What type of study is this?

This is a Quantitative Study study.

September 16, 2025

Assessment of quality of reporting meta-analysis and systematic review of pharmacotherapy of COVID-19 using preferred reporting items for systematic reviews and meta-analyses 2020 statement

Key Points

On average, the PRISMA score for meta-analyses was 91.20 ± 7.14, revealing good reporting quality.
Statistical analysis showed no significant difference between scoring methods, indicating consistent adherence across methods.
Twelve out of 14 meta-analyses applied the Cochrane Handbook, highlighting its importance in evaluating quality and bias.
The study emphasizes the need for improved compliance with protocol and registration items in systematic reviews.

Abstract

This research aimed to evaluate how well systematic reviews and meta-analyses (MAs) on COVID-19 pharmacotherapy adhered to the Preferred Reporting Items for Systematic Reviews and MAs (PRISMA) 2020 reporting guidelines. Fourteen studies published between January 2022 and June 25, 2023, in PubMed were analysed. Reporting quality in both abstracts and full texts was analysed using the PRISMA 2020 checklist. Two types of scores were performed: one evaluating the 26-item score with merged subgroups, and another evaluating the 41-item score with unmerged subgroups. The characteristics of these MAs revealed that the number of studies included ranged from 6 to 34, while the journal impact factors varied from 1.8 to 76.2. Most MAs (12/14) applied the Cochrane Handbook to evaluate quality and risk of bias. On average, each meta-analysis had 8.43 authors and 8010.93 participants. The mean score of PRISMA items was 91.20 ± 7.14 for the 26 score and 87.80 ± 8.81 for the 41 score in 14 MAs. Regarding the PRISMA items percentage score, 17 PRISMA items were 100% reported. However, certain items, such as registration and protocol (50%) and certainty assessment (57.14%) were less reported. For abstracts, three sub-items were 100% reported, while limitations of evidence (14.28%) and funding (21.42%) were the least reported. There was no statistically significant difference between the mean scores from the two scoring methods ( P = 0.27. A positive correlation was observed between the PRISMA score and the number of authors ( P < 0.05). Linear regression analysis was performed for 26 items ( P = 0.051) and for 41 items ( P = 0.042), and linearity was established. No significant differences were found between MAs with higher PRISMA scores (≥24/26 and ≥36/41) and lower PRISMA scores (≤23/26 and ≤35/41), MAs with and without PRISMA endorsement, and among those using different bias assessment tools for both models. There was no significant correlation between the PRISMA score and the guidance tool for quality and risk of bias assessment ( P = 0.45 < 0.05) for both scores. A Chi-square test also indicated no significant correlation between the use of risk assessment tools and PRISMA scores ( P = 0.45). The compliance is good for the 26-item score analysis as compared to 41 items. The protocol and registration item have suboptimal compliance. The study supports mandating PRISMA adherence during article submissions. The number of authors significantly influences reporting quality, highlighting the need for universal adherence to PRISMA guidelines to meet essential reporting standards. Greater emphasis should be placed on the risk of bias assessment and outcomes as critical components of reporting quality.

问 AI

Bookmark