3-HMG-CoA reductase inhibitors (statins) block the growth of malignant cells in vitro. A meta-analysis of randomized controlled trials failed to show reduced risk of cancers in statin users. Case-control studies, however, have the advantage of examining remote exposures. This study determined the association between statins and breast cancer, colon cancer, lung cancer, prostate cancer, or any cancer in case-control studies. A comprehensive search for studies published through November 2006 was performed. Twenty case-control studies (100 129 incident cancer cases) were combined to obtain a pooled odds ratio using an inverse variance method. A funnel plot did not suggest a significant absence of unpublished data. The studies were significantly heterogeneous (P<0.01), thus a random effects model was used. The pooled OR and 95% confidence intervals for statin users and cancer were as follows: any cancer 0.71 (0.56-0.89), breast cancer 0.86 (0.60-1.23), colon cancer 0.89 (0.82-0.97), lung cancer 0.75 (0.50-1.11), and prostate cancer 0.74 (0.45-1.20). In this meta-analysis of case-control studies, we found a significant association between statin usage and any cancer, but when stratified by cancer type, only the association with colon cancer remained. On the basis of these results, randomized control trials with longer follow-up times than previously used are warranted.