Determinants of household safe drinking water practices in Kabul, Afghanistan: New insights from behavioural survey data

•Household water treatment technology can mitigate water-borne disease risks.•COM-B behaviour change model was employed to examine HWT determinants.•Reflective & automatic motivation, and physical opportunity influenced behaviour.•Socio-economic, psychosocial, and contextual factors have key roles in water policy. More than 2 billion people worldwide lack access to safe drinking water. Household water treatment (HWT) is an interim option for reducing the risk of water born disease. Understanding the factors that influence HWT behaviour is crucial for delivering successful interventions aimed at scaling relevant technologies, but the literature tends to emphasise psychological determinants with little consideration of socioeconomic and contextual factors. This article responds to this literature by using the COM-B model to examine the determinants of HWT practices through a comprehensive and context-sensitive behaviour definition. We informed this model through a cross-sectional survey design in which we collected data from 913 households in two peri‑urban neighbourhoods of Kabul, Afghanistan. Our findings from descriptive statistical and regression analysis highlight the importance of not only psychological but also socio-economic and contextual determinants of HWT behaviour: Especially the COM-B dimensions of reflective and automatic motivation, and physical opportunity – which are heavily influenced by local context and economic circumstances – had statistically significant associations with performing HWT. The practical significance of these dimensions was similarly pronounced. For example, an increase in the physical opportunity index by 0.1 units from an average value of 0.7 to 0.8 would be associated with a 7.7 percentage-point higher likelihood of HWT performance. These results suggest that the COM-B model can be utilised to systematically design interventions aimed at promoting HWT practices, while highlighting the need to broaden behavioural analyses of HWT and consider contextual factors to develop interventions that are tailored to the specific needs and obstacles of different communities. [Display omitted]