An epidemiologic study of trends in prevalence of rheumatoid factor seropositivity in Pima Indians - Evidence of a decline due to both secular and birth-cohort influences

Objective. Previous population studies have suggested that both rheumatoid factor (RF) production and rheumatoid arthritis (RA) may be declining in occurrence, and. both secular and birth-cohort influences have been implicated. Since Pima Indians have a very high incidence of RA and also have shown recent evidence or a decline in RA, this study evaluated the relative contributions of age, secular, and birth-cohort influences on RF seropositivity in the Pima Indian population. Methods. RF data, as assayed by both the bentonite flocculation test (BFT) and the sheep cell agglutination test (SCAT), were available on 5,345 Pima Indians born between 1886 and 1975, who were surveyed at biennial intervals between 1966 and 1995. An age-period-cohort analysis was conducted using data on 18,295 examinations undertaken during the period of study. Results. There was a decline in the proportion of positive test results for RF (titer greater than or equal to1:32) by both BFT and SCAT, in both male and female subjects from 1966-1975 to the later decades of the study (1976-1985 and 1986-1995). Across all periods, by both assays, the crude proportion of positive titers increased with increasing age of the subjects. There was a very clear birth-cohort effect: the highest likelihood of seropositivity was in those individuals born around the end of the nineteenth century, with continuing decline in seropositivity up to the most recent birth cohort. A logistic regression analysis, adjusting for Pima heritage and sex, demonstrated a substantially greater influence of birth cohort than of calendar year on the frequency of RF positivity. Conclusion. In the Pima Indian population, environmental influences in early life are important predictors of the lifelong likelihood of RF positivity. This may have implications for understanding the epidemiology and etiology of RA.