Biometrika Advance Access published online on February 28, 2007
Biometrika, doi:10.1093/biomet/asm019
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
Estimating a treatment effect with repeated measurements accounting for varying effectiveness duration
Program in Biostatistics, Public Health Sciences Division, Fred Hutchinson Cancer Research Center, Seattle, Washington 98109, U.S.A.
Divisions of Biostatistics & Epidemiology, School of Public Health, University of California, Berkeley, California 94720, U.S.A.
Department of Biostatistics & Epidemiology, University of California, San Francisco, California 94143, U.S.A.
Department of Pathology, School of Medicine, Johns Hopkins University, Baltimore, Maryland 21205, U.S.A.
yqchen{at}scharp.org
jingrong{at}stat.berkeley.edu
scheng{at}biostat.ucsf.edu
bjackso{at}jhmi.edu
Received for publication 1 November 2005.
Revision received 1 June 2006.
 |
Abstract |
|---|
To assess treatment efficacy in clinical trials, certain clinical outcomes are repeatedly measured over time for the same subject. The difference in their means may characterize a treatment effect. Since treatment effectiveness lag and saturation times may exist, erosion of treatment effect often occurs during the observation period. Instead of using models based on ad hoc parametric or purely nonparametric time-varying coefficients, we model the treatment effectiveness durations, which are the time intervals between the lag and saturation times. Then we use some mean response models to include such treatment effectiveness durations. Our methodology is demonstrated by simulations and analysis of a landmark HIV/AIDS clinical trial of short-course nevirapine against mother-to-child HIV vertical transmission during labour and delivery.
Key Words: effect erosion • HIV/AIDS clinical trial • longitudinal study • mean response model • time-varying coefficient