How to handle censoring and the need to consider competing/semi-competing risks already increases the complexity of designing clinical trials with time-to-event outcomes. Additional complexity is added when using a time-to-event outcome in a cluster randomized trial because the correlation between individuals within a cluster also needs to be considered. Using the Strategies to Reduce Injuries and Develop Confidence in Elders (STRIDE) trial, a cluster randomized trial with the primary outcome of time to first serious fall injury, as a motivating example, this talk will discuss considerations in the design and analysis of cluster randomized trials with time-to-event outcomes in the presence of censoring and competing/semi-competing risks. It will also discuss the methods available for clustered time-to-event data and weigh the pros and cons of those methods under varying design considerations. This talk will conclude by outlining the gaps in knowledge that still exist.

Dr. Denise Esserman is a Professor of Biostatistics at the Yale School of Public Health (YSPH), the Co-Director of the Yale Center for Analytical Sciences (a collaborative center housed in YSPH), and the Director of the Biostatistics Core for the Yale Program on Aging. Her work, both collaborative and methodological, primarily focuses on the design, coordination, conduct, and analysis of pragmatic clinical trials, including cluster randomized trials with time-to-event data and competing risk, patient preference trials, stepped-wedge designs, and individually randomized group-treatment trials. She was a senior statistician on the STRIDE trial.