Affiliation: Gillings School of Global Public Health, Department of Environmental Sciences and Engineering
Sanitation is a global priority: 1 billion people lack access to any sanitation facility and practice open defecation, which contributes to child mortality, stunting, and decreased school attendance. Community-led total sanitation (CLTS) is a sanitation promotion approach implemented in over 50 countries, in which a facilitator visits a village, and “triggers” a collective desire to eliminate open defecation (OD). Implementing CLTS has challenges: it requires frequent follow-up visits by facilitators, depends on collective action by communities, and the costs are not well understood. I conducted an operational research project collaboratively with Plan International to investigate the role of CLTS in addressing global sanitation challenges, and how to optimize implementation. Chapter 1 of this dissertation is an evaluation of training “natural leaders” (NLs—motivated community members) during a CLTS intervention in Ghana, using a multi-site, randomized field trial. Chapter 2 is an evaluation of teacher-facilitated CLTS in Ethiopia, using a multi-site, quasi-experimental study design. Chapter 3 is a bottom-up, activity-based cost analysis of the Ghana and Ethiopia interventions. Training NLs in Ghana caused a 19.9 percentage point decrease in OD. The impact was greatest in small, remote, socially cohesive villages. Teacher-facilitated CLTS in Ethiopia was associated with a 9.8 percentage point smaller decrease in OD than health worker-facilitated CLTS. Neither approach was effective in villages with low baseline OD. The implementation cost in Ghana and Ethiopia ranged from $14.15 to $81.56 per household targeted, and generated community activity and latrine construction. Latrines built during CLTS tended to be made of local, low-durability materials. CLTS should be targeted to villages with high OD, where potential for impact is higher. Training NLs can reduce OD, provided they are from cohesive villages. CLTS should be part of a broader sanitation strategy, as it is not applicable everywhere, and low quality latrines may not last. The multi-site evaluations revealed variation of outcomes across settings. Bottom-up costing enabled greater disaggregation than any prior sanitation study, revealing the burden participatory approaches place on local actors, and potential for improved cost-efficiency. These findings and tools are also applicable to other environmental health behavior interventions.