Quality review of therapeutic meta-analyses in the glaucoma literature
Mon statut pour la session
Authors: Douglas S. M. Iaboni, ErditCelo, Jayme Vianna. Dalhousie University.
Author Disclosure Block:D.S.M. Iaboni: None. E. Celo: None. J. Vianna: None.
Abstract Body:
Purpose: Systematic reviews with meta-analyses are an excellent resource in ophthalmology and
medicine in general. Unfortunately, the quality of meta-analyses can vary greatly; therefore, it is important to routinely assess the available evidence. The goal of this study was to analyzet he quality of meta-analyses in therapeutic studies in glaucoma.
Study Design: Literature review and quality assessment.
Methods: After searching the published reports from MEDLINE, EMBASE, and the Cochrane library for meta-analyses in glaucoma between 2008 and 2020, articles were first screened by two authors for relevance then full-text review was performed. The AMSTAR-2 checklist was applied to the meta- analyses that met our inclusion criteria in order to determine article quality. Quality was both described qualitatively and by using a quantitative scale. This quality measure was then compared to multiple other variables including year of publication, scientific journal rating (SJR), and journal of publication.
Results: Electronic searches imported 2506 studies for screening. Of these, 2102 were deemed to not meet inclusion criteria. Afterwards, 394 articles were subjected to full-text review where another 282 articles were excluded. The remaining 114 articles were included in this quality review. In total,7, 7, 39, and 61 studies were found to be of high, moderate, low, and critically low quality, respectively. There did appear to be a positive relationship between SJR and study quality, as all articles with a score of moderate quality or higher were in journals with an SJR greater than 1.00. Articles in the Cochrane Database had the highest chance of being high quality. There was no relationship between year of publication and study quality. The most common deficits of the reviewed meta-analyses were not providing a list of excluded studies, not including sources of funding for included studies, and not including an explicit statement of having a pre-registered protocol.
Conclusions: The majority of meta-analyses in therapeutic glaucoma are of poor quality. As such, practitioners should exercise caution when relying on study results. The AMSTAR-2 tool is a useful tool in evaluating quality and may be used to assist clinicians in assessing evidence. Authors could considerably improve study quality by addressing critical domains on the AMSTAR-2tool in their meta- analyses.