Objective: To investigate smoking for its association with the incidence of seropositive and seronegative rheumatoid arthritis (RA).
Methods: A cohort of adult Finns was examined by the Social Insurance Institution's Mobile Clinic in 1966-72. The 24,445 women and 28,364 men who had neither arthritis nor a history of it at the start of the study were followed until the end of 1989 using record linkage with the Institution's population register to identify patients entitled to free antirheumatic medication. Sufficient information was obtained on 512 incident cases of RA, of whom 119 men and 229 women were seropositive and 42 men and 122 women seronegative.
Results: There was a close association between smoking and the incidence of seropositive RA in men. It was not due to confounding by age, geographical location of residence, marital status, social class, self-perceived general health, or body mass index, although these factors correlated with smoking history. As adjusted for these factors, the relative risk of seropositive RA was 2.6 (95% confidence interval, 1.3-5.3) in male ex-smokers and 3.8 (95% confidence interval, 2.0-6.9) in current smokers, in comparison with the men who had never smoked. The association persisted throughout the entire followup period, but it was most distinct after the first 14 years of followup. Smoking did not predict seropositive RA in women, nor was it predictive of seronegative RA in men or women.
Conclusion: Exposure to tobacco smoke, or some factor or cluster of factors associated with smoking, may trigger the production of rheumatoid factors and, in interaction with the male sex, subsequently contribute to the development of clinically manifest RA.