Horace J. Dodge

Estimates of the prevalence of rheumatic diseases in the population of Tecumseh, Michigan, 1959–60

Abstract Over 90 per cent of the more than 9000 residents of Tecumseh, Michigan participated in a program of comprehensive health examinations in 1959–1960. Included in the examination were inquiries regarding rheumatic symptoms, physical examination of the spine and peripheral joints, and the latex fixation test for rheumatoid factor and serum uric acid measurement. Joint pain or aching, joint swelling, morning stiffness, and past arthritis or rheumatism were common complaints, occurring in one-eighth to one-third of the population age 6 yr and over. The age-sex specific prevalence rates for each of these historical items rose with increasing age. Male and female rates were quite similar during the first four decades of life, but thereafter female rates were somewhat higher. The prevalence of positive latex fixation tests for rheumatoid factor was essentially the same for male and female subjects, 3.4 per cent for males and 3.35 per cent for females. The rates rose progressively from approximately one per cent in the 6–16 yr age group to almost 14 per cent for males and 9.4 per cent for females in the oldest age groups. The latex fixation test performed poorly as a case detection tool, only one-third of those respondents with positive tests having any other evidence to suggest a diagnosis of rheumatoid arthritis. Relatively high rates of latex positively occurred in individuals with a history of jaundice or infectious hepatitis and in those with evidence of emphysema or right heart failure. None of the respondents with psoriasis or pregnancy at the time of examination had positive latex tests. The prevalence of latex positivity showed a rise with increasing systolic blood pressure in persons over 30 yr of age, but appeared to be unrelated to diastolic blood pressure, serum cholesterol or serum uric acid. Prevalence rates for “definite” rheumatoid arthritis, based on the diagnostic criteria proposed by the American Rheumatism Association, were 0.4 per cent for all subjects age 6 yr and over and 0.5 per cent for all subjects age 16 yr and over. Prevalence rates for “definite” plus “probable” rheumatoid arthritis were 1.3 per cent in the age group 6 yr and over and 1.7 per cent in the age group 16 yr and over. In all diagnostic categories prevalence rates for females exceeded those for males. In the age group 16 yr and over the female to male ratio was 2.3: 1 for “definite” disease and 2.4: 1 for total suspected cases of rheumatoid arthritis. The prevalence rates for rheumatoid arthritis rose with increasing age; rates for “definite” disease rose from 0.44 per cent in the fourth decade to 0.79 per cent in the eighth decade for males and from 1.69 per cent in the fifth decade to 2.47 per cent in the eighth decade for females. The over-all prevalence rates for suspected ankylosing spondylitis, not confirmed by X-ray examination of the spine or sacroiliac joints, were 0.4 per cent for males and 0.05 per cent for females age 6 yr and over. Prevalence rates for osteoarthritis, diagnosed on the basis of physical rather than radiological examination were 2.2 per cent for males and 5.0 per cent for females age 6 yr and over. Rates were highest in the older age groups, being 20.3 per cent for males and 40.8 per cent for females in the age group 60 yr and over. The prevalence of Heberdens nodes was greater in female than in male respondents and demonstrated a similar rise with increasing age. The prevalence of “probable” gout, based on available clinical information was 0.5 per cent for male and 0.3 per cent for female subjects age 4 yr and over having serum uric acid determinations. The highest rates were observed in subjects in the age range 40–59 yr. The prevalence rates for a history of rheumatic fever and / or chorea were similar for male and female subjects age 6 yr and over, 0.8 per cent in the case of rheumatic fever and 0.1 per cent in the case of chorea. A history of chorea was not obtained among respondents under age 20. Physical evidence of rheumatic heart disease was recorded in all age groups and was slightly more prevalent in females, 0.6 per cent as compared to 0.4 per cent in males. Comparisons of the results of this investigation with those of other population studies have been made although the interpretation of any differences or similarities is inherently limited by methodological problems and observer variation.

The relationship of serum uric acid to risk factors in coronary heart disease

Abstract Analysis of the uric acid data from the initial examination cycle of the Tecumseh Community Health Study reveals that not only age and sex but also body weight or build should be considered when evaluating serum uric acid levels in disease or relating serum uric acid levels to other physiologic variables. It was found that serum uric acid levels of persons with coronary heart disease were not significantly different from the mean of the population studied and hence cannot be considered an attribute associated with the disease. Furthermore, there appears to be no evident relationship between serum uric acid levels and serum cholesterol or blood sugar, both of which are regarded as risk factors for coronary heart disease. Although there is no consistent pattern of a relationship between uric acid levels and casual blood pressure levels, some trends are suggestive. Of the conditions and disease states studied, only gout and pregnancy differed significantly from the means of the entire population with respect to serum uric acid scores; the levels in gouty subjects, as expected, were higher and in pregnant women lower, confirming reports in the literature.

AGE-SEX SPECIFIC PREVALENCE OF RADIOGRAPHIC ABNORMALITIES OF THE JOINTS OF THE HANDS, WRISTS AND CERVICAL SPINE OF ADULT RESIDENTS OF THE TECUMSEH, MICHIGAN, COMMUNITY HEALTH STUDY AREA, 1962-1965*

Abstract Radiographs of the hands and wrists and cervical spine of 4415 residents of Tecumseh, Michigan, aged 16 yr and over, have been reviewed. Evidence of rheumatoid arthritis was observed in the hands and wrists of 0.19 per cent of males and 0.34 per cent of females. The prevalence of cervical spine changes, which could be attributed to rheumatoid arthritis or ankylosing spondylitis, was insignificant. Degenerative changes in both examined regions occurred much more frequently and showed a marked positive correlation with advancing age. Radiographic evidence of osteoarthrosis of the hands and wrists was observed in 19.5 per cent of males and 23 per cent of females. In the cervical spine, degenerative disc disease (16–19 per cent) was noted somewhat more frequently than osteoarthrosis (11 per cent). Among the other radiographic lesions which were observed, those of post traumatic changes, including amputation of digits, in the hands and wrists and fusion of cervical vertebral bodies, thought to represent a failure of segmentation in embryonic life, were most frequent.

Biologic variation of human serum immunoglobulin concentrations: Sex-age specific effects

Abstract Concentrations of serum immunoglobulins provide clinically useful parameters of immunologic or inflammatory diseases. Valid interpretations of these data require recognition of biologic variations which exist throughout the life-span of the individual. The most important of these variables are sex, age and race. The present study reports the sex-age specific changes in serum immunoglobulin concentrations in 3213 unselected individuals from 5 to 94 yr of age from a community study. Evaluation of biologic differences was accomplished by multiple linear regression on log e transformed data. IgG and IgA concentrations increased with age with slight but significant differences between the sexes. Females had higher serum levels of IgG and lower levels of IgA than males. IgM levels did not change with age and were significantly higher in females than in males. None of the immunoglobulin concentrations decreased with age, nor was there a change in variability. These observations document significant biologic variations which occur in immunoglobulin concentrations and are the first data available based on an unbiased community sample.

OBSERVATIONS ON THE DISTRIBUTION OF SERUM URIC ACID LEVELS IN PARTICIPANTS OF THE TECUMSEH, MICHIGAN, COMMUNITY HEALTH STUDIES A COMPARISON OF RESULTS OF ONE METHOD USED AT TWO DIFFERENT TIMES AND OF TWO METHODS USED SIMULTANEOUSLY

Abstract Individuals participating in the Tecumseh Community Health Study, in 1959–1960 (TCHS I) and 1962–1965 (TCHS II), had serum uric acid determinations by an enzymatic spectrophotometric method, SUA(ES), as part of a comprehensive health examination. Among the 1633 male and 1725 female subjects, the mean serum uric acid concentration had increased by 0.32 mg per 100 ml for males, and by 0.18 for females, over the the average of 4.1 yr between determinations. It was found that, approximately, 40 per cent of subjects had a decrease, and 56 per cent, an increase, in serum uric acid concentration. The change was greater than plus or minus 2.0 mg per 100 ml for 28.7 per cent of males and 19.4 per cent of females. Decreases of as much as 8.2, and increases of as much as 9.9 mg per 100 ml were observed. How much of the average increase, in an average 4.1 yr interval between tests, is due to ageing, except for males in the age range of 10–19 yr, is problematical. It seems probably that most of the average increase was due to technical factors in performance of the test. In the study of 1962–1965, serum uric acid determinations were done by two methods. An automated colorimetric method was compared with an enzymatic spectrophotometric method. Both methods showed the same types of distribution curves of serum uric acid concentrations and curves of age-sex specific mean serum uric acid. The values obtained by the automated colorimetric method were slightly, but, consistently, higher to a statistically significant degree than those of the enzymatic spectro-photometric method. For 2092 male subjects, the mean SUA(AC) was 0.21 mg per 100 ml greater than the mean SUA(ES), while for 2104 females subjects, this difference was 0.14. Of 4196 study subjects, whose sera were tested by both methods, about 90.0 per cent showed an agreement of the two values within ± 0.9 mg per 100 ml. However, 1.9 per cent of males and 1.6 per cent of females, had ES values which exceeded AC values by 2.0 mg per 100 ml, while 2.8 per cent of male and 2.3 per cent of female subjects had AC values greater than ES values by 2.0 mg per 100 ml. The extremes of differences range from 4.4 to 7.3 mg per 100 ml. The data in this paper gives us increased confidence in the essential validity of the mean age and sex distributions of serum uric acid values, from the Tecumseh Community Health Studies, published in an earlier paper [2]. It is concluded that, for the purposes of population studies, the automated colorimetric method is more advantageous than the enzymatic spectrophotometric method, because of its lesser cost. Comparison of the results of the two studies emphasizes the problems of definition of hyperuricemia. Further mathematical analysis of the distribution curves obtained in such studies may lead to definitions of normo- and hyperuricemia of greater precision for clinical purposes.