Sample Size and Power Caculation for Cluster Randomized Trial

Seminar Series

Wednesday, January 8, 2020 - 12:00
Hock Plaza, 8th Floor #8065
Hrishikesh Chakraborty, DRPH

Abstract: Cluster randomized trial (CRT) generate correlated observations and researchers need to account for the correlated observations during sample size calculation. Intra-cluster correlation (ICC) is a measure of the average Correlation within clusters and precisee ICC estimates are key to calculate sample size for CRT.  There have been methods developed to esimate ICC for continuous, binary, and survival outcomes.  I will present new ICC esimation method for categorical outcome.  I will also illustrate methodology for the sample size calculation of a hierarchical endpoint in the pragmatic trial of higher vs lower serum phosphate targets in patients undergoing hemodialysis (HiLo), a cluster randomized trial.  

Speaker: Hrishikesh Chakraborty, DRPH
Associate Director
DCRI Biostatistics
Duke University, School of Medicine