Developing Methods to Analyze Data Missing Due to Death in Cluster Randomized Trials

This project focuses on the development of new methods, software, and guidance for researchers on how to analyze cluster randomized trials data when data are missing due to death. Cluster randomized trials, or CRTs, are research studies that compare treatments among different groups of patients, or clusters. An example of a cluster is a group of people who receive care at a single clinic. CRTs often examine patient-centered outcomes, such as quality of life or physical function. Sometimes people with serious illness who are enrolled in CRTs may die during the study. When data are missing due to death, it is hard to measure how the treatment affected patient-centered outcomes. Missing data due to death can also bias results.

In this study, the research team is creating and testing new methods to account for CRT data that are missing due to death. The team is comparing results from the new methods with those from existing methods. Further, the research team will review methods examined in the study with patients, caregivers, researchers, and funders. With their input, the team is creating guidance for addressing CRT data that are missing due to death. Finally, the team is creating software for analyzing patient-centered outcomes in CRTs.

Patient-Centered Outcomes Research Institute