Piped water supply interruptions and acute diarrhea among under-five children in Addis Ababa slums, Ethiopia: A matched case-control study

BACKGROUND The problem of intermittent piped water supplies that exists in low- and middle-income countries is particularly severe in the slums of sub-Saharan Africa. However, little is known about whether there is deterioration of the microbiological quality of the intermittent piped water supply at a household level and whether it is a factor in reducing or increasing the occurrence of acute diarrhea among under-five children in slums of Addis Ababa. This study aimed to determine the association of intermittent piped water supplies and point-of-use (POU) contamination of household stored water by Escherichia coli (E. coli) with acute diarrhea among under-five children in slums of Addis Ababa. METHODS A community-based matched case-control study was conducted from November to December, 2014. Cases were defined as under-five children with acute diarrhea during the two weeks before the survey. Controls were matched by age and neighborhood with cases by individual matching. Data were collected using a pre-tested structured questionnaire and E. coli analysis of water from piped water supplies and household stored water. A five-tube method of Most Probable Number (MPN)/100 ml standard procedure was used for E. coli analysis. Multivariable conditional logistic regression with 95% confidence interval (CI) was used for data analysis by controlling potential confounding effects of selected socio-demographic characteristics. MAIN FINDINGS During the two weeks before the survey, 87.9% of case households and 51.0% of control households had an intermittent piped water supply for an average of 4.3 days and 3.9 days, respectively. POU contamination of household stored water by E. coli was found in 83.3% of the case households, and 52.1% of the control households. In a fully adjusted model, a periodically intermittent piped water supply (adjusted matched odds ratio (adjusted mOR) = 4.8; 95% CI: 1.3-17.8), POU water contamination in household stored water by E. coli (adjusted mOR = 3.3; 95% CI: 1.1-10.1), water retrieved from water storage containers using handle-less vessels (adjusted mOR = 16.3; 95% CI: 4.4-60.1), and water retrieved by interchangeably using vessels both with and without handle (adjusted mOR = 5.4; 95% CI: 1.1-29.1) were independently associated with acute diarrhea. CONCLUSION We conclude that provision of continuously available piped water supplies and education of caregivers about proper water retrieval methods of household stored water can effectively reduce POU contamination of water at the household level and thereby reduce acute diarrhea among under-five children in slums of Addis Ababa. Promotion of household water treatment is also highly encouraged until the City's water authority is able to deliver continuously available piped water supplies.