Defining Analytic Populations in Randomized Trials: Revisiting the Role of Intention to Treat

The primary goal of randomized controlled trials is typically to understand the impact of a treatment on an outcome(s) of interest. Randomization is a critical design component for making these types of cause-and-effect inferences, as it ensures balance between the treatment groups in terms of both known and unknown confounders (factors related to both exposure and outcome), which can introduce bias when imbalanced across exposure (treatment) groups. When comparing treatment groups in the primary analysis of a randomized controlled trial, participants are typically included with the group to which they were randomized, following a framework referred to as intention to treat (ITT). ITT intentionally does not account for postrandomization events (eg, protocol nonadherence) in order to maintain the integrity of the randomization. There are other non-ITT frameworks for randomized controlled trial analyses, including the common per-protocol, as-treated, and modified ITT analyses. This article summarizes ITT and non-ITT frameworks and the research questions which they address, discusses their trade-offs, and provides information about operationalizing randomized controlled trial analyses, including discussion of the available topical guidance, considerations for conducting analyses in the face of missing data, and evolving statistical methods for non-ITT analyses.

Authors

Institutions

Publication Details

Journal
Stroke
Published
2026-11-19
DOI
https://doi.org/10.1161/strokeaha.126.053678
Primary Topic
Statistical Methods in Clinical Trials
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Defining Analytic Populations in Randomized Trials: Revisiting the Role of Intention to Treat

Gary Cutter, Mathew J. Reeves, Xiaoxiao Zhou, Megan E. McCabe et al.
Stroke
Statistical Methods in Clinical Trials
article

Defining Analytic Populations in Randomized Trials: Revisiting the Role of Intention to Treat

Gary Cutter, Mathew J. Reeves, Xiaoxiao Zhou, Megan E. McCabe, George Howard
article en

Abstract

The primary goal of randomized controlled trials is typically to understand the impact of a treatment on an outcome(s) of interest. Randomization is a critical design component for making these types of cause-and-effect inferences, as it ensures balance between the treatment groups in terms of both known and unknown confounders (factors related to both exposure and outcome), which can introduce bias when imbalanced across exposure (treatment) groups. When comparing treatment groups in the primary analysis of a randomized controlled trial, participants are typically included with the group to which they were randomized, following a framework referred to as intention to treat (ITT). ITT intentionally does not account for postrandomization events (eg, protocol nonadherence) in order to maintain the integrity of the randomization. There are other non-ITT frameworks for randomized controlled trial analyses, including the common per-protocol, as-treated, and modified ITT analyses. This article summarizes ITT and non-ITT frameworks and the research questions which they address, discusses their trade-offs, and provides information about operationalizing randomized controlled trial analyses, including discussion of the available topical guidance, considerations for conducting analyses in the face of missing data, and evolving statistical methods for non-ITT analyses.

Stroke
University of Alabama at Birmingham (US), Michigan State University (US)
Statistical Methods in Clinical Trials
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon