Sample size calculation for survival time data in cluster randomized trials using the NIATx 200 trial design as an example

• Activity Number: 253
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Antje Jahn-Eimermacher*+ and Jim Robinson and Andrew Quanbeck and Dennis McCarty
• Companies: University Medical Center, Mainz and University of Wisconsin-Madison and University of Wisconsin-Madison and Oregon Health and Science University
• Keywords: sample size ; cluster randomized ; survival ; time to event

Abstract

In several trials with a time to event endpoint, the interventions to be compared are randomized not to individuals but to groups of individuals (clusters), which might be patients of the same hospital. One example is the NIATx 200 trial, a cluster randomized trial investigating methods of disseminating quality improvement to addiction treatment centers in the U.S.. One of the primary endpoints to be compared between the different interventions is the time patients have to wait for their first treatment. Members of the same cluster tend to be more similar than members of different clusters causing intra-cluster correlation. Correlation affects the power of a trial and thus has to be considered when planning the sample size. We illustrate how to plan the sample size for clustered time to event data using the NIATx 200 trial as an example.