Network Meta-Analysis

The Bayesian approach is preferred because it does not require large-sample normality, handles multiarm trials and between-study heterogeneity better, and expresses treatment comparisons as probabilities. Treatment ranking will use p scores from 0 to 1, with higher values indicating a greater probability of being most efficacious. The adult network meta-analysis found no evidence of different clinical resolution between amoxicillin and amoxicillin-clavulanate. Certainty of evidence will be appraised using the CINeMA framework, which applies GRADE principles to NMA, across six domains including within-study bias, reporting bias, indirectness, imprecision, heterogeneity, and incoherence. The review performed an unplanned frequentist random-effects network meta-analysis comparing amoxicillin with amoxicillin-clavulanate in adult trials only. The network was sparse, limiting the power of inconsistency tests. No clear transitivity violation was found, although the number of studies per comparison was small. This review is prospectively registered in PROSPERO and follows PRISMA-P 2015 and PRISMA-NMA reporting guidelines. Transitivity will be assessed by comparing distributions of potent…