RICHARD D. GELBER, Ph.D.; ARON GOLDHIRSCH, M.D.
To the editor: The use of meta-analysis to formally combine results from several clinical trials is growing rapidly. The article by L'Abbé and associates (1) reviews important design and statistical issues related to the process of meta-analysis. Unfortunately, the authors do not address issues regarding the interpretation of results; aspects that are medically relevant for patient care decision making are neglected.
Three points should be made. First, as we recently described for cancer medicine (2), the result of a meta-analysis is based on an "arithmetic construct." The null hypothesis may be tested despite this construct because of the assumption of
GELBER RD, GOLDHIRSCH A. Meta-Analysis in Clinical Research. Ann Intern Med. ;108:158–159. doi: 10.7326/0003-4819-108-1-158_2
Download citation file:
Published: Ann Intern Med. 1988;108(1):158-159.
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use