Intention-to-Treat Analysis
An analytical strategy in which all randomised participants are included in the group to which they were assigned, regardless of adherence or protocol deviations.
Full Definition
Intention-to-treat (ITT) analysis is a statistical principle applied in randomised controlled trials whereby data from all participants are analysed according to their original allocation, preserving the integrity of randomisation and minimising bias from post-randomisation dropouts or protocol violations. In dental research trials, ITT analysis is particularly relevant when attrition is anticipated, such as in multi-visit interventional studies. Editors should verify that authors distinguish between ITT, modified ITT (mITT), and per-protocol analyses, as these carry different implications for result interpretation. Missing data handling methods (e.g., multiple imputation, last observation carried forward) should be explicitly stated. Confusing ITT with per-protocol analysis is a frequent and important error in trial reporting.
Usage
Usage note: Editors should flag when authors describe an 'ITT analysis' but exclude randomised participants; this may constitute a modified ITT and requires clarification.
In Context
- "The primary analysis was performed on an intention-to-treat basis, with multiple imputation used for missing outcome data." — Statistical methods section of an RCT
- "The authors conducted only a per-protocol analysis; peer reviewers requested an intention-to-treat analysis to accompany it." — Peer review decision letter