Per-Protocol, Intention-To-Treat, And Complier Average Causal Effects Analyses In Randomized Controlled Trials: Linking Evidence To Practice

This article describes the differences between common analyses for randomized controlled trials (RCTs). Intention-to-treat (ITT) and per-protocol (PP) analyses are common approaches, but readers may also encounter complier average causal effects (CACE) analysis, a newer method that is gaining popularity. Because these analyses answer slightly different research questions, the aim of this article is to help readers of RCTs understand why researchers conduct these different types of analyses and how to interpret the findings. .