Incidence of rheumatoid arthritis in a genetically predisposed population.

A follow-up study (mean duration five years) was undertaken on 370 previously unaffected first degree relatives from multicase rheumatoid arthritis (RA) families. The objectives were to determine the incidence of RA in this group and the possible predictors of disease development. In all, 14 individuals developed RA, equivalent to an incidence of 8/1000 person-years of observation. There was no control group included in this study as the intensive-follow-up required substantial compliance from highly motivated families. Population-based estimates, however, from a number of sources would suggest an annual incidence below 0.5/1000 per year, substantially less than the rate obtained in this study. The small number of incident cases precluded definitive conclusions about risk factors within these families but there were no important effects of age or sex. Possession of HLA-DR1 or DR4 explained only some of the increased risk. The increased incidence observed in the previously unaffected relatives of such families would suggest that this familial clustering did not for the most part arise by chance and that other shared genetic or environmental influences are relevant.