Tools for meta-analysis of proportions and prevalence from studies reporting event counts and sample sizes. Provides transformed and untransformed inverse-variance models, random-effects estimation, heterogeneity statistics, prediction intervals, subgroup analysis, meta-regression, leave-one-out sensitivity analysis, influence diagnostics, forest plots, funnel plots, and an optional binomial generalized linear mixed model interface. The package is designed for epidemiological, veterinary, medical, and One Health applications, including antimicrobial resistance prevalence studies.
Meta-analysis of proportions and prevalence for epidemiological, veterinary, medical, and One Health studies.
Main functions: meta_prop(), prop_transform(), prop_heterogeneity(), forest_prop(), funnel_prop(), subgroup_prop(), metareg_prop(), loo_prop(), influence_prop(), bias_prop(), predict_prop(), and optional meta_prop_glmm().
The package focuses on transparent inverse-variance methods and supports logit, arcsine, raw-proportion, and Freeman-Tukey transformations, with DL, REML, and Paule-Mandel random-effects estimators.