Editorial Commentary: Clinical Trial Subgroup Analyses and Investigation of Secondary Outcome Measures Should Be Limited in Number to Avoid False Findings.

Abstract

Large randomized clinical trials are our highest quality evidence. Performing subgroup analyses and analysis of secondary outcome measures can and should be done to increase the scientific yield of these expensive and difficult studies. However, caution must be maintained when dramatic or unexpected outcomes are achieved. If we perform a large number of subgroup analyses and these are independent of each other, the risk of false-positive findings increases.