‘Modernity’ and Toilets: Formative and Implementation Research to Design and Evaluate a Theory-Based Behavioural Intervention to Improve Toilet Use in Rural India

Toilet use for safe disposal of faeces is an effective practice for prevention of diseases. However, despite improved toilet coverage, use remains a persistent problem in many low-and middle-income country contexts. This thesis describes the formative and implementation research to design and evaluate a theory-driven, behavioural intervention to promote toilet use among rural households with a government or contractor-built toilet in Bhavnagar district, Gujarat, India. The research reported in this thesis is guided by the Behaviour Centred Design (BCD) framework and includes a scoping review, formative research, a creative process to design an intervention for toilet use promotion, and a process evaluation to understand the implementation of, and response to, the intervention and the appropriateness of its theory of change. The scoping review aimed to list and classify known determinants of toilet use. This was followed by formative research on determinants of toilet use or non-use among the target population in rural Bhavnagar. Insights gained from this process were used to design an intervention drawing on the concept of ‘modernity’ and including the motives of comfort and convenience, status, affiliation and hoard. The intervention’s impact evaluation showed a small increase in toilet use, which was not statistically significant (unadjusted difference + 6.3%, 95%CI 1.1, 11.4, adjusted difference + 5.0%, 95%CI -0.1, 10.1.). A mixed methods process evaluation guided by the UK Medical Research Council’s framework for conducting process evaluations and based on the intervention’s theory of change was conducted post hoc to understand how the overall context, inputs, implementation, as well as participant response and acceptance influenced the behavioural outcome . We found that the concept of ‘modernity’ resonated with the target population. However, due to short implementation timeframe, seasonality and contextual factors such as topography and high toilet coverage in study villages, it was difficult to reach target households within the stipulated timeframe. Thus, while the intervention was able to psychologically ‘trigger’ some in the target population, the intervention’s relatively low intensity and short duration were insufficient to cause behaviour change in favour of toilet use. A detailed assessment of contextual dimensions may have resulted in better coverage, leading to greater implementation intensity. This study, based on the BCD framework, presents lessons for the design and evaluation of behaviour change programmes. Future behavioural research on the science of intervention delivery is required to determine the minimum ‘quantity and quality’ of intervention required to cause behaviour change, conduct detailed assessments of contextual factors that may influence uptake of intervention and plan inclusive targeting of intervention components to sub-populations with the greatest need.