Assessing imprecision in Cochrane systematic reviews: a comparison of GRADE and Trial Sequential Analysis

Ref ID 505
First Author G. Castellini
Journal SYSTEMATIC REVIEWS
Year Of Publishing 2018
URL https://air.unimi.it/retrieve/handle/2434/595908/1086420/Castellini_Assessing imprecision.pdf
Keywords • Cochrane
• Imprecision
• General medical
• Certainty
Problem(s) • Weaknesses identified in some Cochrane reviews
• Interpreted without considering certainty or overall quality of the evidence base
Article Type Empirical
Article Subtype Cross-sectional survey/Methodological systematic review
First Author Country Italy
Aim To assess the reporting of and adherence to GRADE and to compare the assessment of imprecision assessed by GRADE or by trial sequential analysis in a sample of 100 systematic reviews published in the Cochrane Library in 2016 and 2017.
Level of Investigation Analytical
Summary of Findings Very few (13.0%) of the reviews that graded the evidence reported the criteria they applied to assess imprecision. The most common imprecision components were width of 95% confidence interval (8.0%) and optimal information size referred to participants (4.0%). Almost half of the outcomes (48.0%) were downgraded for imprecision. In 30.0% of reviews, judgement of outcomes differed between the review authors and the authors of this article who followed the GRADE Handbook. A significant lack of reporting of and adherence to GRADE was observed in Cochrane systematic reviews. Trial sequential analysis was useful for validating the appropriate implementation of GRADE.
Number of systematic reviews included 100
Number of eligible systematic reviews assessed 100
Treatment impacted No
Treatment impacted description
Interpretation impacted Not Applicable
Interpretation impacted description