Comparing the behavioural impact of a nudge-based handwashing intervention to high-intensity hygiene education: a cluster-randomised trial in rural Bangladesh.

OBJECTIVE: To determine the impact of environmental nudges on handwashing behaviours among primary school children as compared to a high-intensity hygiene education intervention. METHODS: In a cluster-randomised trial (CRT), we compared the rates of handwashing with soap (HWWS) after a toileting event among primary school students in rural Bangladesh. Eligible schools (government run, on-site sanitation and water, no hygiene interventions in last year, fewer than 450 students) were identified, and 20 schools were randomly selected and allocated without blinding to one of four interventions, five schools per group: simultaneous handwashing infrastructure and nudge construction, sequential infrastructure then nudge construction, simultaneous infrastructure and high-intensity hygiene education (HE) and sequential handwashing infrastructure and HE. The primary outcome, incidence of HWWS after a toileting event, was compared between the intervention groups at different data collection points with robust-Poisson regression analysis with generalised estimating equations, adjusting for school-level clustering of outcomes. RESULTS: The nudge intervention and the HE intervention were found to be equally effective at sustained impact over 5 months post-intervention (adjusted IRR 0.81, 95% CI 0.61-1.09). When comparing intervention delivery timing, the simultaneous delivery of the HE intervention significantly outperformed the sequential HE delivery (adjusted IRR 1.58 CI 1.20-2.08), whereas no significant difference was observed between sequential and simultaneous nudge intervention delivery (adjusted IRR 0.75, 95% CI 0.48-1.17). CONCLUSION: Our trial demonstrates sustained improved handwashing behaviour 5 months after the nudge intervention. The nudge intervention's comparable performance to a high-intensity hygiene education intervention is encouraging.

A qualitative comparative analysis of well-managed school sanitation in Bangladesh.

BACKGROUND: Continued management of sanitation and hygiene services, post-intervention, is a global challenge, particularly in the school-setting. This situation threatens anticipated impacts of school sanitation and hygiene investments. To improve programming and policies, and increase the effectiveness of limited development resources, we seek to understand how and why some schools have well-managed sanitation post-intervention, while others do not. METHODS: Based on in-depth qualitative data from 16 case schools in Meherpur, Bangladesh, we employ fuzzy-set qualitative comparative analysis to identify the necessary and sufficient conditions, or combinations of conditions (referred to as pathways), that lead to either well-managed or poorly managed school sanitation. We include posited sustainability determinants from the literature and factors that emerged from the cases themselves in the analysis. RESULTS: We identified three distinct pathways sufficient to support well-managed services, providing multiple options for how well-managed school sanitation could be encouraged. Two of these are applicable to both government and non-government schools: (1) quality construction, financial community support and a champion; and (2) quality construction, financial government support, a maintenance plan and school management committee involvement. On-going financial support for operations and maintenance was identified as a necessary condition for continued service management, which was absent from many schools with poorly managed services. However, financial support was insufficient alone and other conditions are needed in conjunction, including quality construction and incentivizing conditions, such as school management committee involvement in sanitation specifically, a sanitation champion, and/or one teacher clearly responsible for toilet maintenance. Surprisingly, the number of students per toilet (ranging from 18-95 students) and toilet age (ranging from 8-32 months) had no significant effect on sanitation conditions. CONCLUSIONS: Findings corroborate those from a similar study in Belize, and comparison suggests the need for financial community support and the possibly tenuous reliance on local champions in the absence of adequate government support for operations and maintenance. Sub-determinants to the necessary conditions are also discussed which have implications for school sanitation in Bangladesh and may have broader relevance for other low-income countries though further research is needed.

Social Influence on Handwashing with Soap: Results from a Cluster Randomized Controlled Trial in Bangladesh.

We analyzed data from a cluster-randomized controlled trial conducted among 20 schools in Rajshahi, Bangladesh, to explore the role of social influence on handwashing with soap (HWWS) in a primary school setting. Using data collected through covert video cameras outside of school latrines, we used robust Poisson regression analysis to assess the impact of social influence-defined as the presence of another person near the handwashing location-on HWWS after a toileting event. In adjusted analyses, we found a 30% increase in HWWS when someone was present, as compared with when a child was alone (Prevalence ratio 1.30; 95% confidence interval: 1.14-1.47, P < 0.001). The highest prevalence of HWWS was found when both child(ren) and adult(s) were present or when just children were present (64%). Our study supports the conclusion that the presence of another individual after a toileting event can positively impact HWWS in a primary school setting.

Behavior Change without Behavior Change Communication: Nudging Handwashing among Primary School Students in Bangladesh.

Behavior change communication for improving handwashing with soap can be labor and resource intensive, yet quality results are difficult to achieve. Nudges are environmental cues engaging unconscious decision-making processes to prompt behavior change. In this proof-of-concept study, we developed an inexpensive set of nudges to encourage handwashing with soap after toilet use in two primary schools in rural Bangladesh. We completed direct observation of behaviors at baseline, after providing traditional handwashing infrastructure, and at multiple time periods following targeted handwashing nudges (1 day, 2 weeks, and 6 weeks). No additional handwashing education or motivational messages were completed. Handwashing with soap among school children was low at baseline (4%), increasing to 68% the day after nudges were completed and 74% at both 2 weeks and 6 weeks post intervention. Results indicate that nudge-based interventions have the potential to improve handwashing with soap among school-aged children in Bangladesh and specific areas of further inquiry are discussed.

Assessing new patient access to mental health providers in HMO networks.

OBJECTIVE: This study examined access to mental health providers in health maintenance organization (HMO) networks. METHODS: A telephone survey was conducted with a stratified random sample of mental health providers listed as being in a network for at lease one of six HMOs operating in Connecticut (response rate=72%; N=366). Data were collected between December 2006 and March 2007. Measures included the accuracy of network listings, acceptance rates of new patients, and reasons for not accepting new patients. Acceptance of new patients was defined as scheduling an appointment within two weeks from the time of the initial contact. Logistic regression was used to examine acceptance rates of new patients while controlling for type of provider (social worker, nurse, psychologist, or psychiatrist) and practice characteristics. RESULTS: Findings indicate that 17% of sampled HMO network listings were inaccurate. Among the providers with an accurate listing, 73% were accepting new HMO patients and 76% were accepting new self-pay patients. These aggregate acceptance rates of new patients mask differences among providers, with psychiatrists significantly less likely than other providers to accept new patients (55% of psychiatrists were accepting new patients). The most common reason for not accepting new patients was the lack of available appointments. CONCLUSION: Results indicate that access to mental health providers in HMO networks varied by type of provider. For HMO enrollees seeking treatment for mental health problems from a provider with a master's degree in social work (M.S.W. degree), network access was not a major problem. Scheduling an appointment with a psychiatrist, particularly a psychiatrist treating children only, was more difficult.