Spatial Distribution and Determinants of NHIS Out-of-Pocket Health Expenditure in Ghana: A 261-District Machine Learning and Geospatial Analysis
**Author:** Valentine Golden Ghanem | Ghana COCOBOD Cocoa Clinic, Accra, Ghana
**ORCID:** 0009-0002-8332-0220
**Affiliation:** Ghana COCOBOD Cocoa Clinic, Accra, Ghana
**Reporting standard:** STROBE · RECORD-Spatial · TRIPOD+AI
**Date:** 2026
**Status:** Manuscript in preparation
---
## 1. Overview
This repository contains the complete reproducible analytical pipeline for a cross-sectional ecological study of National Health Insurance Scheme (NHIS) non-enrolment across all 261 administrative districts of Ghana. The analysis integrates Ghana Population and Housing Census 2021 data with Demographic and Health Survey data (2019, 2022) to characterise the spatial distribution and structural socioeconomic determinants of district-level NHIS uninsurance.
**Key finding:** NHIS non-enrolment is strongly spatially concentrated (Global Moran's I = 0.461, z = 11.94, p 30.2%) | 130 / 261 (49.8%) |
| Global Moran's I (uninsurance) | 0.461 (z=11.94, p **Tip:** The dashboard works fully offline once downloaded. The poster is print-ready at A0 (841 × 1189 mm).
## 9. Reporting Standard
This study follows the **STROBE** (Strengthening the Reporting of Observational Studies in Epidemiology) reporting guideline for observational ecological studies. Machine learning components follow **TRIPOD+AI**; spatial statistical components follow **RECORD-Spatial**.
## 10. Ethics
Retrospective analysis of publicly available, de-identified aggregate administrative and survey data. Ethical oversight: Ghana Health Service Ethics Review Board. No individual-level data were accessed.
## 11. Citation
**APA:**
Ghanem, V. G. (2026). *Spatial distribution and socioeconomic determinants of NHIS non-enrolment across 261 districts in Ghana: a geographically weighted regression and machine learning analysis.* GitHub.
github.com
**BibTeX:**
```bibtex
@misc{ghanem2026nhis,
author = {Ghanem, Valentine Golden},
title = {Spatial distribution and …