On Cluster Randomized Trials with the Desirability of Outcome Ranking (DOOR) Endpoints

TL;DR

This paper develops new statistical methods to extend the DOOR framework for evaluating patient outcomes in cluster randomized trials, accommodating various cluster configurations and sizes.

Contribution

It introduces a suite of methods based on U-statistics and influence functions to analyze cluster trials with DOOR endpoints, applicable in diverse scenarios.

Findings

Methods perform well in simulations across different cluster sizes and numbers.

Application to a neonatal trial illustrates practical utility.

Extensions handle mixed cluster treatment groups and small sample sizes.

Abstract

Cluster randomized trials are widely used when individual randomization is logistically infeasible or when correlations between observations cannot be ignored, especially in fields such as ophthalmology, infectious disease, vaccine research, and sociology. The desirability of outcome ranking (DOOR) framework evaluates patient-centric benefit-risk using an ordinal outcome and a Wilcoxon-Mann-Whitney statistic-based approach to compare outcome distributions between interventions. We propose a suite of new methods to extend DOOR to cluster trials based on properties of U-statistics and influence functions to estimate within-cluster and between-cluster treatment effects. These approaches can be applied in different scenarios, including mixtures of clusters with two treatment groups and clusters with only one group, and both small and large numbers of clusters. Simulations demonstrate that the proposed methods perform well under various scenarios regarding the number of clusters and cluster sizes. As an illustration, we apply the proposed methods to a cluster randomized crossover trial comparing delayed cord clamping and umbilical cord milking for newborns.