Up-to-dateness of reviews is often neglected in overviews: a systematic review

Abstract

Background and objective: As systematic reviews may run out of date, it might be necessary to update them. Out-of-date reviews may jeopardize the comparability when used in the context of overviews (review of reviews).

Methods: Seven electronic databases were searched for overviews up to November 2012. We first aimed to analyze whether the authors of overviews additionally searched for primary studies or alternatively explained why they did not. Second, we sought to analyze the actual publication lag (publication date of the overview - publication date of the review) in overviews and to develop recommendations for authors of overviews.

Results: We included 147 overviews. The mean publication lag in overviews was more than 5 years. A median of 36% of the reviews were published more than 6 years ago. Only one in four reviews considered up-to-dateness. The methods for updating reviews were heterogeneous. We found no overview that systematically investigated whether an update was necessary.

Conclusion: The issue of up-to-dateness when conducting overviews seems to be neglected by most authors of overviews. Authors should assess the quality of evidence, based on their included reviews first.

Keywords: Evidence-based medicine; Information science; Meta-analysis; Methods; Research design; Systematic review.