Article type
Abstract
Background
High-quality systematic reviews and meta-analysis (SRMAs) can strengthen the evidence base for prevention and health promotion. A Measurement Tool to Assess Systematic Reviews 2 (AMSTAR 2) is a tool for methodological assessment of SRMAs. However, the current AMSTAR 2 might have some challenges, for example, a SRMA without reporting of a protocol might be rated as low or very low quality. We revised the original AMSTAR 2 and rating criteria, and applied this modification to some SRMAs.
Objective
To assess and compare two approaches to appraising SRMAs using AMSTAR 2.
Methods
Data sources for this study were SRMAs included in our umbrella review on dietary sodium and patient important outcomes in adults, which searched in MEDLINE, EMBASE, Cochrane Database of Systematic Reviews and Web of Science until December 2023. We assessed methodological quality using the original and modified AMSTAR 2, respectively. Our modification added two items to address concerns we perceive as critical in the conduct of SRMAs, including absolute effect estimates for dichotomous outcome, and certainty of evidence. We also rated the overall quality of each SRMA based on modification as high, moderate, low or critically low. All data were summarized using descriptive statistic.
Results
We included 55 SRMAs. Nineteen SRMAs assessed the certainty of evidence, and four of 26 SRMAs reported the absolute effect estimates. Item 2 (protocol) and item10 (funding of included studies in SRMA) were not fulfilled in over half of SRMAs. According to original AMSTAR 2, the methodological quality was mainly low (n=25) and critically low (n=19), followed by moderate (n=9) and high (n=2). According to modified AMSTAR 2, nineteen SRMAs were rated as moderate quality, followed by low quality (n=15), very low quality (n=14), and high quality (n=7). Thirty-two SRMAs had the same quality rating based on the above two approaches. The remaining SRMAs had higher quality rating by modified AMSTAR 2 (improved from low to moderate in 13 SRMAs).
Conclusions
The absolute effect estimates were rarely used for dichotomous outcome in SRMAs. The quality of SRMAs by modified AMSTAR 2 is better than that of the original version.
High-quality systematic reviews and meta-analysis (SRMAs) can strengthen the evidence base for prevention and health promotion. A Measurement Tool to Assess Systematic Reviews 2 (AMSTAR 2) is a tool for methodological assessment of SRMAs. However, the current AMSTAR 2 might have some challenges, for example, a SRMA without reporting of a protocol might be rated as low or very low quality. We revised the original AMSTAR 2 and rating criteria, and applied this modification to some SRMAs.
Objective
To assess and compare two approaches to appraising SRMAs using AMSTAR 2.
Methods
Data sources for this study were SRMAs included in our umbrella review on dietary sodium and patient important outcomes in adults, which searched in MEDLINE, EMBASE, Cochrane Database of Systematic Reviews and Web of Science until December 2023. We assessed methodological quality using the original and modified AMSTAR 2, respectively. Our modification added two items to address concerns we perceive as critical in the conduct of SRMAs, including absolute effect estimates for dichotomous outcome, and certainty of evidence. We also rated the overall quality of each SRMA based on modification as high, moderate, low or critically low. All data were summarized using descriptive statistic.
Results
We included 55 SRMAs. Nineteen SRMAs assessed the certainty of evidence, and four of 26 SRMAs reported the absolute effect estimates. Item 2 (protocol) and item10 (funding of included studies in SRMA) were not fulfilled in over half of SRMAs. According to original AMSTAR 2, the methodological quality was mainly low (n=25) and critically low (n=19), followed by moderate (n=9) and high (n=2). According to modified AMSTAR 2, nineteen SRMAs were rated as moderate quality, followed by low quality (n=15), very low quality (n=14), and high quality (n=7). Thirty-two SRMAs had the same quality rating based on the above two approaches. The remaining SRMAs had higher quality rating by modified AMSTAR 2 (improved from low to moderate in 13 SRMAs).
Conclusions
The absolute effect estimates were rarely used for dichotomous outcome in SRMAs. The quality of SRMAs by modified AMSTAR 2 is better than that of the original version.