OBJECTIVE: To evaluate the association between coffee consumption and the risk of prostate cancer. METHODS: We searched PubMed, EMBASE, and the bibliographies of relevant articles in August 2009. Two evaluators independently reviewed and selected articles based on predetermined selection criteria. RESULTS: Twelve epidemiological studies (eight case-control studies and four cohort studies) were included in the final analysis. In a meta-analysis of all included studies, when compared with the lowest level of coffee consumption, the overall relative risk (RR) of prostate cancer for the highest level of coffee consumption was 1.16 (95% confidence interval [CI] 1.01-1.33). In subgroup meta-analyses by study design, there was a significant positive (harmful) association between coffee consumption and prostate cancer risk in seven case-control studies using both crude and adjusted data (RR 1.20, 95% CI 1.02-1.40; and RR 1.21, 95% CI 1.03-1.43, respectively), whereas there was no significant association in four cohort studies using crude or adjusted data (RR 0.97, 95% CI 0.68-1.38; and RR 1.06, 95% CI 0.83-1.35, respectively). CONCLUSION: Given that a cohort study gives a higher level of evidence than a case-control study, there is no evidence to support a harmful effect of coffee consumption on prostate cancer risk. Further prospective cohort studies are required.