Methodological issues from peer reviewed Chinese systematic reviews

Tags: Poster
Li J

Background: Doctors in China are more and more aware of the importance of systematic review (SR) and involved in the conduction of SR. Editors in many Chinese Journals accept and publish SR.

Objectives: To summarize the main methodological issues in Chinese systematic reviews, based on the peer review experience of the author.

Methods: To check the manuscripts of systematic review sent to the author for peer review, according to the QUOROM STATEMENT.

Results: The author received 11 Chinese systematic reviews during past year. Among, ten are related to the SR of interventions, one is related to causation. The main methodological issues included: (1) Rationale of some SRs was not clearly stated; (2) Searching: Information sources were not adequately searched, especially for SR related to Traditional Chinese Medicine. Full text of studies published in English journals was not adequately identified; (3) Selection: The selection criteria of some SRs were not defined based on population, intervention, principal outcomes, and study design; (4) Validity assessment: Both checklists and scales were used for quality assessment, but the usefulness of the critical appraisal results was not stated, especially when checklists were used; (5) Data synthesis: Every SR used meta-analysis for data combination. Once the heterogeneity existed, the reviewers always selected random effect model to combine the results, without carefully exploring the possible causes and considering whether the meta-analysis was appropriate or not; (6) Forest plot: RevMan was always used to produce meta-analysis results, namely forest plot. Unfortunately, some reviewers ignored that the RevMan default settings were focused on unfavorable outcomes. When the benefit outcomes were used, the labels on the horizontal axis should be changed in order to ensure that the results displayed were consistent with what was described in the text.

Conclusions: Methodological problems are common in Chinese systematic reviews. As more and more doctors, postgraduates and undergraduates are involving in SR conduction, the quality of the SR should be guaranteed via systematically methodological training including RevMan. Peer reviewers and editors of journals should use the QUOROM STATEMENT to check the quality of SR.