The Impact of Drinking Water Quality and Sanitation on Child Health: Evidence from Rural Ethiopia

Abstract

This paper examines the association between stored drinking water quality and sanitation on diarrhoea incidence among under-five children in Ethiopia. Using primary household survey data and microbiological water quality testing for Escherichia coli, our results show that uncontaminated stored drinking water and safe child stool disposal are associated with decreased child diarrhoea incidences of 18 and 20 percentage points, respectively. In contrast, neighbourhood concentration of pit latrine shows an increased incidence of child diarrhoea of 16 percentage points. To protect rural children from the risk of contracting diarrhoea, improving household drinking water quality and changing people’s behaviour towards safe sanitation practices is needed.

Acknowledgements

The authors express their gratitude to Bill and Melinda Gates Foundation for funding this research under the WATSAN-AGICULTURE Nexus project – guiding pro-poor investments in the nexus among domestic water quality and quantity, sanitation and hygiene, and agriculture from the bottom-up. They would like to acknowledge the support from the Center for Development Research (ZEF), the Ethiopian Economics Association (EEA), the Welthungerhilfe, and the Organization for Rehabilitation and Development in Amhara (ORDA). They also thank the anonymous reviewers who helped improve the paper substantially. Any remaining mistakes and inconsistencies are the responsibility of the authors. The data and the Stata code used in this analysis will be made available to researchers upon request to the authors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary Materials

Supplementary Materials are available for this article which can be accessed via the online version of this journal available at https://doi.org/10.1080/00220388.2018.1493193

Notes

1. WHO/UNICEF Joint Monitoring Programme (JMP) for WATSAN defines an improved drinking water source as one that is adequately protected from outside contamination, particularly from contamination with faecal matters, and an improved sanitation facility is a technology that hygienically separates human excreta from human contact.

2. Multiple-use water systems refer to cases where households/communities use a given water source for more than one economic activity: water supply for domestic purpose and for productive use are not exclusive, for instance where a single water source is used for drinking and irrigation.

3. Because of their weak immune systems, young children are more prone to diarrheal diseases than adults. Diarrhoea is also one of the leading causes of under-five child morbidity and mortality, and contaminated drinking water is considered as the major cause of diarrheal diseases.

4. The CSA and ICF International (2012) report indicated that there is a marked difference between urban and rural areas in the way children’s stools are disposed of. Nationally, a higher proportion of urban children’s stools (63%) are disposed of safely than of rural children’s stools (31%).

5. The survey result indicated that the level of diarrhoea prevalence with high-risk category based on the intervention variables were 24 per cent (contaminated stored water quality), 17 per cent (with pit latrine), and 20 per cent (with unsafe child stool disposal).

6. The treatment variable may be an endogenous regressor due to unobservable heterogeneity among households (for example, some households may have unobserved preferences that cause them to have better household water quality than other similar households). Such unobservable covariates could affect the health outcome too. Measurement error also poses another endogeneity concern: microbiological test protocols require repeating water sample tests to reduce measurement errors. Here, we performed the test one-time and it may not reflect the actual water quality.

7. Nichols (2011) generally showed that in the case of a binary regression with a binary endogenous variable, linear IV generates robust consistent estimates of the ATT (average treatment on the treated) while BP produces efficient estimates of the ATE.

8. Child age-squared term was excluded in the regression as it is not statistically significant, but children between six and 24 months old were the most affected age group in this study.

Additional information

Funding

This work was supported by the Bill and Melinda Gates Foundation [OPP1044771].