In a double-blind randomized controlled trial of a new oral anticoagulant versus warfarin, 8% of patients assigned to the new drug stopped taking it after one month. At the end of follow-up, the investigators analyse every patient in the group to which they were originally allocated, irrespective of adherence. The chief advantage of this approach is that it:

• A Provides the best estimate of efficacy under ideal adherence conditions
• B Allows detection of adverse events occurring after discontinuation of the drug
• C Eliminates the dilution caused by non-adherent patients
• D Preserves the prognostic balance created by randomization ✓

Explanation

Intention-to-treat analysis retains every participant in their original randomized arm, preserving the baseline equivalence in known and unknown prognostic factors that randomization created. Excluding dropouts or protocol violators breaks this balance and reintroduces selection bias. Per-protocol analysis, not intention-to-treat, estimates efficacy under ideal adherence, and analysing as randomized actually includes any dilution from non-adherence rather than eliminating it.

Reference: Gordis Epidemiology, 6th ed.

High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP

Written and medically reviewed by the StethoPrep medical team.