Hotspots of unimproved sources of drinking water in Ethiopia: mapping and spatial analysis of Ethiopia demographic and health survey Data 2016

Abstract

Abstract Background More than 35% of the Ethiopian population are using drinking water from unimproved sources. As per the United Nations’ Sustainable Development Goals, Ethiopia is aspiring to achieve universal and equitable access to safe and affordable drinking water for all by 2030. For these goals to be accomplished, it is important to map the country’s hotspot areas of unimproved source of drinking-water so that resource allocation and disease control can be optimized there. Therefore, the objective of this study is to map and identify hotspot areas of unimproved sources of drinking water in Ethiopia. Methods A population based cross-sectional study was conducted in Ethiopia from January 18 to June 27, 2016. Data were collected from 10,064 households using a pretested and structured questionnaire. A stratified two-stage cluster sampling was employed where the enumeration areas were primary sampling units and households were secondary sampling units. Systematic sampling with probability proportional to size was employed to select samples. Datasets were cleaned and entered into SaTScan and ArcGIS software for mapping and analysis. The Global Moran’s I and spatial scan statistical tests (Bernoulli model) were done to explore the presence of clustering in the study area and local spatial clusters (hotspots) of unimproved sources of drinking water using ArcGIS version 10.3 and Kuldorff’s SaTScan version 9.4 software, respectively. Results Unimproved sources of drinking water were spatially clustered in the study area (Moran’s I: 0.35, p < 0.05). A total of 143 significant clusters was identified. Of which, eight were most likely (primary) clusters and the other 135 were secondary clusters. The first spatial window which contains primary clusters was located in Amhara and Afar regions (LLR: 78.89, at p < 0.001). The other 33 spatial windows which contain secondary clusters were found in all regions, except Gambela region and Addis Abeba city administration (with a range of LLR: 10.09–78.89, p < 0.001). Conclusions This study allowed the identification of important non-random clusters and hotspots of unimproved sources of drinking water. Therefore, these results will be determinant to help decision makers in their geographical interventions to combat problems related to drinking water quality.