Big data joins fight against childhood obesity
Increasing rates of childhood obesity mean that chronic conditions such as diabetes and cardiovascular disease are becoming ever more prevalent. “Being overweight in childhood is associated with being overweight in adulthood,” explains BigO project coordinator Anastasios Delopoulos from Aristotle University of Thessaloniki in Greece. “This is why childhood obesity is such an important health challenge for the future.” Complexities surrounding this issue, however, can make public health interventions difficult to monitor, evaluate and predict. Delopoulos notes that childhood obesity is linked not only to biology and behaviour, but also to broader social contexts. For example, a child’s economic environment, school meal or how they commute to school can all have an impact on their health.
Citizen-driven science
A multifaceted approach is therefore needed to properly address this major health issue. The BigO project set out to achieve this through the development of an innovative platform that brings together big data sets on childhood obesity. “We wanted to show how advanced technology and analytics can assist in the design and implementation of effective intervention programmes and policies,” adds Delopoulos. “This was achieved through intensive research and collaboration between software developers, data engineers and data scientists, as well as behaviour and health scientists, paediatric clinicians and teachers.” These positive interactions between public health officers and policymakers were critical. “In the later stages of the project, we were able to focus on case studies that highlighted the usefulness of our platform,” says Delopoulos. “For example, collected data was used to investigate the effect of COVID restrictions on the behaviour of kids. We were able to illustrate differences in behaviour between areas with high- and low-income populations.” In addition, over 5 500 schoolchildren from six different European cities were involved in the project as citizen scientists. These kids collected data on their behavioural patterns and local environment, using the myBigO app. “We were able to collect over 107 000 pictures, as well as huge amounts of other data,” notes Delopoulos. “We were very impressed by the efforts of these citizen scientists. The project demonstrated that kids themselves want to contribute to science, and help their peers.”
Data-led innovations
All sourced data, together with data from national statistics and Google maps, has since been made accessible via portals that are aimed at specific end users. The BigO for Health Policy Makers platform, amongst other things, provides evidence for local, regional or national health authorities. The goal is to help them design interventions that foster healthier lifestyles, through encouraging more physical activity, better dietary choices or better sleep. The BigO for Clinicians platform on the other hand is targeted at supporting health experts – clinicians, paediatricians and dieticians – by providing detailed information on children’s everyday behaviours. This can be useful for evaluating treatment plans and for optimising interventions for children who are overweight or obese. “Our hope now is that these resources will be integrated into the decision-making procedures of local governments and health experts,” says Delopoulos. “We also hope that the data collected will help us to better understand the complex causal mechanisms that determine obesity.” Furthermore, the citizen science model adopted through the project has been shown to be a novel and effective way of collecting scientific evidence and encouraging participation. This model has the potential to be taken up by other initiatives.
Keywords
BigO, obesity, childhood, diabetes, biology, economic, data, analytics, paediatric
