|
Activity Number:
|
428
- Contributed Poster Presentations: Health Policy Statistics Section
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Health Policy Statistics Section
|
|
Abstract #328532
|
|
|
Title:
|
Optimal Sample Size for Cluster Randomized Trials: a Simulation-Based Search Algorithm
|
|
Author(s):
|
Ruoshui Zhai* and Roee Gutman
|
|
Companies:
|
Brown University and Brown University
|
|
Keywords:
|
Cluster Randomization Trial;
Sample Size Calculation;
Randomization Distribution
|
|
Abstract:
|
Cluster randomization trials (CRTs) is an experiment designed to randomize groups rather than individual units to treatments. Some advantages of CRTs include simplicity of design and reduction of experimental contamination. However, when the goal is to estimate individual level effects CRTs could be less efficient than randomizing individuals directly because of the within cluster correlation. The statistics literature has described several closed-form formulas for approximating the required number of clusters for a pre-defined effect size and power. The derivations completely rely on asymptotic approximation. However, in many CRTs only a small number of clusters are actually randomized and asymptotic approximation may not be appropriate. We propose an approach to identify the required sample size for CRTs by approximating the distribution of the test statistic with simulations. The approach is non-parametric and can incorporate any type of outcome or test statistic. It can also take in data from previous studies and adjust for different budgets of enrolling a cluster to the control and the intervention arm. We demonstrate the approach using data from a recent CRT in nursing homes.
|
Authors who are presenting talks have a * after their name.
