Effect of a Behaviour Change Intervention on the Quality of Peri-Urban Sanitation in Lusaka, Zambia: A Randomised Controlled Trial

"It is possible to improve the structural quality and cleanliness of shared sanitation by targeting landlords with a scalable, theory-driven behaviour change intervention without subsidy or provision of the relevant infrastructure."

Data show that the number of people using shared toilets increased from 204 million to 465 million over the period 1990 to 2015 in Africa alone. This study investigated to what extent sanitation could be improved by the residents of an informal settlement in Zambia themselves, through behaviour change promotion alone, in the absence of institutional change or financial subsidy.

The study took place in Bauleni, an informal settlement in southeast Lusaka with approximately 4,000 plots. Formative research showed that most toilets were shared by multiple households, with a resident landlord responsible for toilet provision for the households living on each plot. The poor quality of toilet provision appeared to relate to the fact that landlords undervalued their tenants' willingness to pay for quality improvements. Landlords were unaware of how improving sanitation could reduce the burden of managing a plot both financially and socially.

Working with a creative agency, the researchers used the behaviour-centred design (BCD) approach to design an intervention designed to reach the landlords called the Bauleni Secret. BCD is based on a reinforcement learning paradigm and includes a design process that assesses existing knowledge, fills gaps in knowledge through formative research, and then creates, delivers, and evaluates the intervention.

Bauleni Secret was marketed as a secret-society-style intervention in which landlords were invited to participate in a selective programme that would share secrets about how to build their wealth and bring peace to their plots. Participants attended a series of meetings at 4 venues in Bauleni that promoted one of 4 outcomes: sealed toilets, locks on the inside of toilet doors, locks on the outside of toilet doors, and a pamodzi cleaning rota. A pamodzi (meaning "togetherness" in Nyanja) rota was a specific kind of cleaning system, designed and branded for the intervention that, in part, was meant to encourage all individuals to participate by making their failure to participate more noticeable.

Led by 4 pairs of facilitators, the meetings featured a trained actor and a neighbourhood health committee member - a pairing that was designed to allow entertaining non-health messages to be presented. Participants were shown videos of tenants' perspectives on the outcome of focus to provoke discussion; demonstrations ("emo-demos" - see photo above) and games were used to re-evaluate the benefits of solving the sanitation issues being discussed; and practical sessions were run that featured commitments to a so-called improvement buddy. Follow-up monitoring visits helped participants troubleshoot any barriers. Landlords were instructed to get tenant signatures on a card handed out at the end of each meeting, verifying that the relevant improvement had been made. To stimulate attendance, a prize draw for one of three smartphones was offered to landlords who attended all 4 meetings and had 4 improvement cards signed by their tenants.

Between June 9 and July 6 2017, 1,085 landlords were enrolled and randomly assigned to the intervention (n=543) or the control group (n=542). The intervention was delivered from August 1 2017, and evaluated from February 15 to March 5 2018. Analysis was based on the 474 intervention and 454 control landlords surveyed at study end. The 4 primary outcomes were: having a rotational cleaning system in place (to improve hygiene); having a solid door on the toilet used by tenants with an inside lock (for privacy); having an outside lock (for security); and having a sealed toilet (to reduce smell and contamination).

The intervention was associated with improvements in the prevalence of cleaning rotas (relative risk (RR) 1.16, 95% confidence interval (CI) 1.05-1.30; p=0.0011), inside locks (1.34, 1.10-1.64; p=0·00081), outside locks (1.27, 1.06-1.52; p=0·0028), and toilets with simple covers or water seals (1.25, 1.04-1.50; p=0.0063).

Put another way, the researchers found that plots in the intervention group had significantly better quality toilets compared with those in the control group across all 4 dimensions of quality improvement, with an approximate 10 percentage-point increase in the intervention group in each. In addition, landlords in the intervention group were more slightly likely to have saved money, purchased materials, or begun construction of toilet improvements. These improvements were generally not expensive to implement.

The researchers point to several reasons for the intervention's apparent success: the use of a systematic and theory-based process to understand the problem, the isolation of key behaviours to be changed, and the creative design and careful evaluation of the intervention. Delivering surprising messages to landlords, incorporating social learning and influence through landlord group meetings, and accountability mechanisms to facilitate behaviour were also probably key in ensuring that desired improvements were carried out.

Although "future work is needed on the question of sustainability, as well as on the potential health impact of a range of toilet improvements", the researchers conclude: "The private sector...has a part to play in providing appropriate products, which will hopefully be encouraged by increasing acknowledgment of willingness to pay for sanitation. The results of this trial suggest that in conjunction with such efforts, a theory-based, creatively designed demand-side-only intervention could play an important part in improving the quality of peri-urban sanitation."