I get height with a little help from my friends: herd protection from sanitation on child growth in rural Ecuador

Abstract

Background: Infectious disease interventions, such as vaccines and bed nets, have the potential to provide herd protection to non-recipients. Similarly, improved sanitation in one household may provide community-wide benefits if it reduces contamination in the shared environment. Sanitation at the household level is an important predictor of child growth, but less is known about the effect of sanitation coverage in the community.

Methods: From 2008 to 2013, we took repeated anthropometric measurements on 1314 children under 5 years of age in 24 rural Ecuadorian villages. Using mixed effects regression, we estimated the association between sanitation coverage in surrounding households and child growth.

Results: Sanitation coverage in the surrounding households was strongly associated with child height, as those with 100% coverage in their surroundings had a 67% lower prevalence of stunting [prevalence ratio (PR) 0.32, 95% CI 0.15-0.69] compared with those with 0% coverage. Children from households with improved sanitation had a lower prevalence of stunting (PR 0.86, 95% CI 0.64-1.15). When analysing height as a continuous outcome, the protective effect of sanitation coverage is manifested primarily among girls during the second year of life, the time at which growth faltering is most likely to occur.

Conclusions: Our study highlights that a household's sanitation practices can provide herd protection to the overall community. Studies which fail to account for the positive externalities that sanitation provides will underestimate the overall protective effect. Future studies could seek to identify a threshold of sanitation coverage, similar to a herd immunity threshold, to provide coverage and compliance targets.

Keywords: Herd Protection; Malnutrition; Sanitation; Stunting.

Figure 1.

Distribution of coverage of improved sanitation within 500 m of the household in rural northern Ecuador, 2008-11.

Figure 2.

Predicted height in cm among females and males by coverage of sanitation in the 500 m surrounding the household, northern Ecuador, 2008-11. Multilevel model includes a three-way interaction between child’s age, child’s sex and sanitation coverage, and is adjusted for the following time-varying covariates: household sanitation, household education, household wealth and wealth in the surrounding households. Age was included as a continuous variable using a restricted cubic spline with knots at 0.5, 1, 1.5, 2.5 and 4 years. Model also includes random intercept and random age slope for each child.

Figure 3.

Predicted prevalence of moderate or severe stunting (and 95% confidence limits) by level of sanitation coverage within 500 m of the household, rural northern Ecuador, 2008-2011. Predictions are adjusted for age, household sanitation, household education, household wealth, and neighborhood wealth. Model also includes random intercept for each child.