In 2013, preventable and treatable illnesses such as diarrhoea and pneumonia claimed the lives of nearly a half a million Indian children under the age of five, the study said.
The initiative -- world health partners (WHP) sky programme -- which combined franchising business models and telemedicine, was funded with grants exceeding $23 million from the Bill and Melinda Gates Foundation and other donors.
The programme sought to train and enrol thousands of informal providers and create a massive network of franchisees to deliver more effective healthcare in rural areas of Bihar in India.
"Children in rural Bihar do not receive the correct treatment, even though treatment for common illnesses like simple diarrhoea is well known, inexpensive and widely available," said lead author Manoj Mohanan, Assistant Professor at the Duke University, US.
"The WHP model of social franchising aimed to significantly improve rates of appropriate treatment of childhood diarrhoea and pneumonia," Mohanan added.
Between 2011 and 2014, the WHP Sky programme developed a network of about 9,000 franchisees, most of whom were informal sector practitioners.
However, the programme's original goal was to build a network of 20,000 practitioners.
The research showed that WHP Sky providers constituted a small share of available health care providers -- just 6 per cent of private providers.
Among all child visits to health providers in the study clusters, only 2.9 per cent of children with diarrhoea and 2.7 per cent of children with symptoms of pneumonia were brought to WHP Sky providers.
The net result was that the programme had a zero effect on the likelihood children would receive proper treatment for the targeted ailments, or on health care utilisation patterns, the researchers said.
"There are a number of likely reasons why the program failed to have an impact. Most importantly, the business model was based on untested assumptions about demand -- not only practitioners' willingness to pay to join the network, but also the level of demand for care provided by members in the network," Mohanan said.
In order to evaluate the impact of the programme, researchers collected data on 36,315 randomly selected children aged five and under in 2011 and another 31,635 children in 2014.
The study controlled for several variables, including literacy of the child's mother, the number of other children in the household, income level and caste.
The findings were published in the journal October Health Affairs.
