Bibliometric analysis of ‘overviews of systematic reviews’ from 2000 to 2018

Article type
Authors
Lunny C1, Zheng E2, Ngsee P2, Lin L2, Yang N2, Shinger G2, Sadeghipouya S2, Dosanjh M2, Chen A2, Tasnim S1, Neelakant T3, Stevens A4, Shea B5, Ellis U3, Reid E3, Wright J6
1Cochrane Hypertension Group, University of British Columbia
2Pharmaceutical Sciences, University of British Columbia
3University of British Columbia
4Ottawa Research Insititute
5Ottawa Research Institute
6Cochrane Hyptertension Group, University of British Columbia
Abstract
Background: Overviews of systematic reviews aim to primarily search for, retrieve, and synthesise the results of systematic reviews. For topic areas with a large volume of literature and breadth of scope, overviews serve as an efficient way to synthesise the evidence. Research is needed to understand the scope and trend in overviews published annually and to identify factors that may influence the incidence of publication.
Objectives: The purpose of this study is to evaluate the epidemiology and characteristics of overviews of systematic reviews (overviews) including the number of overviews published per year and the annual proportion of overviews among all published studies, and evaluate trends over time.
Methods: We searched MEDLINE (Ovid), Epistemonikos and the Cochrane library databases from 2000 to December 2018. Inclusion criteria: a) overviews that had a methods section and described a systematic literature search in the body of the paper, b) examined the effects of a health intervention, c) published in any language and d) aimed to primarily synthesise systematic reviews. Articles were extracted independently by multiple reviewers, compared, and then triple checked by the lead investigator. We extracted the medical field of the overview; year of publication; country and continent of the corresponding author; number of authors (1, 2-3, 4-5, 6-10, more than 10); journal of publication; number of included studies; number of patients total in the systematic reviews; the time required to conduct each overview; inclusion of meta-analysis; funding status; and funding type.
To determine the global trend in overview publications between 2000 and 2018 the annual number of overviews published worldwide was calculated and presented in a line graph. Number of overviews published per year were stratified by journals and continents. We also assessed the number of duplicate overviews on similar topics.
Results: From a total of 9917 records retrieved from Ovid MEDLINE, Cochrane Database of Systematic Reviews, and Epistemonikos databases, as well as 237 records retrieved from other sources, 1688 duplicates were removed. Of the remaining 8220 records, 7663 were excluded. In total, 557 overviews were included. Over half of publications were from European authors. In total, 65% of the overviews were published after the start of 2013, and 53% between 2015 and 2018. Only 43% of the overviews reported funding sources. The most common medical field of focus overall was diseases of the musculoskeletal system. The overviews were most frequently published in the Cochrane Database of Systematic Reviews and the Sao Paulo Medical Journal.
Conclusions: Our findings demonstrate a rapid increase in the number of overviews published between 2000 and 2018. Overviews should be registered publicly, and authors should avoid duplication of overviews on similar topics to avoid contributing to research waste.