Intention to Treat vs Per-protocol Analysis

Background Story:

We often use different populations (Safety, Efficacy, ITT, Per-Protocol) to generate Clinical Trial Reports, but what are the differences between the populations? 
Today we will go over the basics of Intention to Treat and Per-protocol Analysis

We can estimate the Average Causal Effect of the treatment due to the Randomization process.( Treatment is exchangeable)

The Counterfactuals 

  • What is the outcome if all patients had been treated compared to all patients had not been treated?

Ideal Scenario:
Intent to Treat:Mean in the treated to Mean in the Placebo

Average Casual Effect CE= E[Y(tr)] -E[Y(pl)]= E(Y|Group A)- E(Y|Group B)
E[Y(tr)]=E[Y(tr)|A]-E[Y(tr)|B]

Reality: Some subjects did not take the medication.
Per-Protocol: Conly compare the compliers

Happy Studying!

Thanks 77 for sharing the Havard Catalyst material !

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Social media & sharing icons powered by UltimatelySocial
%d bloggers like this: