For numerous years, health professionals have been worried about “food deserts,” locations where inhabitants lack superior nutritional choices. Presently, a study spearheaded by MIT focusing on three major global metropolises employs a novel, detailed approach to scrutinize the matter, concluding that a scarcity of nutritious dining options nearby correlates with obesity and various health outcomes.
Instead of merely charting geographical locations, the scientists assessed the dietary worth of millions of food items across approximately 30,000 restaurant menus, yielding a more exact analysis of the relationship between communities and nutrition.
“We demonstrate that what is offered in a restaurant directly relates to individuals’ health,” states MIT researcher Fabio Duarte, co-author of a recently released paper detailing the study’s findings. “The food landscape is significant.”
The open-access publication, “Data-driven nutritional assessment of urban food landscapes: insights from Boston, London, Dubai,” was issued this week in Nature: Scientific Reports.
The co-authors include Michael Tufano, a PhD candidate at Wageningen University in the Netherlands; Duarte, associate director of MIT’s Senseable City Lab, which employs data to analyze cities as dynamic systems; Martina Mazzarello, a postdoctoral researcher at the Senseable City Lab; Javad Eshtiyagh, a research fellow at the Senseable City Lab; Carlo Ratti, a professor of practice and director of the Senseable City Lab; and Guido Camps, a senior researcher at Wageningen University.
Analyzing the menu
To perform the study, the researchers scrutinized menus from Boston, Dubai, and London during the summer of 2023, establishing a database of millions of items available from popular food-delivery services. The team then assessed the food items according to ratings from the USDA’s FoodData Central database, an information repository containing 375,000 types of food products. The study utilized two main metrics: the Meal Balance Index and the Nutrient-Rich Foods Index.
The researchers evaluated around 222,000 menu items from over 2,000 restaurants in Boston, approximately 1.6 million menu items from about 9,000 restaurants in Dubai, and nearly 3.1 million menu items from roughly 18,000 restaurants in London. In Boston, around 71 percent of the items were included in the USDA database; in Dubai and London, those figures were 42 percent and 56 percent, respectively.
The team then assessed the nutritional quality of the menu items and correlated the food data with health outcome statistics from Boston and London. In London, they identified a clear association between neighborhood menu selections and obesity rates, with a somewhat less distinct link observed in Boston. Areas offering food choices rich in dietary fibers, sometimes coupled with fruits and vegetables, typically present better health metrics.
In Dubai, the researchers lacked similar health data but observed a strong link between rental prices and the nutritional quality of food at the neighborhood level, indicating that wealthier residents enjoy superior nutritional options.
Revisiting the food landscape
By executing the study in this manner, the researchers added an analytical layer to previous investigations of food deserts. While earlier research has paved the way by pinpointing neighborhoods with inadequate food access, this analysis provides a more detailed evaluation of actual consumption patterns. The research aims to assess the intricate mix of food available within a given area, even in locations with limited choices.
“We were not content with the notion that if fast food predominates, it’s a food desert, but if you have a Whole Foods, it’s not,” Duarte articulates. “The reality is more nuanced.”
For the researchers at the Senseable City Lab, this study introduces a new approach that aids in comprehending urban dynamics and the implications of the urban environment on health. Previous lab investigations often concentrated on topics such as urban mobility while extending to issues including mobility and air pollution, among others.
Examining food and health at the neighborhood level exemplifies the multitude of ways in which data-rich environments can be scrutinized in detail.
“When we initially began our work on cities and data, the data resolution was exceedingly low,” Ratti reflects. “Today, the volume of data is vast, creating a significant opportunity to analyze cities and understand the impact of the urban environment as a major determinant of health. We regard this as a new frontier for our laboratory. It’s remarkable how we can now examine this with such precision in urban areas.”