rheumatoid arthritis in adults - centers for disease ...rheumatoid arthritis in adults arnold engel,...
TRANSCRIPT
Data from theNATIONAL HEALTH SURVEY
Series 11Number 17
Rheumatoid Arthritis
in Adults
United States -1960 = 1962
U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
Public Health Service
Health Resources Administration
National Center for Health Statistics
Rockville, Maryland
First issued in the Public Health Services publication No. 1000- Series No. 1 l-No. 17
September 1966
NATIONAL CENTER FOR HEALTH STATISTICS
EDWARD B. PERRIN, Ph.D., Director
PHILIP S. LAWRENCE, SC.D., Deputy Director
GAIL F. FISHER, Associate Direct;rfor the Cooperative Health Statistics SystemELIJAH L. WHITE, Amociate Director for Data Systems
IWAO M. MORIYAMA, Ph. D., Associate Director for International Statistics
EDWARD E. MINTY, Associate Directorfor Management
ROBERT A. ISRAEL, Associate Directorfor Operations
QUENTIN R. REMEIN, Associate Directorfor Pro@am Development
ALICE HAYWOOD, Irzj&maticm Officer
DIVISION OF HEALTH EXAMINATION STATISTICS
COOPERATION OF THE EUREAU OF THE CENSUS
In accordance with specifications established by the NationalHealth Survey, the Bureau of the Census, under a contractualagreement, participated in the design and selection of thesample, and carried out the first stage of the field interview-ing and certain parts of the statistical processing
Library o/ Congress Catalog Card Number 66-60079
CONTENTSPage
Examination and Criteria ---------------------------------------------Diagnostic Criteria and Classification of Rheumatoid Arthritis --------- -
The Joint Examination ----------------------------------------------
Distribution of Diagnostic Criteria -------------------------------------Findings by Age and Sex-------------------------------------------Demographic Variables ---------------------------------------------
Discussion -----------------------------------------------------------
Summary -------------------------------------------------------------
References -----------------------------------------------------------
Detailed Tables -------------------------------------------------------
Appendix I. Items on the Medical History Relating to Arthritis -------------
Appendix II. Form Used in Recording Findings on the Physical Examinationof the Joints ---------------------------------------------------------
Appendix HI. Reference X-ray Plates -----------------------------------
Appendix IV. Effectiveness of Dia~ostic Criteria ------------------------
Appendix V. Demographic Terms ---------------------------------------
Appendix VI. Distribution of Dia~ostic Findings -------------------------
Appendix VII. Statistical Notes -----------------------------------------~e Survey Design --------------------------------------------------Reliability --------------------------------------------------------Sampling and Measurement Error ------------------------------------Expected Values ----------------------------------------------------Small Numbers -----------------------------------------------------Tests of Si~ificance ------------------------------------------------
1
1
2
369
14
14
15
16
26
28
29
31
34
38
40
4040
40414141
IN THIS REPOR T are pyesented findings on the prevalence of ?’heuma-toid arthritis (RA) obtainedfrom Cycle I of the Health Examination SuY-vey. Cycle I consisted of examinations of a nationwide p~obability sampleof persons 18-79 yeavs of age selected from the U.S. civilian, noninsti-tutional population.
This report describes the steps taken in diagnosing RA, presents the datacollected, and compares the information obtained in this suvvey withthat obtained in’ other suvveys. The relationship of the prevalence ofRA to the demographic vayiables of age, ?’ace, sex, family income, edu-cation, family size, place description, marital status, usual activity, oc-cupation, and industry aye examined.
Some 3.6 million adults had RA. 8A was more prevalent in women thanin men. The likelihood of having RA was about the same for both whiteand Neg~o adults. The prevalence of RA varied by certain othev demo-graphic factors. Among the differentials noted was a lower than expectedRApvevalence for persons with more education anda lower than expectedprevalence for men in the professions and in technical and managerialfields.
SYMBOLS
Data notavailable ----------------------- ---
Category nonapplicable ------------------ . . .
Quantity zero -------------------------- -
Quantity more than Obut less than 0.05---- 0.0
Figure does not meet standards ofreliability or precision ----------------- *
RHEUMATOID ARTHRITIS IN ADULTS
Arnold Engel, M.D., and Jean Roberts, Division of Health Examination StatisticsThomas A. Burch, M.D., National Institute of Artlwitis and Metabolic Diseases
Rheumatoid arthritis is a chronic inflamma-tory disease of connective tissue manifestedby varying degrees of joint and constitutionalsymptoms. Permanent joint deformities, disa-bility, and chronic invalidism are not infrequentconsequences of this disease. It is hoped that theM.a furnished by the Health Examination Surveynmy prove useful to practitioners and to investi-gators working with this puzzling disease ofunknown etiology.
The National Health Survey uses three meth-ods for obtaining information about the health ofthe U.S. population. The first is a household inter-view in which persons are asked to give informa-tion relating to their health or to the health of otherhousehold members. The second is the collectionof data from available records, such as hospitalforms. The third is direct examination. The HealthExamination Survey (HES) was organized to use thethird procedure, drawing samples of the popu-lation of the United States and, by medical exami-nation and with various tests and measurements,undertaking to characterize the population understudy.
The first goal of the Health ExaminationSurvey was to examine a nationwide probabilitysample of the civilian, noninstitutionalized popu-lation aged 18-79 years to obtain information onthe prevalence of cardiovascular diseases andcertain other chronic diseases, dental health, andthe distribution of a number of anthropometric andsensory characteristics. Altogether 6,672 of asample of 7,710 persons were examined in the firstsurvey, which began in October 1959 and ended in
December 1962. Medical and other HES staffmembers gave these sample persons a standardexamination, lasting about 2 hours, in speciallydesigned mobile clinics.
This report discusses the prevalence ofrheumatoid arthritis by age, race, and sex andaccording to certain other demographic factors.It also compares the findings of this survey withthat of other surveys. The HES findings in regardto osteoarthritis have been published separately. *The rheumatoid arthritis report is one of a seriesdescribing and evaluating the plan, conduct, andfindings of the first cycle of the Health ExaminationSurvey. A description of the general plan and ofthe sample population and response have beenpublished, 293providing a general background forall the reports of findings. In this report the entireexamination of the joints is outlined “and thoseparts of the examination relating to the diagnosisof rheumatoid arthritis are discussed.
EXAMINATION AND CRITERIA
Diagnostic Criteria and Classification of
Rheumatoid Arthritis
The diagnostic criteria used in the surveywere essentially those of the American Rheuma-tism Association (ARA)4 for active rheumatoidarthritis classified according to the number ofcriteria fulfilled. Diagnostic criteria with theweight assigned to criteria according to the
1
present somewhat arbitrary system were asfollows :
Symptoms
1. Morning stiffness— 1 point.2. Pain on motion or tenderness in at least
one joint (observed by physician)— 1point.
Clinical Signs (Observed by Physician)
3. A total allotment of 1-3 points wasallocated to joint swelling according to thefollowing classification:a.
b.
c.
Swelling (soft tissue thickening orfluid, not bony growth ‘alone) in atleast one joint— 1 point.Swelling as above in at least one otherjoint—2 points.Swelling as above with simultaneoussymmetrical involvement on both sidesof the body (terminal phalangeal jointsdo not qualif y)—3 points.
4. Subcutaneous nodules over bony promi-nences on extensor surfaces or in juxta-articular regions—1 point.
Radiological Signs
5. X-ray changes; the minimum requirementin the Survey was an erosion of the bonein the involved joint— 1 point.
Serology
6. The demonstration of rheumatoid factor byany method which in two laboratories hasbeen positive in not over 5 percent ofnormal controls— 1 point.
Cases were classified as follows:
3 or 4 points— Probable case of rheumatoidarthritis.
5 or 6 points— Definite case of rheumatoidarthritis.
7 or 8 points—Classical case of rheumatoidarthritis.
The rates for rheumatoid arthritis as deter-mined by the Health Examination Survey referonly to those cases diagnosed by the single exami-nation administered by the survey. Due to theremittent nature of the disease there maybe casesthat might not be labeled as rheumatoid arthritisin a single examination but might be so classifiedin a subsequent examination. 5
The Joint Examination
The medical history. —After a brief interviewby a receptionist the examinee was given a medicalhistory form to complete. Included among thequestions were some pertaining to rheumatoidsymptoms or diseases. The questions dealt withmorning stiffness, joint pain, swelling and tender-ness, and the presence or absence of a previoushistory of arthritis. Of all these questions, onlythe one on morning stiffness was included in thedetermination of the diagnoses of rheumatoidarthritis.
The physical examination .—This was a
standardized physical examination for which themethod of examination had been specifically out-lined to the examining physician, who was a fellowor resident in internal medicine. Joints includedin this examination were the hip, knee, ankle,feet, cervical and lumbar spine, elbow, wrist,shoulder, metacarpophalangeal, proximal inter-phalangeal, distal interphalangeal, sternoclavicu-lar, and sacroiliac. Among the manifestationsof joint involvement to be looked for were painon motion, limitation of motion, tenderness oncompression of joints, swelling, deformity, subcu-taneous nodules, and atrophy. The form used inthe physical examination of the joint is shown inAppendix II. The diagnostic criteria obtained fromthe physical examination were pain on motion,tenderness, joint swelling, and subcutaneous nod-ules. Since the subjects were only examined once,no information is available concerning the durationof any of the findings. Unfortunately informationobtained from the patient on symptom durationwas too unreliable for use.
Serology. —Serum for the bentonite floccu-lation test was refrigerated and shipped to theNational Institute of Arthritis and MetabolicDiseases (NI,IMD) for analysis for rheumatoid
2
Table A. Percent of men and women with specified findings and relative prevalence bysex: United States, 1960-62
Diagnostic ‘finding
Symmetricaljoint swelling--------------------------
Subcutaneousnodules--------------------------------
Tenderness------------------------------------------
Pain on motion--------------------------------------
Swelling, one joint---------------------------------
Positive bentonite flocculationtest----------------
Swelling, two joints--------------------..------------
PositiveX-ray--------------------------------------
Morning siAffness -----------------------------------
Severe and frequent -------------------------------
Mild and frequent ---------------------------------Severe and infrequent -----------------------------
Mild and infrequent -------------------------------
=
Percent
0.9 3.1* *
9.4 17.5
1.9 3.41.7 . 1.8
3.4 3.5
0.3 0.31.0 0.6
22.1 32.2
4.3 8.4
3.5 4.91.3 2.4
13.1 16.5
Women
Men
Relativeyevalenc~
3.3*
1.9
1.81.1
1.0
0.8
0.7
1.5
2.0
1.41.9
1.3
factor, In the test bentonite, anaturally occurring clinical cases at the NIAMD utilizing standardsclay in powder form was suspended undistilledwater, Particles of optimum size for the testwere obtained by differential centrifugation. Theparticles were coated with gamma globulin anddyed with methylene blue. Rheumatoid factor,if present in an examinee’s serum, flocculatedcoated particles. Findings were considered posi-tive when at least 50 percent of the bentoniteparticles were clumped into compact massesbya serum diluted 1:32 ormore.G
Radiological examination.—X-rays weretaken of the hands and feet of each examineeby the survey technicians. A 10 by 12 inchfilm was takenof both hands with abone standardplaced between them. Similarly a 10 by12 inchfilm was taken of the feet with a bone standardplaced between them. The X-rays were readindependently by three expert specialists inarthritis from the NIAMD. None of the readershad any knowledge of the age, sex, or clinicalstate of any individual. A series of referenceX-ray plates (Appendix III) w?re prepared from
set forth by Kellgren and Lawrence.7 The filmswere read for the presence of osteoporosis,cartilage destruction, bone destruction, andanky-losis. A 5-point score was used with ascoreof2-4 being indicative of rheumatoid arthritis.A high degree of agreement was found among thereaders. Rating disagreements were resolved inconference by the three readers. A descriptionof the process of resolving disagreements inratings is given in the National Center for HealthStatistics report on osteoarthritis.1
DISTRIBUTION OF
DIAGNOSTIC CRITERIA
Table A gives the percent distribution ofeach diagnostic finding inthepopulation accordingto sex. The diagnostic findings tenderness andpain on motion are listed separately, and aseparate tabulation of morning stiffness brokendown into fourdegrees ofseverityisalso included.
3
The data for morning stiffness were obtained fromthe self -administered history (for exact form seeAppendix I). The four possible combinations ofanswers to the question were as follows:
Symptoms occurred and Bothered the examinee
Every few days Quite a bit
Every few days Just a little
Less often Quite a bit
Less often Just a little
(Severe and frequent)
(hiild and frequent)
(Severe and infrequent)
(Mild and infrequent)
Symmetrical swelling, joint tenderness, painon motion, and severe morning stiffness weremuch more common in women than men. Swellingin one joint, positive bentonite flocculation test,swelling in two joints, and a positive X-ray wereabout as common or more common in men thanwomen. It should also be noted that the moremarked the morning stiffness the greater the ex-
isoo
1,200 —MEN X-ray
L
‘i
Ipoo –
u
kE 800 —J~p
i.D 600 —.zuoKu
L 400 —
200 —
Symmetrical
o20 30 40 50 60 70 so
AGE IN YEARS
Figure 1. Prevalence of positive X-ray,symmetrical swell ing, and rheumatoid ar-
thritis in men, divided by the total
prevalence for all ages multipl ied by
100, by age.
(100 percent equals average prevalence for
all ages)
4
WOMEN
X-r:y
01 I20 30 40 50 60 70
1
AGE IN YEARS
Figure 2. Prevalence of positive X-ray,symmetr ical swell inq, and rheumatoid ar -thritis in women, divided by the total
prevalence for all ages multiplied by
100, by age.
(IOC percent equals average prevalence forall ages)
cess in women. The general pattern ofthedis-tribution of diagnostic findings by sex appears toshow a greater relative prevalence in women forthose findings accompanying an acute episodeofjoint disease.
The distribution by age of the diagnosticfindings in the male and female populations areoutlined in figures 1-8. Each finding is plottedusing its own mean rate for all age groupsas a standard. Data are presented as the per-centage that each age group rate forms of thetotal rate for all age groups. For example, ifthe rate at age 60 for a particular finding wastwice the total mean rate the graph would bemarked at the 200 percent level. The actual ratesfor the diagnostic findings by age are given intable 1. On each graph a similar curve forrheumatoid arthritis is included for reference.
In general all the diagnostic findings show agradient with increasing age. The curves followmore or less closely the curve for rheumatoidarthritis, depending to a large extent on the
./ 1MEN
800 iII
700I
I
IJm I
i Ic
JI.
: 500>
;
; 400
i>J
“ 300‘l-
1-ruun on moll
100 &
III
A 1
20 30 40 50 60 70 80
AGE IN YEARS
Figure 3, Prevalence of tenderness, painon motion, and rheumatoid arthritis’ inmen, divided by the total prevalence forall ages multiplied by 100, by age.
(100 percent equals average prevalence forall ages)
relative specificity of the diagnostic finding tothe diagnosis of rheumatoid arthritis. It isinteresting to note that mild, infrequent morningstiffness shows no age gradient and so lacks anycorrelation with the rheumatoid arthritis curve.This lack of correspondence reflects the verylow true positive rateformild, infrequent morningstiffness. It is also interesting to note that allthe diagnostic findings either decrease or leveloff in males for the age group 65-74. Thisparallels the small decrease in the rheumatoidarthritis rates for males in this age group.
Table B. Percent with positiveflocculation tests accordingExamination Survey, 1960-62
bentoniteto Health
Rheumatoid arthritisstatus
No rheumatoid arthritis-
Probable rheumatoidarthritis --------------
Definite rheumatoidarthritis --------------
Classical rheumatoidarthritis-------------”-
—
Pointvalue
Percentpositive
BFT
2.6
23.03.5
12.260.0
100.0100.0
Of considerable interest is the distributionof positive bentonite flocculation tests (13FT)in the Health Examination Survey. Table B givesthe percentof positive bentonite flocculationtestspresent in the nonrheumatoid examinees and in
100
0
. I
wOMEN
,.
IAGE IN YEARS
I
Figure k. Prevalence of tenderness, painon motion, and rheumatoid arthritis in
women, divided by the total prevalence
for all ages multiplied by 100, by age.
(100 percent equals average prevalence forall ages)
5
800
700
600
,1
400
1
I
300
200
100
0
MEN
i
IIIIIIIi
R #IIIIII
Morning stiffness (severe, frequentl
,,,”.-,,,,,,,,wt~’’’””
I *-L
20 30 40 50 60 70
AGE IN YEARS
Figure 5. Prevalence of morning stiffness
I
severe, frequent), morning stiffnessmild, infrequent), and rheumatoid ar-thritis in men, divided by the totalprevalence for all ages multiplied by100, by age.
([00 percent equals average prevalence forall ages)
the rheumatoid examinees according tothe ARApoint count. It can be seen that apositive BFTispresent in low frequencies inprobablerheumatoidarthritis and in definite rheumatoid arthritiswhich has a point value of 5. Thus, only iftheARA point count is60rmore isthe percent withpositive BFT very high. Some of the cases witha positive BFT found in the probable groupwith apoint valueof3 areprobablythe resultof achancecombination of individuals with other positivediagnostic findings of point value 2 coupled with a
6
I WOMEN
500
t
Ill1- Ia 400aJ I~o I1-~300 )o RA//1-2w /
~ 200n Morning stiffness
100 —
,,,,s,mmw(mild, infrequent)
o . l--
20 30 40 50 60 70 80
AGE IN YEARS
Fitiure 6. Prevalence of mornin9 stiffness
[severe,
1
frequent , morning- stiffnessmild, infrequent , and rheumatoid ar-
thritis in women, divided by the totalprevalence for all ages multiplied by100, by age.
(100 percent equals average prevalence forall ages)
positive BFTdue to factorsother thanrheumatoidarthritis. Theoverallprevalenceofapositive 13FTcan therefore be expected to be more in aclinical group than in a population survey sincethe clinical group contains a much higher pro-portion of definite and classical cases ofrheuma-toid arthritis.
Findings by Age and Sex
On the basis of the survey findings it isestimated that of the 111.1 million adults inthe United States aged 18-79 in 1960-62, 3.6million had rheumatoid arthritis.This represents3.2 percent of the adults in this age range.As classifiedby ARA criteria, 30percentofthesecases were definite or classical arthritis and70 percent were probable cases of arthritis(table 2).
800
700
200
100
0
AGE IN YEARS
Figure 7. Prevalence of positive benton-ite flocculation test, swelling in onejoint, and rheumatoid arthritis in men,divided by the total prevalence for allages multiplied by 100; by age.
(100 percent equals average prevalence forall ages)
Rates for rheumatoid arthritis that arereferred to in the remainder of this report arethe sum of the rates for all cases designatedas classical, definite, orprobablebyARA classifi-cation.
In the Health Examination Survey theprev-alence rate for females was 4.6 percent and thatfor males was 1.7 percent, giving asex ratio of2.7/1.
In females the prevalence of rheumatoidarthritis rose sharply with age, with the gradientbecoming steeper in each older age group. Intheage group 18-24 nearly 0.3 percenthadrheuma-
600
500
IIzK 400
J<5300
L>
i
5 200KIIL
100
0
i-,.”.,oin,)#/Swelling
/ ~
\...-
●*...’/
/.,,,.
— ‘(To+*,.,.**”.>..,s,’’” ,
,.*’
20 30 40 50 60 70
;,,,,,s. I I I I
AGE IN YEARS.
Figure 8. Prevalence of positive benton-ite flocculation test, swelling in onejoint, and rheumatoid arthritis in women,divided by the total prevalence forallages multiplied by 100, by age.
(100 percent equals average prevalence forall ages)
toid arthritis. By age75-79 years 23.5 percentofthe women had rheumatoid arthritis” (tables2 and 3).
In males the prevalence of rheumatoid ar-thritis also rose sharply with age. In the agegroup 18-24 only 0.2 percent had rheumatoidarthritis, By age 75-79 years 14.1 percent ofthe men had rheumatoid arthritis. In contrasttofemales, however, there wasoneagegroup(65-74 years) among males in which there was anactual decrease inprevalence whencompared withthe preceding age group. Although this decreaseis not statistically significant, data from otherstudies (see page 8) and individual analysis ofthe diagnostic criteria (see page 6) support thelikelihood ofat least aleveling offoftherheuma-toid arthritis rates for this age group.
Comparison of the HES findings in regardto distribution of rheumatoid arthritis by age andsex with those found in other surveys is of someinterest. One of these was asurveymade of the
7
population of the town of Tecumseh, Mich., inwhich 90 percent of all inhabitants over the ageof 6 were examined.g From this source, onlydata on persons aged 18-79 years—a total of4,796 examinees—will be used for comparativepurposes.9 The second was a survey of a sampleof the population over the age of 15 in the townsof Leigh and Wensleydale, England, in which atotal of 2,234 people were examined.lo The thirdwas a large-scale survey of a sample of thepopulation of 68 municipalities in the Netherlands.A total of 141,845 persons were first screened bya questionnaire, and the ones with rheumaticsymptoms were examined.ll The age groupsurveyed was from 15-64 years old. In all threesurveys ARA criteria were employed in makingthe diagnoses of rheumatoid arthritis, However,the survey in the Netherlands counted onlyprobable cases “with four criteria at hand”instead of the three criteria used in the othertwo surveys and in the Health ExaminationSurvey. Even with the use of ARA criteria, ob-server differences may account for a major partof the differences in prevalence rates betweendifferent surveys. Sex and age differentials,however, are much less likely to be affected byobserver differences or by variation in methodsof examination.
The female/male sex ratio in the othersurveys was as follows; Tecumseh, 3.8/1;Netherlands, 2.5/1; Leigh and Wensleydale, 2.4/1.The corresponding HES ratio was 2.7/1.
The trends by age reported by the threesurveys were in general similar to those reportedfor the United States. In figures 9 and 10, ratesfrom all four surveys are plotted as percentageof the total rate for male and female, respectively,at different age levels. For example, if the rateat ages 65-74 years was twice the total rate forall age groups combined, the rate for 65-74would be plotted at 200 percent on the graph.Since the data for the Netherlands only went upto age 64, the total rate was less than it wouldhave been had ages over 64 been included. Thisresulted in slightly increased percentages in theearly age groups in the Netherlands as comparedwith the other groups.
For the males, three out of four surveysshowed a decrease in prevalence starting justbefore the age of 60. This decrease took placeabout a decade earlier in the Tecumseh study
1,200
1,100
900
8001
700II600
)
j 500II400
300
MEN
Tacums(USA)
200 —
100 —
,u .1 I
10 20 30 40 50 60 70 8(
AGE IN YEARS
Figure 9. Prevalence of rheumatoid ar -~hr itis in men as reported in four sur-veys, divided by the total prevalencefor all ages multiplied by 100, by age,
( 100 percent equals average prevalence forall ages)
than in the others. As noted previously, theprevalence of all diagnostic findings eitherdecreased or leveled off for men in the age group65-74. Therefore in males it is probable thatthere was no increase in prevalence of rheuma-toid arthritis for the 65-74 age group. All threesurveys with relevant data showed an increasein prevalence in the oldest age group. It mustbe noted, however, that the numbers of peopleexamined in the oldest age group of each of these
8
three surveys (HES, Tecumseh, Leigh andWensleydale) were not large.
In females the trend by age was quitesimilar in all study groups until about the ageof 60. Then the prevalence either leveled off(Leigh and Wensleydale), decreased slightly andthen increased slightly (Tecumseh), or increasedsteadily (HES). In the HES data it can be calcu-lated that the rate for females aged 75-79 wassignificantly higher than that for ages 55-64. Sinceonly a small number of people were examined inall three surveys in these older age groups, it
is quite possible that differences in prevalencefound in these groups among the three surveyswere due to chance variation,
Demographic Variables
In the discussion that follows, the populationis classified in a variety of ways—for instance,by race, by family income, by education—and the
.
600
WOMEN
500 - :—:
;-
4z
? HES #~ 400 —
\:: -Ja
$~-
; :
- 300 —1.3
(USA).Ku$ 200 —J1a.
100 —
10 20 30 40 50 60 70 80
AGE IN YEARS
Figure 10. Prevalence of rheumatoid ar-thritis in women as reported in four
surveys, divided by the total prevalence
for all ages multiplied by 100, by age.
prevalence of rheumatoid arthritis in different
groups is compared. If the population is classifiedby family income, for example, the prevalenceof rheumatoid arthritis in different income groupsis examined to determine whether prevalence
rates vary from one income group to another.In making these comparisons, allowances must bemade for the differences from one group to another
in the distribution of people by age and sex.
Because the sampling variability of age-sex
specific values for any group is usually large,a summary comparison by sex was thoughtpreferable to the presentation of prevalencerates specific by age and sex. For this reasonthe actual prevalence rate for each group iscompared with an expected rate.
The expected value of a particular groupis obtained by weighting age- and sex-specificrates for the total United States by the age-sexdistribution for that group. The obvious meaning
can be attached to differences between actualand expected rates with the understanding thatdifferences may arise by chance. A positivedifference, for example, indicates that the preva-
lence rate for the group is higher than expected.Alternatively the data can be presented as aratio of actual to expected rates. If the ratiois greater than 1.0 the actual rate is higherthan expected. If the ratio is less than 1.0the actual rate is less than expected. In general,where there is no statistically significant dif-
ference between the actual and expected valuesfor a group, differences for individual age-specific groups exhibit only random fluctuations.
Race. — So far as can be judged from the data,
the prevalence of rheumatoid arthritis was thesame in white and Negro adults (table 4). Therates for nonwhite males other than Negroes weresuggestively higher. In our sample the bulk of thisgroup was composed of American Indians inArizona. No reliable estimates can be made from
this small sample as to the prevalence of rheuma-toid arthritis in American Indians as a whole orin the other nonwhite non-Negro group. Of someinterest are prevalence data for rheumatoid
arthritis from a survey by Dr. Burch of 2,005subjects in two Indian tribes, the Blackfoot andthe Pima. In this survey the combined female/male
( 100 percent equals average prevalence for
all ages)
15
inz
2= 10IIIn
0g
aw0.
5Id5c
o
:?..,~a
Indian women~;~”””””””
&20 30 40 50 60 70 80
AGE IN YEARS
Figure Il. Prevalence of rheumatoid ar-thritis in Pima and Blackfoot Indians ascompared with U.S. men and women, by age.
sex ratio was only l.3/l(fig.ll).Atpresentthereis no explanation available to account for therelatively high rheumatoid arthritis prevalencein males of this group. 12
Residence. —There appears to be some-what less rheumatoid arthritis in “other verylarge metropolitan areas’’ (population 500,000-3,000,000) and somewhat more in “other urbanareas.” These differences are statistically sig-nificant in females and are in the same directionbut not significant in males (table 6). Therewere no other significant differences in rateswhether classified by geographic region (table5)or by place description (table 7). These various,somewhat confusing groupings are discussed inAppendix V, where all the demographic terms usedin this report are defined.
The only other data available for comparisonare provided by a survey of 68 municipalities inthe Netherlands.ll In it the lowest rates werefound for both sexes in the largest towns(Amsterdam, Rotterdam, Hague, and Utrech).
One must of necessity take into account theproblem of observer differences when oneevaluates place differences in rates. This is ofespecial importance in such diseases as rheuma-
toid arthritis, where the diagnosis is basedlargely on criteria directly observed by thephysician. Significant differences in prevalencecan conceivably be produced in different areasdue to some consistent pattern of observers’differences. This possibility has been greatlyminimized in the Health Examination Survey bythe use of a large number of physicians (62)for conducting the examinations.
Another complicating factor in the surveywas that the itinerary for HEX was designedto avoid the South in summer and the North inwinter. Thus if there are significant seasonalfluctuations in the prevalence of rheumatoidarthritis. the pattern of scheduling used in thesurvey fluctuations may have acted to cancelout regional differences.
Education and family income. —In men therewas a distinct trend toward decreasing prevalenceof rheumatoid arthritis with a higher level ofeducation (table 8 and fig. 12). Men who hadless than a fifth grade education had the highestrates, and men at the highest educational levelhad the lowest rates. For females no such trendexisted, and the only significant finding was anelevation of the rheumatoid arthritis rate in thosewomen with less than a fifth grade education.
2.0
1-
MEN WOMEN
n 1
Undtr 5 5-8 9-12 13 years Under 5 5-8 9-12 13 y4Cir9years years years :::r yaars years Y@ar~ :Jfr
EOUCATION
Figure 12. Ratio of actual to expectedrate of rheumatoid arthritis in adults,by education.
10
For all the diagnostic findings in men, withthe exception of the bentonite flocculation tests,an overall trend existed toward decreasing prev-alence with increasing education (tables 9 and 10).A similar but less marked trend was also presentfor women. A positive BFT at dilutions of 1:512or more (which are felt to be somewhat morehighly diagnostic of rheumatoid arthritis thanlower dilutions)ls had a low prevalence for thehigher educational groups, It should also be notedthat severe frequent morning stiffness showed adefinite trend toward decreased prevalence withhigher educational levels but mild, infrequentmorning stiffness showed none. ln general whenboth sexes are combined, diagnostic findings ofgreater diagnostic effectiveness for rheumatoidarthritis showed a more’ pronounced trend witheducation.
RATlO OF ACTUAL TO EXPECTE O
“~
amlly. IncomeMEN
$*%O. .
$y);-,“
$?000.9,999
$10,000and over
$.. .
I. .
WOMEN
$!5:0
$:;;;
$4Qoo-6,999
$7,000-9,s99
$Io,opoandOVW
Figure 13. Ratio of actual to expectedrate of rheumatoid arthritis in adults,by family income,
There appeared to be somewhat higher ratesthan expected for males with less than $2,000family income and somewhat lower rates thanexpected for those with $4,000-$6,999 familyincome. The distribution of rates in “femaleswas close to that expected, with exactly the ex.~ected rate for the income group under $2,000 andeven a slight increase (insignificant) in ratefor the $4,000-$6,999 income group, which wasthe lowest for males (table 11 and fig. 13).
Findings for HES by education and incomediffered somewhat from the findings of King andCobb in their survey of the Arsenal Health Districtin Pittsburgh. 14 Although they reported anincreased rate for females with a fifth gradeeducation or lower, they found no differencein rates by education in males. As in the HESstudy the distribution of rates by income in fe-males was close to that expected. For males therewas a trend for lower rates with higher income.However, the data were not strictly comparablewith those of HES, since only three income groupswere reported in King and Cobb’s study.
Of some interest is Cobb’s finding ofrelatively higher rates when the educationallevel was low and the income high or whenthe educational level was high and the incomewas low. The HES data, when similarly cross-classified for education and income, show nosuch elevation in rates for either high educationand low income or low education and high income.Using the same sort of indirect age adjustment asCobb, the rates for the HES study were as follows:
Education
Under 5 years ----------
5-8 years --------------
9 yea~s ‘and over -------
Income
e
Rate per100 adults
J__
5.8 4.9
3.8 3.2
2,1 2.6
Some of the differences in findings in the twosurveys may have resulted from the use of an
11
MEN WOMEN
EXCESS RATES PER 100 ADULTS EXCESS RATES PER 100 ADULTS-2.0 -1,0 0.0 1.0 2.0 -2.0 -1,0 0.0 1.0 2.0
Occupofion
~~
Professional, technical,other kindred workers, andmonagar ial work ors .................................................
%
Farm ●rs and form managers f...............................................
Clerical and sole!workers ...................................................0.,..,.,,.,.,,!.............,,.,.,.,.,
b aCroftsmcn,foromcn, andKindrad workers 1 ......................................................................
1
Optfntlves and kindred
a
workors ....................................................................................
I
privota housahold andservice w orkors .......................................................................
Form and othor laborers(*.ctpt mine) I ............................................................
A IIDafa for ~ome” dO “Ot ~eat standards of reliability Or pfecisiOn.
Figure 14. Excess of actual over expected prevalence of rheumatoid arthritis, ty occupationand sex.
MEN WOMEN
EXCESS RATES PER 100 AOULTSEXCESS RATES PER 100 AOULTS
-20 -1.0 ao MY w -2.0 -1.0 Cto Ill 20
Industry II I
I tI I
I
Agriculture, foro$try, ondfishwias ..,,.,,., ..........................!...................................................
Mimng and construction ' ........................................................
Monufocturinq .............................................................................
Transportation com municotion,and othtr public utilities 1.............................
Whois$alc and retail
Flnanca, insurance, anreal estats ..................
Service and miscellon
Government 1 ........... ....4
lDotQ for women do not meet standards of reliability Or precisiOn,
Figure 15. Excess of actual over expected prevalence of rheumatoid arthritis, by industry andsex.
12
“index of rheumatoid arthritis” by Cobb forthe study of associations of the disease and en-vironmental factors of interest. Cobb reportedthat his index had a sensitivity of 0.65 and aspecificity of 0.95 for the diagnosis of rheuma-toid arthritis. His Rheumatoid Arthritis Index isdefined as a “yes” answer to all three of thefollowing questions:
1.
2.3.
Have you ever had arthritis or rheuma-tism?Have you ever had swelling in any joints?CO you wake up with stiffness or achingin your joints and muscles?
As noted previously the HES technique wasbased on the use of the ARA diagnostic criteria,which may have produced somewhat differentresults from Cobb’s diagnostic method.
Occupation, industry, and usual activitystatus.—Men engaged in managerial and technicaloccupations and in the professions had a lowerthan expected prevalence of rheumatoid arthritis(table 12 and fig. 14). Although there were nosignificant differences in particular industriestaken individually, there was a strong suggestionof lower prevalence. for men in industries with,in general, a lower proportion of persons workingat jobs involving more than minimal physicalactivity. Thus there were no cases of rheumatoidarthritis found in men employed in government,finance, insurance, real estate, transportation,communication, and other public utilities (table 13and fig. 15).
In regard to occupational data it is necessaryto consider the importance of health status andin particular arthritis as a factor involved inchange of occupational or retirement. Thus lowerrates in working people may result from theremoval of people with a disability from thelabor force. On the other hand an individualwith some degree of disability may be shiftedto a “light job” and remain employed in the sameindustry. The rates for both men and womenemployees were suggestive of a low prevalence;however, these rates had too large a samplingvariability for the differences among them to bedeemed significant (table 14).
RATIO OF ACTUAL TO EXPECTEO
0.0 0.5 1.0 1.5 2.0 2.5
%&l&
Married
Widowed
Oivorced
Separated
Nev.rmarried
Married
Widowed
Oivorced
Sepdro$ed
N.v@rmarried
I WOMEN
Figure 16. Ratio of actual to expectedrate of rheumatoid arthritis. kY marital. .status.
The only other data available for comparisonon occupation and industry are provided byde Graaff’s study in the Netherlands. In his study,male workers in agriculture had the highest rates,while professionals had the lowest rates. Therewere no significant differences in rates amongemployed females. 1*
Marital status and numbey of childven. —Widowed men and unmarried females had signifi-cantly less rheumatoid arthritis than expected(table 15 and fig. 16). After the age of 45 thereappeared to be a significantly lower rate for womenwithout children than for those with children(table 16). Contrary to Cobb’s findings 14 therate for women with four or more children wasgenerally not higher and was never significantlyhigher than for those with one to three children.
13
DISCUSSION
In the report on HES rheumatoid arthritisfindings, data were presented indicating thatrheumatoid arthritis prevalence rates by ageand sex differentials are comparable to thosereported for several other populations. TheHES findings also indicate no significant differ-ences in rates according to race or region of theUnited States.
In males there appeared to be significantvariations in rates with regard to the demographicvariables of education and, more equivocally,income, occupation, and industry. One possiblefactor acting to produce these variations may havebeen the relative frequency of minor trauma tojoints associated with occupation and socioeco-nomic status. Although not definitely established,it is the belief of some observers that injury toa joint not infrequently appears to precipitatethe onset of an inflammatory arthritis whichlater spreads to involve other joints and finallypresents the features of rheumatoid arthritis. 15
Conclusions drawn from this survey or anypopulation survey using ARA criteria for diag-noses must be qualified by the extent that thesecriteria do in fact provide a valid diagnosis ofrheumatoid arthritis. This is especially importantfor the category “probable rheumatoid arthritis; ’16in which fall 70 percent of all cases in the HESstudy. Moreover, it is probable that further studyof rheumatoid arthritis will show that this diag-nosis includes a variety of diseases which are ofdifferent etiology but which cannot be distin-guished in the light of our present knowledge.
The HES findings in regard to the specificARA criteria tend to show that they are not infact of equal value in making a diagnosis ofrheumatoid arthritis in a general population.Hence there would appear to be merit to the sug-
gestions of othermade to establish
investigators that efforts beappropriate weights for each
one of the diagnostic criteria. 17Some final caveats are in order. 1he sample
size used for the Health Examination Survey leadsto numerous statistics with high samplingvariability. Thus, many of the demographic dif-ferentials indicated in this report should beregarded as suggestive rather than proved. Mostdemographic labels are crude indexes, only thefirst steps to an investigation. Furthermore, thevariables on which the data have been classifiedare more or less correlated (education, forinstance, is highly correlated with occupation).If the sample had been larger, more detailedcross-classification would have been appropriate.
SUMMARY
Some 3.6 million adults aged 18-79 yearshad rheumatoid arthritis. Rheumatoid arthritisis rare among young adults in the United Statesbut becomes increasingly common at older ages.It is more prevalent in women than in men.Negro and white persons are about equally likelyto have the disease.
Men with more education have lower ratesthan men with less education. In occupationalterms, men in the professions and in technicaland managerial fields have relatively low rheuma-toid arthritis rates.
There are no well-defined differences inprevalence evident by place of residence except fora somewhat lower rate for metropolitan areas withpopulations of 500,000 to 3,000,000 and a somewhatincreased rate for urban areas not associated withstandard metropolitan statistical areas.
Rheumatoid arthritis is less prevalent inwidowed men and unmarried females than inpersons of other marital status.
14
lNat,ional Center for Health Statistics: Prevalence of
osteoarthritis in adults, by age, sex, race, and geographic
area, United States, 1960-62. V’ital and Health ,Watistics.
PHS Pub, No. 1000-Series 11-No. 15. Public Health Service.
Washington. U.S. Government Printing Office, June 1966.
2National Center for Health Statistics: Plan and initial
program of tbe Health Examination Survey. Vital and Health
Statistics. PHS Pub. No, 1000-Series l-No. 4. Public Health
Service. Washington. U.S. Government Printing Office, July1965.
‘National Center for Health Statistics: Cycle I of the
Health Examination Survey, sample and response, UnitedStates, 1960.62. vital and Health Statistics. PHS Pub. No.
1000-Series 11-No, 1. Public Health Service. Washington.U.S. Government Printing Office, Apr. 1964.
4Ropes, M., and others: Revision of diagnostic criteria
for rheumatoid arthritis. Buil.Rheumat.Dis. 9:175-176, Dec.1958.
‘Cobb, S,: A method for the epidemiological study of re-mittent disease. Am, J,.Pub.Health 52:1119-1125, July 1962.
6Bozicevich, J., and others: Bentonite flocculation testfor rheumatoid arthritis. Prac.Soc.Ezper. Bio2.&Med. 97:180-
1!3:3,Jan. 1958.
7Kellgren, J. H., and Lawrence, J. S.: Radiological as-
sessment of rheumatoid arthritis. Ann. Rheumat.Dis. 16:485-
493, Dec. 1957.8Mikkelsen, W. hf., and others: Clinical and serological
estimates of the prevalence of rheumatoid arthritis in the pop-
ulation of Tecumseh, Michigan, 1959-60, in J. H. Kellgren,
cd., The Epidemiology of Chronic Rheumatism, Vol. I. Oxford,
England. Blackwell Scientific Publications, 1963. p. 239.
‘Dodge, H, J.: Personal communication, Oct. 1965.
10Lawrence, J. S.: Prevalence of rheumatoid arthritis.
.4nn.Rheumat.Dis. 20:11, Mar. 1961.
llde Graaff, R.: Proceedings of the [sra Symposium on the
Social Aspects of Chronic Rheumatic Joint Affections” Es-
pecially Rheumatoid Arthritis. International Congress Series
No. 23. New York. 3?xcerpta Medics Foundation, 1959. pp.
7-14.
12 Burch, T. .4.: Unpublished data, 1965.
13 Burch, T. A., and O’Brien, W. M.: Evaluating diagnostic
criteria for rheumatoid arthritis. h!ilbank Mem .Fund Quart.
XLIII (2):161-168, Apr. 1965.14King, S. H., and Cobb> S.: Psychological factors in tbe
epidemiology of rheumatoid arthritis. J. Chronic Dis. 7(6):466,Jhne 1958.
15Kelley, M.: hlonoarticular trauma and rheumatoid arthri-
tis. Ann. Rheumat.Dis. 10:307, Sept. 1951.
16Solings, A. J., and Cats, A.: Assessment of the ARA
criteria for rheumatoid arthritis in a university hospital, inJ. H. Kellgren, ed., The Epidemiology of Chronic Rheumatism,Vol. L oxford, England. Blackwell Scientific Publications,
1963. p. 217.
17Acheson, R.kf.: The.American Rheumatism Association
criteria. Milbanki fem. Fund Quart. XLIII (2):127, Apr. 1965.
18 Ropes, hl. A., andothers: Proposed diagnostic criteria for
rheumatoid arthritis, report of a study conducted by a commit-
tee of the American Rheumatism Association. Ann. Rheumat.
Dis. 16:118, Mar. 1957.
19Abramson, J. H., and others: Study ing the epidemiology
of rheumatoid arthritis in Israel, methodological considera-tions. Arthritis Rheum. 7(2):153, Apr. 1964.
000
15
DETAILED TABLES
Page
Table 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15,
16.
Prevalenceof diagnostic findingsfor rheumatoidarthritis in adults, by sex andage: United Statea, l96O-62-------------------------------------------------------17
Prevalenceof classical,definite,and probablerheumatoidarthritisin adults, bysex and age: United States, l96O-62-----------------------------------------------18
Prevalenceof rheumatoidarthritis in adults, by sex and age: United States, 1960-62--------------------------------------------------------------------------------19
Actual and expectedprevalencerates of rheumatoidarthritisin adults, by sex andrace: United States, l96O-62----------------------------'--------------------------19
Actual and expectedprevalencerates of rheumatoidarthritisin adulta, by sex andgeographicregion: United States, 1960-62-----------------------------------------20
Actual and expectedprevalencerates of rheumatoidarthritisin adults, by sex andpopulation-sizegroups:United States, 1960-62----------------.-------------------- 20
Actual and expectedprevalence rates of rheumatoid arthritis,by sex and placedescription:United States, l96O-62-----------------------------------------------20
Actual and expectedprevalencerates of rheumatoidarthritis,by sex and education:United States, l96O-62------------------------------------------------------------21
Ratio of actual to expectedrates of positivediagnosticfindingsin adults,by sexand education:United States, l96O-62---------------------------------------------21
Percent of adults having positivediagnosticfindings,bysex and education:UnitedStates, l96O-62-------------------------------------------------------------------22
Actual and expectedprevalencerates of rheumatoidarthritisin adults, by sex andfamily income:United States, l96O-62---------------------------------------------22
Actual and expectedprevalencerates of rheumatoidarthritisin adults, by sex andoccupation:United States, l96O-62------------------------------------------------23
Actual and expectedprevalencerates of rheumatoidarthritisin adults, by sex andindustry:United States, l96O-62--------------------------------------------------24
Actual and expectedprevalencerates of rheumatoidarthritisin adults, by sex andusual activity status:United Statea, 1960-62-------------------------------------24
Actual and expectedprevalencerates of rheumatoidarthritisin adults, by sex andmarital status:United States, l96O-62--------------------------------------------25
Prevalencerates of rheumatoid arthritisin women, by number of childrenand age:United Statea, l96O-62------------------------------------------------------------25
16
Table 1. Prevalence of diagnostic findings for rheumatoid arthritis in adults, by sex and age:
United States, 1960-62
Morningstiff-ness
(mild, in-frequent)
Painon
motion 1Posi- Posi-tive tiveX-ray BFT 1SweU- sym-
ing,one metrical
joimt swelling
— —
Morningstiff-ness
(severe,frequent)
Tender-ness
Sex and age
Men
.
Total, 18-79 years-
years--------------
years--------------
years--------------
years--------------
years--------------
years--------------
years--------------
Women
Total, 18-79 years-
years--------------
years--------------
years--------------
years--------------
years--------------
years--------------
years--------------
Percent of specified population group
4:25
1.46
1.95
2.78
4.17
8.78
7.68
11.14
8.43
13.12
15.24
10.89
13.78
13.14
12.71
13.36
14.64
16.46
9.36 1.92 0.97 3.41 1.70 0.93
0.47
0.38
2.27
1.52
10.58
3,06
18-24
25-34
35-44
45-54
55-64
65-74
75-79
3.07
5.95
8.30
8.64
15.85
15.66
22.81
17.50
0.15
0.16
1.90
2.15
4.22
4.38
1.12
3.38
0.52
1.01
2.23
5.41
5.54
4.89
13.94
3.48
0.88
0.74
0.91
1.53
3.79
3.01
4.68
1.79
0.17
0.54
2.36
1.61
12.39
0.64
18-24
25-34
35-44
45-54
55-64
65-74
75-79
3.98
4.31
6.98
10.58
12.40
16.29
7.19
12.10
16.25
19.81
17.35
16.06
15.14
16.51
3.02
6.43
14.29
23.58
32.45
29.85
34.43
0.55
0.77
2.52
4.66
5.74
5.89
14.36
0.89
0.94
3.56
3.75
3.72
8.67
12.33
0.14
0.69
0.78
2.48
3.76
3.58
4.93
0.27
0.33
1.43
2.05
5.29
10.59
17.13
0.46
0.94
0.43
2.07
2.34
17
Table 2. Prevalenceof classical,definite,and probablerheumatoidarthritisin adults, by sexand age: United States, 1960-62
Sex and age
Both sexes
Total, 18-79 years------------
Men
Total, 18-79 years------------
18-24
25-34
35-44
45-54
55-64
65-74
75-79
18-24
25-34
35-44
45-54
55-64
65-74
75-79
years-------------------------
years-------------------------
years-------------------------
years-------------------------
years-------------------------
years-------------------------
years-------------------------
Women
Total, 18-79 years------------
yeara-------------------------
years-------------------------
years-------------------------
years-------------------------
years-------------------------
years-------------------------
years-------------------------
Total II Classical I Definite I Probable
3.2
1.7
0.2
0.5
1.5
4.2
3.1
14.1
4.6
0.3
0.6
2.1
4.4
8.3
14.1
23.5
Rate per 100 adults
0.2
0.2
0.4
6.7
0.2
0.2
0.2
0.3
0.5
0.8
0.3
.
.
0.2
1.6
0.2
1.5
1.2
0.1
0.7
0.8
2.1
4.1
6.2
2.3
1.1
0.2
0.5
1.3
2.2
2.9
5.9
3.3
0.3
0.6
1.1
3.4
5.9
9.5
17.2
NOTE: Age-specificrates are subject to high samplingvariabilitybut are included to suggesttrends and differentials.
18
Table 3. Prevalence of rheumatoid arthritis in adults, by sex and age: United States, 1960-62
Age
18-24
25-34
35-44
45-54
55-64
65-74
75-79
18-24
25-34
35-44
45-54
55-64
65-74
75-79
Total, 18-79 years------------------------------------------------
years-------------------------------------------------------------
years-------------------------------------------------------------
years-------------------------------------------------------------
years-------------------------------------------------------------
years-------------------------------------------------------------
years-------------------------------------------------------------
years-------------------------------------------------------------
Total, 18-79 years------------------------------------------------
years.........------------------.-----—. -------------------------
years-------------------------------------------------------------
years-------------------------------------------------------------
years---------......-----------------------.----------------------
years---------------------------------------------------------------
years----------------------------------------------------------------
years........-----------------------------------------------------
Bothsexes Men Women
Number of adultsin thousands
3,591
39
71
314
613
988
1,026
540
895
16
61
149
315
152
202
Rate per 100
3.2
0.3
0.3
1.3
3.0
6.3
9.2~
18.8
1.7
0.2
0.5
1.5
4.2
3.1
14.1
2,696
23
71
253
464
673
874
338
Ults
4.6
0.3
0.6
2.1
4.4
8.3
14.1
23.5
Table 4. Actual and expected prevalence rates of rheumatoid arthritis in adults, by sex and race:United States,,1960-62
* 1 1
Race
t--Actual
MenI
Women
Expected DifferenceActua 1
IActual IExpected Difference
Expected
Rate per 100 adults
Actual
Expected
White------- 1.7 1.7 0.0 0.97 4.6 4.7 -0.1 0.98
Negro ------- 1.5 1.6 -0.1 0.94 4.7 4.0 0.7
Other-------
1.17
4.2 0.8 3.4 5.00 5.3 3.2 2.1 1.65
19
Table 5. Actual and expected prevalence rates of rheumatoid arthritis in adults, by sex andgeographic region: United States, 1960-62
Men Women
Region
Actual Expected DifferenceActua 1
Actual Expected DifferenceActual
Expected Expected
I Rate per 100 adults
Northeast--- 1.4 1.6 -0.2 0.87 4.7 4.7 0.0 0.99
South------- 1.6 1.6 - 1.00 4.8 4.3 0.5 1.11
West-------- 2.1 1.9 0.2 1.11 4.4 4.8 -0.4 0.92
Table 6. Actual and expected prevalence rates of rheumatoid arthritis in adults, by sex and p~pu-lation-size groups: United States, 1960-62
Population-size group
Giant metropolitan areas------
Other very large metropolitanareas------------------------
Other standardmetropolitanstatisticalareas------------
Other urban areas-------------
Rural areas-------------------
Actua1
1.7
1.2
1.3
2.5
1.7
Ex-pected
1.7
1.7
1.6
1.6
1.8
Men I Women
r ActualDiffer- ~xence petted
ActualEx- Differ- ~x-
‘ctual petted ence pecLed
Rate per 100 adults
-0.5
-0.3
0.9
-0.1
1.00
0.70
0.80
1.53
0.97
4.4
2.6
4.6
8.2
3.4
4.7
4.5
4.5
4.2
5.1
-0.3
-1.9
0.1
4.0
-1.7
-0.92
0.58
1.01
1.96
0.67
Table 7. Actual and expectedprevalencerates of rheumatoidarthritis,by sex and place descrip-tion: United States, 1960-62
Place description
SMSA-in central city----------
SMSA-outsidecentral city-----
Urban, noc SMSA---------------
Rural farm--------------------
Rural nonfann-----------------
20
Men
=
1.2 1.9
1
1.6 1.5
2.0 1.5
2.4 2.1
2.1 1.8
Women
ActualEx- Differ- Ex-
‘ceual petted ence pecced
Rate per 100 adults
-0.7 0.63 4.4 4.9 -0.5
0.1 1.13 3.9 4.3 -0.4
0.5 1.37 6.9 4.3 2.6
0.3 1.15 5.9 4.6 1.3
0.3 1.17 3.9 5.0 -1.1
0.89
0.92
1.61
1.28
0.78
Table 8. Actual and expectedprevalence rates of rheumatoid arthritis,by sex and education:United States, 1960-62
Men I Women
Education Actua1 Actual
Actual ‘x- Differ- Ex- Ex- Differ- ~petted ence petted ‘ctual petted ence
petted
Under 5 years----------------- 5.5
5-8 years--------------------- 3.9
9-12 years-------------------- 0.513 years and over------------- 0.3
Rate per 100 adults
3.3 2.2 1.65 L3.3 8.2 5.1 1.61
2.7 1.2 1.46 6.3 6.7 -0.4 0.94
1.1 -0.6 0.45 2.7 3.3 -0.6 0.831.1 -0.8 0.26 3.8 3.7 0.1 1.02
Table 9. Ratio of actual to expectedrates of positivediagnosticfindingain adults, by sex and “education:United Statea, 1960-62
Sex andeducation
Both sexes
Under 5 years--
5-8 years------
9-12 years-----
13 yearsand over------
~
Under 5 years--
5.8 yea~s ------
9-12 years-----
13 yearsand over------
Women
Under 5 years--
5-8 yeara------
9-12 yeara-----
13 yearaand over------
1 I I I 1
Ratio of actual to expected
1.52
1.17
0090
0.62
1.95
1.26
0.77
0.36
1.30
1.11
0.94
0,75
1.16
1.08
1.01
0.70
1.24
1.16
0.94
0.68
1.12
1.04
1.04
0.70
1.25
1.25
1.00
0.45
1.11
1.12
1.20
0.43
1.32
1.32
0.91
0.46
1.52
0.97
1.01
0.61
1.17
0.96
1.10
0.60
1.80
0.99
0.94
0.61
2.02
1.14
0.35
0.54
2.24
1,02
0.39
0.00
1.68
1.34
0.33
1.23
1059
1.09
0.79
0.82
1.91
1.44
0.29
0.25
1.49
‘0.97
0.89
0.97
Posi-tiveX-ray
Yates
1.36
1.21
0.84
0.37
0.84
1.41
0.78
0.35
2.38
0.84
0.94
0.37
ILMorning
BFT BFT stiff-1:32 1:512 nessor or (mm::,
more morequent)
0.85
1.22
0.75
1.26
0.69
1.10
0.80
1.34
1.00
1.34
0.72
1.17
1.28
1.79
0.39
0.20
1.51
1.63
0.13
0.33
0.77
2.16
0.66
0.00
1.00
0.83
1.07
1.02
0.92
0.86
1.15
0.88
1.07
0.81
1.03
1.15
●
21
Table 10. Percent of adults having positive diagnostic findings,by sex and education: UnitedStates, 1960-62
EPain Swell-ing,
on
‘ot=on j%e
m~f;g.
ness(s;~e,
quent)1
Swell-ing, I ‘%:-1 :,::-1R I lR2 l%%gSex and
educationTender-ness two
joints
Both sexes
Under 5 years--
5-8 years------
9-12 years-----
13+ years------
Men
Under 5 years--
5-8 years------
9-12 years-----
13+ years------
Women
Under 5 years--
5-8 years------
9-12 years-----
13+ years------
Percent of adults
13.1
9.1
5.1
3.4
12.1
7.0
2.6
1,3
14.1
11.3
7.0
5.8
4.9
4.4
2.2
1.0
3.1
2.9
1.8
0.7
6.6
5.9
2.5
1.3
1.0
0.5
0.1
0.1
1.3
0.5
0.1
0.0
0.7
0.5
0.1
0.3
6.2
3.3
1.1
1.2
3.7
2.2
0.2
0.1
8.5
4.4
1.9
2.3
2.2
1.6
0.4
0.2
1.8
2.3
0.4
0.2
2.6
0.7
0.4
0.2
4.4
5.5
2.1
3.5
3.6
5.0
2.2
3.5
5.2
6,0
2.1
3.5
1.5
1.8
0.2
0.1
2.6
2.3
0.1
0.2
0.5
1.2
0.3
0.0
14.8
12.4
16.1
14.9
12.1
11.3
15.1
11.4
17.4
13.5
17.0
18.8
21.4
18.3
11.8
8.2
15.7
13.4
7.4
5.6
26.6
23.1
15.3
11.1
3.9
2.3
1.4
0.9
2.9
2.1
1.5
0.8
4.9
2.4
1.3
1.0
Table 11. Actual and expected prevalence rates of rheumatoid arthritis in adults, by sex andfamily income: United States, 1960-62
Women
Family income
IWte per 100 adults
Under $2,000------------------
$2,000-$3,999-----------------
$4,000-$6,999-----------------
$7,000-$9,999-----------------
$10,000 and over--------------
Unknown----------------------- L6.6 3.3
1.5 2.0
0.6 1.2
0.9 1.1
1.4 1.51
0.7 1.7
3.3
-0.5
-0.6
-0.2
-0.1
-1.0
2.01
0.78
0.44
0.77
0.92
0.39 J-7.2 7.2
4.3 4.8
4.1 3,4
3.1 3.6
3.6 4.1
6.0 5.7
0.0-0.5
ci*7
-0.5
-0.5
0.3
1.000,90
1.22
0.85
0.87
1.06
22
Table 12. Actual and expectedprevalence rates of rheumatoid arthritisin adults, by sex andoccupation:United States, 1960-62
Sex and occupation
&
Professional,technical,and managerial-----------------
Farmers and farm managers-------------------------------
Clericaland salesworkers------------------------------
Craftsmen,foremen,and kindredworkers-----------------
Operativesand kindredworkers--------------------------
Private householdand serviceworkers-------------------
Farm and other laborers (exceptmine)-------------------
Women
Professional,technical,and managerial-----------------
Clericaland sales workers------------------------------
Operativesand kindred workers--------------------------
Privatehouseholdand serviceworkers -------------------
Actual Expected I Actua1
Expected
0.5
2.7
1.4
1.1
0.6
2.8
0.5
2.5
2.0
3.1
5.1
Rate per 100 adulta
1.1
1.8
1.0
1.1
0.8
1.3
1.1
3.0
2.4
3.2
3.9
-0.6
0.9
0.4
-0.2
1.5
-0.6
-0.5
-0.4
-0.1
1.2
0.47
1.50
1.38
1.00
0.72
2.25
0.50
0.82
0.81
0.95
1.31
23
Table 13, Actual and expectedprevalencerates of rheumatoidarthritisin adults,by sex and in-dustry:United States, 1960-62
Sex and industry
Men
Agriculture,forestries,and fisheries------------------
Mining and construction---------------------------------
Manufacturing-------------------------------------------
Transportation,communication,and other publicutilities----------------------------------------------
Wholesale and retail trade------------------------------
Finance, insurance,and real estate---------------------
Serviceand miscellaneous-------------------------------
Government----------------------------------------------
Women
Agriculture,forestries,and fisheries------------------
Manufacturing--------........----------------------------
Wholesale and retail trade------------------------------
Finance, insurance,and real estate---------------------
Serviceand miscellaneous-------------------------------
Actua1 Expected Difference
Rate per 100 adults
2.0
1.5
1.2
0.0
1.0
0.0
1.2
0.0
3.0
3.7
2.9
2.1
3.3
1.5
1.1
1.0
1.0
1.0
1.0
1.2
0.8
3.0
2.9
2.9
2.8
3.5
0.5
0,4
0.2
-1.0
0.0
-1.0
0.0
-0.8
0.0
0.8
-0.7
-0.2
Actua1
Expected
1.29
1.40
1.20
0.00
1.06
0.00
1.06
0.00
1.00
1.28
1.00
0.75
0.94
Table 14. Actual and expectedprevalence rates of rheumatoid arthritisin adults, by sex andusual activity status:United States, 1960-62
Sex and usual activitystatus
Men
Usually working-----------------------------------------
&tired -------------------------------------------------
Other---------------------------------------------------
Women
Usuallyworking-----------------------------------------
Keepinghouse-------------------------------------------
Other---------------------------------------------------
Actua1Actual Expected Difference Expected
Rate per 100 adults
1.0 1.2
6.8 5.9
2.9
2.8
5.5
1.3
3.6
5.1
3.1 2.1
-0.2 0.77
0.9 1.161.6 2.20
-0.8 0.76
0.4
1.0
1.07
1.46
24
Table 15. Actual and expected prevalence rates of rheumatoid arthritis in adults, by sex andmarital status: United States, 1960-62
Men Women
Marital status
Married---------
Vlidowed---------
Divorced--------
Separated-------
Never married---
Actua1 ActuslActual Expected Difference *xpecte~ Actual Expected Difference ~xpecte~
Rate per 100 adults
-L0.99
0.17
1.14
0.71
, 2.02
3.8
13.9
3.5
1.6
0.6
3.9 -0.1 0.99
11.4 2.5 1.22
4.7 -1.2 0.73
3.0 -1.4 0.54
2.1 -1.5 0.26
Table 16, Prevalence rates of rheumatoid arthritis in women,by number of children and age: UnitedStates, 1960-62
h
Number of children
Age
o 1-3 4 or more
18-24 years----------------------------------------------
25-34 years----------------------------------------------
35-44 years----------------------------------------------
45-54 years----------------------------------------------
55-64 years----------------------------------------------
65-74 years----------------------------------------------
75-79 years----------------------------------------------
Rate per 100 women
0.9
2.9
2.4
4.0
11.5
14.3
0.4
0.52.3
4.7
9.3
15.5
28.1
1.1
1.8
5.6
11.4
15.8
25.0
#
25
APPENDIX I
ITEMS ON THE MEDICAL HISTORY RELATING TO ARTHRITIS
4
27. a. Have you ever had morning stiffness, or weakness when You
get up? mmml
If YES b. How often? ~-C. Does it bother you ~@mml
I28. a. Ho# about swelling of the joints? Have YOIJnoticed anything
like that? mmml
If YES b. How often? Every few days1-
c. Does it bother You 1-1 ~
I
29. a. How about pain in the joints? Have you noticed anything like
that? l“m~
If YES b. How often? Every few days 1-C. Does it bother you
@!!mzIl~
I[30. a. Howabout tenderness of thejointsl’ I-laveyo unoticedanything
like that? mmzl
If YES b. How often? ~lz=Glc. Does it bother you [-] [-l
rlls 3032 lrag* 91
REV. 4-61
26
;1, a, tlave you ever had any reason
or arthritis?
If YES or ? b. Did a doctor
arthritis?
to think you may have rheu~tism
Bmm
tell you it was rheumatism or
mm
c. How long ago did you first start having it?
m-’~
d. Have you had it in the past 12 months?B@IIllIl
e. Do you take any pills or medicine for it? PIIIIIKJ
64. Has a doctor ever said you had gout? mm
IwPHS-3032 \Paga 16 I
REV. II-61
27
APPENDIX II
FORM USED IN, RECORDING FINDINGS ON THE
PHYSICAL EXAMINATION OF THE JOINTS
SUMMARY OF JOINT i NVOLVEIIENT
Joints
MAN IFESTAT 10HS
lender S.aalllllg Oaformlty Llm!tat!On Othorl Coda
;1. Shoulder
;2. Elbow
;3. Wrist
,U. Metacarpo-
phalangeal
5. Proxirral -
inter-
phalangeal
jb. Distal-
i nter-
phalangeal
j7. Hip
ja. Knee
19. Ankle
,0. Feet
,1, Cervical
spine
;2. Lumbar
spine
j~. other”
Record positive findings as R for zight, L for left, RL for both, except for spine (Items 61 and 62) which should becheck m&kcd.
Fingers (Items 54, 55, and 56): Record total number of joints involved on right or left.1c!Other> * ~anif=~tation~ incIude Hebecden> .S nodes, subcutaneous nodules, ulnas deviation, Pain On mOtiOn, heat)
arrophy, and funnel fist.
*‘‘Other” joints include temporomandibular, srcrnoclavicular, sacroiliac, and specific joints of the feet.
28
APPENDIx III
REFERENCE X-RAY PLATES
The following X-rays show gradings of rheumatoidarthritis bone destruction in the metacarpophalangeal(MC P), proximal interphalangeal (PIP), and metatar-sophalangeal (MTP) joints of the hands and feet. Theradiograph series is from the Clinical Center, NationalInstitutes of Health.
RHEUMATOID ARTHRITISBONE DESTRUCTION
I 2 3 4
GRADES
29
RHEUMATOID ARTHRITIS BONE DESTRUCTION
n
n
Tn
I 2 3 4
GRADES
30
APPENDIX IV
EFFECTIVENESS OF DIAGNOSTIC CRITERIA
Some estimates of the effectiveness of the diagnosticcriteria are given in figure I and table I. Figure I showsthe true positive rates for the various diagnosticfindings, A “true positive” is defined in this survey asany individual who is included in the classical, definite,or positive rheumatoid arthritis group. Table I givesthe prevalence rate in percentages of each diagnosticfinding in both the rheumatoid and nonrheumatoid
populations. The prevalence rates in the rheumatoidpopulation represent the sensitivity of the diagnosticfindings.
%ecificities and true positive rates for the
diagnostic “ndings are also given in the table. Sinceextra weight had already been assigned to symmetricalswelling and swelling in two joints, no true positive rateor specificity was calculated for either of these findings.Sensitivity is the chance of finding a true case, and
TRuE POSITIVE RATE
0.0 0.1 0!2 0.3 0,4 0.5 0.6 0.7 0,8 0.9 1.0
Morning stiffness, total,, .....
Morning stiffness(8cvcra, freq.cnt) ................
Morning stiffness(mild, freq.rnnt ).......................
Morning stiffness(severe, infrequent) ..............
Morning stiffness(mild, Inf rc q ant ) ...................
Ta nd arne as .............................
Elthar Ianderness orpain on motion . ......... ........ ...
Pdn on motion . ....... ...... ........
Both tenderness andpain on moflon ...................
Swallinq (one Joint] ..... ....... ..
Posit Ive BAT..,, ......................
Posit Iva X-ray ....................
Subcutaneous nodules ........
Figure 1. True positive rates, by diagnostic findings.
31
Table I. Prevalence rate in percentagesof each diagnosticfinding in both the rheumatoidandnonrheumstoidpopulations:Heath ExaminationSurvey, 1960-62
Diagnosticfinding
Mo~~ng atiffnes~,total--------------------
Severe and frequent-----------------------
Mild and frequent-------------------------
Severe and infrequent---------------------
Mild and infrequent-----------------------
Tenderness----------------------------------
Paf.non motion------------------------------
Either tendernessor pain on moti,on---------
Both tendernessand pain on motion----------
Swelling,one joint-------------------------
Swelling,two joints------------------------
Symmetricalswelling------------------------
Postive bentoniteflocculationtest---------
PositiveX-ray------------------------------
Subcutaneous nodules------------------------
:::o;t
rheumatoidpopulation
26.8
6.2
4.4
1.7
14.6
11.8
2.2
12.8
1.3
1.2
0.1
0.0
2.6
0.3
*
Percentin
rheumatoidpopulation(sensitivity)
65.7
30.4
9.3
5.9
20.1
72.1
17.6
77.9
11.8
18.1
8.8
62.3
19.6
11.3
*
Truepositiverate
(percent)
7.2
13.4
6.3
9.8
4.2
16.1
20.0
16.1
22.9
33.0.....
...
19.4
54.8
*
Specifhf.ty(percent)
73.2
...
...
● .,
● ,.
88.2
97.8
?37.2
98.7
98.8
● ..
● . .
97.4
99.7
*
specificityis the chanceof correctlyidentifyinga resultin some alterationof thevaluesforthethreehealthyperson.
As canbe seenfromtableII,allthreeindexes,i.e.,sensitivity,specificity,and true positiverate,arederivedfrom twofactors.FactorIinvolvesthecorrectdiagnosisofa caseasrheumatoidarthritis,andfactor
IIinvolvesthedeterminationofthepresenceorabsenceofa particulardiagnosticfinding.DiagnosticerrorsinfactorIandlabelingerrorsinfactorH willofnecessity
indexes.Neverthelessitislikelythatthevaluesforthe indexesdetailedin tableI giveatleasta roughapproximationofthetruepicture.
Itisinterestingtocomparesome ofthesefindingswiththespecificitiesfoundintheclinicalmaterialon
which the originalformulationof the ARA criteriawere based18(tableIH).Naturally,inapopulationstudy,
Table 11. Derivationof aensi.tivity,apeci.fi,c-i.ty,and true positiveratea
Df.agnoati.cfinding
Df.agnoatf.cfindingpresent--------------
Sensitivity=&
a c
b d
Specificity= &
Truepositiverate= -&
Table 111. Specifici.tiesi.npercent
Diagnosticfinding
Morning stiffness,cotal-
Mornf.ngstiffness (severeand frequentonly)------
Ei.thertendernessorpain on motion----------Joint swelling-----------X-ray changes------------Subcutaneousnodules-----
THEs
clinical populationcases Study
38 73.1
--- 93.8
_!_l_-E32
where there are many people entirely free from jointcomplaints, the specificities are higher than in aclinical practice, where everyone coming under obser-vation thinks he has some joint complaint. Thereforethe ciiugtmstic effectiveness of diagnostic findings ofhigh specificity in population studies might more readilyhe ascertained by comparing their true positive rates(table I). With this difficulty in the use of specificitiestaken into account, the specificities reported by Ropes,and others, *6 still roughly correspond in rank orderof magnitude to those found in the HES.
The major discrepancy is the finding of a greaterspecificity in the clinical group for morning stiffnessthan for pain on motion and/or tenderness, which is areversal of the HES finding. This reversal may beexplained by the fact that the Health Examination Survey
did not specify any particular time interval for theduration of morning stiffness. Burch found in a populationstudy that morning stiffness lasting iess than 10 minuteshad no association with rheumatoid arthritis. 13 In astudy by Abramson and others 324 out of 758 answered“yes” to the interview question “Do you wake Up withstiffness or aching in joints or muscles?” Of theseanswering yes only 19 percent were recorded by aclinician as having genuine morning stiffness.19 Thusevaluation by a clinician probably acted to increasethe specificity for morning stiffness in the originalARA clinical group. One can also note that more thanhalf of all the morning stiffness found by HES was mildand infrequent. As noted previously this minimalmorning stiffness had a very low true positive ratefor rheumatoid arthritis.
ooo —
33
APPENDIX V
DEMOGRAPHIC TERMS
Age. —The age recorded for each person is the ageat last birthday. Age is recorded in single years.
Race. —Race is recorded as “white,” “Negro,”or “other. “ “Other” includes American Indian, Chinese,Japanese, and so forth. Mexican persons are includedwith “white” unless definitely known to be Indian orof another nonwhite race.
Population size. —The five classes comprising thischaracteristic were derived from the design of thesample, which accomplished a stratification of theprimary sampling units by population size in each ofthree broad geographic locations. Because the surveywas started in 1960, the primary sampling units withineach of the five population-size classes were necessarilybased on populations and definitions of the 1950 census.The name of each selected primary sampling unit withineach population-size class and geographic location,along with other selected sample data, is presented inan earlier report. 3
The definitions for each ~f the five population-sizeclasses are as follows:
Giant metropolitan. weas. —This class includesprimary sampling units defined in the census asstandard metropolitan statistical areas (SMSA’S)having a population of 3,000,000 persons or more.
OtheY very large metyopolitaz ayeas. —Included inthis class are standard metropolitan statisticalareas with a population of 500,000 to 3,000,000 asdefined by the 1950 census.
Othev standayd met?’opolitan statistical ayeus. -
This class includes other SMSA’S.
Other urban areas.— This includes primary sam-pling units which are highly urban in compositionbut are not defined as SMSA’S.
Rwal ayeas.—llis includes primary samplingunits which are primarily rural in compositionaccording to census definitions,
Region.—For the purpose of classifying the popula-tion by geographic area, the United States was dividedinto three major regions. This division was especially
made for the design of the HES sample. The regionsand the States included are as follows:
Region
Northeast ---------
South ------------
West -------------
States Included
Maine, Vermont, New Hampshire,Massachusetts, Connecticut,Rhode Island, New York, NewJersey, Pennsylvania, Ohio, andMichiganDelaware, Maryland, District ofColumbia, West Virginia, Virginia,North Carolina, South Carolina,Georgia, Florida, Kentucky,Tennessee, Alabama, Mississippi,Arkansas, Louisiana, Oklahoma,and TexasWashington, Oregon, California,Idaho, Nevada, Montana, Utah,Arizona, Wyoming, Colorado,New Mexico, North Dakota, SouthDakota, Nebraska, Kansas,Minnesota, Iowa, Missouri, Wis-consin, Illinois, and Indiana
f.khan and rwal. —For the first six primarysampling units where examinations were conducted, thedefinition of urban and rural is the same as that usedin the 1950 census. These locations are Philadelphia,Pa., Valdosta, Ga., Akron, Ohio, Muskegon, Mich.,Chicago, Ill., and Butler, Mo. For the remainder of thesampling units the 1960 census definitions are used.
The change from 1950 to 1960 definitions is ofsmall consequence in the survey, since only sixlocations are affected. The major difference is thedesignation in 1960 of urban towns in New England andof urban townships in New Jersey and Pennsylvania.
According to the 1960 definition, the urban popula-tion comprises all persons living in (a) places of 2,500inhabitants or more incorporated as cities, boroughs,villages, and towns (except towns in New England, NewYork, and Wisconsin); (b) the densely settled urban
34
fringe, whether incorporated or unincorporated, ofurbanized areas; (c) towns in New England and town-ships in New Jersey and Pennsylvania which containno incorporated municipalities as subdivisions andhave either 25,000 inhabitants or more or a populationof 2,500-25,000 and a density of 1,500 persons or moreper square mile; (d) counties in States other than theNew England States, New Jersey, and ,Pennsylvania thathave no incorporated municipalities within their bound-aries and have a density of 1,500 persons or more persquare mile; and (e) unincorporated places of 2,500inhabitants or more not included in any urban fringe.The remaining population is classified as rural,
Place description. -In this survey the urban popula-tion is classified as living “in the central city” or“outside the central city” “of an SMSA. The remainingurban population is classified aa “not in SMSA.”
The definitions and titles of standard metropolitanstatistical areas are established by the U.S. Bureau ofthe Budget with the advice of the Federal Committee onStandard Metropolitan Statistical Areas.
The definition of an individual standard metropolitanstatistical area involves two considerations: first, acity or cities of specified population to constitute thecentral city and to identify the county in which it islocated as the central county; and, second, economicand social relationships with contiguous counties whichare metropolitan in character so that the periphery ofthe specific metropolitan area may be determined.
Persons “in the central city” of an SMSA aretherefore defined as those whose residency is in thecity appearing in the stand and metropolitan statisticalarea title. Persons residing in an SMSA but not in thecity appearing in the SMSA title are considered to beresiding “outside the central city.”
The remaining population is allocated into rural-farm and rural-nonfarm groups. The farm populationincludes all persons living in rural territory on placesof 10 acres or more from which sales of farm productsamounted to $50 or more during the previous 12 monthsor on places of less than 10 acres from which sales offarm products amounted to $250 or more during thepreceding 12 months. Other persons living in ruralterritory are classified as nonfarm. Persons are alsoclassified as nonfarm if their household paid rent forthe house but their rent did not include any land usedfor farming.
Employment status. — This term applies to theemployment status of persons during the 2-week periodprior to the week of interview. It is not intended thatthis term define the labor force or provide estimatesof the employed or unemployed population at the timeof the survey.
Persons who report that they either worked at orhad a job or business at any time during the 2-weekperiod prior to the week of interview are consideredemployed. This includes paid work as an employee ofsomeone else, self-employment in business, farming, or
professional practice, and unpaid work in a familybusiness or farm. Persons on ]qyoff from a job andthose absent from their job or business because oftemporary illness, vacation, strike, or bad weatherare considered employed if they expect to work as soonas the particular event causing their absence no longerexists. Free-lance workers are considered as currentlyemployed if they have a definite-arrangement with oneor more employers to work f~r pay according to aweekly or monthly schedule either full time or parttime. Excluded are such persons who have no definiteemployment schedule but work only when their servicesare needed. Also excluded are (1) persons receivingrevenue from an enterprise in whose operation they donot participate, (2) persons doing housework or charitywork for which they receive no pay, and (3) seasonal
fworkers durin the portion of the year they are notworking. (It sho ld be noted that these data were notcollected for Philadelphia.)
Occupation. —A person’s occupation maybe definedas his principal job or business. For the purposes ofthis survey the principal job or business of a respondentis defined in one of the following ways. If the personworked during the 2-week-reference period of theinterview or had a job or business, the questionconcerning his occupation (or what kind of work he wasdoing) applies to his job during that period. If therespondent held more than one job, the question isdirected to the one at which he spent the most time.When equal time is spent at each job, the questionrefers to tbe one he considers most important. Aperson who has not begun work at a new job, is lookingfor work, or is on layoff from work is questioned abouthis last full-time civilian job. A full-time job is definedas one at which the person spent 35 hours or more perweek and which lasted 2 consecutive weeks or more.A person who has a job to which he has not yet reportedand has never had a previous job or business is clas-sified as a “new worker. ”
Fe occupational groups are shown below with theappropriate census code categories.
Occupational title Census code
Professional, technical,and managerial workers ----- R,000-195, 250-285
Farmers and farm managers-- N,222Clerical and sales workers --- S, Y, Z, 301-395Craftsmen, foremen, and
kindred workers ------------ Q, 401-545Operatives and kindred
workers ------------------- T, W, 601-721Private household and
service workers ------------ P, 801-803, 81O-89OFarm and other laborers
(except mine) -------------- U, V, X, 901, 905,960-973
Unknown (including newworkers) ------------------ 995 and all other codes
35
(U.S. Bureau of Census, 1960 Census of Population,Classified Index of Occupation and Industries, U.S.Government Printing Office, Washington, D.c.,1960.)This information was not collected for Philadelphiaand Valdosta.
Industry. -The industry in which a person wasreportedly working is classified by the major activityof the establishment in which he worked.
The only exceptions to the above are those fewestablishments classified according to the major activityof the parent organization, and they are as follows:Iaimratories, warehouses, repair shops, and places forstorage.
The industry groupings are shown below. (Dataon industries were not collected for Valdosta andPhiladelphia.) The census code (the Classified Indexof Occupation and Industries) and the Standard Indus-trial Classification (SIC) code components are alsolisted.
lndusty title Censuscode
Agriculture, forestry, andfisheries ------------------------- A, 017, 018
Mining and construction ------- C, 126-156
Manufacturing -------------------- B, M, 206-459
Transportation, communi-cation, and other publicutili ties -------------------------- L, 507-579
Wholesale and retail trade ----- D, F, G, 606-696
Finance, insurance, and realestate ---------------------------- 706-736
Service and miscellaneous---- E, H, K, 806-898
Government ----------------------- J, 906-936
Unknown (including newworkers) ------------------------- 999
SIC code
01,02, 07(excludes0713), 06, 09
10-14, 15-17
19-39, 0713
40-49
50, 52-59
60-67
70, 72, 73, 75, 76,78,82,84,86,88,89
91-94
99
The industry “government” differs somewhat fromthe usual industrial classification of government, sinceit is limited to the postal service and to Federal, State,and local public administrations. This category includesonly uniquely governmental functions and excludes thoseactivities which may also be carried out by privateenterprise. For example, teachers in public educationalfacilities and nurses engaged in medical services ofgovernmental agencies are included with the “serviceand miscellaneous” group.
UsuaL activity status.— All persons are classifiedaccording to their usual activity status during the 12-month period prior tu the week of interview. The usualactivity status, in case more than one is reported, isthe one at which the person spent the most time duringthe 12-month period.
The categories of usual activity status are usuallyworking, usually keeping house, retired, and other. Forseveral reasons these categories are not comparable
with somewhat similarly named categories in officialFederal labor force statistics, First, the responsesconcerning usual activity status are accepted withoutdetailed questioning, since the objective of the questionis not to estimate the numbers of persons in laborforce categories but to identify crudely certain popula-tion groups which may have differing health problems.Second, the figures represent the usual activity statusover the period of an entire year, whereas officiallabor force statistics relate to a much shorter period,usually 1 week. Finally, in the definitions of specificcategories which follow, certain marginal groups areclassified differently to simplify procedures.
Usually workiw? includes persons who are paidemployees; self-employed in their own business,profession, or in farming; or unpaid employeesin a family business or farm. Work around thehouse or volunteer or unpaid work, such as fora church, is not counted as working.
Usually keeping house includes women whose majoractivity is described as “keeping house” and whocannot he classified as “working.”
Retired includes persons 45 years of age and overwho consider themselves retired. In case of doubta person 45 years of age and over is counted asretired if he or she has either voluntarily orinvoluntarily stopped working, is not looking for
work, and is not described as “keeping house. ”A retired person may or may not be unable towork.
OtheY in this report includes men not classifiedas “working” or “retired” and women not classifiedas “working, ” “keeping house,” or “retired.”Persons who are going to school are included inthis group,
Education. —Each person is classified by educationin terms of the highest grade of school completed. Onlygrades completed in regular schools, where persons aregiven a formal education, are included, A “regular”school is one which advances a person toward anelementary or high school diploma or a college,university, or professional school degree, Thus, educa-tion in vocational, trade, or business schools outsidethe regular school system is not counted in determiningthe highest grade of school completed.
Income of family OY unrelated individuals. —Eachmember of a family is classified according to thetotal income of the family of which he is a member.Within the household all persons related to each otherby blood, marriage, or adoption constitute a familY.Unrelated individuals are classified according to theirown income.
The income recorded is the total of all incomereceived by members of the family in the 12-month
36
period preceding the week of interview. Income from considered married. Separated refers to marriedall sources is included, e.g., wages, salaries, rents persons who have a legal separation, those livingfrom properties, pensions, and help from relatives. apart with intentions of obtaining a divorce, and other
Marital statzm. —The categories of marital status persons permanently or temporarily estranged fromare mavried, widowed, divorced, sepa~ated, and neve?’ their spouse because of marital discord.married. Persons with common-law marriages are
37
APPENDIX VI
DISTRIBUTION OF DIAGNOSTIC FINDINGS
This appendixcontainsa listingof allthepatterns The frequencyofoccurrenceofeach patternisincludedof distributionof the diagnosticfindingsfound in the intableIV.survey arrangedaccordingtothetotalARA pointcount.
Table IV.Distribution of diagnostic findings by ARA point count:Health E~amination Survey,1960-62
Number of cases
Point count: 8
2-----------------------1-----------------------
Point count: 7
4 -----.--- --------- ----1 -- -- - --- - - - - - -- - - - ----
Point count: 6
4----------------------4----------------------1----------------------1----------------------
Point count: 5
27---------------------8----------------------2----------------------1--------.-------------1----------------------1-----------------------1-------------------.--
Point count: 4
22---------------------:9-------------------------------------------
3----------------------3--.-------------------2----------------------1-----------------------1----------------------1----------------------
symmet-rical;well-ing
++
++-
-1-
:+
+-1-i-+
-1-++
+
+
Swell-ing,twojoints
+
++
+
h7ell-ing,oneioint
+
+
4orn-ing;tiff-ness
+i-
++
++
+
++
++
+
-1-
++
++
tender-ness
++
+-+
-1-++
+++
+
+
+-I-+
Painon
notion
+
+
-1-
+-!-
-!-
+-1-
++
?osi-tiveBFT
++
-1-+
-1-
++
+
-1-++
++
Posi-tiveX-ray
++
+-!-
++
-1-++
Subcu-taneousxodules
++
-4
+
38
Table IV. Distribution of diagnostic findings by ARA point count: Health Examination Survey,1960-62—Con.
*
Number of cases
Point count: 3
26---------------------.--...............---
;:-------------------------------------------
3--------------------------------------------
L--------------------
2. . . . . . . . . . . . . . . . . . ----2----------------------2........-.........----2.---------------------........--------------
L---------------------1..................----1----------------------1----------------------1---------.........----
Point count: 2
332--------------------45---------------------33----------” ----------31---------------------27---------------------14---------------------10---------------------6. ------- -....-.. ------3--------------------------------------------
L--------------------3..................-.-.3----------------......-----------------...-.
L---------------------1---------.-.?---------1----------------------1--------......-.------
Point count: 1
1 284------------------hi---------------------305--------------------31---------------------20---------------------;;3--------------------
.-...-.. . . . . . . . . -----
ymnet-ricalwell-ing
+
hrell-ing,twooints
+
++
+
)well-ing,oneioint
+
+
+-1-++
+
+
+
+
+
-1-
+
forn-ing;tiff-ness
+
-!-
++
+
+
+
++++
++
+++
1-
+
‘tende-rness
+
+++++
+-1-+
+++
+
++
+
++
Painon
lotion
+
++
+
+++
+
+
+
+
++
+
+
osi-iveBFT
+
++
+
+
+
+
++++
+
osi-ive-ray
+
+
+
+
+
+++
+
;ubcu-:aneous!odules
++
+
—0 o o—
39
APPENDIX
STATISTICAL
The Survey Design
The first cycle of the Health Examination Surveyemployed a highly stratified multistage probabilitydesign in which a sample of the civilian, noninstitutionalpopulation of the conterminous United States 18-79years of age was selected. At the first stage, a sampleof 42 primary sampling units (PSU’S) was drawn fromamong the 1,900 geographic units into which the UnitedStates was divided. Random selection was controlledwithin regional and size-of-urban-place strata intowhich the units were classified. As used here a PSUis a standard metropolitan statistical area or one tothree contiguous counties. Later stages result in therandom selection of clusters of typically about fourpersons from a neighl.mrhood within the PSU. The totalsample included some 7,700 persons in 29 differentStates. The detailed structure of the design and theconduct of the survey have been described in previousreports. Z3
Reliability
The methodological strength of the survey derivesespecially from its use of scientific probability sam-pling techniques and highly standardized and closelycontrolled measurement processes. This does not implythat statistics from the survey are exact or withouterror. Data from the survey are imperfect for threemajor reasons: (1) results are subject to samplingerror, (2) the actual conduct of a survey never agreesperfectly with the design, and (3) the measurementprocesses themselves are inexact even though standard-ized and controlled.
The first-stage evaluation of the survey was re-ported in reference 3, which dealt principally withan analysis of the faithfulness with which the samplingdesign was carried out. This study notes that out of the7,700 sample persons the 6,670 who were examined—aresponse rate of over 86 percent—gave evidence thatthey were a highly representative sample of the civilian,noninstitutional population of the United States. Impu-tation of nonrespondents was accomplished by attributingto nonexamined persons the characteristics of compa-rable examined persons as described in reference 3.
Vll
NOTES
The specific procedure used amounted to inflating thesampling weight for each examined person in order tocompensate for sample persons at that stand of thesame age-sex group who were not examined.
Sampling and Measurement Error
In the present report, reference has been madeto efforts to minimize bias and variability of themeasurement techniques.
The probability design of the survey makes pos-sible the calculation of sampling errors. Traditionallythe role of the sampling error has been the determinationof how imprecise the survey results may be becausethey come from a sample rather than from the measure-ment of all elements in the universe.
The estimation of sampling errors for a studyof the type of the Health Examination Survey is diffi-cult for at least three reasons: (1) measurement errorand “pure” sampling error are confounded in thedata—it is not easy to find a procedure which willeither completely include both or treat one or the otherseparately, (2) the survey design and estimation pro-cedure are complex and, accordingly, require compu-tationally involved techniques for the calculation of
Table V. Standard errors in prevalence ratesfor rheumatoid arthritis in adults, by sex andage: United Statea, 1960-62
Age Men Women
Total, 18-79 years -----
18-24 yeara ------------------25-34 years ------------------
35-44 years ------------------
45-54 years ------------------55-64 years ------------------65-74 years ------------------
75-79 yeara ------------------
+0.2 0.3
0.2
0.4 0.5
0.6 0.7
1.3 1.0
1.0 3.2
7.2 6,4
40
variances, and (3) from thesands of statistics, many fortion for which there are a
survey are coming thou-subclasses of thepopula-small number of sample
cases. Estimates of sampling error are obtained fromthe sample data and are themselves subject tosamplingerror when the number of cases in a cell is small or,even occasionally, when the number of cases is sub-stantial.
Estimates of approximate sampling variabilityfor selected statistics used in this report are presentedin tables V and VI. These estimates have been pre”-pared by a replication technique which yields overallvariability through observation of variability amongrandom subsamples of the total sample. The methodreflects both “pure” sampling variance and a part ofthe measurement variance.
In accordance with usual practice, the intervalestimate for any statistic may be considered the rangewithin one standard error of the tabulated statistic, with68 percent confidence; or the range within two standarderrors of the tabulated statistic, with 95 percent conf i-dence.
Expected Values
In tables 4-15 the actual prevalence rates for thevarious demographic variables are compared with theexpected. The computation of expected rates was doneas follows:
Suppose that in an area (say, the Northeast) theHealth Examination Survey estimates that there areNi persons in the ith age group (i= 1,2 . ..7. sum ofNi = N).
Suppose the Health Examination Survey estimatesthat the RA prevalence rate for the United States inthe ithage group is Xl.
Then the expected RA rate for the area is
Comparison of an actual value for, say, a region,with the expected value for that region is undertakenon the assumption that a meaningful statement can bemade which holds, in some average way, for all personsin the region. This may or may not be true. The spec-ified region may have higher values for young persons
and lower values for old persons than are found in otherregions. In that case an average comparison will ob-literate one or both of these differentials. A similarremark may be made with respect to values computedfor all races together, since relationships found in onerace may not be found in another. In arriving at thegeneral conclusions expressed in the text, an effort wasmade to consider all the specific data, including data notpresented in this report; but it must be recognized thatbalancing such evidence is a qualitative rather than aquantitative exercise. The standard error of the differ-ence between an actual and an expected value may beapproximated by the standard error of the actual value.
Small Numbers
In some tables magnitudes are shown for cells forwhich the sample size is SQ small that the samplingerror may be several times as great as the statisticits elf. Obviously in such instances the statistic hasno meaning in itself except to indicate that the truequantity is small. Such numbers, if shown, have beenincluded to convey an impression of the overall storyof the table.
Tests of Significance
Tests of significance for the demographic variableswere performed in two ways. The first was to dividethe difference between the actual and expected valuesby the standard error of the actual value. For example,for unmarried women the actual value was 1.6 percentlower than expected, and the standard error was 0.4percent. Since the difference was four times its stand-ard error, it may be deemed statistically significant,
The second method was to examine the age-specific differences (not published) between the prev-alence for the specified group and the prevalence forall persons. Thus, for widowed men the RA prevalencefor all six age groups was less than the overall prev-alence for these age groups. One of the seven agegroups (25-34 years) had no cases of RA recordedand thus could not be used for purposes of comparison.The probability of such an occurrence is 0.02, and thedifference is considered statistically significant.
41
Table VI. Standard errors in prevalence rates for rheumatoid arthritis in adults, by sex andselected characteristics: United States, 1960-62
Characteristic
Race
White ---------.........-------------------------------------------------------------
Negro-------------------------------------------------------------------------------
Other--------------------------------------------------------------------------------
Re~ion
Northeast---------------------------------------------------------------------------
South.........---------.............................--------------------------------
West --------------------------------------------------------------------------------
Population-size group
Giant
Other
Other
Other
Rura1
metropolitan areas-----------.------------------------------------------------
very large metropolitan areas------------------------..-------...-...---------
standard metropolitan statistical areas---------.........---------------------
urban areas................--------........------------------------------------
areas-------------------------------------------------------------------------
Place description
SkisA-incentral city----------------------------------------------------------------
SM8A-outside central city-----------------------------------------------------------
Urban, not SMSA......---------------------------------------------------------------
Rural farm--------...--...--------.-----------------------..-..-..------------------
Rural nonfam -----------------------------------------------------------------------
Usual activity status
Usually working -......-.-...-..........................----------------------------.-
Keeping house-----------------------------------------------------------------------
Retired.........---------.........------------------------------------.-------------
Other-------------------------------------------------------------------------------
Industry
Agriculture, forestry, and fisheries---.------------------------------------------.-
Mining and constmction -------------------------------------------------------------
Manufacturing -----------------------------------------------------------------------
Transportation, communication, and other public utilities ---------------------------
Wholesale and retail txade----------------------------------------------------------
Finance, insurance, and real estate-------------------------------------------------
Service and miscellaneous -----------------------------------------------------------
Government--------------------------------------------------------------------------
0.4
0.8
3.0
0.4
0,6
0.6
0.6
0.6
0.5
0.9
0.7
0.4
0.5
0.81.2
0.8
0.3
2,0
1.2
1.1
0.8
0.5
0.6
0.7
0.9
1.4
3.5
1.01.1
1.0
1.1
0.9
1.1
1.9
1.0
1.00.9
1.72.2
1.1
0.7
1.0
7.9
1.6
2.9
1.4
1.2
2.0
1.0
2.9
42
Table VI. Standard errors inselected
prevalence rates for rheumatoid arthritis in adults, by sex andcharacteristics: United States, 1960-62—Con.
Characteristic
Occupation
Professional$ technical, and managerial ---------------------------------------------
Farmers and fammanagers -----------------------------------------------------------
Clerical and sales workers----------------------------------------------------------
Craftsmen, foremen, and kindred workers ---------------------------------------------
Operatives and kindred workers ------------------------------------------------------
Private household and service workers -----------------------------------------------
Farm and other laborers (except mine)-----------------------------------------------
Education
Under 5 years-----------------------------------------------------------------------
5-8 years--.---------------.--------------------------------------------------------
9-12 years--------------------------------------------------------------------------
13 years and over -------------------------------------------------------------------
Familv income
Under $2,000 . . . . . . . . . . . . . . . . . . . . . . . . . . . --------- --------- --------- . . . . . . . . . ---------
$2,OOO-$3,999-----------------------------------------------------------------------
$4, OOO-$6,999-----------------------------------------------------------------------
$7, ooo-$9,999-----------------------------------------------------------------------
$10,000 and over--------------------------------------------------------------------
Marital status
Married ........---------------------------------------------------------------------
Widowed -----------------------------------------------------------------------------
Divorced-.....-.--------------------------------------------------------------------
Separated....-.-.-----------------------------------------------------------------..
Never mrried -----------------------------------------------------------------------
Men
0.3
1.6
0.8
0.5
0.4
1.4
0.5
2.0
1.0
0.2
0.2
1.8
0.60.3
0.5
0.7
0.4
1.2
1.9
1.5
0.7
Women
1.1
9.0
0.8
1.4
1.6
2.5
3.3
1.5
0.6
1.1
1.7
1.2
1.0
1.0
1.2
0.8
3.1
1.8
1.6
0.4
43
* U. S. GOVERNMENT PRINTING OFFICE : 1967 0 - 255-816
VITAL AND HEALTH STATISTICS PUBLICATION SERIES
Originally Public Health Service Publication No. 1000
SeTies 1.
Series 2.
Series 3.
Series 4.
SeVies 10.
Series 11.
Series 12.
Se7ies 13.
Series 14.
Series 20.
Sm”es 21.
Series 22.
PYogm?ns and collection fivoceduves. —Reports which describe the general programs of the National
Center for Health Statistics and its offices and divisions, data collection methods used, definitions,and other material necessary for understanding the data.
Data evaluation and methods research. — Studies of new statistical methodology including: experi-
mental tests of new survey methods, studies of vital statistics collection methods, new analyticaltechniques, objective evaluations of reliability of collected data, contributions to statistical theory.
Analytical studies —Reports presenting analytical or interpretive studies based on vital and health
statistics, carrying the analysis further than the expository types of reports in the other series.
Documents and committee Yeports. - Final reports of major committees concerned with vital and
health statistics, and documents such as recommended model vital registration laws and revisedbirth and death certificates.
Data $rom the Health Interview Swmev. —Statistics on illness, accidental injuries, disability, useof hospital, medical, dental, and other services, and other health-related topics, based on data
collected in a continuing national household interview survey.
Data from the Health Examination Survey. —Data from direct examination, testing, and measure-
ment of national samples of the civilian, noninstitutional population provide the basis for two typesof reports: (1) estimates of the medically defined prevalence of specific diseases in the UnitedStates and the distributions of the population with respect to physical, physiological, and psycho-
logical characteristics; and (2) analysis of relationships among the various measurements withoutreference to an explicit finite universe of persons.
Data from the Ins titutional Pojnda tion Surveys — Statistics relating to the health characteristics ofpersons in institutions, and their medical, nursing, and personal care received, based on nationalsamples of establishments providing these services and samples of the residents or patients.
DatQ j?om the Hospital Discharge Survey. —Statistics relating to discharged patients in short-stayhospitals, based on a sample of patient records in a national sample of hospitals.
Data on health resources: manpower and facilities. —Statistics on the numbers, geographic distri -bution, and characteristics of health resources including physicians, dentists, nurses, other healthoccupations, hospitals, nursing homes, and outpatient facilities.
Data on mortality. —Various statistics on mortality other than as included in regular annual ormontnly reports —special analyses by cause of death, age, and other demographic variables, also
geographic and time series analyses.
DOta on natality, ma?%age, and divorce. —Various statistics on natality, marriage, and divorce
other than as included in regular annual or monthly reports ~pecial analyses by demographicvariables, also geographic and time series analyses, studies of fertility.
Data porn the National Natulity and Mortality Surveys. — Statistics on characteristics of birthsand deaths not available from the vital records, based on sample surveys stemming from theserecords, including such topics as mortality by socioeconomic class, hospital experience in thelast year of life, medical care during pregnancy, health insurance coverage, etc.
For a list of titles of reports published in these series, write to: Gffice of Information
Nat ional Center for Health StatisticsPublic Health Service, HSMHARockville, Md. 20852
U.S. DEPARTMENT OF
HEALTH, EDUCATION, AND WELFARE
POSTAGE AND FEES PAID
U.S. DEPARTMENT OF H. E.W,
HEW 390
Public Health Service THIRD CLASS
HEALTH RESOURCES ADMINISTRATIONBLK. RATE
5600 Fishers Lane
Rockville, Md. 20652
OFFICIAL BUSINESS
Penalty Private Use, $300