BIOS Seminar: The sps3 experience

Leslie McClure, PhD, professor and chair of epidemiology and biostatistics at Drexel University’s Dornsife School of Public Health, will present “A Post-Hoc Evaluation of a Sample Size Re-estimation: The SPS3 Experience,”

Methodologically, Dr. McClure is interested in the design and analysis of randomized clinical trials with multiple outcomes, and methodology for re-estimating sample size in on-going randomized trials. She is the principle investigator for the Coordinating Center for the CDC-sponsored Diabetes LEAD (Location, Environmental Attributes, and Disparities) Network, with a primary goal of furthering our understanding of the role of community-level factors and geographic differences in diabetes incidence and prevalence across the US and across demographic groups.

Abstract:

The use of adaptive designs has been increasing in randomized clinical trials. Sample size re-estimation is a type of adaptation in which nuisance parameters are estimated at an interim point in the trial and the sample size re-computed based on these estimates. However, post-hoc evaluations of adaptations, including sample size re-estimations, have not
frequently been performed, leaving questions about the gains that come from utilizing these procedures in practice. In this talk, I describe the statistical impact of implementing a sample size re-estimation in the Secondary Prevention of Small Subcortical Strokes (SPS3) study and describe the effect of the adaptation on the practical aspects of the study.