrespective othe duration of pregnancy; the death is indicated byfact that after such separation, the fetus does notbreathe or show any other evidence of life, such asbeating of the heart, pulsation of the umbilical cord,definite movement of voluntary muscles. For statisticpurposes, fetal deaths are classified according togestational age. In this report tabulations are shownfetal deaths with stated or presumed gestation of 20weeks or more and of 28 weeks or more, the lattergestational age group also known as late fetal deathSee relatedGestation; Live birth; Rate: Death andrelated rates.

First-listed diagnosis—In the National HospitalDischarge Survey this is the first recorded finaldiagnosis on the medical record face sheet (summasheet).

General hospitals—SeeHospital.

General hospitals providing separatepsychiatric services—SeeMental health organization.

Geographic region and division—The 50 Statesand the District of Columbia are grouped for statisticpurposes by the U.S. Bureau of the Census into 4geographic regions and 9 divisions. The groupings aas follows:

NortheastNew England

Maine, New Hampshire, Vermont,Massachusetts, Rhode Island,Connecticut

Middle AtlanticNew York, New Jersey,Pennsylvania

MidwestEast North Central

Ohio, Indiana, Illinois, Michigan,Wisconsin

West North CentralMinnesota, Iowa, Missouri, NorthDakota, South Dakota, Nebraska,Kansas

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

he

rl

for

.

y

l

e

SouthSouth Atlantic

Delaware, Maryland, District ofColumbia, Virginia, West Virginia,North Carolina, South Carolina,Georgia, Florida

East South CentralKentucky, Tennessee, Alabama,Mississippi

West South CentralArkansas, Louisiana, Oklahoma,Texas

WestMountain

Montana, Idaho, Wyoming,Colorado, New Mexico, Arizona,Utah, Nevada

PacificWashington, Oregon, California,Alaska, Hawaii

Gestation—For the National Vital StatisticsSystem and the Centers for Disease Control andPrevention’s Abortion Surveillance, the period ofgestation is defined as beginning with the first day ofthe last normal menstrual period and ending with theday of birth or day of termination of pregnancy. SeerelatedAbortion; Fetal death; Live birth.

Gross domestic product (GDP)—GDP is themarket value of the goods and services produced bylabor and property located in the United States. Aslong as the labor and property are located in theUnited States, the suppliers (that is, the workers and,for property, the owners) may be either U.S. residentor residents of the rest of the world. See relatedConsumer Price Index; Health expenditures, national.

Health expenditures, national—See relatedConsumer Price Index; Gross domestic product.

Health services and supplies expenditures—Theseare outlays for goods and services relating directto patient care plus expenses for administeringhealth insurance programs and government publihealth activities. This category is equivalent tototal national health expenditures minusexpenditures for research and construction.

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


National health expenditures—This measureestimates the amount spent for all health servicesand supplies and health-related research andconstruction activities consumed in the UnitedStates during the calendar year. Detailed estimatesare available by source of expenditures (forexample, out-of-pocket payments, private healthinsurance, and government programs), type ofexpenditures (for example, hospital care, physicianservices, and drugs), and are in current dollars forthe year of report. Data are compiled from avariety of sources.

Nursing home expenditures—These cover carerendered in skilled nursing and intermediate carefacilities, including those for the mentally retarded.The costs of long-term care provided by hospitalsare excluded.

Personal health care expenditures—These areoutlays for goods and services relating directly topatient care. The expenditures in this category aretotal national health expenditures minusexpenditures for research and construction,expenses for administering health insuranceprograms, and government public health activities.

Private expenditures—These are outlays forservices provided or paid for by nongovernmentalsources—consumers, insurance companies, privateindustry, philanthropic, and other nonpatient caresources.

Public expenditures—These are outlays forservices provided or paid for by Federal, State,and local government agencies or expendituresrequired by governmental mandate (such as,workmen’s compensation insurance payments).

Health insurance coverage—National HealthInterview Survey (NHIS) respondents were askedabout their health insurance coverage at the time of theinterview in 1984, 1989, and 1997 and in the previousmonth in 1993–96. Questions on health insurancecoverage were expanded starting in 1993 comparedwith previous years. In 1997 the entire questionnaire

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

was redesigned and data were collected using acomputer assisted personal interview (CAPI).

Respondents are covered by private healthinsurance if they indicate private health insurance or ifthey are covered by a single service hospital plan,except in 1997 when no information on single serviceplans was obtained. Private health insurance includesmanaged care such as health maintenanceorganizations (HMO’s).

Until 1996 persons were defined as havingMedicaid or other public assistance coverage if theyindicated that they had either Medicaid or other publicassistance, or if they reported receiving Aid to Familieswith Dependent Children (AFDC) or SupplementarySecurity Income (SSI). After welfare reform in late1996, Medicaid was delinked from AFDC and SSI. In1997 persons were considered to be covered byMedicaid if they reported Medicaid or aState-sponsored health program.

Medicare or military health plan coverage is alsodetermined in the interview, and in 1997 othergovernment-sponsored program was determined.

If respondents do not report coverage under one ofthe above types of plans and they have unknowncoverage on either private health insurance orMedicaid then they are considered to have unknowncoverage.

The remaining respondents are considereduninsured. The uninsured are persons who do not havecoverage under private health insurance, Medicare,Medicaid, public assistance, a State-sponsored healthplan, other government-sponsored programs, or amilitary health plan. Persons with only Indian HealthService coverage are considered uninsured. Estimatesof the percent of persons who are uninsured based onthe NHIS (table 129) are slightly higher than thosebased on the March Current Population Survey (CPS)(table 146). The NHIS asks about coverage at the timeof the survey (or in some survey years, coverageduring the previous month), whereas the CPS asksabout coverage over the previous calendar year. Thismay result in higher estimates of Medicaid and otherhealth insurance coverage and correspondingly lower

....................................................................... 147

in

elatce

dhn

th

pa

it

ther

o

p

l

r

148 ........

Appendix II ............................................................................................................G

loss

ary

estimates of persons without health care coveragethe CPS compared with the NHIS. In addition, theCPS estimate is for persons of all ages whereas thNHIS estimate is for persons under age 65. See reFee-for-service health insurance; Health maintenanorganization; Managed care; Medicaid; Medicare.

Health maintenance organization (HMO)—AnHMO is a prepaid health plan deliveringcomprehensive care to members through designateproviders, having a fixed monthly payment for healtcare services, and requiring members to be in a plafor a specified period of time (usually 1 year). PureHMO enrollees use only the prepaid capitated healservices of the HMO’s panel of medical careproviders. Open-ended HMO enrollees use the preHMO health services but in addition may receivemedical care from providers who are not part of theHMO’s panel. There is usually a substantialdeductible, copayment, or coinsurance associated wthe use of nonpanel providers. These open-endedproducts are governed by State HMO regulations.HMO model types are:

Group—An HMO that delivers health servicesthrough a physician group that is controlled byHMO unit or an HMO that contracts with one omore independent group practices to providehealth services.

Individual practice association (IPA)—An HMOthat contracts directly with physicians inindependent practice, and/or contracts with onemore associations of physicians in independentpractice, and/or contracts with one or moremultispecialty group practices. The plan ispredominantly organized aroundsolo-single-specialty practices.

Mixed—An HMO that combines features of grouand IPA. This category was introduced inmid-1990 because HMO’s are continuallychanging and many now combine features ofgroup and IPA plans in a single plan.

See relatedManaged care.

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

ed

id

h

r

Health services and supplies expenditures—SeeHealth expenditures, national.

Health status, respondent-assessed—Healthstatus was measured in the National Health InterviewSurvey by asking the respondent, ‘‘Would you say____________’s health is excellent, very good, good,fair, or poor?’’

Hispanic origin—Hispanic origin includes personsof Mexican, Puerto Rican, Cuban, Central and SouthAmerican, and other or unknown Latin American orSpanish origins. Persons of Hispanic origin may be ofany race. See relatedRace.

HIV —SeeHuman immunodeficiency virusinfection.

Home health care—Home health care as definedby the National Home and Hospice Care Survey iscare provided to individuals and families in their placeof residence for promoting, maintaining, or restoringhealth; or for minimizing the effects of disability andillness including terminal illness.

Hospice care—Hospice care as defined by theNational Home and Hospice Care Survey is a programof palliative and supportive care services providingphysical, psychological, social, and spiritual care fordying persons, their families, and other loved ones.Hospice services are available in home and inpatientsettings.

Hospital—According to the American HospitalAssociation, hospitals are licensed institutions with atleast six beds whose primary function is to providediagnostic and therapeutic patient services for medicaconditions by an organized physician staff, and havecontinuous nursing services under the supervision ofregistered nurses. The World Health Organizationconsiders an establishment to be a hospital if it ispermanently staffed by at least one physician, can offeinpatient accommodation, and can provide activemedical and nursing care. Hospitals may be classifiedby type of service, ownership, size in terms of numberof beds, and length of stay. In the National HospitalAmbulatory Medical Care Survey (NHAMCS)

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

o

e

n

fhd

e,d

f

el


hospitals include all those with an average length ofstay for all patients of less than 30 days (short-stay)hospitals whose specialty is general (medical orsurgical) or children’s general. Federal hospitals andhospital units of institutions and hospitals with fewerthan six beds staffed for patient use are excluded. SrelatedAverage length of stay; Bed; Days of care;Emergency department; Outpatient department;Patient.

Community hospitalstraditionally included allnon-Federal short-stay hospitals except facilitiesfor the mentally retarded. In the revised definitiothe following additional sites are excluded:hospital units of institutions, and alcoholism andchemical dependency facilities.

Federal hospitalsare operated by the FederalGovernment.

For profit hospitalsare operated for profit byindividuals, partnerships, or corporations.

General hospitalsprovide diagnostic, treatment,and surgical services for patients with a variety omedical conditions. According to the World HealtOrganization, these hospitals provide medical annursing care for more than one category ofmedical discipline (for example, general medicinspecialized medicine, general surgery, specializesurgery, and obstetrics). Excluded are hospitals,usually in rural areas, that provide a more limitedrange of care.

Nonprofit hospitalsare operated by a church orother nonprofit organization.

Psychiatric hospitalsare ones whose major type oservice is psychiatric care. SeeMental healthorganization.

Registered hospitalsare hospitals registered withthe American Hospital Association. About98 percent of hospitals are registered.

Short-stay hospitalsin the National HospitalDischarge Survey are those in which the averaglength of stay is less than 30 days. The Nationa

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

r

e

Health Interview Survey defines short-stayhospitals as any hospital or hospital department inwhich the type of service provided is general;maternity; eye, ear, nose, and throat; children’s; orosteopathic.

Specialty hospitals, such as psychiatric,tuberculosis, chronic disease, rehabilitation,maternity, and alcoholic or narcotic, provide aparticular type of service to the majority of theirpatients.

Hospital-based physician—SeePhysician.

Hospital days—SeeDays of care.

Human immunodeficiency virus (HIV)infection—Mortality coding: Beginning with data for1987, NCHS introduced category numbers *042–*044for classifying and coding HIV infection as a cause ofdeath. HIV infection was formerly referred to ashuman T-cell lymphotropic virus-III/lymphadenopathy-associated virus (HTLV-III/LAV) infection. Theasterisk before the category numbers indicates thatthese codes are not part of theNinth Revision of theInternational Classification of Diseases(ICD-9).Before 1987 deaths involving HIV infection wereclassified to Deficiency of cell-mediated immunity(ICD-9 No. 279.1) contained in the title All otherdiseases; to Pneumocystosis (ICD-9 No. 136.3)contained in the title All other infectious and parasiticdiseases; to Malignant neoplasms, including neoplasmsof lymphatic and hematopoietic tissues; and to anumber of other causes. Therefore, before 1987, deathstatistics for HIV infection are not strictly comparablewith data for 1987 and later years, and are not shownin this report.

Morbidity coding: The National HospitalDischarge Survey codes diagnosis data using theInternational Classification of Diseases, NinthRevision, Clinical Modification(ICD–9–CM).Discharges with diagnosis of HIV as shown inHealth,United States, have at least one HIV diagnosis listedon the face sheet of the medical record and are notlimited to the first-listed diagnosis. During 1984 and

...................................................................... 149

y

v

ge

y

o

d

au

e

.

.

Table VI. Codes for industries, according to the Standard IndustrialClassification (SIC) Manual

Industry Code numbers

Agriculture, forestry, and fishing . . . . . . . . . . . . . . 01–09Mining. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–14Construction . . . . . . . . . . . . . . . . . . . . . . . . . . 15–17Manufacturing . . . . . . . . . . . . . . . . . . . . . . . . . 20–39Transportation, communication, and public utilities . . 40–49Wholesale trade. . . . . . . . . . . . . . . . . . . . . . . . 50–51Retail trade . . . . . . . . . . . . . . . . . . . . . . . . . . 52–59Finance, insurance, and real estate. . . . . . . . . . . . 60–67Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70–89Public administration . . . . . . . . . . . . . . . . . . . . . 91–97

150 .........

Appendix II ............................................................................................................G

loss

ary

1985 only data for AIDS (ICD–9–CM 279.19) wereincluded. In 1986–94, discharges with the followingdiagnoses were included: acquired immunodeficiencsyndrome (AIDS), human immunodeficiency virus(HIV) infection and associated conditions, and positiserological or viral culture findings for HIV(ICD–9–CM 042–044, 279.19, and 795.8). Beginninin 1995 discharges with the following diagnoses werincluded: human immunodeficiency virus (HIV)disease and asymptomatic human immunodeficiencvirus (HIV) infection status (ICD–9–CM 042 andV08). See relatedAcquired immunodeficiencysyndrome; Cause of death; International Classificatiof Diseases, Ninth Revision; InternationalClassification of Diseases, Ninth Revision, ClinicalModification.

ICD; ICD codes—SeeCause of death;International Classification of Diseases, NinthRevision.

Incidence—Incidence is the number of cases ofdisease having their onset during a prescribed perioof time. It is often expressed as a rate (for example,the incidence of measles per 1,000 children 5–15 yeof age during a specified year). Incidence is a measof morbidity or other events that occur within aspecified period of time. See relatedPrevalence.

Individual practice association (IPA)—SeeHealth maintenance organization (HMO).

Industry of employment—Industries areclassified according to theStandard IndustrialClassification (SIC) Manualof the Office of

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

e

n

rsre

Management and Budget. Three editions of the SIC arused for coding industry data inHealth, United States:the 1972 edition; the 1977 supplement to the 1972edition; and the 1987 edition.

The changes between versions include a fewdetailed titles created to correct or clarify industries orto recognize changes within the industry. Codes formajor industrial divisions (table VI) were not changedbetween versions.

The category ‘‘Private sector’’ includes allindustrial divisions except public administration andmilitary. The category ‘‘Civilian sector’’ includes‘‘Private sector’’ and the public administration division.The category ‘‘Not classified’’ is comprised of thefollowing entries from the death certificate: housewife,student, or self-employed; information inadequate tocode industry; establishments not elsewhere classified

Infant death—An infant death is the death of alive-born child before his or her first birthday. Deathsin the first year of life may be further classifiedaccording to age as neonatal and postneonatal.Neonatal deaths are those that occur before the 28thday of life; postneonatal deaths are those that occurbetween 28 and 365 days of age. SeeLive birth; Rate:Death and related rates.

Inpatient care—SeeMental health service type.

Inpatient days—SeeDays of care.

Insured—SeeHealth insurance coverage.

Intermediate care facilities—SeeNursing home.

International Classification of Diseases, NinthRevision (ICD-9)—The International Classification ofDiseases(ICD) classifies mortality information forstatistical purposes. The ICD was first used in 1900and has been revised about every 10 years since thenThe ICD-9, published in 1977, is used to code U.S.mortality data beginning with data year 1979. (Seetables IVandV.) See relatedCause of death;International Classification of Diseases, NinthRevision, Clinical Modification.

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

leth

r

a

inrs

ta;d

o

e

here

Table VII. Codes for diagnostic categories from the International Classification of Diseases, Ninth Revision, Clinical Modification

Diagnostic category Code numbers

Females with delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V27Human immunodeficiency virus (HIV) (1984–85) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279.19

(1986–94) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 042–044, 279.19, 795.8(Beginning in 1995) . . . . . . . . . . . . . . . . . . . . . . . 042, V08

Malignant neoplasms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140–208Large intestine and rectum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153–154, 197.5Trachea, bronchus, and lung. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162, 197.0, 197.3Breast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174–175, 198.81Prostate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

Diabetes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250Psychoses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293–299Diseases of the nervous system and sense organs . . . . . . . . . . . . . . . . . . . . . . . . . . . 320–389Diseases of the circulatory system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390–459

Diseases of heart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391–392.0, 393–398, 402, 404, 410–416, 420–429Ischemic heart disease. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410–414

Acute myocardial infarction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410Congestive heart failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 428.0

Cerebrovascular diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430–438Diseases of the respiratory system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460–519

Bronchitis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466.0, 490–491Pneumonia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466.1, 480–487.0Asthma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493

Hyperplasia of prostate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600Decubitus ulcers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 707.0Diseases of the musculoskeletal system and connective tissue . . . . . . . . . . . . . . . . . . . . 710–739

Osteoarthritis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715Intervertebral disc disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 722

Injuries and poisoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 800–999Fracture, all sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 800–829

Fracture of neck of femur (hip) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 820


International Classification of Diseases, NinthRevision, Clinical Modification (ICD–9–CM)—TheICD–9–CM is based on and is completely compatibwith the International Classification of Diseases, NinRevision. The ICD–9–CM is used to code morbiditydata and the ICD-9 is used to code mortality data.Diagnostic groupings and code number inclusions foICD–9–CM are shown intable VII; surgical andnonsurgical operations, diagnostic procedures, andtherapeutic procedures and code number inclusionsshown intable VIII.

ICD-9 and ICD–9–CM are arranged in 17 mainchapters. Most of the diseases are arranged accordto their principal anatomical site, with special chaptefor infective and parasitic diseases; neoplasms;endocrine, metabolic, and nutritional diseases; mendiseases; complications of pregnancy and childbirthcertain diseases peculiar to the perinatal period; anill-defined conditions. In addition, two supplementalclassifications are provided: the classification of fact

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

re

g

l

rs

influencing health status and contact with healthservice and the classification of external causes ofinjury and poisoning. See relatedCondition;International Classification of Diseases, NinthRevision; Mental health disorder.

Late fetal death rate—SeeRate: Death andrelated rates.

Leading causes of death—SeeCause-of-deathranking.

Length of stay—SeeAverage length of stay.

Life expectancy—Life expectancy is the averagnumber of years of life remaining to a person at aparticular age and is based on a given set ofage-specific death rates, generally the mortalityconditions existing in the period mentioned. Lifeexpectancy may be determined by race, sex, or otcharacteristics using age-specific death rates for thpopulation with that characteristic. See relatedRate:Death and related rates.

....................................................................... 151

e

ee

s

Table VIII. Codes for procedure categories from the International Classification of Diseases, Ninth Revision, Clinical Modification

Procedure category Code numbers

Extraction of lens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1–13.6Insertion of prosthetic lens (pseudophakos) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.7Myringotomy with insertion of tube . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.01Tonsillectomy, with or without adenoidectomy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28.2–28.3Coronary angioplasty (Prior to 1997) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36.0

(Beginning in 1997). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36.01–36.05, 36.09Coronary artery bypass graft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36.1Cardiac catheterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37.21–37.23Pacemaker insertion or replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37.7–37.8Carotid endarterectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38.12Endoscopy of large or small intestine with or without biopsy . . . . . . . . . . . . . . . . . . . . . . . . . . 45.11–45.14, 45.16, 45.21–45.25Cholecystectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.2Prostatectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60.2–60.6Bilateral destruction or occlusion of fallopian tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66.2–66.3Hysterectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68.3–68.7, 68.9Cesarean section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74.0–74.2, 74.4, 74.99Repair of current obstetrical laceration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75.5–75.6Reduction of fracture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76.7,79.0–79.3Arthroscopy of knee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80.26Excision or destruction of intervertebral disc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80.5Total hip replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.51Lumpectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85.21Mastectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85.4Angiocardiography with contrast material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88.5

152 ........

Appendix II ............................................................................................................G

loss

ary

Limitation of activity —In the National HealthInterview Survey limitation of activity refers to along-term reduction in a person’s capacity to performthe usual kind or amount of activities associated withhis or her age group. Each person is classifiedaccording to the extent to which his or her activitiesare limited, as follows:

Persons unable to carry on major activity;

Persons limited in the amount or kind of majoractivity performed;

Persons not limited in major activity but otherwislimited; and

Persons not limited in activity.

See relatedCondition; Major activity.

Live birth —In the World Health Organization’sdefinition, also adopted by the United Nations and thNational Center for Health Statistics, a live birth is thcomplete expulsion or extraction from its mother of aproduct of conception, irrespective of the duration ofthe pregnancy, which, after such separation, breatheshows any other evidence of life such as heartbeat,

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

or

umbilical cord pulsation, or definite movement ofvoluntary muscles, whether the umbilical cord hasbeen cut or the placenta is attached. Each product ofsuch a birth is considered live born. See relatedGestation; Rate: Birth and related rates.

Live-birth order —In the National Vital StatisticsSystem this item from the birth certificate refers to thetotal number of live births the mother has had,including the present birth as recorded on the birthcertificate. Fetal deaths are excluded. See relatedLivebirth.

Low birthweight —SeeBirthweight.

Major activity (or usual activity)—This is theprincipal activity of a person or of his or her age-sexgroup. For children 1–5 years of age, the majoractivity refers to ordinary play with other children; forchildren 5–17 years of age, the major activity refers toschool attendance; for adults 18–69 years of age, themajor activity usually refers to a job, housework, orschool attendance; for persons 70 years of age andover, the major activity refers to the capacity forindependent living (bathe, shop, dress, or eat without

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

alt

d

ive

s

of.

dieds.

as

,

r.

lth

es

te

65

g

,of

d

ny


needing the help of another person). See relatedLimitation of activity.

Managed care—Managed care is a health careplan that integrates the financing and delivery of hecare services by using arrangements with selectedhealth care providers to provide services for covereindividuals. Plans are generally financed usingcapitation fees. There are signifcant financial incentfor members of the plan to use the health careproviders associated with the plan. The plan includeformal programs for quality assurance and utilizationreview. Health maintenance organizations (HMO’s),preferred provider organizations (PPO’s), and pointservice (POS) plans are examples of managed careSee relatedHealth maintenance organization;Preferred provider organization.

Marital status—Marital status is classifiedthrough self-reporting into the categories married anunmarried. The term married encompasses all marrpeople including those separated from their spouseUnmarried includes those who are single (nevermarried), divorced, or widowed. The AbortionSurveillance Reports of the Centers for DiseaseControl and Prevention classified separated peopleunmarried before 1978.

Maternal mortality rate —SeeRate: Death andrelated rates.

Medicaid—This program is State operated andadministered but has Federal financial participation.Within certain broad federally determined guidelinesStates decide who is eligible; the amount, duration,and scope of services covered; rates of payment foproviders; and methods of administering the programMedicaid provides health care services for certainlow-income persons. Medicaid does not provide heaservices to all poor people in every State. Itcategorically covers participants in the Aid to Familiwith Dependent Children program and in theSupplemental Security Income program. In most Stait also covers certain other people deemed to bemedically needy. The program was authorized in 19

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

h

s

s

by Title XIX of the Social Security Act. See relatedHealth expenditures, national; Health maintenanceorganization; Medicare.

Medical specialties—SeePhysician specialty.

Medical vendor payments—Under the Medicaidprogram, medical vendor payments are payments(expenditures) to medical vendors from the Statethrough a fiscal agent or to a health insurance plan.Adjustments are made for Indian Health Servicepayments to Medicaid, cost settlements, third partyrecoupments, refunds, voided checks, and otherfinancial settlements that cannot be related to specificprovided claims. Excluded are payments made formedical care under the emergency assistanceprovisions, payments made from State medicalassistance funds that are not federally matchable,disproportionate share hospital payments, cost sharinor enrollment fees collected from recipients or a thirdparty, and administration and training costs.

Medicare—This is a nationwide health insuranceprogram providing health insurance protection topeople 65 years of age and over, people entitled tosocial security disability payments for 2 years or moreand people with end-stage renal disease, regardlessincome. The program was enacted July 30, 1965, asTitle XVIII, Health Insurance for the Aged of theSocial Security Act, and became effective on July 1,1966. It consists of two separate but coordinatedprograms, hospital insurance (Part A) andsupplementary medical insurance (Part B). See relateHealth expenditures, national; Health maintenanceorganization; Medicaid.

Mental health disorder—The Center for MentalHealth Services defines a mental health disorder as aof several disorders listed in theInternationalClassification of Diseases, Ninth Revision, ClinicalModification (ICD–9–CM) orDiagnostic andStatistical Manual of Mental Disorders, Third Edition(DSM-IIIR). Table IX shows diagnostic categories andcode numbers for ICD–9–CM/DSM-IIIR andcorresponding codes for theInternational

...................................................................... 153

is

erle,

teajor

.

c

lnn,

154 ....

Appendix II ............................................................................................................G

loss

ary

Classification of Diseases, Adapted for Use in theUnited States, Eighth Revision(ICDA-8) andDiagnostic and Statistical Manual of MentalDisorders, Second Edition(DSM-II). See relatedInternational Classification of Diseases, ClinicalModification.

Mental health organization—The Center forMental Health Services defines a mental healthorganization as an administratively distinct public orprivate agency or institution whose primary concernthe provision of direct mental health services to thementally ill or emotionally disturbed. Excluded areprivate office-based practices of psychiatrists,psychologists, and other mental health providers;psychiatric services of all types of hospitals oroutpatient clinics operated by Federal agencies oththan the Department of Veterans Affairs (for exampPublic Health Service, Indian Health Service,Department of Defense, and Bureau of Prisons);general hospitals that have no separate psychiatricservices, but admit psychiatric patients tononpsychiatric units; and psychiatric services ofschools, colleges, halfway houses, communityresidential organizations, local and county jails, Staprisons, and other human service providers. The mtypes of mental health organizations are describedbelow.

Freestanding psychiatric outpatient clinicsprovideonly outpatient services on either a regular oremergency basis. The medical responsibility forservices is generally assumed by a psychiatrist

General hospitals providing separate psychiatriservicesare non-Federal general hospitals thatprovide psychiatric services in either a separate

Table IX. Mental health codes, according to applicable revision of the DiagClassification of Diseases

Diagnostic category D

Alcohol related . . . . . . . . . . . . . . . . . . . . . . . . 291, 303, 309.13Drug related . . . . . . . . . . . . . . . . . . . . . . . . . . 294.3, 304, 309.14Organic disorders (other than alcoholism and drug) . . 290, 292, 293, 294 (eAffective disorders . . . . . . . . . . . . . . . . . . . . . . 296, 298.0, 300.4Schizophrenia . . . . . . . . . . . . . . . . . . . . . . . . . 295

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

psychiatric inpatient, outpatient, or partialhospitalization service with assigned staff andspace.

Multiservice mental health organizationsdirectlyprovide two or more of the program elementsdefined under Mental health service type and arenot classifiable as a psychiatric hospital, generalhospital, or a residential treatment center foremotionally disturbed children. (The classificationof a psychiatric or general hospital or a residentiatreatment center for emotionally disturbed childretakes precedence over a multiservice classificatioeven if two or more services are offered.)

Partial care organizationsprovide a program ofambulatory mental health services.

Private mental hospitalsare operated by a soleproprietor, partnership, limited partnership,corporation, or nonprofit organization, primarilyfor the care of persons with mental disorders.

Psychiatric hospitalsare hospitals primarilyconcerned with providing inpatient care andtreatment for the mentally ill. Psychiatric inpatientunits of Department of Veterans Affairs generalhospitals and Department of Veterans Affairsneuropsychiatric hospitals are combined into thecategory Department of Veterans Affairspsychiatric hospitals because of their similarity insize, operation, and length of stay.

Residential treatment centers for emotionallydisturbed childrenmust meet all of the followingcriteria: (a) Not licensed as a psychiatric hospitaland primary purpose is to provide individuallyplanned mental health treatment services in

nostic and Statistical Manual of Mental Disorders and International

SM-II/ICDA-8 DSM-IIIR/ICD-9-CM

291, 303, 305.0292, 304, 305.1–305.9, 327, 328

xcept 294.3), 309.0, 309.2–309.9 290, 293, 294, 310296, 298.0, 300.4, 301.11, 301.13295

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

,

ar

f

e


conjunction with residential care; (b) Include aclinical program that is directed by a psychiatristpsychologist, social worker, or psychiatric nursewith a graduate degree; (c) Serve children andyouth primarily under the age of 18; and (d)Primary diagnosis for the majority of admissionsis mental illness, classified as other than mentalretardation, developmental disability, andsubstance-related disorders, according toDSM-II/ICDA-8 or DSM-IIIR/ICD–9–CM codes.See relatedTable IX.Mental health codes.

State and county mental hospitalsare under theauspices of a State or county government oroperated jointly by a State and countygovernment.

See relatedAddition; Mental health service type.

Mental health service type—refers to thefollowing kinds of mental health services:

Inpatient careis the provision of 24-hour mentalhealth care in a mental health hospital setting.

Outpatient careis the provision of ambulatorymental health services for less than 3 hours at asingle visit on an individual, group, or familybasis, usually in a clinic or similar organization.Emergency care on a walk-in basis, as well as cprovided by mobile teams who visit patientsoutside these organizations are included.‘‘Hotline’’ services are excluded.

Partial care treatmentis a planned program ofmental health treatment services generallyprovided in visits of 3 or more hours to groups opatients. Included are treatment programs thatemphasize intensive short-term therapy andrehabilitation; programs that focus on recreation,and/or occupational program activities, includingsheltered workshops; and education and trainingprograms, including special education classes,therapeutic nursery schools, and vocationaltraining.

Residential treatment careis the provision ofovernight mental health care in conjunction with

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

e

an intensive treatment program in a setting otherthan a hospital. Facilities may offer care toemotionally disturbed children or mentally illadults.

See relatedAddition; Mental health organization.

Metropolitan statistical area (MSA)—Thedefinitions and titles of MSA’s are established by theU.S. Office of Management and Budget with theadvice of the Federal Committee on MetropolitanStatistical Areas. Generally speaking, an MSA consistsof a county or group of counties containing at leastone city (or twin cities) having a population of 50,000or more plus adjacent counties that are metropolitan incharacter and are economically and socially integratedwith the central city. In New England, towns and citiesrather than counties are the units used in definingMSA’s. There is no limit to the number of adjacentcounties included in the MSA as long as they areintegrated with the central city. Nor is an MSA limitedto a single State; boundaries may cross State lines.Metropolitan population, as used in this report inconnection with data from the National HealthInterview Survey, is based on MSA’s as defined in the1980 census and does not include any subsequentadditions or changes.

Multiservice mental health organizations—SeeMental health organization.

National ambient air quality standards—TheFederal Clean Air Act of 1970, amended in 1977 and1990, required the Environmental Protection Agency(EPA) to establish National Ambient Air QualityStandards. EPA has set specific standards for each ofsix major pollutants: carbon monoxide, lead, nitrogendioxide, ozone, sulfur dioxide, and particulate matterwhose aerodynamic size is equal to or less than 10microns (PM-10). Each pollutant standard represents amaximum concentration level (micrograms per cubicmeter) that cannot be exceeded during a specified timinterval. A county meets the national ambient airquality standards if none of the six pollutants exceedthe standard during a 12-month period. SeerelatedParticulate matter; Pollutant.

...................................................................... 155

i

g

te

,the

is

n

t

il

h

d

gr

156 .........

Appendix II ............................................................................................................G

loss

ary

Neonatal mortality rate—SeeRate: Death andrelated rates.

Non-Federal physicians—SeePhysician.

Nonpatient revenue—Nonpatient revenues arethose revenues received for which no direct patientcare services are rendered. The most widely recognsource of nonpatient revenues is philanthropy.Philanthropic support may be direct from individualsor may be obtained through philanthropic fund raisinorganizations such as the United Way. Support mayalso be obtained from foundations or corporations.Philanthropic revenues may be designated for direcpatient care use or may be contained in an endowmfund where only the current income may be tapped.

Nonprofit hospitals—SeeHospital.

Notifiable disease—A notifiable disease is onethat, when diagnosed, health providers are requiredusually by law, to report to State or local public healofficials. Notifiable diseases are those of public interby reason of their contagiousness, severity, orfrequency.

Nursing care—The following definition ofnursing care applies to data collected in NationalNursing Home Surveys through 1977. Nursing carethe provision of any of the following services:application of dressings or bandages; bowel andbladder retraining; catheterization; enema; full bedbath; hypodermic, intramuscular, or intravenousinjection; irrigation; nasal feeding; oxygen therapy; atemperature-pulse-respiration or blood pressuremeasurement. See relatedNursing home.

Nursing care homes—SeeNursing home.

Nursing home—In the Online Certification andReporting database, a nursing home is a facility thacertified and meets the Health Care FinancingAdministration’s long-term care requirements forMedicare and Medicaid eligibility. In the NationalMaster Facility Inventory and the National NursingHome Survey a nursing home is an establishment wthree or more beds that provides nursing or personacare services to the aged, infirm, or chronically ill. T

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

zed

nt

st

d

is

th

e

following definitions of nursing home types apply todata collected in National Nursing Home Surveysthrough 1977.

Nursing care homesmust employ one or morefull-time registered or licensed practical nurses anmust provide nursing care to at least one-half theresidents.

Personal care homes with nursinghave some butfewer than one-half the residents receiving nursincare. In addition, such homes must employ one omore registered or licensed practical nurses ormust provide administration of medications andtreatments in accordance with physicians’ orders,supervision of self-administered medications, orthree or more personal services.

Personal care homes without nursinghave noresidents who are receiving nursing care. Thesehomes provide administration of medications andtreatments in accordance with physicians’ orders,supervision of self-administered medications, orthree or more personal services.

Domiciliary care homesprimarily providesupervisory care but also provide one or twopersonal services.

Nursing homes are certified by the Medicareand/or Medicaid program. The following definitions ofcertification levels apply to data collected in NationalNursing Home Surveys of 1973–74, 1977, and 1985.

Skilled nursing facilitiesprovide the mostintensive nursing care available outside of ahospital. Facilities certified by Medicare provideposthospital care to eligible Medicare enrollees.Facilities certified by Medicaid as skilled nursingfacilities provide skilled nursing services on adaily basis to individuals eligible for Medicaidbenefits.

Intermediate care facilitiesare certified by theMedicaid program to provide health-relatedservices on a regular basis to Medicaid eligibleswho do not require hospital or skilled nursing

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

e

o

o

e

e

e


facility care but do require institutional care abovthe level of room and board.

Not certified facilitiesare not certified as providersof care by Medicare or Medicaid.

See relatedNursing care; Resident.

Nursing home expenditures—SeeHealthexpenditures, national.

Occupancy rate—The American HospitalAssociation defines hospital occupancy rate as theaverage daily census divided by the average numberhospital beds during a reporting period. Average dailycensus is defined by the American Hospital Associatias the average number of inpatients, excludingnewborns, receiving care each day during a reportingperiod. The occupancy rate for facilities other thanhospitals is calculated as the number of residentsreported at the time of the interview divided by thenumber of beds reported. In the Online SurveyCertification and Reporting database, occupancy is thtotal number of residents on the day of certificationinspection divided by the total number of beds on theday of certification.

Office—In the National Health Interview Survey,an office refers to the office of any physician in privatpractice not located in a hospital. In the NationalAmbulatory Medical Care Survey, an office is anylocation for a physician’s ambulatory practice otherthan hospitals, nursing homes, other extended carefacilities, patients’ homes, industrial clinics, collegeclinics, and family planning clinics. However, privateoffices in hospitals are included. See relatedOffıcevisit; Outpatient visit; Physician; Physician contact.

Office-based physician—SeePhysician.

Office visit—In the National Ambulatory MedicalCare Survey, an office visit is any direct personalexchange between an ambulatory patient and aphysician or members of his or her staff for thepurposes of seeking care and rendering health servicSee relatedOutpatient visit; Physician contact.

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

f

n

s.

Operations—SeeProcedure.

Outpatient department—According to theNational Hospital Ambulatory Medical Care Survey(NHAMCS), an outpatient department (OPD) is ahospital facility where nonurgent ambulatory medicalcare is provided. The following are examples of thetypes of OPD’s excluded from the NHAMCS:ambulatory surgical centers, chemotherapy, employeehealth services, renal dialysis, methadone maintenance,and radiology. An outpatient department visit is adirect personal exchange between a patient and aphysician or other health care provider working underthe physician’s supervision for the purpose of seekingcare and receiving personal health services. See relatedEmergency department; Hospital.

Outpatient surgery—According to the AmericanHospital Association, outpatient surgery is performedon patients who do not remain in the hospitalovernight and occurs in inpatient operating suites,outpatient surgery suites, or procedure rooms within anoutpatient care facility. Outpatient surgery is a surgicaloperation, whether major or minor, performed inoperating or procedure rooms. A surgical operationinvolving more than one surgical procedure isconsidered one surgical operation. See relatedAmbulatory surgery.

Outpatient visit—The American HospitalAssociation defines outpatient visits as visits forreceipt of medical, dental, or other services by patientswho are not lodged in the hospital. Each appearanceby an outpatient to each unit of the hospital is countedindividually as an outpatient visit. See relatedOffıcevisit; Physician contact.

Partial care organization—SeeMental healthorganization.

Partial care treatment—SeeMental healthservice type.

Particulate matter—Particulate matter is definedas particles of solid or liquid matter in the air,including nontoxic materials (soot, dust, and dirt) andtoxic materials (for example, lead, asbestos, suspended

...................................................................... 157

e

e

y

0rp

.

ve

158 .......

Appendix II ............................................................................................................G

loss

ary

sulfates, and nitrates). See relatedNational ambient airquality standards; Pollutant.

Patient—A patient is a person who is formallyadmitted to the inpatient service of a hospital forobservation, care, diagnosis, or treatment. See relatAdmission; Average length of stay; Days of care;Discharge; Hospital.

Percent change—SeeAverage annual rate ofchange.

Perinatal mortality rate, ratio —SeeRate: Deathand related rates.

Personal care homes with or withoutnursing—SeeNursing home.

Personal health care expenditures—SeeHealthexpenditures, national.

Physician—Physicians, through self-reporting, arclassified by the American Medical Association andothers as licensed doctors of medicine or osteopathas follows:

Active (or professionally active) physiciansarecurrently practicing medicine for a minimum of 2hours per week. Excluded are physicians who ainactive practicing medicine less than 20 hoursweek, have unknown addresses, or specialties nclassified (when specialty information ispresented).

Federal physiciansare employed by the FederalGovernment; non-Federal or civilian physiciansare not.

Hospital-based physiciansspend the plurality oftheir time as salaried physicians in hospitals.

Offıce-based physiciansspend the plurality of theirtime working in practices based in private offices

Data for physicians are presented by type ofeducation (doctors of medicine and doctors ofosteopathy); place of education (U.S. medicalgraduates and international medical graduates); actistatus (professionally active and inactive); employm

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

d

,

eerot

itynt

setting (Federal and non-Federal); area of specialty;and geographic area. See relatedOffıce; Physicianspecialty.

Physician contact—In the National HealthInterview Survey, a physician contact is defined as aconsultation with a physician in person or bytelephone, for examination, diagnosis, treatment, oradvice. The service may be provided by the physicianor by another person working under the physician’ssupervision. Contacts involving services provided on amass basis (for example, blood pressure screenings)and contacts for hospital inpatients are not included.

Place of contact includes office, hospital outpatientclinics, emergency room, telephone (advice given by aphysician in a telephone call), home (any place inwhich a person was staying at the time a physicianwas called there), clinics, HMO’s, and other placeslocated outside a hospital.

In the National Health Interview Survey, analysesof the annual number of physician contacts and placeof contact are based upon data collected using a2-week recall period and are adjusted to produceannual estimates. Analyses of children without aphysician contact during the past 12-month period arebased upon a different question that uses a 12-monthrecall period. Analyses of the interval since lastphysician contact are based upon the length of timebefore the week of interview in which the physicianwas last consulted. See relatedOffıce; Offıce visit.

Physician specialty—A physician specialty is anyspecific branch of medicine in which a physician mayconcentrate. Data are based on physician self-reports oftheir primary area of speciality. Physician data arebroadly categorized into two general areas of practice:generalists and specialists.

Generalist physiciansare synonymous withprimary care generalists and only includephysicians practicing in the general fields offamily and general practice, general internalmedicine, and general pediatrics. They specificallyexclude primary care specialists.

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

s

S

r


Primary care specialistspractice in thesubspecialties of general and family practice,internal medicine, and pediatrics. The primary carsubspecialties for family practice include geriatricmedicine and sports medicine. Primary caresubspecialties for internal medicine includediabetes, endocrinology and metabolism,hematology, hepatology, cardiac electrophysiologyinfectious diseases, diagnostic laboratoryimmunology, geriatric medicine, sports medicine,nephrology, nutrition, medical oncology, andrheumatology. Primary care subspecialties forpediatrics include adolescent medicine, criticalcare pediatrics, neonatal-perinatal medicine,pediatric allergy, pediatric cardiology, pediatricendocrinology, pediatric pulmonology, pediatricemergency medicine, pediatric gastroenterology,pediatric hematology/oncology, diagnosticlaboratory immunology, pediatric nephrology,pediatric rheumatology, and sports medicine.

Specialist physicianspractice in the primary carespecialties, in addition to all other specialist fieldsnot included in the generalist definition. Specialistfields include allergy and immunology, aerospacemedicine, anesthesiology, cardiovascular diseasechild and adolescent psychiatry, colon and rectalsurgery, dermatology, diagnostic radiology,forensic pathology, gastroenterology, generalsurgery, medical genetics, neurology, nuclearmedicine, neurological surgery, obstetrics andgynecology, occupational medicine,ophthalmology, orthopedic surgery, otolaryngologypsychiatry, public health and general preventivemedicine, physical medicine and rehabilitation,plastic surgery, anatomic and clinical pathology,pulmonary diseases, radiation oncology, thoracicsurgery, urology, addiction medicine, critical caremedicine, legal medicine, and clinicalpharmacology.

ee relatedPhysician.

Pollutant—A pollutant is any substance thatenders the atmosphere or water foul or noxious to

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

e

,

,

,

health. See relatedNational ambient air qualitystandards; Particulate matter.

Population—The U.S. Bureau of the Censuscollects and publishes data on populations in theUnited States according to several different definitions.Various statistical systems then use the appropriatepopulation for calculating rates.

Total populationis the population of the UnitedStates, including all members of the Armed Forcesliving in foreign countries, Puerto Rico, Guam,and the U.S. Virgin Islands. Other Americansabroad (for example, civilian Federal employeesand dependents of members of the Armed Forcesor other Federal employees) are not included.

Resident populationincludes persons whose usualplace of residence (that is, the place where oneusually lives and sleeps) is in one of the 50 Statesor the District of Columbia. It includes membersof the Armed Forces stationed in the United Statesand their families. It excludes internationalmilitary, naval, and diplomatic personnel and theirfamilies located here and residing in embassies orsimilar quarters. Also excluded are internationalworkers and international students in this countryand Americans living abroad. The residentpopulation is usually the denominator whencalculating birth and death rates and incidence ofdisease. The resident population is also thedenominator for selected population-based ratesthat use numerator data from the National NursingHome Survey.

Civilian populationis the resident populationexcluding members of the Armed Forces.However, families of members of the ArmedForces are included. This population is thedenominator in rates calculated for the NCHSNational Hospital Discharge Survey.

Civilian noninstitutionalized populationis thecivilian population not residing in institutions.Institutions include correctional institutions,detention homes, and training schools for juvenile

...................................................................... 159

e

y

9

o

l

ot

,

160 ......

Appendix II ............................................................................................................G

loss

ary

delinquents; homes for the aged and dependent(for example, nursing homes and convalescenthomes); homes for dependent and neglectedchildren; homes and schools for the mentally orphysically handicapped; homes for unwedmothers; psychiatric, tuberculosis, and chronicdisease hospitals; and residential treatment centThis population is the denominator in ratescalculated for the NCHS National HealthInterview Survey; National Health and NutritionExamination Survey; National AmbulatoryMedical Care Survey; and the National HospitalAmbulatory Medical Care Survey.

Postneonatal mortality rate—SeeRate: Deathand related rates.

Poverty level—Poverty statistics are based ondefinitions originally developed by the Social SecuritAdministration. These include a set of money incomethresholds that vary by family size and composition.Families or individuals with income below theirappropriate thresholds are classified as below thepoverty level. These thresholds are updated annuallyby the U.S. Bureau of the Census to reflect changesthe Consumer Price Index for all urban consumers(CPI-U). For example, the average poverty thresholdfor a family of four was $16,036 in 1996 and $13,35in 1990. For more information, see U.S. Bureau of thCensus:Money Income of Households, Families, andPersons in the United States, 1996.Series P-60.Washington. U.S. Government Printing office. SeerelatedConsumer Price Index.

Preferred provider organization (PPO)—Healthplan generally consisting of hospital and physicianproviders. The PPO provides health care services toplan members usually at discounted rates in return fexpedited claims payment. Plan members can use Por non-PPO health care providers, however, financiaincentives are built into the benefit structure toencourage utilization of PPO providers. See relatedManaged care.

Prevalence—Prevalence is the number of casesa disease, infected persons, or persons with some o

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

rs.

in

e

rPO

fher

attribute present during a particular interval of time. Itis often expressed as a rate (for example, theprevalence of diabetes per 1,000 persons during ayear). See relatedIncidence.

Primary admission diagnosis—In the NationalHome and Hospice Care Survey the primary admissiondiagnosis is the first-listed diagnosis at admission onthe patient’s medical record as provided by the agencystaff member most familiar with the care provided tothe patient.

Primary care specialties—SeePhysicianspecialty.

Private expenditures—SeeHealth expenditures,national.

Procedure—The National Hospital DischargeSurvey (NHDS) and the National Survey ofAmbulatory Surgery (NSAS) define a procedure as asurgical or nonsurgical operation, diagnostic procedureor therapeutic procedure (such as respiratory therapy)recorded on the medical record of discharged patients.A maximum of four procedures per discharge inNHDS and up to six procedures per discharge inNSAS were recorded and coded to theInternationalClassification of Diseases, Ninth Revision, ClinicalModification. Previous editions ofHealth, UnitedStatesclassified procedures into surgical and diagnosticand other nonsurgical procedures. The distinctionbetween surgical and diagnostic and nonsurgicalprocedures has become less meaningful due to thedevelopment of minimally invasive and noninvasiveprocedures thus the practice of classifying procedureshas been discontinued. See relatedAmbulatorysurgery; Outpatient surgery.

Proprietary hospitals—SeeHospital.

Psychiatric hospitals—SeeHospital; Mentalhealth organization.

Public expenditures—SeeHealth expenditures,national.

Public health activities—Public health activitiesmay include any of the following essential services of

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

,dt

o

s

e

the

e

s

f

f

ntsal,

to

for


public health—surveillance, investigations, education,community mobilization, workforce training, research,and personal care services delivered or funded bygovernmental agencies.

Race—Beginning in 1976 the FederalGovernment’s data systems classified individuals intothe following racial groups: American Indian or AlaskaNative, Asian or Pacific Islander, black, and white.Depending on the data source, the classification byrace may be based on self-classification or onobservation by an interviewer or other persons fillingout the questionnaire. Starting in 1989, data from theNational Vital Statistics System for newborn infantsand fetal deaths are tabulated according to race ofmother, and trend data by race shown in this reporthave been retabulated by race of mother for all yearsbeginning with 1980. Before 1980 data were tabulateby race of newborn and fetus according to race of boparents. If the parents were of different races and oneparent was white, the child was classified according tthe race of the other parent. When neither parent waswhite, the child was classified according to father’srace, with one exception: if either parent wasHawaiian, the child was classified Hawaiian. Before1964 the National Vital Statistics System classified allbirths for which race was unknown as white.Beginning in 1964 these births were classifiedaccording to information on the previous record.

In Health, United States, trends of birth rates,birth characteristics, and infant and maternal mortalityrates are calculated according to race of mother unlespecified otherwise. Vital event rates for the AmericanIndian or Alaska Native population shown in this bookare based on the total U.S. resident population ofAmerican Indians and Alaska Natives as enumeratedby the U.S. Bureau of Census. In contrast the IndianHealth Service calculates vital event rates for thispopulation based on U.S. Bureau of Census countydata for American Indians and Alaska Natives whoreside on or near reservations. See relatedHispanicorigin.

Rate—A rate is a measure of some event, diseasor condition in relation to a unit of population, along

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

h

s

,

with some specification of time. See relatedAgeadjustment; Population.

Birth and related rates

Birth rate is calculated by dividing the number oflive births in a population in a year by themidyear resident population. For census years,rates are based on unrounded census counts ofresident population, as of April 1. For thenoncensus years of 1981–89 and 1991, rates arbased on national estimates of the residentpopulation, as of July 1, rounded to 1,000’s.Population estimates for 5-year age groups aregenerated by summing unrounded populationestimates before rounding to 1,000’s. Starting in1992 rates are based on unrounded nationalpopulation estimates. Birth rates are expressed athe number of live births per 1,000 population.The rate may be restricted to births to women ofspecific age, race, marital status, or geographiclocation (specific rate), or it may be related to theentire population (crude rate). See relatedCohortfertility; Live birth.

Fertility rate is the total number of live births,regardless of age of mother, per 1,000 women oreproductive age, 15–44 years.

Death and related rates

Death rateis calculated by dividing the number odeaths in a population in a year by the midyearresident population. For census years, rates arebased on unrounded census counts of the residepopulation, as of April 1. For the noncensus yearof 1981–89 and 1991, rates are based on nationestimates of the resident population, as of July 1rounded to 1,000’s. Population estimates for10-year age groups are generated by summingunrounded population estimates before rounding1,000’s. Starting in 1992 rates are based onunrounded national population estimates. Ratesthe Hispanic and non-Hispanic white populationsin each year are based on unrounded Statepopulation estimates for States in the Hispanicreporting area. Death rates are expressed as the

...................................................................... 161

at,s

e

res,

re

l

s

a.f

ee

trtin

e

ts

162 .........

Appendix II ............................................................................................................G

loss

ary

number of deaths per 100,000 population. The rmay be restricted to deaths in specific age, racesex, or geographic groups or from specific causeof death (specific rate) or it may be related to thentire population (crude rate).

Fetal death rateis the number of fetal deaths withstated or presumed gestation of 20 weeks or modivided by the sum of live births plus fetal deathstated per 1,000 live births plus fetal deaths.Latefetal death rateis the number of fetal deaths withstated or presumed gestation of 28 weeks or modivided by the sum of live births plus late fetaldeaths, stated per 1,000 live births plus late fetadeaths. See relatedFetal death; Gestation.

Infant mortality ratebased on period files iscalculated by dividing the number of infant deathduring a calendar year by the number of livebirths reported in the same year. It is expressedthe number of infant deaths per 1,000 live birthsNeonatal mortality rateis the number of deaths ochildren under 28 days of age, per 1,000 livebirths.Postneonatal mortality rateis the numberof deaths of children that occur between 28 daysand 365 days after birth, per 1,000 live births. SrelatedInfant death.

Birth cohort infant mortality ratesare based onlinked birth and infant death files. In contrast toperiod rates in which the births and infant deathsoccur in the same period or calendar year, infandeaths comprising the numerator of a birth cohorate may have occurred in the same year as, orthe year following the year of birth. The birthcohort infant mortality rate is expressed as thenumber of infant deaths per 1,000 live births. SerelatedBirth cohort.

Perinatal relates to the period surrounding thebirth event. Rates and ratios are based on evenreported in a calendar year.Perinatal mortalityrate is the sum of late fetal deaths plus infantdeaths within 7 days of birth divided by the sumof live births plus late fetal deaths, stated per

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

e

s

1,000 live births plus late fetal deaths.Perinatalmortality ratio is the sum of late fetal deaths plusinfant deaths within 7 days of birth divided by thenumber of live births, stated per 1,000 live births.

Feto-infant mortality rateis the sum of late fetaldeaths plus all infant deaths divided by the sum oflive births plus late fetal deaths, stated per 1,000live births plus late fetal deaths. See relatedFetaldeath; Gestation; Infant death; Live birth.

Maternal deathis one for which the certifyingphysician has designated a maternal condition asthe underlying cause of death. Maternal conditionsare those assigned to Complications of pregnancy,childbirth, and the puerperium, ICD-9 codes630–676. (See relatedtable V.) Maternal mortalityrate is defined as the number of maternal deathsper 100,000 live births. The maternal mortalityrate is a measure of the likelihood that a pregnantwoman will die from maternal causes. The numberof live births used in the denominator is a proxyfor the population of pregnant women who are atrisk of a maternal death.

Region—SeeGeographic region and division.

Registered hospitals—SeeHospital.

Registered nursing education—Registerednursing data are shown by level of educationalpreparation. Baccalaureate education requires at least 4years of college or university; associate degreeprograms are based in community colleges and areusually 2 years in length; and diploma programs arebased in hospitals and are usually 3 years in length.

Registration area—The United States hasseparate registration areas for birth, death, marriage,and divorce statistics. In general, registration areascorrespond to States and include two separateregistration areas for the District of Columbia and NewYork City. All States have adopted laws that requirethe registration of births and deaths and the reportingof fetal deaths. It is believed that more than 99 percentof the births and deaths occurring in this country areregistered.

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

d

h93

d

ne

t

in

e

e

he

m

fo

9


The death registration areawas established in1900 with 10 States and the District of Columbia, anthe birth registration areawas established in 1915,also with 10 States and the District of Columbia. Botareas have covered the entire United States since 1Currently, Puerto Rico, U.S. Virgin Islands, and Guamcomprise separate registration areas, although theirare not included in statistical tabulations of U.S.resident data. See relatedReporting area.

Relative survival rate—The relative survival rateis the ratio of the observed survival rate for the patiegroup to the expected survival rate for persons in thgeneral population similar to the patient group withrespect to age, sex, race, and calendar year ofobservation. The 5-year relative survival rate is usedestimate the proportion of cancer patients potentiallycurable. Because over one-half of all cancers occurpersons 65 years of age and over, many of theseindividuals die of other causes with no evidence ofrecurrence of their cancer. Thus, because it is obtainby adjusting observed survival for the normal lifeexpectancy of the general population of the same agthe relative survival rate is an estimate of the chanceof surviving the effects of cancer.

Reporting area—In the National Vital StatisticsSystem, the reporting area for such basic items on tbirth and death certificates as age, race, and sex, isbased on data from residents of all 50 States in theUnited States and the District of Columbia. Thereporting area for selected items such as Hispanicorigin, educational attainment, and marital status, isbased on data from those States that require the iteto be reported, whose data meet a minimum level ofcompleteness (such as, 80 or 90 percent), and areconsidered to be sufficiently comparable to be usedanalysis. In 1993–96 the reporting area for Hispanicorigin of decedent on the death certificate included 4States and the District of Columbia. See relatedRegistration area; National Vital Statistics SysteminAppendix I.

Resident—In the Online Certification andReporting database, all residents in certified facilities

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

3.

ata

t

o

d

,

r

are counted on the day of certification inspection. Inthe National Nursing Home Survey, a resident is aperson on the roster of the nursing home as of thenight before the survey. Included are all residents forwhom beds are maintained even though they may beon overnight leave or in a hospital. See relatedNursinghome.

Resident population—SeePopulation.

Residential treatment care—SeeMental healthservice type.

Residential treatment centers for emotionallydisturbed children—SeeMental health organization.

Self-assessment of health—SeeHealth status,respondent-assessed.

Short-stay hospitals—SeeHospital.

Skilled nursing facilities—SeeNursing home.

Smoker—SeeCurrent smoker.

Specialty hospitals—SeeHospital.

State health agency—The agency or departmentwithin State government headed by the State orterritorial health official. Generally, the State healthagency is responsible for setting statewide publichealth priorities, carrying out national and Statemandates, responding to public health hazards, andassuring access to health care for underserved Stateresidents.

Substance abuse treatment clients—In theSubstance Abuse and Mental Health ServicesAdministration’s Uniform Facilities Data Set substanceabuse treatment clients have been admitted totreatment and have been seen on a scheduledappointment basis at least once in the month before thesurvey reference date or were inpatients on the surveyreference date. Types of treatment include 24-hourdetoxification, 24-hour rehabilitation or residentialcare, and outpatient care.

Surgical operations—SeeProcedure.

Surgical specialties—SeePhysician specialty.

...................................................................... 163

o

nsy

d

ee

en

l0

ar

o

6or,

g

fr

er

ge

,

f

164 ......

Appendix II ............................................................................................................G

loss

ary

Uninsured—SeeHealth insurance coverage.

Urbanization—In this report death rates arepresented according to level of urbanization of thedecedent’s county of residence. Metropolitan andnonmetropolitan counties are categorized intourbanization levels based on an NCHS-modificationthe 1993 rural-urban continuum codes. The 1993rural-urban continuum codes were developed by theEconomic Research Service, U.S. Department ofAgriculture using the 1993 U.S. Office of Managemeand Budget definition of metropolitan statistical area(MSA’s). The codes classify metropolitan counties bpopulation size and level of urbanization andnonmetropolitan counties by level of urbanization anproximity to metropolitan areas. NCHS modified the1993 rural-urban continuum codes to make thedefinition of core and fringe metropolitan countiescomparable to the definitions used for the 1983 codFor this report, the 10 categories of counties have bcollapsed into 5 categories (a) core metropolitancounties contain the primary central city of an MSAwith a 1990 population of 1 million or more; (b) fringmetropolitan counties are the noncore counties of aMSA with 1990 population of 1 million or more; (c)medium or small metropolitan counties are in MSA’swith 1990 population under 1 million; (d) urbannonmetropolitan counties are not in MSA’s and have2,500 or more urban residents in 1990; and (e) ruracounties are not in MSA’s and have fewer than 2,50urban residents in 1990. See relatedMetropolitanstatistical area (MSA).

Usual source of care—Usual source of care wasmeasured in the National Health Interview Survey(NHIS) in 1991 by asking the respondent, ‘‘Is thereparticular clinic, health center, doctor’s office, or otheplace that you usually go to if you are sick or needadvice about your health?’’ In 1993 and 1994 therespondent was asked, ‘‘Is there a particular personplace that __ usually goes to when ___ is sick orneeds advice about __ health?’’ In the 1995 and 199NHIS, the respondent was asked ‘‘Is there one doctperson, or place that __ usually goes to when __ is

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

f

t

s.en

r

sick or needs advice about ___ health?’’ Persons whoreported multiple sources of care are defined as havina usual source of care. Additionally, persons whoreported the emergency room as their usual source ocare are defined as having no usual source of care fothe purposes of this report.

Wages and salaries—SeeEmployer costs foremployee compensation.

Years of potential life lost—Years of potentiallife lost (YPLL) is a measure of premature mortality.Starting withHealth, United States, 1996–97, YPLL ispresented for persons under 75 years of age becausethe average life expectancy in the United States is ov75 years. YPLL-75 is calculated using the followingeight age groups: under 1 year, 1–14 years, 15–24years, 25–34 years, 35–44 years, 45–54 years, 55–64years, 65–74 years. The number of deaths for each agroup is multiplied by the years of life lost, calculatedas the difference between age 75 years and themidpoint of the age group. For the eight age groupsthe midpoints are 0.5, 7.5, 19.5, 29.5, 39.5, 49.5, 59.5and 69.5. For example, the death of a person 15–24years of age counts as 55.5 years of life lost. Years opotential life lost is derived by summing years of lifelost over all age groups. InHealth, United States, 1995and earlier editions, YPLL was presented for personsunder 65 years of age. For more information, seeCenters for Disease Control.MMWR. Vol 35 no 25S,suppl. 1986.

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

t

of

M

.............................................................................................. Appendix IIIAdditionalData

Detailed Tables With Additional Years of DataAvailable in Electronic Spreadsheet Files

Many of the detailed tables in this report presendata for extended time periods. Because of spacelimitations on the printed page, only selected yearsdata are shown to highlight major trends. For thetables listed below, additional years of data areavailable in electronic spreadsheet files that may beaccessed through the internet and on CD-ROM.

Tablenumber Table topic

1 Resident population2 Poverty3 Fertility rates and birth rates5 Live births6 Prenatal care7 Teenage childbearing8 Nonmarital childbearing9 Maternal education

11 Low birthweight15 Abortions16 Abortions17 Contraception19 Infant mortality rates20 Infant mortality rates21 Infant mortality rates22 Infant mortality rates28 Life expectancy29 Age-adjusted death rates by State30 Age-adjusted death rates for selected causes31 Years of potential life lost36 Death rates for all causes37 Diseases of heart38 Cerebrovascular diseases39 Malignant neoplasms40 Malignant neoplasms of respiratory system41 Malignant neoplasm of breast42 Chronic obstructive pulmonary diseases43 Human immunodeficiency virus (HIV) infection44 Maternal mortality45 Motor vehicle-related injuries46 Homicide47 Suicide48 Firearm-related injuries49 Occupational diseases50 Occupational injury deaths53 Notifiable diseases61 Cigarette smoking

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

To access the files on the internet, go to the FTPserver on the NCHS homepage atwww.cdc.gov/nchswwwand select ‘‘Data Warehouse’’andHealth, United States.

Spreadsheet files are also available on a CD-ROentitled ‘‘Publications from the National Center forHealth Statistics,’’ featuringHealth, United States,1999, vol 1 no 5, 1999. The CD-ROM may bepurchased from the Government Printing Office or theNational Technical Information Service.

Additional data years available

1981–89, 1991–951986–891981–84, 1986–891971–74, 1976–79, 1981–84, 1986–89, 1991–931981–84, 1986–891981–84, 1986–891981–84, 1986–891981–84, 1986–891981–84, 1986–891981–84, 1986–881981–84, 1986–8819901984, 1985–86, 1988–891984, 1985–86, 1988–8919841981–84, 1986–881975, 1981–841992–94, 1993–95, 1994–961991–931985, 1991–961981–84, 1986–89, 1991–941981–84, 1986–89, 1991–931981–84, 1986–89, 1991–931981–84, 1986–89, 1991–931981–84, 1986–89, 1991–931981–84, 1986–89, 1991–931981–84, 1986–8919881981–84, 1986–89, 1991–931981–84, 1986–89, 1991–931981–84, 1986–89, 1991–931981–84, 1986–89, 1991–931981–84, 1986–87 1989, 1991–921979, 1981–84, 1986–881981–84, 1986–881985, 1988–89, 1991–931987–88

...................................................................... 165

http://www.cdc.gov/nchswww/

Tablenumber Table topic Additional data years available

62 Cigarette smoking 1987–8864 Use of selected substances 1982, 198865 Use of selected substances 1981–84, 1986, 1987–8866 Cocaine-related emergency room episodes 1986–8974 Occupational injuries 1981–84, 1986–8875 Physician contacts 1985–8677 Physician contacts 1990–92, 1993–9583 Ambulatory care visits 1993–9487 Additions to mental health organizations 1986, 198890 Discharges 1988–8991 Discharges 1989, 199496 Hospital admissions 1991–93

101 Persons employed 1983–84, 1986–89, 1991103 Physicians 1970, 1987, 1989, 1992–94106 Staff in mental health organizations 1986, 1988110 Hospitals 1991–93112 Community hospital beds 1985, 1988–89, 1995–96113 Occupancy rates 1985, 1988–89, 1995–96117 Consumer Price Index 1965, 1975122 Employers’ costs and health insurance 1992–93, 1995–96123 Hospital expenses 1991–92, 1994127 Funding for health research 1984, 1986–89, 1991–92132 Health maintenance organizations 1984, 1986–87, 1989, 1991134 Medicare 1988–89, 1991–94135 Medicare 1967, 1991–94136 Medicaid 1986–89, 1991–93137 Medicaid 1986–89, 1991–93138 Department of Veterans Affairs 1985, 1988–89, 1991143 Medicare 1994–95

166 ..................................................................................................................................

Appendix III ...........................................................................................................Ad

ditio

nalD

ata

DHHS Pub No. (PHS) 99-1232-19-0477 (7/99)

DEPARTMENT OFHEALTH & HUMAN SERVICES

Centers for Disease Control and PreventionNational Center for Health Statistics6525 Belcrest RoadHyattsville, Maryland 20782-2003

OFFICIAL BUSINESSPENALTY FOR PRIVATE USE, $300