|
Session Title: Bayesian Mixture Modeling Versus Traditional Meta-analysis: Examining the Treatment Advantage Research Using Three Meta-analytic Approaches.
|
|
Multipaper Session 326 to be held in Lone Star E on Thursday, Nov 11, 3:35 PM to 4:20 PM
|
|
Sponsored by the Quantitative Methods: Theory and Design TIG
|
| Chair(s): |
| Karen Larwin,
University of Akron, Wayne, drklarwin@yahoo.com
|
|
Bayesian Mixture Modeling: Also Known as Bayesian Meta-analysis
|
| Presenter(s):
|
| James Michael Menke, University of Arizona, menke@email.arizona.edu
|
| Abstract:
Comparative effectiveness research (CER) tackles the decisions as to whether a treatment or intervention should be used, by comparing its effectiveness and cost to some standard. CER is an initiative brought forth by the Obama administration as a way of better informing health choices and improving health care system efficiencies by increasing transparency. CER methods have been around for years in Australia and the UK. Their arrival in the US is timely with respect to the current convergence of a health care, health insurance, and economical crisis. Bayesian Mixture Modeling (BMM) offers a way to conduct CER. BMM is a method of data synthesis that allows studies from different specialties and programs to be compared as if they are all arms of a single study, with certain convenient advantages. There are four general steps in CER under Bayesian and decision analysis that will be presented and discussed: 1) convert effect sizes to probabilities, 2) simulate or model a direct comparison between at least two treatments to estimate relative effect sizes, 3) estimate stability of findings via sensitivity analyses, and 4) interpret findings.
|
|