Mortality of rheumatoid arthritis in Japan: a longitudinal cohort study.

OBJECTIVE To determine the mortality risk of Japanese patients with rheumatoid arthritis, taking into account lifestyle and physical factors, including comorbidity. METHODS 91 individuals with rheumatoid arthritis were identified during screening a cohort of 16 119 Japanese atomic bomb survivors in the period 1958 to 1966. These individuals and the remainder of the cohort were followed for mortality until 1999. Mortality risk of the rheumatoid patients was estimated by the Cox proportional hazards model. In addition to age and sex, lifestyle and physical factors such as smoking status, alcohol consumption, blood pressure, and comorbidity were included as adjustment factors for the analysis of total mortality and for analysis of mortality from each cause of death. RESULTS 83 of the rheumatoid patients (91.2%) and 8527 of the non-rheumatoid controls (52.9%) died during mean follow up periods of 17.8 and 28.0 years, respectively. The age and sex adjusted hazard ratio for mortality in the rheumatoid patients was 1.60 (95% confidence interval, 1.29 to 1.99), p < 0.001. Multiple adjustments, including for lifestyle and physical factors, resulted in a similar mortality hazard ratio of 1.57 (1.25 to 1.94), p < 0.001. Although mortality risk tended to be higher in male than in female rheumatoid patients, the difference was not significant. Pneumonia, tuberculosis, and liver disease were significantly increased as causes of death in rheumatoid patients. CONCLUSIONS Rheumatoid arthritis is an independent risk factor for mortality. Infectious events are associated with increased mortality in rheumatoid arthritis.