Risk of Bias

Including unpublished datasets is expected to reduce publication bias. The protocol favors JBI over AMSTAR-2 because the expected evidence base is methodologically heterogeneous and not primarily intervention-focused. The protocol explicitly avoids the Newcastle-Ottawa Scale and STROBE checklist for prevalence-focused research. Risk of bias will be assessed using the Newcastle-Ottawa Scale. Risk of bias is assessed at the jurisdiction level with a modified 8-item tool. Overall evidence quality is assessed with the GRADE framework. Publication bias is assessed with funnel plots and Egger's test. Each included study will be independently assessed using the revised Cochrane risk-of-bias tool for randomised trials. Risk of bias will be independently assessed using QUADAS-2. Only one of the 13 primary studies had low risk of bias. Risk-of-bias results will inform evidence quality characterisation but will not determine study inclusion. QUADAS-2 will be used for diagnostic accuracy studies. Heterogeneity in systems, populations, instruments, cost definitions, currencies, and analyses prevented meta-analysis. COSMIN Risk-of-Bias will be used for measurement-property studies. Study validi…