Date of Award

Summer 8-9-2016

Degree Type


Degree Name

Master of Public Health (MPH)


Public Health

First Advisor

Dora Il’lyasova, PhD

Second Advisor

Christine Elizabeth Stauber, PhD


Introduction: Millions of children worldwide die before they reach their fifth birthday. Approximately 50% of all deaths in children are associated with malnutrition. Although remarkable improvements have been seen in the past few decades, child malnutrition remains a major public health problem in Ethiopia. Malnutrition has been linked to various morbidities and as the underlying cause of 57% of mortality in the country. It is caused by complex and multidimensional biological, social economic, and environmental factors. There are scarce literatures examining the environmental factors, such as access to water and sanitation, on the likelihood of child malnutrition as measured by stunting, wasting, and underweight in Ethiopia, which the current study sought to investigate.

Aim: The primary aim of this study was to examine the relationship between households’ access to water and sanitation facilities and the likelihood that a child will become stunted or wasted or underweight.

Methods: The study utilized a nationally representative data from 2011 Ethiopia Demographic and Health Survey. The sample size was 9,611 children aged 0-59 months. A weighted descriptive statistical analysis was performed to examine the frequency distribution of the study’s primary independent variables (sanitation and water), dependent variables (childhood stunting, wasting, and underweight), and all other variables included in the study. Weighted bivariate analysis was conducted using logistic regression to quantify association between stunting, wasting, and being underweight and different independent variables. Weighted multivariate logistic regression analysis was performed to control for potential confounders while examining the relationship between the primary independent and dependent variables. Odds ratios, 95% confidence limits, and p-value were calculated. We considered three sets of potential confounders: child’s (child’s gender, child’s age, and child’s size at birth diarrheal disease, fever), maternal (maternal education) and household characteristics (maternal BMI, place of residence, wealth index, stool disposal, time to get water). Only variables that showed significant association (p-value

Main results: Approximately 44%, 10%, and 29% of the children under-five years of age were stunted, wasted, and underweight, respectively. About 54% of the study population used unimproved source of drinking water and about 82% used unimproved sanitation facility. Our bivariate logistic analysis revealed that children in households with unimproved source of drinking water had higher odds of stunting compared to children in households with improved drinking water source (OR: 1.2; 95% CL 1.02-1.34). Adjustment for child’s characteristics yielded AOR: 1.2; 95% CL 1.0-1.4. Addition of maternal characteristics attenuated this association (AOR: 1.1; 95% CL 1.0-1.3 1.0; 95% CL 0.8-1.2). Finally, inclusion of household characteristics showed stunting was not associated with unimproved source of drinking water (AOR: 1.0; 95% 0.8-1.2). The bivariate analysis revealed household access to unimproved source of drinking water was not significantly associated with wasting (OR: 1.0; 95% CL 0.8-1.3) and underweight (OR: 1.2; 95% CL 1.0-1.4). Adjustment of child, maternal, and household characteristic showed an inverse association between source of drinking water and wasting (AOR: 0.7; 95% CL 0.6-0.9). In the bivariate analysis, access to unimproved sanitation was significantly associated with stunting (OR: 1.3; 95% CL 1.02-1.74) and underweight (OR: 1.5; 95% CL 1.1-2.1). Compared to children living in homes with access to improved sanitation facility, children in household with unimproved sanitation facility had 1.4 increased odds of being stunted (95% CL 1.1-1.9) after adjustment for child’s characteristics. Adjustment of child, maternal, and household characteristics attenuated this association (AOR: 1.1; 95% CL 0.8-1.5). Children in household with unimproved sanitation facility had higher odds of being underweight after adjusting for child characteristics (AOR: 1.6; 95% CL 1.2-2.2). Addition of maternal characteristic reduced the association (AOR: 1.5; 95% CL 1.1-2.0). Finally, the addition of household characteristics further attenuated this association (AOR: 1.4; 95% CL 1.1-1.9). Children from households with improved water but unimproved sanitation had higher odds of wasting and being underweight compared to children living in household with both services: AORs adjusted for child’s characteristics were 2.3 (95% CL 1.3-4.3) for wasting and 2.4 (95% CL 1.6-3.6) for underweight; when maternal characteristics were included, AORs were 2.2 (95% CL 1.2-4.1) and 2.1 (95% 1.4-3.3) for wasting and underweight, correspondingly; finally, when household characteristics were included AORs were 2.0 (95% CL 1.1-3.9) and 1.9 (95% CL 1.2-3.0), respectively.

Conclusion: Our results suggest that household access to unimproved source of drinking water and sanitation increase the likelihood of malnutrition. Therefore, initiatives to increase access to improved sources of drinking water and sanitation facilities along with nutritional intervention could help alleviate the high burden of malnutrition in Ethiopia.