Poor nutrition among children has long been understood to stunt their growth – but it is now widely recognized that this is not the full explanation for the rates of stunting across the globe. This column highlights the importance of other factors driving the childhood growth process, in particular practices related to water, sanitation and hygiene (WASH). The challenge for policy-makers is to find ways to improve both nutrition and WASH practices simultaneously.
Over one in five children under the age of five has stunted growth, technically measured as low height for a given age. Research has linked stunting with poor human capital development, health and labor market outcomes in later years, implying huge losses of human and economic potential.
The underlying causes of poor early childhood health and stunted growth are multifaceted, and there is a growing recognition that the rates of stunting observed across the globe cannot be explained by under-nutrition alone.
The role of hygienic environments, shaped through water, sanitation, and hygiene (WASH) practices, in particular has been increasingly emphasized. A hygienic environment can improve child outcomes through reduced incidence of illnesses such as diarrhoea, which is not only a primary cause of mortality but also diverts nutrients intake away from growth processes.
Furthermore, a hygienic environment can prevent the development of conditions such as environmental enteric dysfunction (EED), an asymptomatic infection that limits the absorption of nutrients, thereby constraining growth despite eating well.
Evidence from medical research indicates that nutrition and WASH practices should interact with one another in the formation of child health. But recent randomized control trials (RCTs) exploring this link have had mixed results.
For example, two large studies – the WASH benefits trials in rural Kenya and Bangladesh, and the SHINE trial in rural Zimbabwe – examined the interaction between nutrition and WASH by providing combinations of nutritional supplements and elementary WASH interventions, such as building toilets, providing handwashing stations and soap, and using chlorine to purify drinking water.
All three trials find no impacts of the WASH interventions on child height, when implemented as standalone interventions (for example, building toilets only), or in combination. Further, none of the trials find evidence that nutritional supplements and WASH interventions interact with one another in shaping children’s height.
Additional analysis of the data from the trials highlights the difficulty of quantifying this interaction through RCTs: implemented interventions must first generate a sufficiently large improvement in a child’s hygiene environment. The elementary WASH interventions implemented in the trials may not have achieved this.
In a new study, we take a different approach to the same question, this time using high frequency longitudinal data from the Philippines from 1983 to 1986 to estimate the individual and combined contribution of nutrition and WASH to the formation of child height and weight in the first two years of life. Our analysis accounts for the fact that nutritional choices and WASH practices might be based on shocks and preferences that are unobserved by the researcher, which would lead to incorrect estimates if ignored.
We find that protein intake and better WASH individually lead to better child height and weight outcomes. Furthermore, improved WASH generates a small but robust and statistically significant improvement in the effect of protein on child height and weight.
A child at the 10th percentile of the distribution of WASH practices would be 2.57 cm taller by the age of two years if he or she were given an additional egg a day between the ages of six months and two years. For a child at the 90th percentile of the WASH distribution, the increase would be 2.73cm.
This result provides one explanation for the puzzle of stubbornly high stunting rates in some countries, despite significant income growth.
A question remains though, of why our study finds evidence of a positive interaction between WASH and nutrition, whereas the RCTs do not. There are additional features of our study that are likely to explain some of this difference.
- First, households in our sample predominantly live in urban rather than rural areas. In these more densely populated contexts, better elementary WASH practices, and interventions of the type implemented in the RCTs, could reduce pathogen exposure sufficiently to translate into measurable impacts on child health.
- Second, the interaction effects that we identify are small in magnitude. Detecting such an effect size requires extremely large sample sizes in an RCT framework, which can be difficult to achieve if interventions are randomized at the village level, as was the case in the earlier trials.
- Third, for ethical reasons, RCTs can only encourage, rather than compel, participants to change their behavior. Imperfect compliance with interventions compounds the problem of a lack of statistical power.
Taking this research together, our study suggests that better child health can come from a combination of improvements in both nutrition and WASH behaviours. These might come from a variety of innovative approaches to encourage sustained improvements in WASH infrastructure and behaviour, and reduce children’s exposure to pathogens, in both rural and urban contexts.
Laura Abramovsky is a Research Associate of the Institute for Fiscal Studies. Her current work focuses on tax and social protection policy and programme evaluation in developing countries.
Britta Augsburg is Associate Director at the Institute for Fiscal Studies. Her research concentrates on understanding the effectiveness of programs and policies that tackle constraints to productivity of poverty affected individuals and households, with a particular focus on credit and technology adoption constraints.
Pamela Jervis is an Assistant Professor of Economics at the Department of Industrial Engineering, University of Chile and a Research Associate of the Institute for Fiscal Studies in the UK.
Bansi Malde is a Lecturer in Economics at the University of Kent and a Research Associate at the Institute for Fiscal Studies.
Angus Phimister is researcher economist at IFS and works in the Centre for the Evaluation of Development Policies (EDePo).