Zoonotic and Environmental Sources of Infant Enteric Pathogen Infections Identified with Longitudinal Sampling

Abstract

Many enteric pathogens that infect young children can be zoonotic, yet the exposure risk of domestic animals living in close proximity to young children is poorly understood. Here, we longitudinally measured 33 enteric pathogens in child stool, animal feces, and the household environment (n = 28,743 pathogen-sample observations) to investigate pathogen transmission between animals and children under two in pastoralist communities in rural Northern Kenya. Children were typically infected with 1 enteric pathogen by 3 months of age, and pathogen burden increased with age; 85% of enteric pathogens detected in child stool were also detected in animal feces. New infections in children were associated with preceding household detection of the same pathogen in soil (Odds ratio: 8.8, 95% confidence interval: 3.3-23) and on child hands (odds ratio: 5.0, 95% confidence interval 1.1-17). Regression modeling revealed transmission of pathogens from poultry, dog, and ruminant feces to household soil, and between child hands and child stool. Our results provide new evidence that domestic animals in the household environment contribute to early life enteric pathogen exposure, and that child hand hygiene could substantially prevent animal-child transmission.

Keywords: environmental microbiology; exposure assessment; zoonotic pathogen transmission.

1

Conceptual diagram of study design. Twenty-five children were enrolled within each target age group (0 to <3, 3 to <6, 6 to <12, and 12 to <24 months). Children <6 months of age and a subset (n = 10) of children between 6 and 12 months of age were sampled longitudinally across four total visits with 7 days between each visit; all other children >6 months were sampled once. The sampling symbol represents sample collection of soil, drinking water, child hand rinse, fomite rinse, food, animal feces, and child and caregiver stool.

2

Enteric pathogen profiles across sample types: (A) proportion of samples positive for each pathogen across sample types (left) and in child stool by age (right). Sample sizes for each group are indicated at the top of the figure. (B) Number of pathogens (out of a maximum of 33 pathogens) present by sample type and child age. * p < 0.05 ** p < 0.01 *** p < 0.001

3

Distribution of quantities among positive samples measured by TAC by sample type (n = 732 samples).

4

Odds ratios of pathogen detection in animal feces or environmental samples and subsequent new infection with the same pathogen in children. Bars represent 95% confidence intervals. The dotted red line indicates an odds ratio of 1. The environmental samples category includes hands, soil, water, and food. Food alone is not displayed because of large uncertainty around the point estimate.

5

Visualization of temporal associations of the same pathogen present in hosts and environmental samples. Circles represent the sample type. Arrow widths represent the number of pathogens for which detection in the sample type was associated with detection in the receiving sample type on the next visit within the same household (logistic regression models, association is included if odds ratio is significant p < 0.05; p-values not adjusted for multiple comparisons).