An increasing number of studies have investigated the association between SLCO1B1 −521T>C and −388A>G polymorphisms and the risk of statin-induced adverse drug reactions (ADRs), but the results have been inconsistent. This meta-analysis was performed to gain more insight into the relationship. PubMed, Embase, Cochrane Library and Web of Science were searched for relevant articles published before March 5th, 2015. The quality of included studies was evaluated by the Newcastle-Ottawa Quality scale. Pooled effect estimates (odds ratios [ORs] or hazard ratios [HRs) and corresponding 95 % confidence intervals (CIs) were calculated to assess the association in overall and subgroup analyses for various genetic models. Begg’s rank correlation test and Egger’s linear regression test were used to examine the publication bias. A total of nine cohort and four case–control studies involving 11, 246 statin users, of whom 2, 355 developing ADRs were included in the analysis. Combined analysis revealed a significant association between the SLCO1B1−521T>C polymorphism and increased risk for ADRs caused by various statins, but the synthesis heterogeneity was generally large (dominant model: pooled effect estimate = 1.85, 95 % CI 1.20–2.85, P = 0.005; I2 = 80.70 %, Pheterogeneity < 0.001). Subgroup analysis by statin type showed that the ADRs risk was significantly elevated among simvastatin users (dominant model: pooled effect estimate = 3.43, 95 % CI 1.80–6.52, P = 0.001; I2 = 59.60 %, Pheterogeneity = 0.060), but not among atorvastatin users. No significant relationship was found between the −388A>G polymorphism and ADRs caused by various statins (dominant model: pooled effect estimate = 0.94, 95 % CI 0.79–1.13, P = 0.526; I2 = 40.10 %, Pheterogeneity = 0.196). The meta-analysis suggests that SLCO1B1 −521T>C polymorphism may be a risk factor for statin-induced ADRs, especially in simvastatin therapy. Conversely, there may be no significant association for −388A>G polymorphism.