How should covariates be handled in randomized trials? Empirical evidence from 50 trials and recommendations for...

Covariate adjustment can improve precision and power in randomized clinical trials and is recommended by major regulatory agencies. However, there is limited empirical evidence on how different adjustment strategies perform across diverse real-world trials, leaving uncertainty about which methods and covariates should be prespecified in statistical analysis plans. We aim to address this gap and provide practical recommendations.

We conducted a large-scale empirical study using individual-level data from 50 publicly available randomized trials (29,094 participants; 574 treatment-outcome comparisons). We compared commonly used covariate-adjusted estimators, including analysis of covariance, inverse-probability weighting, g-computation, and machine learning-based approaches, combined with 3 covariate selection strategies. Performance was evaluated using precision gains, changes in point estimates, computational reliability, and the probability that covariate adjustment altered statistical significance relative to an unadjusted analysis.

Covariate adjustment improved precision in most settings, with a median variance reduction of 13.3% for continuous outcomes and 4.6% for binary outcomes. Adjustment was substantially more likely to produce a gain in statistical significance than a loss. Parsimonious regression approaches using a small prespecified set of prognostic covariates performed as well as or better than more complex methods, particularly in small- to medium-sized samples. Machine learning-based estimators did not provide additional precision and were more prone to computational failure for binary outcomes.

Across a wide range of randomized trials, parsimonious covariate adjustment provided consistent efficiency gains without introducing systematic bias. These findings support routine covariate adjustment in primary trial analyses and provide practical guidance for writing statistical analysis plans, selecting covariates, and planning sample size. All curated datasets and analysis code are openly released as a reproducible resource to support future clinical research.