Abstract
BACKGROUND: Systematic reviews, such as those prepared by Cochrane, are the strongest evidence on which to base recommendations in clinical guidelines. Network meta-analysis (NMA) can be used to combine the results of studies to compare multiple treatments, which is advantageous over pair-wise meta-analysis (PW-MA) that compares two treatments. We aimed to summarise which, when, where, who, and why Cochrane Reviews that applied NMA were cited in guidelines; and to compare the citation of NMA reviews in guidelines with PW-MA reviews. METHODS AND FINDINGS: We carried out a systematic review of Cochrane reviews that applied NMA and we summarised their citation in guidelines. The Cochrane Database of Systematic Reviews was searched (15th January 2024). Additionally, a cohort of Cochrane reviews that applied PW-MA was matched to the NMA reviews. Two authors assessed eligibility and extracted data. We summarised review and guideline characteristics, and the use of the review in guidelines. RESULTS: Of the 60 included NMA reviews, 26 reviews (43%) were cited in 89 guidelines (1-13 per review). 15 NMA reviews (58%) were first cited within two years of publication, with the remaining 11 reviews (42%) cited 2-6 years later. 52 guideline developers authored the guidelines. The number of citations was higher for NMA than PW-MA reviews (rate ratio 1.53 (1.08 to 2.19), p = 0.02). The number of times reviews were commissioned or cited alongside a recommendation was also higher for NMA than PW-MA reviews (rate ratio 4.40 (1.80 to 13.14), p = 0.003). NMA reviews were more likely to be cited in the text surrounding a recommendation or used for NICE guideline development (1.94 (1.08 to 3.63), p = 0.03). CONCLUSIONS: Cochrane NMA reviews appear to have more impact than PW-MA reviews, but many are not cited in guidelines. Further work is needed to explore the barriers to use of NMAs and promote their use.