Hindus are, on average, richer and more educated than Muslims. But oddly, the child mortality rate for Hindus is much higher. All observable factors say Hindus should fare better, but they don't. Economists refer to this as the Muslim mortality puzzle.

In a new study, researchers believe that they may have found a solution to the puzzle. And, surprisingly, the solution lies in a single factor – open defecation.

Asian enigma

Today more than 60 crore Indians – more than half the population – defecate in the open. Recent research has shown that open defecation leads to higher child mortality rates and stunted growth. Some regions in India do worse than sub-Saharan Africa on those parameters.

Dean Spears at the Delhi School of Economics had been studying the effects of open defecation in India, which led him to a solution to the "Asian engima" – why are Indian children shorter, on average, than African children, even though people are poorer, on average, in Africa. The height of children is one of the most important measures of their wellbeing, and Spears found that it is severely affected by open defecation.

Poor children play in the same fields where their friends and parents relieve themselves. Disease-causing bacteria and parasites then end up on these children's hands and feet, and eventually in their food and drink. While some of these germs make people visibly sick, the symptoms of others may not be visible until later in life. Many years of exposure to such bugs can cause enteropathy – a chronic intestinal problem that prevents proper absorption of calories and nutrients, leading to stunted growth.

Religious issues

When Spears, along with Arabinda Ghosh and Oliver Cumming published, their work on stunted growth, they relied on data derived from the HUNGaMA (hunger and malnutrition) survey, which looked at thousands of people in a few representative districts throughout India.

For the new study, however, a more reliable data source is the government-run National Family Health Survey, which collects information from women aged 13 to 49. Apart from personal information, the survey asks them to report birth and death histories, as well as information about household assets, infrastructure, and health habits, including information about toilet facilities and their usage. The dataset Spears and Geruso selected had about 310,000 Hindu and Muslim children from three survey rounds conducted in 1992/3, 1998/9, and 2005/6.

Using all those factors, statistical analysis revealed that open defecation may contribute to the difference in child mortality rates. "We show that the entire gap between Muslim and Hindu child mortality can be accounted for by a particular kind of sanitation externality," wrote Spears and Michael Geruso, professor of economics at the University of Texas at Austin, in their working paper. They found that Muslims, regardless of income, were 20% more likely to use toilets than Hindus.

More convincingly, the analysis showed that moving from a locality where everybody defecates in the open to a locality where no one does is associated with a larger difference in child mortality than moving from the bottom quintile of asset wealth to the top quintile.

Could differences in open defecation be just one factor among other differences in general hygiene practices? Probably not. Spears and Geruso found no systematic differences among the religious groups when it came to hand washing or water purification techniques.

It was important to be sure that there were no other systematic differences among religious practices of Hindus and Muslims that could have contributed. In "the rare places where Hindus are less likely to defecate in the open than Muslims", Spears and Geruso found that the advantage reverses – child mortality among Hindus turns out to be lower than that among Muslims.

The analysis also showed that Hindu households residing in villages with majority Muslim population experienced lower child mortality than Hindus living among other Hindus. The reverse also held true – Muslims living among Hindus had higher mortality rates than if they lived among Muslims.

"This is consistent with the findings that it is not your own hygiene practices but that of your neighbours which matter most," said Sandy Cairncross, professor of environmental health also at the London School of Hygiene and Tropical Medicine, who has worked in India to increase toilet use.

A positive sign is that Spears and Geruso found that breastfeeding could counteract the negative effects of poor sanitation. The reason is that breastfed children are less exposed to disease-causing germs directly through contaminated water and food, even if the mother was exposed to those germs directly.

This study only draws a correlational link between open defecation and child mortality rates among religious groups. "The causes could lie in the differences among Hindu and Muslim religious codes," said Jeroen Ensink, lecturer in public health engineering at the London School of Hygiene and Tropical Medicine, who has worked on sanitation practices in India.

Behavioural change

What is clear is reducing the number of those who defecate in the open will have a large impact on the health of individuals and the communities they live in. However, changing behaviour, even if the benefits are so obvious, is easier said than done. The Indian government has spent crores of rupees to build toilet facilities through programmes such as Total Sanitation Campaign, but they often remain unused or are repurposed.

"Just because they have a latrine, people don't stop defecating in the open," said Cairncross. His work in Asia and Africa has revealed that people don't install toilet facilities for health reasons. Instead the common reasons cited are security, social prestige and dignity.

While education can help change behaviour and hygiene practices, it is a slow process. The problem is so large and so urgent that Ensink thinks that policies need to be set in place to drive behavioural change.

This article has been edited to remove inaccuracies.