Rheumatoid arthritis in twins: a study of aetiopathogenesis based on the Australian Twin Registry.

The 1980 cohort of the Australian Twin Registry contains 3808 pairs of twins, 258 of whom self reported a diagnosis of rheumatoid arthritis (RA) in one or both subjects. Seventy two pairs were lost to follow up by 1990. The remaining 186 pairs received a self administered questionnaire, followed, if necessary, by telephone interviews to them, their general practitioners, and their specialists. Twenty discordant and three concordant pairs of twins were verified as having RA. The prevalence of RA in this sample was 0.40%. There was an 89% false positive rate for the self reported diagnosis of RA. Pairwise concordance percentages for RA were as follows: monozygotic 21% (95% confidence interval (CI) = 6 to 44), dizygotic 0% (95% CI = 0 to 25). It was concluded that: (a) there is a high false positive rate in self reporting RA; (b) the prevalence of RA in Australia may be less than the 0.8-1.0% often quoted; and (c) genetic factors play some part in the aetiopathogenesis of RA but do not account entirely for its determination.