Session number: 1159
Workshop title: Exploring Study Heterogeneity in Quantitative Systematic Reviews
Author(s):
JE Cornell Veteran's Evidence-Based Research, Dissemination, and Implementation Center
Objective(s): It is important in quantitative systematic reviews to move beyond simple statistical adjustments for study heterogeneity and identify clinical, study design and patient factors that contribute to heterogeneity in outcomes observed in published research on health care interventions. The objectives of this workshop are 1. To provide an introduction to the problem of study heterogeneity in quantitative systematic reviews and 2. To illustrate graphical and analytic techniques used to explore sources of study heterogeneity in systematic reviews on medical interventions. We begin with a presentation of the clinical and design characteristics that contribute to study heterogeneity in systematic reviews, and we examine some graphical and statistical methods for detecting heterogeneity within a set of randomized controlled trials. After briefly considering random-effect estimators, we discuss issues related to the selection of stratification factors for subgroup analyses and the use of meta-regression to identify study-level characteristics that may contribute to the observed heterogeneity among the studies. Emphasis will be placed on the use of prior knowledge, experience, and empirical evidence to identify potential sources of heterogeneity and to generate a priori working hypotheses for these exploratory investigations.
Activities: Didactic presentation, with participant discussion, of the clinical and statistical issues associated with study heterogeneity in systematic reviews.
Target audience: New or experienced researchers who are conducting, or are interested in conducting, a quantitative systematic review.
Audience familiarity: This workshop is designed for researchers who have a basic knowledge of elementary meta-analysis procedures and elementary statistics.