The role of social identity in reducing unhealthy sanitation practice: a cluster-randomized trial in Ghana

Abstract

Background: Worldwide 2.4 billion people practice open defecation (OD), which can lead to diarrheal diseases – a major cause of child mortality. To change this practice, behaviour change theories like the RANAS model suggest developing interventions based on psycho-social determinants, such as risk beliefs, attitudes, norms, ability beliefs and self-regulation. The theory of social identity furthermore proposes different dimensions (e.g. centrality) to explain why people act towards a common goal like a more hygienic environment for the whole community. This study aimed at testing the effectiveness of a behaviour change campaign on the reduction of OD, mediated by different factors from the RANAS model and moderated by social identity within communities. Methods: A cluster-randomized trial was implemented comparing different randomly assigned intervention groups with one control group in rural Ghana. 3125 households in 132 communities were interviewed by structured questionnaires. OD frequencies and all variables of the RANAS model as well as social identity were assessed before and after the implementation of a behaviour change campaign. Findings: Moderation analyses revealed, that people in the intervention communities reported significantly less OD behaviour (-53.2%) compared to the control group (-21.4%). The effect was moderated by social identity. Communities with higher levels of social identity showed larger decreases in OD than those with lower social identity. Discussion: Results imply that campaigns focusing on OD should consider the social context and not be implemented as a one-for-all approach. Analysing pre-existing social conditions could improve the effect of this intervention.