Methodological review showed that time-to-event outcomes are often inadequately handled in cluster randomized trials

Objectives To estimate the prevalence of time-to-event (TTE) outcomes in cluster randomized trials (CRTs) and to examine their statistical management. Study design setting We searched PubMed identify primary reports CRTs published six major general medical journals (2013–2018). Nature and, for TTE outcomes, methods sample size, analysis, measures intracluster correlation were extracted. Results A analysis was used 17% (32/184) either as a or secondary outcome sensitivity analysis. Among five with outcome, two accounted both nature size calculation; one reported measure 32 least 44% (14/32) clustering all analyses. identified 12 additional which there at not analyzed might have been preferred. Conclusion are uncommon but appropriate infrequently used. Our results suggest that further methodological development explicit recommendations needed.