The field of early childhood development is one of extremes.
In rich countries with bountiful resources, policy-makers and experts debate the merits of universalizing pre-kindergarten, or how play can benefit toddlers. In many poor countries, where resources are scarce, the discussion often revolves around reducing infant deaths, or providing kids access to basic services like healthcare. Cheap, scalable technology has a huge potential to narrow this gap—but in many cases, the cutting-edge innovations are found in the first category of countries rather than the second.
A notable exception is Malawi. The East African country has become a hub for projects that use technological innovation to try meet public health or humanitarian goals. One example is an ambitious experiment to use wearable devices to remotely capture high-frequency data on children’s development and use it to improve those children’s outcomes.
The project, currently being piloted by the Center for Child Well-Being & Development (CCWD) at the University of Zurich and UNICEF, is in its infancy. But if it works, it has the potential to revolutionize the way children’s health and wellbeing are measured in rural settings. It could also help politicians and international organizations create customized interventions that can help children thrive based on their individual needs.
The problem
Malawi faces incredible barriers to health care. Half of its nearly 18 million people have to travel one hour, often in animal-drawn carts, to get to their nearest community health center, and two-and-a-half hours to get to a hospital. Over half of Malawian women, and 39% of men, say they can’t afford to go to a hospital. There are fewer than 0.1 doctors (pdf) for every 1,000 people and 6.5% of children die before turning five.
These barriers make it nearly impossible to gather reliable data on children’s health, get ahead of disease outbreaks, and design effective prevention programs to tackle some of the country’s major health problems, like malaria. “In this context,” writes the CCWD research team, “decision-making can be like driving through a dense fog in the middle of the night.”
New technologies represent a huge untapped opportunity to change this context. Malawi recently became the first African country to create a “drone corridor,” an area in which the government and UNICEF deploy drones to collect and monitor topographical data during natural disasters, extend internet coverage during emergencies, deliver medical supplies, and more. Building on this success, the team at CCWD partnered with UNICEF last year (pdf) to adapt existing wearable devices to a Malawian setting.
The (possible) solution
This longitudinal study, which CCWD says is the first of its kind (pdf), aims to monitor children in 6,000 rural households over 18 years. It’s a massive undertaking that involves multiple phases of development. Right now, it’s just in the piloting and set-up phase.
The researchers want to collect “the smallest set of markers that can predict the largest set of critical health and child development conditions.” Community health workers and selected individuals known as “community watchers” will start by collecting biomarker data from wearable devices worn by two groups of kids: those who get a check-up at their local health center and those selected to receive home visits.
The researchers hope they can apply machine learning to data collected from the wearables, as well as diagnostic data from health workers, to predict and send warnings about potential disease outbreaks at the community level. They also hope to create customized child development interventions which could ultimately “predict the impacts of each service for each child profile at each development stage.”
The wearables will collect basic indicators like skin temperature, heart and respiratory rates. They’ll also measure more “specific” and complex indicators like oxygen saturation, electro-encephalogram (EEG), and facial coding (figuring out people’s emotions through their facial expressions), which could theoretically detect psychological pathologies. For example, EEG readings can be used to detect attention deficit and hyperactivity disorder, and oxygen saturation can help spot sleep apnea syndrome, which impacts kids’ cognition.
This step is harder than it sounds. The wearable devices need to work in villages that often don’t have electricity or internet; the kids have to wear the devices; the community health workers have to diligently collect the information each week; and the communities need to support the research.
To tackle these issues, the researchers, in partnership with the government, will give solar battery chargers to the health workers and watchers to recharge the wearable devices and for their personal use. They plan to organize sensitization meetings with participating households, local officials, and village chiefs to get the support of the communities, and are working with a bioethicist to create guidelines for data privacy and consent.
“There are many challenges in actually pulling this off,” says project leader Guilherme Lichand, assistant professor of child wellbeing and development at the University of Zurich. “But I think we have the right partners in place to make it happen.”
If this project accomplishes its goals—and that’s a big if—it could revolutionize the way early childhood development and public health interventions are carried out in developing countries, which currently rely on low-frequency, often poor-quality, data. This project proposes a different model, one that relies on high-frequency, high-quality, and personalized data to target the kids who need it most with the services that will best help them.
The researchers will know more in January 2020, when the results of the pilot test come in. “I think it’s feasible,” says Lichand. But “if it were easy, it would be done.”
Read more from our series on Rewiring Childhood. This reporting is part of a series supported by a grant from the Bernard van Leer Foundation. The author’s views are not necessarily those of the Bernard van Leer Foundation.