Background: In order to provide a balanced viewpoint, systematic reviews should be considering both the beneficial and harmful effects of an intervention. However, reviewers may be hampered by the absence of information about effective methods for searching for studies that contain data on adverse effects.

Objectives: We aimed to review systematic reviews of adverse effects published from 1994 to 2005 in order to determine the methods used to retrieve included studies. We looked specifically at the search strategies used and the data sources examined.

Methods: We searched the Cochrane Database of Systematic Reviews (CDSR) and the Database of Abstracts of Reviews of Effects (DARE) for systematic reviews in which the primary outcomes of interest were adverse effect(s). The search methods reported in these reviews were evaluated independently by two information specialists. Data were extracted regarding databases and other sources used, as well as search terms and limits applied.

Results: We found 277 reviews that met our inclusion criteria. Most reviewers limited themselves to searching MEDLINE and reference checking. One major weakness of the reviews was that the search strategies were not adequately reported. Of those reviews that did report their search strategies, few performed sensitive search strategies as recommended for systematic reviews and only 15 reviews gave enough detail for the search strategies to be reproducible. Only 30 of the 275 reviews that reported on the sources searched had included a source specific to adverse effects. One hundred and fifty-five of the 214 reviews that reported the search terms used had included specific or generic adverse effects terms.

Conclusions: Search strategies within systematic reviews of adverse effects are often poorly reported and of limited scope. Readers may be left wondering about the reproducibility of the search and the comprehensiveness of the review and thus the reliability of the findings.