Data synthesis and analysis
Meta-analysis and meta-regression were performed using Comprehensive Meta-Analysis software. Random-effect pooled odds ratios (ORs) were calculated with the corresponding 95% confidence intervals (CIs) to summarize the results overall and within subgroups. Heterogeneity was assessed by using the I2 statistic. We performed subgroup analyses by study type (RCT, cohort). Random-effects meta-analysis was used to pool analyses. Meta-regression analyses were performed to evaluate whether differences in pravastatin dosage and timing of treatment initiation modified the association between exposure to pravastatin and outcomes, and to explain the heterogeneity in the estimated effect size. The data were analyzed using the Comprehensive Meta-Analysis software.