Centre for Development Economics
Department of Economics, Delhi School of Economics


The Gap between Assumptions and Truth:  Findings from Two Randomized Control Trials of Sanitation Programs in Odisha and Madhya Pradesh


Sumeet Patil

School of Public Health, UC Berkeley and, NEERMAN (Mumbai)

Thursday, 12th February 2015 at 3:00 PM

Venue : Seminar Room (First Floor)
Department of Economics, Delhi School of Economics

All are cordially invited

Poor sanitation is thought to be a major cause of enteric infections and malnutrition among young children. However, can large scale sanitation programs indeed deliver the hypothesized health benefits? We answer this question in the context of India’s Total Sanitation Campaign (TSC) using cluster-randomized, controlled trials in Odisha (2005-06) and Madhya Pradesh (MP; 2009-2011). In both sites, the interventions consisted of Community Led Total Sanitation (CLTS) based behaviour change approaches and subsidies for toilet construction but the intensity and mechanism of the programs differed. We randomized 40 and 80 villages to treatment and control arms equally in Odisha (n~1086 households) and MP (n~3029 households), respectively, and estimated differences in the outcomes between the two groups in an intention-to-treat analysis.
In both trials, the interventions increased percentage of households with IHL in a village (by 19% in MP and by 25% in Odisha) and decreased OD among adults (by ~10% in MP and by ~17% in Odisha). However, the intervention in MP did not improve child health based on multiple outcomes (diarrhoea, HCGI, helminth infections, anaemia, growth). In Odisha, the program increased the height of children under 5 years of age by 1.49 cm compared to the control group (85.56 cm). We also find program effect on child arm circumference, but not on weight and diarrhoea prevalence. By decomposing the treatment effects we further find that the level of OD is strongly correlated with height and arm circumference. Subsidies for toilet construction and behaviour change interventions both are highly complementary to achieve meaningful reductions in OD levels.
The collective evidence suggests that the future refinements of the TSC may immensely benefit by strengthening both the behaviour change and subsidy delivery aspects of the program. However, available evidence also cautions us that that the effect of improved sanitation may depend on other non-sanitation factors (water, handwashing, public health, environmental, etc.). There also has not been much innovation in promoting sanitation (behaviour change) except the anecdotal success of CLTS and its close variants. Unfortunately, the existing epidemiological and economic research in India is too limited to provide meaningful advice about the “How” to policy makers. Therefore, these papers make a call to design and test program refinements to TSC/NBA/SBA in small-scale and short-term pilots across different regions of India before scaling up to the national level.

back to seminars…