The Analysis of Multiple Endpoints in Clinical Trials

TLDR

Randomized clinical trials often compare multiple endpoints, and conventional significance testing can inflate the overall Type I error rate; selecting a single primary endpoint is sometimes infeasible, and Bonferroni correction is another option. The paper investigates a global test statistic that can be applied to any set of asymptotically normal test statistics. The authors develop a general framework that incorporates quantitative, binary, and survival endpoints, present two illustrative examples, and discuss the relative merits of the proposed strategies. The proposed global test is conservative for correlated endpoints in multivariate normal data, and the authors illustrate its performance with two examples, discussing the relative merits of the strategies.

Abstract

Treatment comparisons in randomized clinical trials usually involve several endpoints such that conventional significance testing can seriously inflate the overall Type I error rate. One option is to select a single primary endpoint for formal statistical inference, but this is not always feasible. Another approach is to apply Bonferroni correction (i.e., multiply each P-value by the total number of endpoints). Its conservatism for correlated endpoints is examined for multivariate normal data. A third approach is to derive an appropriate global test statistic and this paper explores one such test applicable to any set of asymptotically normal test statistics. Quantitative, binary, and survival endpoints are all considered within this general framework. Two examples are presented and the relative merits of the proposed strategies are discussed.