Systematic reviews published in higher impact clinical journals were of higher quality

Objectives: To compare the methodological quality of systematic reviews (SRs) published in high- and low-impact factor (IF) Core Clinical Journals. In addition, we aimed to record the implementation of aspects of reporting, including Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flow diagram, reasons for study exclusion, and use of recommendations for interventions such as Grading of Recommendations Assessment, Development and Evaluation (GRADE).

Study design and setting: We searched PubMed for systematic reviews published in Core Clinical Journals between July 1 and December 31, 2012. We evaluated the methodological quality using the Assessment of Multiple Systematic Reviews (AMSTAR) tool.

Results: Over the 6-month period, 327 interventional systematic reviews were identified with a mean AMSTAR score of 63.3% (standard deviation, 17.1%), when converted to a percentage scale. We identified deficiencies in relation to a number of quality criteria including delineation of excluded studies and assessment of publication bias. We found that SRs published in higher impact journals were undertaken more rigorously with higher percentage AMSTAR scores (per IF unit: β = 0.68%; 95% confidence interval: 0.32, 1.04; P < 0.001), a discrepancy likely to be particularly relevant when differences in IF are large.

Conclusion: Methodological quality of SRs appears to be better in higher impact journals. The overall quality of SRs published in many Core Clinical Journals remains suboptimal.