Far East Journal of Theoretical Statistics
Volume 35, Issue 2, Pages 129 - 135
(May 2011)
|
|
STRATIFIED ANALYSIS WITH AN APPLICATION IN CLINICAL TRIALS
Junfeng Shang
|
Abstract: Clinical trials are conducted to allow safety and efficacy data to be collected for health interventions (e.g., drugs, diagnostics, devices, therapy protocols). One of the major interests for clinical trials is to study the association between the treatments or exposure groups and the response. This association may be biased when there is an imbalance in the distribution of an important covariate between the two independent groups. For instance, multicenter trials may vary and the location could dramatically affect the association between the treatments or groups and the response. It is therefore of importance to conduct a stratified analysis to adjust for the association between the treatments and the response with the existence of the stratifying covariate. This paper focuses on separate tables to conduct the association tests of stratified analysis over the strata by a covariate using the association measures such as odds ratio, relative risk, and risk difference. |
Keywords and phrases: Received November 8, 2010 |
|
Number of Downloads: 279 | Number of Views: 814 |
|