Intention-to-treat and per-protocol analysis

In a randomized controlled trial (an RCT), the “intention-to-treat” (ITT) analysis is the analysis done using all the participants according to the group they were originally assigned in the RCT, regardless of what treatment they received. This is looking at the results in an unbiased way as it was originally intended. The issue is when participants drop out of the RCT, or if they stray from the protocol. So in this case, the “per-protocol” analysis can be done, where ONLY participants who completed the study, and followed protocol, are included. The problem with this is it introduces bias, because now the researcher has to choose who makes the cut, which kind of defeats the “randomization” aspect of the RCT.