Abstract
Background
Systematic reviews (SRs) could offer the best evidence supporting interventions, but methodological flaws limit their trustworthiness in decision-making. This cross-sectional study appraised the methodological quality of SRs on atopic dermatitis (AD) treatments.
Methods
We searched MEDLINE, EMBASE, PsycINFO, and Cochrane Database for SRs on AD treatments published in 2019–2022. We extracted SRs’ bibliographical data and appraised SRs’ methodological quality with AMSTAR (A MeaSurement Tool to Assess systematic Reviews) 2. We explored associations between methodological quality and bibliographical characteristics.
Results
Among the 52 appraised SRs, only one (1.9%) had high methodological quality, while 45 (86.5%) critically low. For critical domains, only five (9.6%) employed comprehensive search strategy, seven (13.5%) provided list of excluded studies, 17 (32.7%) considered risk of bias in primary studies, 21 (40.4%) contained registered protocol, and 24 (46.2%) investigated publication bias. Cochrane reviews, SR updates, SRs with European corresponding authors, and SRs funded by European institutions had better overall quality. Impact factor and author number positively associated with overall quality.
Conclusions
Methodological quality of SRs on AD treatments is unsatisfactory. Future reviewers should improve the above critical methodological aspects. Resources should be devolved into upscaling evidence synthesis infrastructure and improving critical appraisal skills of evidence users.
Author contributions
LH: Collection and assembly of data; Data analysis and interpretation; Manuscript writing. YMKC: Collection and assembly of data; Data analysis and interpretation; Manuscript writing. CCCC: Data analysis and interpretation; Manuscript writing. IXYW: Conception and design. CM: Conception and design. VCHC: Conception and design; Manuscript writing; Supervision. All authors read and approved the final manuscript.
Disclosure statement
No potential conflict of interest was reported by the author(s).