amstar2-review-appraisal

Record the review's citation, type, and PICO, pre-specify the critical domains (default items 2, 4, 7, 9, 11, 13, 15) and the Partial Yes convention, then answer all 16 items from the report with quoted evidence. Classify each No in a critical domain as a critical flaw and derive the overall confidence rating from the Box 2 rules.

The seven default critical domains from Box 1 are items 2 (protocol), 4 (search), 7 (excluded studies list), 9 (risk-of-bias assessment), 11 (pooling methods), 13 (RoB in interpretation), and 15 (publication bias). Appraisers may add or substitute domains, but must fix the list before answering.

AMSTAR 2 appraises the methodological quality of one systematic review and yields a confidence rating in that review's conduct. GRADE rates the certainty of a body of evidence per outcome across studies. Use AMSTAR 2 first to decide if a review is trustworthy, then GRADE its findings.

No, AMSTAR 2 appraises systematic reviews, not primary studies. For a single randomized trial use the Cochrane RoB 2 tool, and for non-randomized studies use ROBINS-I; the companion skill assess-study-bias covers both.

No, the authors explicitly recommend against summing item ratings into a score. The instrument produces one of four overall confidence ratings — High, Moderate, Low, or Critically low — derived from counts of critical flaws and non-critical weaknesses.

The paper introduces Partial Yes for items 2, 4, 7, 8, and 9 but never weights it, so the appraisal team must pre-specify the convention. The default here treats it as a non-critical weakness; a strict convention counts it as a critical flaw in critical domains, and the chosen convention must be reported.