Objective: To assess the incidence of chronic persistent rheumatoid arthritis (RA) in a population-based cohort of twins and to determine the impact of smoking. Methods: In a historical cohort study on twins born in 1920 to 1982, we identified 157 cases of RA among 45,280 responders (response rate 80%). Information on smoking was obtained by questionnaire and interview. A mixed-effects Poisson regression model was used to estimate incidence rate ratios with age, sex, smoking duration, and smoking intensity as covariates. We used the SplitLexis procedure in the Epi R package to study a possible effect of period or cohort in addition to age on the variation of the incidence. Results: The annual incidence of chronic persistent RA was 18.8 per 100,000 person-years, ages 15–73 years (females 25.2, males 12.0), increasing with age to a maximum at age 60 years in females and age 70 years in males. The incidence rate ratio among ever-smoking patients was 1.96 (95% confidence interval [95% CI] 1.43–3.76), 1.93 (95% CI 1.00–3.7) after 30 pack-years, and 1.034 (P < 0.001) per year of smoking, implying a doubling of risk after 20 years regardless of sex and smoking intensity. We did not detect significant period or cohort effects. Conclusion: The incidence of chronic persistent RA is lower than the incidence figures reported in inception cohorts. Smoking duration, but not intensity, doubled the risk of RA after 20 years of smoking in both sexes.
- Cohort studies
- Rheumatoid arthritis