One-Third of Systematic Reviews in Rehabilitation Applied the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) System to Evaluate Certainty of Evidence: A Meta-Research Study

Objective: To determine how many systematic reviews (SRs) of the literature in rehabilitation assess the certainty of evidence (CoE) and how many apply the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system to do this. Data Sources: For this meta-research study, we searched PubMed and Cochrane Database of Systematic Reviews databases for SRs on rehabilitation published in 2020. Study Selection and Data Extraction: Two reviewers independently selected the SRs and extracted the data. Reporting characteristics and appropriate use of the GRADE system were assessed. Data Synthesis: The search retrieved 827 records: 29% (239/827) SRs evaluated CoE, 68% (163/239) of which applied the GRADE system. GRADE was used by SRs of randomized controlled trials (RCTs, 88%; 144/163), non-randomized intervention studies (NRIS, 2%; 3/163), and both RCT and NRIS (10%; 16/163). In the latter case, a separate GRADE assessment according to the study design was not provided in 75% (12/ 16). The reasons for GRADE judgment were reported in 82% (134/163) of SRs. Conclusions: One-third of SRs in rehabilitation assessed CoE with the GRADE system. GRADE assessment was presented transparently by most SRs. Journal editors and funders should encourage the uptake of the GRADE system when considering SRs in rehabilitation for publication. The authors should pre-define GRADE assessment in a registered and/or published protocol. Archives of Physical Medicine and Rehabilitation 2023;104:410-7 & COPY; 2022 by the American Congress of Rehabilitation Medicine.

Automated summary

A meta-research study found that about one-third of systematic reviews in the field of rehabilitation assess the certainty of evidence using the GRADE system. Most reviews presented the GRADE assessment transparently, but some had shortcomings in reporting. Journal editors and funders should encourage the use of the GRADE system and require authors to pre-define the GRADE assessment in registered and/or published protocols.