Comparison of intent-to-treat analysis strategies for pre-post studies with loss to follow-up

Abstract

In pre-post studies when all outcomes are completely observed, previous studies have shown that analysis of covariance (ANCOVA) is more powerful than a change-score analysis in testing the treatment effect. However, there have been few studies comparing power under missing post-test values. This paper was motivated by the Behavior and Exercise for Physical Health Intervention (BePHIT) Study, a pre-post study designed to compare two interventions on postmenopausal women's walk time. The goal of this study was to compare the power of two methods which adhere to the intent-to-treat (ITT) principle when post-test data are missing: ANCOVA after multiple imputation (MI) and the mixed model applied to all-available data (AA). We also compared the two ITT analysis strategies to two methods which do not adhere to ITT principles: complete-case (CC) ANCOVA and the CC mixed model. Comparisons were made through analyses of the BePHIT data and simulation studies conducted under various sample sizes, missingness rates, and missingness scenarios. In the analysis of the BePHIT data, ANCOVA after MI had the smallest p-value for the test of the treatment effect of the four methods. Simulation results demonstrated that the AA mixed model was usually more powerful than ANCOVA after MI. The power of ANCOVA after MI dropped the fastest as the missingness rate increased; in most simulated scenarios, ANCOVA after MI had the smallest power when 50% of the post-test outcomes were missing.

Physical fitness in older women with osteoporosis and vertebral fracture after a resistance and balance exercise programme: 3-month post-intervention follow-up of a randomised controlled trial

“…Group, time (i.e., post-intervention and 3-month follow-up) and the interaction between group and time were fixed effects. Mixed models are considered a robust method for missing data in ITT analysis of pre–post studies [ 41 ]. The underlying structure of the model estimates the outcome at each visit, assuming that the missing data have the same correlation structure as observed data [ 42 ].…”

Effectiveness of a web-based computer-tailored intervention promoting physical activity for adults from Quebec City: a randomized controlled trial

“…First, the primary analysis regarding the effectiveness of the intervention was corroborated by the sensitivity analysis. The difference in p -value between the primary analysis (linear mixed model) and sensitivity analysis (multiple imputation) methods could be explained in part by the fact that the former offers greater statistical power (Sullivan, White, Salter, Ryan, & Lee, 2018 ; Xi, Pennell, Andridge, & Paskett, 2018 ). Second, the use of video-delivery tailored feedback in the context of web-based computer-tailored interventions is innovative.…”

A two-component intervention to improve hand hygiene practices and promote alcohol-based hand rub use among people who inject drugs: a mixed-methods evaluation

“…Pre-post hand hygiene frequency change was modelled using a multivariable linear mixed model with a random intercept per subject for all available observations. This type of model is best suited for pre-post intervention studies, as it takes into account correlation between repeated measures within subjects and loss to follow-up, and therefore ensures adherence to the intention-to-treat principle [ 44 ]. The main explanatory variable was the follow-up variable with the baseline visit as a reference.…”