Mapping inequities in safe water access: a spatial and geographically weighted regression analysis of unimproved water source use in sub-Saharan Africa

Gelila Yitageasu; Alemayehu Kasu Gebrehana; Angwach Abrham Asnake; Bizunesh Fantahun Kase; Mitkie Tigabie; Lidetu Demoze

doi:10.1186/s41043-026-01259-0

Mapping inequities in safe water access: a spatial and geographically weighted regression analysis of unimproved water source use in sub-Saharan Africa

J Health Popul Nutr. 2026 Feb 14. doi: 10.1186/s41043-026-01259-0. Online ahead of print.

Authors

Gelila Yitageasu¹, Alemayehu Kasu Gebrehana², Angwach Abrham Asnake³, Bizunesh Fantahun Kase⁴, Mitkie Tigabie⁵, Lidetu Demoze⁶

Affiliations

¹ Department of Environmental and Occupational Health and Safety, Institute of Public Health, College of Medicine and Health Sciences, University of Gondar, Gondar, Ethiopia. gelilayitageasu1@gmail.com.
² Department of Midwifery, College of Health Sciences, Salale University, Fitche, Ethiopia.
³ Department of Epidemiology and Biostatistics, School of Public Health, College of Health Sciences and Medicine, Wolaita Sodo University, Wolaita Sodo, Ethiopia.
⁴ Department of Public Health, College Health Sciences, Mattu University, Mattu, Ethiopia.
⁵ Department of Medical Microbiology, School of Biomedical and Laboratory Science, College of Medicine and Health Sciences, University of Gondar, Gondar, Ethiopia.
⁶ Department of Environmental and Occupational Health and Safety, Institute of Public Health, College of Medicine and Health Sciences, University of Gondar, Gondar, Ethiopia.

PMID: 41691342
DOI: 10.1186/s41043-026-01259-0

Abstract

Background: Access to safe drinking water remains a major challenge in sub-Saharan Africa, where reliance on unimproved water sources exposes millions to waterborne diseases and undermines progress toward Sustainable Development Goal 6 (SDG 6). Understanding spatial patterns and determinants of unimproved water use is essential for targeting WASH interventions.

Method: We analyzed data from 500,845 households across 20,492 clusters in 34 sub-Saharan African countries using Demographic and Health Surveys (2012-2024). Data was processed using Microsoft Excel and STATA version 17. Prevalence estimates accounted for the DHS complex sampling design. Spatial analyses (Global Moran's I, Getis-Ord Gi*, SaTScan™, Anselin Local Moran's I) and Geographically Weighted Regression (GWR) were conducted to identify clusters and key predictors.

Result: Overall, 58% (95% CI: 51%-65%) of households relied on unimproved drinking water sources. This indicates that approximately three in five households utilize unimproved water source. Significant spatial clustering was detected (Global Moran's I = 0.2415, p < 0.001), with major hotspots in Madagascar, Uganda, Rwanda, Mozambique, Malawi, Chad, Nigeria, Togo, Burkina Faso, Liberia, Sierra Leone, Mali, Ethiopia, parts of Angola, the Congo Republic, Zambia, and Zimbabwe. SaTScan™ identified 109 significant clusters, and Local Moran's I revealed mixed high- and low-risk zones. GWR (adjusted R² = 0.917) highlighted key predictors: rural residence, female-headed households, older household heads (50 + years), poverty, limited media exposure, lack of electricity, and poor water access.

Conclusion: Unimproved water use remains widespread and spatially concentrated across sub-Saharan Africa. The determinants identified by GWR poverty, rural residence, limited infrastructure access, and low household media exposure highlight critical inequities driving this burden. Addressing these disparities requires localized, data-driven WASH planning that prioritizes hotspot regions, expands access to affordable improved water systems, strengthens electricity and media outreach infrastructure, and supports vulnerable households. Implementing these targeted and multi-sectoral strategies is essential for accelerating equitable progress toward SDG 6 and improving public health resilience across the region.

Keywords: Cluster; Improved source; Spatial; Sub-Saharan africa; Unimproved source; Water source.