Local capacity for quality evidence syntheses: methodological and reporting quality of systematic reviews from Dalhousie University-affiliated reviewers

Abstract

Background: In order to best support the conduct of systematic reviews (SRs) in our institutional setting, we set out to examine the SRs produced by Dalhousie University affiliates. Initiation of an institutional subscription to Covidence increases contact between the health sciences library and reviewers. This outreach opportunity is a chance to improve SR training, so assessing the degree of adherence to methodological and reporting guidelines will allow development of targeted training and support services for current and future review authors.

Objectives: To assess the methodological and reporting quality of SRs and to describe the characteristics of the teams involved in SR production at one university in Atlantic Canada.

Methods: PubMed, Embase, Cochrane Library, Joanna Briggs Institute, and Scopus were searched for SRs with at least one author affiliated with Dalhousie University. Results were deduplicated and screened using Covidence software. We will assess included reviews for methodological quality using the validated AMSTAR 2 tool and for reporting quality using the PRISMA checklist. We will categorize reviews by subject area and examine review characteristics, including team composition.

Results: A preliminary study of Dalhousie-affiliated SRs published in the past 10 years yielded 163 studies; 35% of SRs (n = 57) included contributions from an information specialist. The current study is ongoing and results will be presented including the percentage of reviews that meet each element of the AMSTAR 2 and PRISMA checklists, composition of review teams, and the distribution of reviews in health disciplines and specialties.

Conclusions: Gaps identified in the methodological and reporting quality of SRs authored by members of the Dalhousie University community will highlight areas where improved outreach and training can make an impact on overall quality of reviews.

Patient or healthcare consumer involvement: Review topics selected in a particular geographic area tend to reflect the health needs of the region and are more likely to be used in immediate care and policy decision making. Although this methodological review did not directly involve patients, ensuring that the syntheses produced by our institution follow accepted standards will impact patient care by improving the quality and reliability of the evidence.