Supplementary data and source register for: AI-Enabled Essential Medicine Stock Management in Resource-Constrained Health Systems - MBA thesis

Here's a description you can paste straight into Zenodo's Description field. It's written for examiners and the university — it explains what the deposit contains, what you did, and (importantly) is transparent about the copyright split so no one thinks you've re-hosted publisher PDFs. Supplementary Data and Source Register This repository accompanies the MBA (Healthcare Management) project report "AI-Enabled Essential Medicine Stock Management in Resource-Constrained Health Systems: A Mixed Secondary Methods Study of Fourteen Selected Low- and Lower-Middle-Income Countries" (Bradford International Alliance, 2026). It provides the materials needed to review and reproduce the study's three analytical layers: (1) a secondary quantitative analysis of country-level health-system, financing, workforce and service-coverage indicators for fourteen countries; (2) a structured literature review with PRISMA-style reporting; and (3) a structured content analysis of national health-policy documents coded against eight readiness themes. Contents Cleaned analytical dataset used in the quantitative layer (fourteen countries; indicators compiled from WHO Global Health Observatory, WHO Global Health Expenditure Database, and World Bank sources). Policy-coding matrix and per-country coded-evidence tables. Literature-review evidence-extraction table and PRISMA screening record. Source register listing every third-party document analysed, with full citation, publisher, official URL, and access date. Data sources and provenance. All quantitative indicators derive from publicly available datasets published by the World Health Organization and the World Bank; original source and reference year are recorded per indicator. National policy documents are official government or agency publications; the study analysed them under fair academic use. Copyright note. Only the author's own derived materials (datasets, coding matrices, extraction tables, code) are distributed here. Copyrighted third-party works — including the journal articles in the literature review — are not redeposited; they are identified in the source register with their original access details so they can be retrieved from the publisher of record. Methods and reproducibility. The quantitative analysis was conducted in Python (pandas, NumPy, SciPy, statsmodels, Matplotlib). Full environment details, the regression specification, and the underlying formulas are documented in Appendix E of the project report. Given the small cross-sectional sample (n = 14), quantitative results are reported as exploratory ecological associations and should not be interpreted causally. Citation. Please cite the associated project report and this deposit (DOI: 10.5281/zenodo.21494069). Licensed under CC BY 4.0.