Diarrheal diseases kill approximately 1.5 million children under age five annually, representing the second leading cause of death and the leading cause of malnutrition for this age group.1 Reducing the prevalence of diarrhea in developing countries requires expanding access to improved water and sanitation facilities and improving hygienic practices that contaminate water. This study evaluated a range of health and non-health benefits of a comprehensive program to expand and improve water and sanitation infrastructure and improve hygienic practices in the Southern Indian state of Tamil Nadu.
Between 2003 and 2007, Water.org and Gramalaya initiated community-level hygiene education courses and water and sanitation infrastructure development programs in 12 rural villages in Tamil Nadu. At the time of the program, the majority of rural households were engaged in rice cultivation and 23 percent lived below the poverty line. At the time of program evaluation, researchers found that diarrheal prevalence was uniformly low in both intervention and comparison villages at less than two percent for children under age five. Primary caretakers in both groups reported less than one medical visit per child per year due to diarrhea, less than one workday lost per year to care for a child with diarrhea, and low levels of school absence due to illness. However, children under age five in both groups exhibited very high prevalence of underweight, stunting, and wasting.
The program interventions comprised village-wide hygiene education, construction of toilets at primary schools, technical support and local training for toilet construction, and renovation of community water pumps. In addition, Gramalaya provided households in eight of the villages with micro-credit to construct private water taps and toilets.
Because Water.org and Gramalaya purposely selected villages to participate in the program, researchers conducted this retrospective evaluation using propensity score matching. Using village-level census data collected in the year preceding the intervention, researchers matched intervention villages with comparison villages in adjacent administrative regions of Tamil Nadu using seven indicators, including the total number of households in the village, the proportion of population belonging to a Scheduled Caste, and the proportion of households with access to tap water. The researchers then conducted a rapid assessment in each of the intervention and potential comparison villages to measure water infrastructure, car and tractor ownership, and school and administrative facilities in order to select a final group of 13 comparison villages.2
The researchers then enrolled up to 50 households per village that contained at least one child under age five and visited them monthly for one year for a total of 12 visits. The study’s total sample size was 900 households with 1,284 children under age five. In each visit, field staff measured caregiver-reported child diarrhea, water and sanitation facilities and behavior, hygienic practices, and socio-economic conditions. In addition, researchers assessed water quality in each village by testing samples from community drinking water sources and stored household drinking water supplies for water quality indicators (E. coli, total coliforms, H2S).
Results and Policy Implications
The study found that the program resulted in a sustained improvement in water and sanitation infrastructure but did not improve health outcomes, hygienic behavior, or household water quality. As a result of the program, households in intervention villages were 8 percentage points more likely to report access to a new water source (26 percent vs. 18 percent) and 33 percentage points more likely to have constructed a new private toilet (48 percent vs. 15 percent). Private toilet construction expanded most significantly for low- and middle-income groups and those from Scheduled Caste households.
Despite improvements in water and sanitation infrastructure, hygienic behavior and health outcomes in intervention villages did not appear to change. Households in villages that received hygiene education exhibited similarly poor hygienic conditions after the program as households in comparison villages. In addition, nearly 40 percent of households with private toilets in intervention villages reported that adults continue to practice open defecation daily. The researchers did not find any difference in reported occurrences of diarrhea among children under five between intervention and comparison villages, but attribute this finding to the uniformly low prevalence of diarrhea in the study area. Furthermore, child growth did not differ between intervention and control villages; stunting, wasting, and malnutrition rates in both groups were very high. The authors suggest that poor child growth in this population likely results from causes other than diarrhea, such as poor nutrition, and bears further research.
Although the study found reduced contamination at village water sources, household water quality did not improve. Significantly fewer samples from intervention village water sources tested positive for E. coli contamination than in comparison villages, and those which were contaminated had significantly lower levels than comparison village water sources. However, by all measures water quality at the household level was worse, and there were no statistical differences in household water quality between intervention households and comparison households.
The study measured significant non-health benefits to improved infrastructure: Women in households with private toilets were 28 percentage points more likely to report feeling safe during defecation (81 percent vs. 53 percent), and installing a private water tap resulted in a household spending 25 fewer minutes per day gathering water compared with households using a public tap (50 minutes vs. 75 minutes).
The researchers conclude that careful assessments of baseline health and socio-economic indicators are required to identify populations in greatest need and that water, sanitation, and hygiene programs should target areas with a high prevalence of diarrhea if the goal is improved health outcomes. The study also highlights the difficulty of changing defecation behavior and the nuanced relationship between toilet construction and actual defecation practice. The researchers believe that addressing the complexity of hygienic behavior may be necessary to fully realize the health benefits of improved water and sanitation infrastructure.
1 World Health Organization. Diarrheal Disease Fact Sheet, August 2009. http://www.who.int/mediacentre/factsheets/fs330/en/index.html