J Evid Based Med: Most Cochrane reviews have not been updated for more than 5 years

29 August, 2021

Below are the citation and abstract of a new paper in the Journal of Evidence Based Medicine, and a comment from me. Unfortunately the full text is restricted access.

CITATION: J Evid Based Med. 2021 Aug 24.

Most Cochrane reviews have not been updated for more than 5 years.

Bodil Dalsgaard Hoffmeyer, Mikkel Zola Andersen, Siv Fonnes, Jacob Rosenberg

DOI: https://doi.org/10.1111/jebm.12447

URL: http://pubmed.ncbi.nlm.nih.gov/34427395

ABSTRACT

Aim: Cochrane reviews are internationally recognized for their high quality, but to reduce the risk of transmitting misleading information, they must be kept up to date. The aim of this study was to quantify the number of reviews in the Cochrane Database of Systematic Reviews (CDSR) that have not been updated for more than five years and to characterize them.

Method: This study was reported closely adapted to the STROBE guidelines. Information about all reviews in the CDSR was extracted in August 2020. Based on a previous study, we defined inactive reviews as reviews with no update in "What's New" or no publication of a new version of the review within the last 5.5 years. The inactive reviews were quantified and characterized and results were visualized through tables and charts.

Results: The study included 7931 reviews from the CDSR. The median age of all reviews in the CDSR was 5.3 years. Fifty-five percent were published for the first time between 1996 and February 2015 and 88% of these had been inactive for 5.5 years or more. Among these, 89% were first publication of the review that had never been updated afterward.

Conclusion: More than half of the Cochrane reviews in CDSR were first published before 2015 and only 12% of these were still active. In order to retain their validity, it would be preferable if Cochrane reviews were kept up to date by the authors either by an update in "What's New" or by publishing a new version of the review.

COMMENT (Neil PW): The authors raise an important point, but the abstract does not explain why they chose 5 years as their benchmark. The necessity to update a systematic review presumably depends on many factors, such as the topic of the review, and the apparent emergence (or otherwise) of new evidence. On HIFA we have previously discussed 'living reviews' whereby authors take responsibility fo ongoing monitoring of the literature for new research, which is then incorporated into review updates.

Neil Pakenham-Walsh, HIFA Coordinator, neil@hifa.org www.hifa.org