Current Developments in Nutrition authors believe big data may reveal new causal associations
According to the World Health Organization (WHO), in 2016 more than 1.9 billion adults, 18 years and older, were overweight. Of these, over 650 million were affected by obesity. These statistics translate into 39% of adults with overweight and 13% with obesity.
Overweight and obesity are risk factors for a broad set of non-communicable diseases, including cardiovascular disease, type 2 diabetes, and some cancers. WHO underscores that “most of the world’s population live in countries where overweight and obesity kill more people than underweight.” In the United States alone, the Centers for Disease Control and Prevention (CDC) estimates that in 2019 “medical costs for adults with obesity were $1,861 higher than medical costs for people with healthy weight.”
The COVID-19 pandemic has exacerbated the problem: obesity complicates COVID-19 treatment, recovery, and vaccine efficacy. As a result, studies estimate that people with obesity were 1.5 times more likely to die of COVID-19.
Given rising obesity rates around the world alongside the persistence of the COVID-19 pandemic, honing in on the precise causes and drivers of obesity is critical. The authors of “Towards Systems Models for Obesity Prevention: A Big Role for Big Data,” published in Current Developments in Nutrition (CDN), believe that big data, the ability to computationally analyze extremely large data sets to reveal patterns, trends and associations, may make that possible.
Despite copious research, the relationship among the various causal factors of obesity is still not well understood. According to the authors of this CDN article, “while recent studies have made advances in capturing a wider range of systemic factors using mixed-methods approaches, static or traditional survey-based approaches fail to appropriately capture the complexity of factors and dynamic interactions between the individual and their environment.” The authors believe that “the collection of big data has begun to allow the exploration of causal associations between behavior, built environment, and obesity-relevant health outcomes.” Eventually, they believe that “big data have the potential to identify intervention points for obesity prevention or management in vivo, across different environmental settings, as well as create data-driven causal maps of obesity determinants.”
The authors point to three areas in particular where big data may be employed to advance our understanding of the underlying causes of obesity, presenting an overview of mobile sensing, citizen science, and artificial intelligence. Rather than replace traditional obesity research methods, these three facets of big data can complement them, improving data quality and making resource-intensive surveillance efforts more efficient. For example, the authors believe big data have the potential to complement single-study-derived insights into food intake, eating rate, and satiety by linking these findings to “relevant bio-social individual characteristics such as history of breastfeeding, health co-morbidities, and subjectively reported satiety and motivations for food choice.”
The authors present BigO, Big Data against Childhood Obesity, as a case study of how big data contribute to evidence-informed obesity policy making. A Horizon2020 project, BigO was launched in 2016, using a mobile health and citizen science approach to capture environmental features and link them to individual behavior and weight status of children. The project involved more than 5,500 children from four European countries who acted as “citizen scientists,” submitting data via a smartwatch or smartphone application. The data collected from BigO is currently available via a public health dashboard and is “intended for eventual public health use as a policy planning tool.”
Despite big data’s potential, the authors note that “several important caveats and limitations surround the use of big data, particularly when tracking multiple individual characteristics or behaviors across time and space.” In particular, there are significant privacy concerns when employing mobile tools, especially among children. Nonetheless, the authors believe that the collection and analysis of big data is critical “to arrive at a more advanced and integrated understanding of the inherently complex etiology of obesity.”