Intention‐to‐treat in randomized controlled trials: Recommendations for a total trial strategy

Abstract

Intention-to-treat (ITT) in randomized controlled trials involves keeping participants in the treatment groups to which they were randomized regardless of whether they withdraw following randomization. Intention-to-treat is a strategy for maintaining the integrity of randomization and strengthening the trial's internal validity. Although ITT is advocated by the Consolidated Standards of Reporting Trials (CONSORT) guidelines, there is confusion about what ITT means and little specific advice on how to achieve it. The purpose of this article is to present definitions of ITT and to suggest strategies for implementing ITT as a total design strategy in nursing clinical trials. Recommendations are offered regarding study planning, study design, subject retention, sampling, data collection, data analysis, and reporting within the context of ITT.

Effects on healthcare utilization of case management for frail older people: A randomized controlled trial (RCT)

“…As the attritions were a threat to internal validity, analyses were conducted with Intention-ToTreat (ITT) (Polit & Gillespie, 2010). The imputations technique LOCF was chosen because of the ease with which it can be accomplished.…”

Treatment with 4Jointz reduces knee pain over 12 weeks of treatment in patients with clinical knee osteoarthritis: a randomised controlled trial

“…Adjustments for multiple comparisons were not used. We used a modified intent to treat (ITT) approach for data analysis 24 , where all patients who were randomised to receive treatment were included in the analysis. Secondly, missing data was imputed using Stata's multiple imputation (MI) functions, using 50 imputations per observation, from non-missing baseline data, using multivariate normal regression to generate outcome data.…”

Reducing risky relationships: a multisite randomized trial of a prison-based intervention for reducing HIV sexual risk behaviors among women with a history of drug use

“…However, 32 women were released but lost to follow-up. Consistent with principles of intention-to-treat analysis (Polit & Gillespie, 2010), their missing follow-up data was imputed. Our use of “mi impute chained” produced 20 datasets with each containing 378 participants.…”