[Skip to Content]
[Skip to Content Landing]
Article
August 12, 1988

Fraudulent Studies and Meta-analysis

Author Affiliations

University of Maryland School of Medicine Baltimore

University of Maryland School of Medicine Baltimore

JAMA. 1988;260(6):791-792. doi:10.1001/jama.1988.03410060061025
Abstract

To the Editor.—  Recently Dr Thacker1 and others2 have discussed meta-analysis as a powerful method for quantitatively analyzing existing research to provide insights not available from individual studies or less rigorous summaries of research findings. Because the quality of a meta-analysis depends on the inclusion of a large sample of welldone studies, a major threat to the validity of meta-analysis is the existence of fraudulent or erroneous published research. While research fraud may be rare, concentration of an investigator's fraudulent articles in one area can present a severe bias in meta-analysis. In recent cases, Slutsky published 60 articles, 12 of which were fraudulent and 48 suspect,3 with six on one topic retracted from one journal4; Darsee published 129 related articles suspected of fraudulence.5Fraudulent articles obviously should be excluded from meta-analyses because they erroneously increase pooled sample size. They may change the results of a

×