Exposure–response relationship of neighbourhood sanitation and children's diarrhoea

Objectives To assess the association of neighbourhood sanitation coverage with under‐five children's diarrhoeal morbidity and to evaluate its exposure–response relationship. Methods We used the Demographic and Health Surveys (DHS) of 29 developing countries in sub‐Saharan Africa and South Asia, conducted between 2010 and 2014. The primary outcome was two‐week incidence of diarrhoea in children under 5 years of age (N = 269014). We conducted three‐level logistic regression analyses and applied cubic splines to assess the trend between neighbourhood‐level coverage of improved household sanitation and diarrhoeal morbidity. Results A significant association between neighbourhood‐level coverage of improved household sanitation and diarrhoeal morbidity (OR [95% CI] = 0.68 [0.62–0.76]) was found. Exposure‐relationship analyses results showed improved sanitation coverage threshold at 0.6. We found marginal degree of association (OR [95% CI] = 0.82 [0.77–0.87]) below the threshold, which, beyond the threshold, sharply increased to OR of 0.44 (95% CI: 0.29–0.67) at sanitation coverage of 1 (i.e. neighbourhood‐wide use of improved household sanitation). Similar exposure–response trends were identified for urban and rural subgroups. Conclusions Our findings suggest that neighbourhood sanitation plays a key role in reducing diarrhoeal diseases and that increase in sanitation coverage may only have minimal impact on diarrhoeal illness, unless sufficiently high coverage is achieved. Objectifs Evaluer l'association entre la couverture de l'assainissement du quartier et la morbidité diarrhéique des enfants de moins de cinq ans et évaluer sa relation exposition‐réponse. Méthodes Nous avons utilisé les enquêtes démographiques et de santé (EDS) de 29 pays en développement en Afrique subsaharienne et en Asie du Sud, menées entre 2010 et 2014. Le résultat principal était l'incidence de diarrhée sur 2 semaines chez les enfants de moins de 5 ans (N = 300426). Nous avons effectué des analyses de régression logistique à trois niveaux et appliqué des splines cubiques pour évaluer la tendance entre la couverture au niveau du quartier de l'assainissement domestique amélioré et la morbidité diarrhéique. Résultats Une association significative entre la couverture au niveau du quartier de l'assainissement domestique amélioré et la morbidité diarrhéique (OR [IC95%] = 0,68 [0,62‐0,76]) a été trouvée. Les résultats des analyses de la relation avec l'exposition ont montré un seuil de couvert