ESTIMATING A RELATIVE RISK ACROSS SPARSE CASE-CONTROL AND FOLLOW-UP STUDIES: A METHOD FOR META-ANALYSIS

Publisher Website
Google Scholar
Abstract
Meta-analysis is the quantitative technique of combining results from different studies. There is a variety of procedures available for combining effect measures across epidemiologic studies. None of these methods provides an overall effect estimate when the data are sparse within studies and come from different study designs. In this paper we discuss the statistical relations between case-control studies and two types of follow-up studies. We use these relations to develop an exact methodology for combining results across study designs. We also use these relations to derive Mantel-Haenszel type formulae for summarizing results across studies. We illustrate these techniques with data pertaining to breast implants and connective tissue disease.

This publication has 0 references indexed in Scilit: