Description

This Master Prompt develops a rigorous framework for organizing, comparing and, where justified, statistically synthesizing quantitative findings from systematic reviews and meta-analyses. It addresses the selection of analytical units and estimands, extraction of effect estimates and uncertainty, assessment of comparability, management of overlapping primary studies, statistical model selection, heterogeneity, prediction intervals, subgroup analysis and sensitivity analysis. Drawing on the supplied intervention, prevalence, exposure and population-health papers, it distinguishes the presentation of existing pooled estimates from recalculation using primary-study data and second-order synthesis of review-level estimates. Particular attention is given to avoiding unjustified pooling, inflated precision, inappropriate effect-measure conversion and overinterpretation of statistical significance. The framework supports transparent quantitative interpretation while preserving differences in study design, populations, outcomes, follow-up periods, methodological quality and the independence of the underlying evidence.