Clinical trials characterizing the effects of an experimental therapy rarely have only a single outcome of interest, but comparing a large number of outcomes using the significance threshold of 0.05 runs a high risk of falsely concluding that a significant effect is present when none exists. In a study with 16 secondary endpoints, the investigators used a serial gatekeeping approach (an alternative to the Bonferroni approach) for statistical testing to minimize the chance of false positive results. (The JAMA Network)