A Quality-Effects Model for Meta-Analysis

Doi, Suhail A. R.^*†; Thalib, Lukman^‡

From the *Division of Endocrinology, Mubarak Al-Kabeer Teaching Hospital; and Departments of †Medicine and ‡Community Medicine (Biostatistics), Kuwait University, Kuwait.

Submitted 17 June 2007; accepted 14 August 2007.

Supplemental material for this article is available with the online version of the journal atwww.epidem.com; click on “Article Plus.”

Correspondence: Suhail A.R. Doi, Department of Medicine, Kuwait University, P.O. Box 24923, 13110 Kuwait. E-mail:[email protected].

Epidemiology19(1):p 94-100, January 2008. |DOI:10.1097/EDE.0b013e31815c24e7

Abstract

We introduce a quality-effects approach that combines evidence from a series of trials comparing 2 interventions. This approach incorporates the heterogeneity of effects in the analysis of the overall interventional efficacy. However, unlike the random-effects model based on observed between-trial heterogeneity, we suggest adjustment based on measured methodological heterogeneity between studies. We propose a simple noniterative procedure for computing the combined effect size under this model and suggest that this could represent a more convincing alternative to the random effects model.