Smoking, drinking, and pancreatitis: A population-based cohort study in Taiwan

Results: During 94,601 person-years of follow-up, 66 incident cases of pancreatitis occurred. Neither current nor ever smoking was associated with the incidence of pancreatitis (adjusted hazard ratio [aHR] 1.13; 95% confidence interval [CI], 0.62-2.06; and aHR, 1.14; 95% CI, 0.63-2.06, respectively). Dose-response analysis also showed no association between smoking and pancreatitis. By contrast, regular (aHR, 3.09; 95% CI, 1.51-6.30) and heavy alcohol drinking (aHR, 6.46; 95% CI, 3.07-13.60) were associated with an increased incidence of pancreatitis compared with abstinence.Conclusions: In this Asian population-based cohort, alcohol was associated with pancreatitis in a dose-dependent way, but smoking was not associated with pancreatitis. These results support for ethnic differences in susceptibility to smoking- and alcohol-related pancreatitis.Objectives: In Western population, smoking is a dose-dependent risk factor for pancreatitis, whereas a threshold of 5 drinks per day may exist for alcohol to increase pancreatitis risk. Given ethnic differences in tobacco and alcohol metabolism, we examined the associations between smoking, alcohol, and pancreatitis in Asians.Methods: A population-based cohort was assembled using participants of the National Health Interview Survey in Taiwan. Information on drinking and smoking was collected by in-person interview at baseline. Incident cases of pancreatitis were identified through data linkage with National Health Insurance claims database. Copyright ? 2014 by Lippincott Williams & Wilkins.