johnfriesen/Accessibility-Sub-Saharan-Africa

Analysis of Access to Health-Care in sub Saharan Africa # Healthcare Accessibility Analysis in Sub-Saharan African Cities This repository contains Python scripts to analyze healthcare accessibility across 19 sub-Saharan African cities. The analysis is based on spatial accessibility (SA) metrics, travel time calculations, and relative risk (RR) evaluations between formal and informal urban settlements. Our study combines various data sources, including population datasets, street networks, healthcare facilities, and land-use classifications, to provide a detailed assessment of healthcare access disparities. ## Methods Overview The code implements a healthcare accessibility model based on travel time to healthcare facilities. Key elements of our methodology, inspired by Levinson and Wu's theory of access assessment, include: - **Location**: Urban areas in sub-Saharan Africa. - **Travel Cost**: Measured in travel time (minutes) via isochrone analysis. - **Healthcare Focus**: All types of healthcare facilities, with a specific analysis on hospitals for comparative purposes. - **Mode of Transport**: Walking, given its primary role in healthcare access in SSA. - **Urban Morphology**: Differentiation between formal and informal settlements, as a proxy for socio-economic factors. The accessibility calculations were performed using transportation network analysis in ESRI ArcGIS Pro 3.2.1, leveraging OpenStreetMap (OSM) data for street networks. Healthcare facility locations were obtained from a geocoded public health facility dataset, while population data were sourced from the WorldPop Constrained UN adjusted dataset. ## Data Sources The data inputs used in this analysis include: 1. **Land Use and Land Cover (LULC)**: Guzder Williams et al. classification (2023), which categorizes residential areas into formal and informal subdivisions. 2. **Population Data**: WorldPop dataset (2020), with a spatial resolution of 100m, providing a baseline for population estimates. 3. **Street Networks**: OpenStreetMap data, allowing for a …