Coffee consumption and risk of pancreatic cancer among non-smokers in the Million Women Study
Session type: Poster sessions
Theme: Epidemiology and prevention
Some studies have suggested that coffee consumption might increase pancreatic cancer risk. Coffee consumption is often associated with smoking, a strong risk factor of pancreatic cancer. Even after adjustment for smoking, any apparent association between coffee consumption and pancreatic cancer risk would still be due to residual confounding.
As expected, in the Million Women Study cohort, the proportion of current smokers among women who reported drinking 5 or more cups of coffee (27%) was much higher than the cohort average (12%). We restricted our analysis to 418,042 women who had never smoked and followed them up for incident pancreatic cancer. Relative risks [RR] of pancreatic cancer were estimated by categories of coffee consumption (0, 1-2, 3-4, >5 cups daily) using Cox regression. The repeatability of self-report coffee consumption was good (correlation=0.8 between reports repeated within 2 years).
There were 1,182 incident cases of pancreatic cancer during 4.9 million person-years of follow up 13 years on average among women who had never smoked in this cohort. With adjustment for body mass index, exercise, alcohol consumption, tea consumption, daily fluid intake, and other factors, there was no significant association between coffee consumption and pancreatic cancer risk (p-for-heterogeneity=0.4). The adjusted relative risks, compared to non-drinkers, were 1.08 (95% confidence interval [95%CI] 0.94-1.25), 1.00 (95%CI 0.82-1.20), and 0.89 (95%CI 0.67-1.19) for women who drank 1-2, 3-4, and 5+ cups daily, respectively. The risks did not change substantially after excluding women who had diabetes or excluding the first 2 years of follow-up.
In the largest prospective cohort of non-smokers to date, we found little evidence suggesting that coffee consumption increased the risk of pancreatic cancer.