Leveraging baseline covariates to analyze small cluster-randomized trials with a rare binary outcome

TL;DR

This paper evaluates covariate adjustment methods like propensity score weighting and multivariable regression in small cluster-randomized trials with rare binary outcomes, providing practical guidance through simulations and a real case study.

Contribution

It systematically compares covariate adjustment strategies in small CRTs with rare outcomes, offering new insights into their efficiency and variance estimation performance.

Findings

Propensity score weighting and multivariable regression have different efficiency advantages depending on scenarios.

Bias-corrected variance estimators perform well in finite samples.

Reanalysis of a pediatric ICU trial demonstrates practical application of methods.

Abstract

Cluster-randomized trials (CRTs) involve randomizing entire groups of participants -- called clusters -- to treatment arms but are often comprised of a limited or fixed number of available clusters. While covariate adjustment can account for chance imbalances between treatment arms and increase statistical efficiency in individually-randomized trials, analytical methods for individual-level covariate adjustment in small CRTs have received little attention to date. In this paper, we systematically investigate, through extensive simulations, the operating characteristics of propensity score weighting and multivariable regression as two individual-level covariate adjustment strategies for estimating the participant-average causal effect in small CRTs with a rare binary outcome and identify scenarios where each adjustment strategy has a relative efficiency advantage over the other to make practical recommendations. We also examine the finite-sample performance of the bias-corrected sandwich variance estimators associated with propensity score weighting and multivariable regression for quantifying the uncertainty in estimating the participant-average treatment effect. To illustrate the methods for individual-level covariate adjustment, we reanalyze a recent CRT testing a sedation protocol in 31 pediatric intensive care units.