This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 422
Type: Contributed
Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
Sponsor: SSC
Abstract - #308055
Title: Imputation Strategies for Missing Binary Outcomes in Cluster Randomized Trials
Author(s): Jinhui Ma*+ and Noori Akhtar-Danesh and Lisa Dolovich and Lehana Thabane
Companies: McMaster University and McMaster University and McMaster University and McMaster University
Address: 50 main street east, room 324, Hamilton, ON, L8N 1E9, Canada
Keywords: missing data ; multiple imputation ; cluster randomized trial ; Markov chain Monte Carlo ; propensity score method ; predictive model
Abstract:

Missing data is a common problem in most trials including cluster-randomized trials (CRTs). Multiple imputation (MI) is considered as the gold standard technique for imputing missing data under the assumption of missing at random. Although most of standard statistical programs have MI procedures, they are applicable only for independent data and, therefore, may not be appropriate for clustered data. We investigated three MI strategies - predictive model, propensity score method, and Markov Chain Monte Carlo method - that take into account the correlation of outcomes within clusters for imputing missing data from CRTs. Using a simulation study based on the Community Hypertension Assessment Trial, we compared the performance of these strategies under different percentages of missing binary outcomes on the basis of kappa statistics and the robustness of the treatment effects.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2010 program




2010 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.