A review of the methodological features of systematic reviews in fetal medicine

Abstract

Systematic reviews of fetal medicine can serve as a tool for translation of research findings from a few expert centres to a wider healthcare specialty. The extent to which reviews of fetal medicine research are systematic and unbiased is not known. In this review of systematic reviews in fetal medicine, we have searched without language restrictions, Medline, Embase, DARE (Database of Abstracts of Reviews of Effectiveness), Cochrane Library (from database inception to 2005) and bibliographies of known reviews, and contacted experts to identify potentially relevant citations of literature for reviews of fetal medicine studies. The selected reviews were assessed for information on framing of questions, literature search and methods of review. The search yielded 659 citations of which 84 reviews met the inclusion criteria. Most of the reviews were in the field of fetal pathology (49/84, 59%). A majority of reviews (58/84, 69%) specified the question to be answered but only half (44/84, 52%) addressed a focussed question. Although 57/84 (68%) reviews had a detailed search description, only 32/84 (38%) searched without language restriction. 45/84 (54%) searched in multiple databases and 27/84 (32%) assessed for the risk of missing studies. There was no difference in quality between reviews of fetal pathology, screening for aneuploidy, fetal growth and fetal therapy, except with respect to specifying the question (p<0.03), search without language restriction (p<0.004), assessment of risk of missing studies (p<0.006) and study quality assessment (p<0.002) where reviews of fetal growth performed better than other domains. Our study reflects the paucity of good quality reviews in fetal medicine research. Existing reviews tend to be poor in reporting methodological features. Particularly, not enough attention is paid to assessment of validity of included studies and means to improving reliability of results through appropriate use of meta-analysis. There is a need for conducting further reviews and for rigour when reviewing fetal medicine research.