Almost 800 million people worldwide lack access to improved drinking water sources. As a consequence, water-borne diseases, including chronic illness and increased mortality from geogenic contaminants in groundwater pose severe threats to human health and well-being worldwide. The situation in developing countries is particularly dire, as central water supply is rare, and mitigation therefore more complex. In rural Bangladesh, millions of people are at risk of drinking water with elevated arsenic, despite the fact that many have gained awareness of the health risks, and over 100’000 safe water options have been installed in recent years. It is being increasingly recognized that without people’s compliance, health risks, including water-related risks cannot be mitigated. The social cognition approach proposes that behavior is best explained by an individual’s perceptions of objective reality. Moreover, these determinants are assumed to be modifiable, wherefore interventions targeting the behavioral determinants should increase behavior change effects of standard interventions that are developed without theoretical considerations. This thesis aims at applying this approach to promote safe water consumption.
To identify potential behavioral determinants of safe water consumption, a review of major health behavior theories was conducted first. The identified determinants were compiled to a theoretical framework of potentially influential behavioral determinants. These were linked to behavior change techniques (BCTs) that are assumed to modify the determinants. Three empirical studies were conducted in Bangladesh that assessed arsenic-safe water consumption and the theory-derived behavioral determinants. In Study 1, a large cross-sectional survey was conducted to gain detailed knowledge about current arsenic-safe water consumption in the target population. Structured face-to-face interviews were conducted with 872 households in six districts of Bangladesh. The structured questionnaire assessed people’s acceptance and use of seven currently available arsenic-safe water options. The most influential behavioral determinants of the habitual use of arsenic-safe water options were identified by multiple linear regression analysis. Studies 2 and 3 aimed at developing theory-based interventions to enhance switching to neighboring arsenic-safe shallow wells (Study 2, N = 370), and promote the use of arsenic-safe deep tubewells (Study 3, N = 340). BCTs were developed that targeted the modification of key behavioral determinants that had been identified from baseline surveys. Thereafter, the developed BCTs were added to a standard informational intervention and compared in cluster-randomized trials regarding their effectiveness to increase the effects of the standard intervention to promote switching to arsenic-safe wells. Moreover, it was determined whether the BCTs would indeed change water consumption by changing the hypothesized behavioral determinants.
Results indicated that overall, one third of the participants of Study 1 did not use the available arsenic-safe water options. Results varied greatly, with piped water supply being the most used option when people had access to it, and rainwater harvesting being the least used. More habitual use of arsenic-safe water options was foremost associated with higher self-efficacy, higher descriptive norms, and higher instrumental attitudes, i.e. when people felt more able to provide as much arsenic-safe water as they needed, when more other people also collected safe water, and when collecting safe water was perceived less time-consuming and effortful. These results were largely confirmed in Studies 2 and 3, but commitment strength emerged as a further important determinant; persons who were more committed to collect arsenic-safe water were more likely to use safe wells. Interventions that targeted increasing commitment strength were developed: reminders, implementation intentions, and public commitment. Confirming most hypotheses, these increased the behavior change effects of the informational intervention by up to 53%, leading to up to 65% well-switching at follow-up. Mediation analyses revealed that the theory-based interventions indeed changed behavior by increasing commitment strength.
The empirical studies demonstrated the usefulness of the social-cognition approach for explaining, predicting, and changing yet another health behavior: safe water consumption. The results of the theory-based interventions provided strong indication that commitment is an important determinant of safe water consumption, and presumably of other behaviors as well. With relevance to health promotion, the increased effects of the theory-based interventions confirm the effectiveness of reminders, implementation intentions, and indicate the importance of more rigorous application of theory in developing and evaluating interventions. Moreover, the here developed BCTs provide promising additions to existing interventions, whereby this research hopes to make a contribution to accelerate the mitigation of arsenic contamination in Bangladesh.