Identifying behavioural determinants for interventions to increase handwashing practices among primary school children in rural Burundi and urban Zimbabwe

Handwashing promotion programmes are increasingly being implemented in developing countries to improve child health and development. Since schools are important settings for disease transmission, school-based interventions aiming at mitigating communicable diseases are likely to reduce the overall community disease burden [1, 2]. According to the WHO/Unicef Integrated Global Action Plan for Pneumonia and Diarrhoea [3], improving access to safe drinking water, providing adequate sanitation, and promoting good hygiene behaviour, such as handwashing with soap, are essential for preventing diarrhoea. In primary schools, interventions promoting handwashing with soap have proven to be effective in reducing infectious diseases in pupils [4–6], Potential constraints include lack of soap and water and the absence of adequate handwashing facilities [7–10]. Increasing the provision of soap and water for handwashing has caused decreases in absenteeism [6, 11, 12], and several studies have reported an association between proper handwashing behaviour and the availability and accessibility of handwashing facilities [13–15].

For handwashing behaviour to be adopted and become a habit, it is not enough to provide proper resources and facilities. Growing evidence suggests that health behaviours such as dietary habits, physical activity patterns, and substance abuse are predicted by such social-cognitive factors as attitude, subjective norms, and self-efficacy beliefs [16–18]. Several studies have indicated that hand hygiene practices depend largely on psychological factors within the individual [19–21]. So far, very few studies have investigated behavioural determinants underlying children’s handwashing practices. Two studies have drawn on the theory of planned behaviour to examine factors affecting proper handwashing. Research by Lopez-Quintero, Freeman, and Neumark [21] in Colombia showed that intentions to perform proper handwashing were determined by perceived control, personal attitudes, and subjective norms. Setyautami, Sermsri, and Chompiku [13] found that students with positive attitudes and perceived behavioural control were twice as likely to wash their hands properly. Several studies have used the knowledge, attitudes, and practices approach to examine the influence of school children’s knowledge, attitudes, and practices on hygiene behaviour; they have reported mixed results concerning the importance of knowledge in determining proper handwashing behaviour [14, 22–24]. Although attitude was mentioned as an important indicator for hygiene behaviour in all of these studies, it was not assessed above and beyond knowledge and practice. More importantly, self-regulatory processes such as action control and feelings of self-efficacy have not yet been investigated.

Researchers urge the use of theories of behaviour change for developing interventions and programmes to change health behaviour [25, 26]. Promoting proper handwashing practices is challenging, and the effectiveness of handwashing interventions have been inconsistent [27]. Applying behaviour change theories to promotion programmes for handwashing may increase their potential for changing behaviour [28]. So far, to the best of our knowledge, no study has used social cognition models from the realm of health psychology to design data-driven handwashing programmes in primary schools in developing countries. In this study, Mosler’s RANAS (risk, attitudes, norms, ability, and self-regulation) approach to behaviour change [29] served as a theoretical framework to measure the behavioural determinants underlying handwashing with soap among primary school children. The model suggests that the behaviour of people is determined by their risk perception, their attitudes toward a behaviour, their beliefs concerning the advantages or disadvantages of adopting or not adopting the behaviour, normative beliefs, perceived self-efficacy, resources, and skills necessary to perform the behaviour. The RANAS blocks assimilate factors from different theories of social and health psychology, such as the theory of planned behaviour [30] and the health action process approach [31], that have been shown to successfully explain and change many types of health behaviour. The RANAS approach provides an analytical tool to analyse the different determinants of behaviour on the basis of quantitative data. Mosler [29] suggests targeting the determinants with the highest intervention potential, that is, determinants with low mean scores and high predictive values on the behaviour within the target population. The corresponding behaviour change techniques are then selected to develop appropriate practical strategies for intervention programmes [32–34]. Several studies have successfully applied the RANAS approach for different health-related behaviours, including handwashing [35], in the water and sanitation sector in developing countries and have shown the added value of implementing data- and theory-based interventions compared to information interventions alone [36–38].

This study uses the RANAS social cognition model of health behaviour to analyse data gathered from surveys of primary school children in two countries regarding the behavioural determinants of the children’s handwashing practices. The aim of the present paper is to describe a psychological approach to designing a handwashing programme using data collected from study participants, theory, and empirical evidence from the literature. The study addresses two main research questions: (1) Which behavioural determinants are related to self-reported handwashing frequencies after using the toilet at school and what is their improvement potential? (2) What theory-based behaviour change techniques can be directed at these behavioural determinants to generate changes in behaviour? Information from this study will serve as baseline data for future campaign development and policy action for an effective school-based handwashing intervention programme.

In this article we describe an application of the RANAS systematic approach to behaviour change for the development of a school handwashing programme for primary school children in a rural and an urban setting in two sub-Saharan African countries. The results of the regression analyses revealed that the RANAS behavioural determinants predicted children’s self-reported handwashing frequencies very well in both countries. In Burundi, high reported handwashing frequencies after using the toilet were best predicted by a high perceived severity of diarrhoea, the perception that many other children wash hands at school too, and a strong confidence in one’s abilities to always perform the behaviour. In Zimbabwe, the behavioural determinants with the highest predictive value proved also to include the perception that other children wash hands at school too, the confidence in one’s abilities to always perform the behaviour, and, moreover, paying a lot of attention to always washing hands after using the toilet at school. The findings in this study are consistent with the results of studies conducted with primary caregivers of young children in Haiti and southern Ethiopia showing that the relevant significant behavioural determinants from the present regression analyses were also predictive of self-reported handwashing [47]. In Bogotá, Colombia, school children also reported higher subjective norms and higher perceived control (akin to self-efficacy) when their intention to wash hands properly was high [21]. School children in Selat sub-district, Indonesia were also more likely to wash hands properly when their perceived behavioural control was high [13]. The results from Burundi and Zimbabwe indicate an overall lack of awareness of hygiene issues in both countries. Low norms for handwashing and the children’s low perceived ability are consistent with the lack of adequate infrastructure at the schools.

The improvement potentials calculated suggest that an intervention targeting social norms and self-efficacy should be most effective in both countries. Additionally, in Burundi, children that do not perceive diarrhoea as severe should be targeted by the intervention. In Zimbabwe, children with less knowledge of diarrhoea and disease transmission should profit from the proposed programme. Based on these results and taking into consideration the observational findings on the school handwashing characteristics, a school handwashing programme was developed that fit the target groups. The interventions of the programme aim to (1) create awareness of the benefits of handwashing through educational activities, (2) raise children’s ability and confidence to wash hands at school through infrastructural improvements, and (3) highlight the commonness of handwashing at school through events and poster creation. Several studies have been able to show that raising awareness for the importance of handwashing and increasing hygiene knowledge leads to an improvement in proper handwashing [4, 10, 55]. Moreover, the presence of handwashing stands at school has been found to be associated with proper handwashing [13–15], and providing soapy water has been shown to raise the frequency of handwashing practices at school [10]. By introducing the new hardware with a big event and because of the continuous use of the handwashing stations by all children, the behaviour should become common practice, increasing the descriptive norm at each school [19, 56] and enhancing the children’s self-efficacy through facilitation of the behaviour [56–58].

The results should be viewed with the caution necessary with self-reported behaviours. Several studies have shown that self-report overestimates handwashing behaviour when compared to observed frequencies [59, 60]. However, collecting observed data on all children included in this study would have been very difficult and costly and extremely time-consuming. In addition, the operationalization of the behavioural determinants can be criticized because they were measured with only one item. Even though we do not have reliability indicators for the survey items, keeping the questionnaire short was necessary to keep the children motivated to participating in the survey. The present study is cross-sectional, so that relationships between variables are descriptive and do not imply causality. However, the results of the regression analyses have been confirmed by previous work focusing on caregivers’ handwashing practices [47].

The RANAS systematic approach to behaviour change allowed us to determine the relative importance of the behavioural determinants underlying school children’s handwashing practices and thus to select appropriate behaviour change techniques. Several reviews of health promotion programmes have concluded that the quality of an intervention is increased by the use of methods derived from social-cognitive theories [28, 61, 62]. The findings of this study strongly suggest that similar handwashing programmes providing education on handwashing issues along with adequate infrastructure could induce behavioural change in rural and urban settings in two different countries.