Dataset for the systematic review and meta-analysis "Global molecular and serological prevalence of Ehrlichia canis in domestic dogs (2016–2026)": screening log, extraction matrix, risk of bias assessment, figures and analysis code

This dataset supports a systematic review and meta-analysis of the global molecular and serological prevalence of Ehrlichia canis in domestic dogs (Canis lupus familiaris), based on studies published between 1 January 2016 and 21 July 2026 and indexed in Scopus. The review protocol (PRISMA-P 2015) is available at https://doi.org/10.5281/zenodo.23919796. The Excel workbook contains five sheets. The first is a data dictionary. The second is the screening log of the 675 records, with the decision and the reason for each one. The third is the extraction matrix of 155 prevalence estimates from 146 included studies (87 molecular and 68 serological), with country, region, latitude, population type, diagnostic method, assay target, number of dogs tested and number of positive dogs. The fourth holds the Joanna Briggs Institute risk of bias ratings for each estimate. The fifth lists the 73 eligible records that were excluded from the quantitative synthesis because they did not report a numerator or denominator. The package also includes the PRISMA 2020 checklist, Figures 1–6 and S1–S2 at 600 dpi, and the Python 3 scripts that reproduce all analyses. The scripts cover binomial-normal generalised linear mixed models, subgroup analyses, mixed-effects meta-regression with the Knapp–Hartung adjustment, sensitivity analyses, and the Egger and Peters tests for small-study effects.