Adaptive Selection of the Optimal Strategy to Improve Precision and Power in Randomized Trials

TL;DR

This paper introduces an adaptive method for selecting the best covariate adjustment strategy in randomized trials, leveraging machine learning and cross-validation to enhance precision while controlling Type-I error, applicable to large trials.

Contribution

It extends adaptive prespecification techniques to large trials, allowing flexible selection among multiple covariate adjustment methods using modern machine learning and cross-validation.

Findings

Maintains Type-I error control in simulations.

Achieves 20-43% sample size reduction for same power.

Demonstrates efficiency improvements in real trial data.

Abstract

Benkeser et al. demonstrate how adjustment for baseline covariates in randomized trials can meaningfully improve precision for a variety of outcome types. Their findings build on a long history, starting in 1932 with R.A. Fisher and including more recent endorsements by the U.S. Food and Drug Administration and the European Medicines Agency. Here, we address an important practical consideration: *how* to select the adjustment approach -- which variables and in which form -- to maximize precision, while maintaining Type-I error control. Balzer et al. previously proposed *Adaptive Prespecification* within TMLE to flexibly and automatically select, from a prespecified set, the approach that maximizes empirical efficiency in small trials (N$<$40). To avoid overfitting with few randomized units, selection was previously limited to working generalized linear models, adjusting for a single covariate. Now, we tailor Adaptive Prespecification to trials with many randomized units. Using $V$-fold cross-validation and the estimated influence curve-squared as the loss function, we select from an expanded set of candidates, including modern machine learning methods adjusting for multiple covariates. As assessed in simulations exploring a variety of data generating processes, our approach maintains Type-I error control (under the null) and offers substantial gains in precision -- equivalent to 20-43\% reductions in sample size for the same statistical power. When applied to real data from ACTG Study 175, we also see meaningful efficiency improvements overall and within subgroups.