What question did this study set out to answer?

This research aims to address the challenges of inference in cluster-randomized trials due to variances in cluster sizes and structures.

June 6, 2026Open Access

Cluster Trials Inference With CARE

Key Points

This research aims to address the challenges of inference in cluster-randomized trials due to variances in cluster sizes and structures.
Performed simulations with real and synthetic data to test current analysis methods.
Analyzed treatment assignment discrepancies in cluster structures.
Proposed the CARE protocol for improved inference methods.
Current methods showed significant limitations in handling cluster size variability.
The CARE protocol provided a more credible approach for trial analysis.
Results indicate increased transparency and comparability across studies when using CARE.

Abstract

ABSTRACT We show that cluster‐randomized trials—especially pragmatic designs—often exhibit substantial (but underappreciated in practice) heterogeneity in cluster sizes and structures, distorting inference. Our simulations—reassigning treatment in real data and varying imbalance in synthetic data—show that currently recommended methods (such as targeted maximum likelihood estimation and small‐sample corrected generalized estimating equations) are not optimized to this challenge. We propose the CARE (Clarify, Apply, Refine, Evaluate) protocol, which anchors inference in a design‐based benchmark and provides a principled pathway for incorporating assumption‐rich methods—making trial analysis more credible, transparent, and directly comparable across studies.

Read Full Paperexternally

Ask AI

Mark Helpful

Bookmark

Relay

View Full Paper