Enable low bandwidth mode Disable low bandwidth mode
children Bangladesh

Is exposure to animal faeces harmful to child nutrition and health outcomes?


Summary of research1

Location: Ethiopia, Bangladesh and Vietnam

What we know: Sub-clinical environmental enteric disorder (EED) is an important causal pathway from poor sanitation and hygiene to stunting; exposure to animal faeces may be an important risk factor for EED.  

What this article adds: ‘Alive and Thrive’ survey data were used for 2,214, 1,750 and 2,104 children age 6-23.9 months of age in Bangladesh, Vietnam and Ethiopia respectively to examine relationships between exposure to animal faeces and stunting, wasting and child morbidity. Animal faeces were visible in 38-42% of household compounds in all three countries. A significant association was found between observed animal faeces and maternal and child cleanliness (odds ratios between 0.52 and 0.71) and between the presence of animal faeces and child height-for-age z (HAZ) scores in Ethiopia (β = -0.22), Bangladesh (β = -0.13) and all countries (β = -0.11). An association between hygiene indicators and diarrhoea was found in Bangladesh only. Findings are consistent with an infection/EED pathway to stunting and suggest that this can be mediated by exposure to animals and their faeces. Behaviour-change communication (BCC) strategies should aim to reduce exposure to animal as well as human excreta and should aspire to “total sanitation”.

It has recently been hypothesised that exposure to livestock constitutes a significant risk factor for diarrhoea and environmental enteric disorder (EED) in young children, which may significantly contribute to undernutrition (Zambrano et al, 2014); however to date very little research has documented the extent of exposure to animal faeces and whether this exposure is associated with child nutritional status in large samples and diverse settings. This study uses data from large-scale nutrition surveys conducted in Ethiopia, Bangladesh, and Vietnam to address three research questions within these quite different socioeconomic contexts: how prevalent are observable animal faeces in household compounds; what factors are most strongly associated with observable animal faeces; and is the presence of animal faeces significantly associated with child height for age, child weight for height, and child morbidity symptoms?

The data for this study are drawn from baseline and endline surveys conducted in rural areas of Bangladesh, Ethiopia, and Vietnam in 2010 and 2014 as part of the ‘Alive and Thrive’ project aimed at reducing undernutrition and death caused by sub-optimal infant and young child feeding (IYCF) practices. The surveys include high-quality data, using large samples on child anthropometry, morbidity symptoms and their determinants, including hygiene spot-checks for the presence of animal faeces in the homestead compound as well as more standard indicators of household hygiene, such as cleanliness of mothers, young children and the homestead environment.2As the primary research interest was infants who are more likely to be left on the ground by themselves and more likely to engage in exploratory mouthing behaviours or geophagy, only the sub-sample of children 6–23.9 months of age was used. Data on anthropometry and hygiene indicators were available for 2,214, 1,750, and 2,104 mother and child dyads in Bangladesh, Vietnam, and Ethiopia, respectively. Statistical models controlled for a wide array of child, maternal and household factors known to be associated with child anthropometry/undernutrition or child illness. In order to explore external validity issues, these data were analysed for each country separately using regression models that pooled data from all three countries.


Results of the study show that animal faeces were visible in 38-42% of household compounds across the three countries and were positively associated with household livestock ownership and negatively associated with maternal and child cleanliness. Results show highly significant and robust associations between observed animal faeces and maternal and child cleanliness (with odds ratios that vary between 0.52 and 0.71), although there are no significant associations with toilet or water facilities (except in the case of hygienic toilets in Vietnam). One-sided tests from multivariate least squares models for children 6-24 months of age indicate that the presence of animal faeces is significantly and negatively associated with child height-for-age z (HAZ) scores in Ethiopia (β = -0.22), Bangladesh (β = -0.13), and in a pooled sample for all three countries (β = -0.11), but not in Vietnam. No significant associations were found between weight-for-height z (WHZ) scores and animal faeces in any of the countries. An association between hygiene indicators and diarrhoea was found in Bangladesh only, the odds ratio for which suggests that the presence of animal faeces increases the risk of diarrhoea by 25%.

Discussion and conclusions

Livestock ownership may have positive effects on child growth through improvements in household socioeconomic status, animal-sourced food consumption and other mechanisms (transportation, social status, women’s empowerment, collateral for credit), but negative impacts through increased risk of infections and/or EED. Consistent with an infection/EED pathway, in Bangladesh and Ethiopia the presence of animal faeces in the compound is negatively associated with child HAZ. In Ethiopia, livestock ownership was also positively associated with child HAZ outcomes. However, there were no similar associations in the Vietnam sample; this may be due to lower prevalence of EED or diarrhoeal infection in general and better care practices for children.

The authors conclude that these findings contribute to a growing body of evidence suggesting that child anthropometry and health outcomes in developing countries may be adversely affected by exposure to animals and their faeces. This risk stems from several factors, including the widespread ownership of livestock and pets in developing countries; the lack of housing and enclosure structures for livestock that separate animals from household members (e.g. scavenging poultry systems); poor hygiene knowledge and practices; the common practice of leaving children to sit or play on homestead floors with little monitoring (especially in warmer climates); and the very high concentration of potentially harmful bacteria in animal faeces.

This literature points to the need to re-evaluate several core features of conventional water, sanitation and hygiene (WASH) strategies. First, most WASH strategies have focused on influencing diarrhoea and morbidity outcomes rather than on sub-clinical EED. Second, most WASH strategies have focused behaviour-change communications on a small number of key messages with emphasis on reducing exposure to human faeces, assuming this poses a greater health burden (to children especially) than exposure to animal faeces. While open defecation remains a major health concern in much of the developing world, exposure to animal faeces is probably more common and potentially also hazardous for child nutrition and health outcomes. Behaviour-change communications strategies should therefore consider reducing exposure to all excreta, human and animal, and aspire to genuine “total sanitation.”


1Headey DD, Nguyen PH, Kim SS, Rawat R, Ruel MT and Menon P. Is exposure to animal feces harmful to child nutrition and health outcomes?: A multicountry observational analysis. The American Journal of Tropical Medicine and Hygiene. Article in Press. First available online December 19, 2016. https://doi.org/10.4269/ajtmh.16-0270

2Interviewers recorded observations of cleanliness of the mother and her child (hair, hands, faces and clothing) on a three-point scale of clean, dusty or dirty and yes/no observations of household cleanliness (based on presence of human faeces, rubbish, dirty clothes and a cover on the main drinking water container).


Zambrano LD, Levy K, Menezes NP, Freeman MC, 2014. Human diarrhea infections associated with domestic animal husbandry: a systematic review and meta-analysis. Trans R Soc Trop Med Hyg 108: 313–325




About This Article

Article type: 
Original articles

Download & Citation

Recommended Citation
Citation Tools