HIV Working Group Seminar

Dr. Linda Harrison, PhD
Research Scientist, Harvard T.H. Chan School of Public Health, Department of Biostatistics

An Evaluation of Confidence Intervals for a Cumulative Proportion to Enable Decisions at Interim Reviews of Single Arm Clinical Trials
Abstract: Clinical trials often include interim analyses of the proportion of participants experiencing an event by a fixed time point. Frequently this cumulative proportion is estimated by the Kaplan-Meier estimator with a Wald approximate CI, which may have coverage issues with small samples or high event proportions. We reviewed four alternative CI methods for cumulative proportions (Beta Product Confidence Procedure (BPCP), BPCP Mid P, Rothman-Wilson, Thomas-Grunkemeier) and two CI methods for simple proportions (Clopper-Pearson, Wilson score). We conducted a simulation study comparing the CI methods and re-analyzed interim data from a HIV trial considering the proportion of participants experiencing virologic failure.