Comparison of the performance of cumulative meta-analytical methods with rare events
View Presentation View Presentation
*Xiao Ding, Merck & Co., Inc. 

Keywords: cumulative meta-analysis

Because of multiple comparisons, regular meta-analysis statistics can not maintain the correct type 1 error rate in the cumulative meta-analysis setting. The cumulative meta-analysis using the law of iterated logarithm (LIL) is proposed by Lan et al to maintain the type 1 error rate, primarily for efficacy endpoints. Simulations are conducted to exam the performance of cumulative meta-analytical methods, including LIL, for safety endpoint with rare events.