Abstract
We evaluated the role of tea and coffee and substances added (sugar/honey, creamers, and milk) on endometrial cancer risk in a population-based case–control study in six counties in New Jersey, including 417 cases and 395 controls. Multivariate odds ratios (OR) and 95% confidence intervals (CI) were computed using unconditional logistic regression. There was a moderate inverse association with coffee consumption, with an adjusted OR of 0.65 (95% CI: 0.36–1.17) for women who reported more than two cups/day of coffee compared to none. Tea consumption appeared to increase risk (OR: 1.93; 95% CI: 1.08–3.45), but after including the variables sugar/honey and cream/milk added to tea in the model, the risk estimate was attenuated and no longer statistically significant (OR: 1.77; 95% CI: 0.96–3.28 for those consuming more than one cup/day of tea compared to nonusers). There was a suggestion of a decreased risk associated with green tea, but the confidence interval included one (adjusted OR for one or more cups/week vs. none: 0.75; 95% CI: 0.48–1.18). We found an association with adding sugar/honey to tea, with those adding two or more teaspoons/cup having an OR of 2.66 (95% CI: 1.42–4.98;
p
for trend <0.01) after adjusting for relevant confounders. For sugar/honey added to coffee the corresponding OR was 1.43 (95% CI: 0.81–2.55). Our results indicate that sugars and milk/cream added to coffee and tea should be considered in future studies evaluating coffee and tea and endometrial cancer risk.