Study found increasing use of core outcome sets in Cochrane systematic reviews and identified facilitators and barriers

Ian J Saldanha,Karen L Hughes,Susanna Dodd,Toby Lasserson,Jamie J Kirkham,Yuhui Wu,Samuel W Lucas,Paula R Williamson,Ian J. Saldanha,Karen L. Hughes,Jamie J. Kirkham,Samuel W. Lucas,Paula R. Williamson
DOI: https://doi.org/10.1016/j.jclinepi.2024.111277
IF: 7.407
2024-02-01
Journal of Clinical Epidemiology
Abstract:OBJECTIVES: In 2019, only 7% of Cochrane systematic reviews (SRs) cited a core outcome set (COS) in relation to choosing outcomes, even though a relevant COS existed but was not mentioned (or cited) for a further 29% of SRs. Our objectives for the current work were to (1) examine the extent to which authors are currently considering COS to inform outcome choice in Cochrane protocols and completed SRs, and (2) understand author facilitators and barriers to using COS.STUDY DESIGN AND SETTING: We examined all completed Cochrane SRs published in the last 3 months of 2022 and all Cochrane protocols published in 2022 for the extent to which they: (a) cited a COS, (b) searched for COS, (c) used outcomes from existing COS, and (d) reported outcome inconsistency among included studies and/or noted the need for COS. One investigator extracted information; a second extractor verified all information, discussing discrepancies to achieve consensus. We then conducted an online survey of authors of the included SRs to assess awareness of COS and identify facilitators and barriers to using COS to inform outcome choice.RESULTS: Objective 1: We included 294 SRs of interventions (84 completed SRs and 210 published SR protocols), of which 13% cited specific COS and 5% did not cite but mentioned searching for COS. A median of 83% of core outcomes from cited COS (interquartile range [IQR] 57% to 100%) were included in the corresponding SR. We identified a relevant COS for 39% of SRs that did not cite a COS. A median of 50% of core outcomes from non-cited COS (IQR 35% to 72%) were included in the corresponding SR. Objective 2: Authors of 236 (80%) of the 294 eligible SRs completed our survey. Seventy-seven percent of authors noted being aware of COS before the survey. Fifty-five percent of authors who did not cite COS but were aware of them reported searching for a COS. The most reported facilitators of using COS were author awareness of the existence of COS (59%), author positive perceptions of COS (52%), and recommendation in the Cochrane Handbook regarding COS use (48%). The most reported barriers related to matching of the scope of the COS and the SR: the COS target population was too narrow/broad relative to the SR population (29%) or the COS target intervention was too narrow/broad relative to the SR intervention (21%). Most authors (87%) mentioned that they would consider incorporating missing core outcomes in the SR/update.CONCLUSION: Since 2019, there is increasing consideration and awareness of COS when choosing outcomes for Cochrane SRs of interventions, but uptake remains low and can be improved further. Use of COS in SRs is important to improve outcome standardization, reduce research waste, and improve evidence syntheses of the relevant effects of interventions across health research.
public, environmental & occupational health,health care sciences & services
What problem does this paper attempt to address?