Chronic diarrhoea and risk of rheumatoid arthritis: findings from the French E3N-EPIC Cohort Study

Objectives. To assess the relationship between gastrointestinal disorders and the risk of further development of RA. Methods. The Etude Epidemiologique aupres des femmes de la Mutuelle generale de ('Education NationaleEuropean Prospective Investigation into Cancer and Nutrition Study is a French prospective cohort including 98 995 healthy women since 1990. Participants completed mailed questionnaires on their lifestyles and health-related information. Gastrointestinal disorders were assessed in the third questionnaire (sent in 1993). Hazard ratios and 95% CIs for incident RA were estimated using Cox proportional hazards regression models with age as the time scale. Models were age adjusted, and then additionally adjusted for known risk factors of RA such as smoking, and for potential cofounders. Results. Among 65 424 women, 530 validated incident RA cases were diagnosed after a mean (s.D.) of 11.7 (5.9) years after study baseline. In comparison with no gastrointestinal disorder, chronic diarrhoea was associated with an increased risk of developing RA during follow-up (hazard ratio= 1.70, 95% CI 1.13, 2.58), independently of dysthyroidism or dietary habits. The association was stronger among ever-smokers (hazard ratio =2.21, 95% CI 1.32, 3.70). There was no association between RA risk and constipation or alternating diarrhoea/constipation. Conclusion. Chronic diarrhoea was associated with an increased risk of subsequent RA development, particularly among ever-smokers. These data fit with the mucosal origin hypothesis of RA, where interaction between intestinal dysbiosis and smoking could occur at an early stage to promote emergence of autoimmunity, followed years later by clinical disease.