ObjectivesTo explore the statistical performance of different methods for meta-analysis of single rates with zero events so as to provide evidence for selecting meta-analysis methods in evidence-based research.MethodsThrough Monte-Carlo simulations, we evaluated the performance of various transformation and correction methods for meta-analysis of single rates with zero events, considering the point estimate bias, confidence interval coverage and width.ResultsWhen the overall event rate was above 30%, all methods showed good statistical performance. As the event rate dropped, log, logit, and arcsine transformations could still maintain good performance. However, when the event rate was less than 5%, only Freeman-Tukey transformation showed excellent performance.ConclusionsThe meta-analysis of single rate based on Freeman-Tukey transformation is robust, and should be recommended as the preferable choice of meta-analysis of single rate with zero events.
ObjectiveTo examine statistical performance of different rare-event meta-analyses methods.MethodsUsing Monte-Carlo simulation, we set a variety of scenarios to evaluate the performance of various rare-event meta-analysis methods. The performance measures included absolute percentage error, root mean square error and interval coverage.ResultsAcross different scenarios, the absolute percentage error and root mean square error were similar for Bayesian logistic regression model, generalized mixed linear effects model and continuity correction, but the interval coverage was higher with Bayesian logistic regression model. The statistical performances with Mantel-Haenszel method and Peto method were consistently suboptimal across different scenarios.ConclusionsBayesian logistic regression model may be recommended as a preferred approach for rare-event meta-analysis.