Tea consumption and leukemia risk: a meta-analysis.

Epidemiologic findings concerning the association between tea consumption and leukemia risk yielded mixed results. We aimed to investigate the association by performing a meta-analysis of all available studies. One cohort studies and six case-control studies with 1,019 cases were identified using PubMed, Web of Science, and EMBASE. We computed summary relative risks (RRs) and 95 % confidence intervals (CIs) using random effect model applied to the relative risk associated with ever, moderate, or highest drinkers vs. non/lowest drinkers. Subgroup analyses were performed based on country (China and USA). Compared with non/lowest drinkers, the combined RR for ever drinkers was 0.76 (95 % CI=0.65-0.89). In subgroup analyses, significant inverse associations were found for both China and USA studies. The summary RR was 0.57 (95 % CI=0.41-0.78) for highest drinkers. Same results were only found in China studies. No significant associations were found for moderate drinkers in overall analysis or in subgroup analyses. There was some evidence of publication bias. In conclusion, this meta-analysis suggests a significant inverse association of high tea consumption and leukemia risk. Results should be interpreted cautiously given the potential publication bias.