Project Summary

One of PCORI’s goals is to improve the methods that researchers use for patient-centered outcomes research. PCORI funds methods projects like this one to better understand and advance the use of research methods that improve the strength and quality of comparative effectiveness research. 

What is the project about?

Groups of people may respond differently to the same treatment. For example, women may, on average, receive more benefit from a treatment than men do. If researchers look at study data for men and women together, they might miss the chance to fully understand the benefits of a treatment. The difference in effect between groups is known as heterogeneity of treatment effect, or HTE. 

Researchers often want to understand HTE for cluster randomized trials, or CRTs. CRTs are a type of research study that compares treatments among clusters of patients, such as people who receive care at a single clinic. Existing guidance for researchers on design and analysis of CRTs has focused on understanding overall treatment effect. There is little guidance on how to plan CRTs to prepare for HTE analyses, for example, to ensure that the sample size is large enough to accurately identify HTE.

In this study, the research team is developing new methods, software, and guidance to help researchers design and plan CRTs to better identify HTE. 

How can this project help improve research methods?

Results may help researchers better design and plan CRTs to detect HTE.

What is the research team doing? 

First, the research team is developing methods for planning two types of CRTs to help detect HTE. The first type is the parallel CRT. In this design, researchers assign clusters by chance to different treatments at the start of the study. The second type is the stepped-wedge CRT. In this type of study, all clusters start with the standard treatment. Then, each cluster switches to the new treatment at a certain time during the study that is chosen by chance. By the end of the study, all clusters are receiving the new treatment. 

For both types of CRT, the research team is creating statistical formulas that help researchers decide how many patients and clusters to enroll in a study to accurately detect HTE. The methods consider whether clusters of patients are receiving care from the same doctor or at the same clinic. Also, the methods account for differences within and between clusters for the outcome and for factors that affect how well treatment works. 

Next, the research team is creating free software to help researchers use the methods for detecting HTE to plan CRTs. Finally, the team is developing guidance to help researchers use the new methods and software.

Research methods at a glance

Design ElementDescription

Develop methods and guidance for planning multilevel parallel CRTs and stepped-wedge CRTs to enable confirmatory HTE analysis

Develop software for planning CRTs to enable confirmatory HTE analysis 

ApproachSample size estimation, power analysis, multilevel modeling, effect modification, Monte Carlo simulations, software development

Journal Citations

Related Journal Citations

Project Information

Fan Li, PhD
Yale School of Public Health
New Methods for Planning Cluster Randomized Trials to Detect Treatment Effect Heterogeneity

Key Dates

July 2021
November 2025


Award Type
State State The state where the project originates, or where the primary institution or organization is located. View Glossary
Last updated: March 14, 2024