Independent and combined effects of nutrition and sanitation interventions on child growth in rural Cambodia: a factorial cluster-randomized controlled trial

BackgroundChildhood exposure to enteric pathogens associated with poor sanitation contributes to undernutrition, associated with adverse effects later in life. This trial assessed the independent and combined effects of nutrition and sanitation interventions on child growth outcomes and enteric pathogen infection in rural Cambodia, where the prevalence of childhood stunting remains high. MethodsWe conducted a factorial cluster-randomised controlled trial of 4,015 households with 4,124 children (1-28 months of age at endline) across three rural provinces in Cambodia. Fifty-five communes (clusters) were randomly assigned to a control arm or one of three treatments: a nutrition-only arm, a sanitation-only arm, and a combined nutrition and sanitation arm receiving both treatments. The primary outcome was length-for-age Z-score (LAZ); other outcomes included weight-for-age Z-score (WAZ), weight-for-length Z-score (WLZ), stunting, wasting, underweight, and caregiver-reported diarrhoea. We assayed stool specimens from a subset of all children (n = 1,620) for 27 enteric pathogens (14 bacteria, 6 viruses, 3 protozoa, and 4 soil-transmitted helminths) and estimated effects of interventions on enteric pathogen detection and density. Analysis was by intention-to-treat. The trial was pre-registered with ISRCTN Registry (ISRCTN77820875). FindingsSelf-reported adherence was high for the nutrition intervention but uptake was low for sanitation. Compared with a mean LAZ of -1.04 (SD 1.2) in the control arm, children in the nutrition-only arm (LAZ +0.08, 95% CI -0.01-0.18) and combined nutrition and sanitation arm (LAZ +0.10, 95% CI 0.01-0.20) experienced greater linear growth; there were no measurable differences in LAZ in the sanitation-only arm (LAZ -0.05, 95% CI -0.16-0.05). We found no effect of any intervention (delivered independently or combined) on either enteric pathogen frequency or pathogen load in stool. Compared with a mean WAZ of -1.05 (SD 1.1) in the control arm, children in the nutrition-only arm (WAZ +0.10, 95% CI 0.00-0.19) and combined intervention arm (WAZ +0.11, 95% CI 0.03-0.20) were heavier for their age; there was no difference in WAZ in the sanitation-only arm. There were no differences between arms in prevalence of stunting, wasting, underweight status, one-week period prevalence of diarrhoea, pathogen prevalence, or pathogen density in stool. InterpretationImprovements in child growth in nutrition and combined nutrition and sanitation arms are consistent with previous efficacy trials of combined nutrition and sanitation interventions. We found no evidence that the sanitation intervention alone improved child growth or reduced enteric pathogen detection, having achieved only modest changes in access and use. FundingUnited States Agency for International Development (USAID), contracts AID-OAA-M-13-00017 and AID-OAA-TO-16-00016. The contents of this publication are the sole responsibility of the authors and do not necessarily reflect the views of USAID or the United States Government.