The study uses nationwide household survey data from India. The primary objective of this research was to identify early warning signs of child mortality that community health workers can use.
“Early identification of risk factors through the help of community health workers can go a long way in helping India reach the Sustainable Development Goals,” said Dr Dweepobotee Brahma, Assistant Professor from School of AI and Data Science, at IIT Jodhpur, in a statement.
Reducing child mortality is a specific goal under the UN-mandated Sustainable Development Goals 2030.
The study, published in the journal Applied Economics, uses a range of machine learning algorithms to assess the relative importance of characteristics such as being first-borns, being born in poorer households, and having a low birth weight.
These early-warning indicators include observable biological characteristics; demographic characteristics; and socio-economic factors of households, mothers and new-borns.
The predictors from the interpretable ML algorithms enable researchers to identify a ‘high-mortality risk’ group of mothers and infants – an important goal of India’s ‘Newborn Action Plan’.
This high-risk group comprises firstborns, mothers with prior deaths or several previous births, newborns suffering from complicated deliveries, small size at birth and unvaccinated infants.
The team said that the future goal is to extend and develop more streamlined screening criteria with the availability of more granular data with a combination of clinical and socio-economic characteristics.
The research also aims to train community health workers to use predictors as a screening mechanism to identify individuals at risk for mortality and refer them to qualified doctors for more rigorous evaluation. Early identification of risk factors will allow women and new-borns to get timely medical care and reduce the child mortality rate in India.
“We identify early newborn care, folic acid supplements and conditional cash transfer (Janani Suraksha Yojana) as the most effective policy interventions,” the researchers wrote in the paper.
“Our analysis sheds light on policy relevance and suggests some new policy prescriptions such as close monitoring of at-risk babies including females and those with small birth-size,” they added.