Behaviour change intervention to improve shared toilet maintenance and cleanliness in urban slums of Dhaka: a cluster-randomised controlled trial

OBJECTIVES Shared toilets in urban slums are often unclean and poorly maintained, discouraging consistent use and thereby limiting impacts on health and quality of life. We developed behaviour change interventions to support shared toilet maintenance and improve user satisfaction. We report the intervention effectiveness on improving shared toilet cleanliness. METHODS We conducted a cluster-randomised controlled trial among users of 1226 shared toilets in 23 Dhaka slums. We assessed baseline toilet cleanliness in January 2015. The six-month intervention included provision of hardware (bin for solid waste, 4 l flushing bucket, 70 l water reservoir), and behaviour change communication (compound meetings, interpersonal household sessions, signs depicting rules for toilet use). We estimated the adjusted difference in difference (DID) to assess outcomes and accounted for clustering effects using generalised estimating equations. RESULTS Compared to controls, intervention toilets were more likely to have water available inside toilet cubicles (DID: +4.7%, 95% CI: 0.2, 9.2), access to brush/broom for cleaning (DID: +8.4%, 95% CI: 2, 15) and waste bins (DID: + 63%, 95% CI: 59, 66), while less likely to have visible faeces inside the pan (DID: -13%, 95% CI: -19, -5), the smell of faeces (DID: -7.6%, 95% CI: -14, -1.3) and household waste inside the cubicle (DID: -4%, 95% CI: -7, -1). CONCLUSIONS In one of few efforts to promote shared toilet cleanliness, intervention compounds were significantly more likely to have cleaner toilets after six months. Future research might explore how residents can self-finance toilet maintenance, or employ mass media to reduce per-capita costs of behaviour change.