The current situation in Cochrane Reviews using GRADE: a cross-sectional study

Article type
Authors
Tian R1, Yao L1, Zhao C1, Zhang X1, Bian Z1
1Chinese Medicine Faculty of Hong Kong Baptist University, Kowloon Tong, Hong Kong
Abstract
Background:
The Cochrane Handbook requires the use of GRADE to assess the evidence in Cochrane Reviews for better transparency and credibility of results. However, no study has assessed the current situation in Cochrane Reviews in terms of the use and reporting of GRADE.

Objectives:
To investigate the current situation regarding the use and reporting of GRADE in Cochrane Reviews.

Methods:
We randomly sampled 5% of all Cochrane Reviews using SPSS 19.0. We captured information about the publication year, usage of GRADE and reporting of the results etc. in a pre-designed table.

Results:
We included 306 Cochrane Reviews, among which 61 (20%) reported grading the evidence using the GRADE system. About 36% of the included reviews were based on GRADE, of which five (1.6%) had no report of the evidence grading results, one (0.3%) reported evidence profiles (EP) and 55 (18%) presented 'Summary of findings' (SOF) tables. Factors that influenced the grading of evidence were reported through EP or SOF tables in 49 reviews (16%). Risk of bias (45, 92%) was the top bias influencing the GRADE results, followed by imprecision (35, 71%) and indirectness (7, 14%).

Conclusions:
Cochrane Reviews increasingly use GRADE, but the proportion among the total number of reviews is low and the GRADE results are inconsistently reported. We therefore suggest that the developers of Cochrane Reviews be aware of the need to adopt GRADE and report this information, which is valuable for Cochrane users in interpreting results.

Patient or healthcare consumer involvement:
None.