Cochrane's COVID-19 Living Systematic Reviews: A Mixed-Methods Study of Their Conduct, Reporting and Currency.

BACKGROUND

Living systematic reviews (LSRs) should provide up-to-date evidence for priority questions where the evidence may be uncertain and fast-moving. LSRs featured prominently during COVID-19 and formed part of Cochrane's response to the pandemic. We conducted a mixed-methods study to describe the characteristics of Cochrane's COVID-19 living reviews, determine the currency of the included evidence, and evaluate authors' experiences in conducting and publishing these reviews.

METHODS

We identified living reviews of COVID-19 from the and extracted data on the number of versions published and publication timelines. We assessed the currency of evidence by comparing studies included in the reviews against a comprehensive list of studies maintained for the Australian living guidelines for COVID-19. The qualitative component involved semi-structured interviews with review authors to identify the barriers and enablers to conducting, reporting and publishing living reviews.

FINDINGS

Cochrane published 25 COVID-19 living systematic reviews. Half of these reviews had not been updated when assessed in June 2023 and only four had been updated more than once. A total of 118 studies were included in the living reviews. We estimated that an additional 119 studies were available and potentially relevant for inclusion. Interviews with six authors indicated that publication timelines were reduced by editorial delays, loss of funding, waning commitment, and the burden of screening search results. An inability to communicate the living status of reviews in the Cochrane Library was a common frustration for many authors. Although authors felt the conclusions of their reviews were still current, only one living review communicated its updated status and made new evidence accessible after the review was published.

CONCLUSIONS

Maintaining and communicating the currency of Cochrane's COVID-19 living systematic reviews was not feasible for many author teams because of author-side, editorial and platform barriers.