Today's news predicts obesity three years from now

When it comes to obesity rates, journalists appear to hold a crystal ball.

A new study describes a surprising, yet reliable, source for predicting obesity rates three years in advance: national news stories.

An analysis by the Cornell Food and Brand Lab and California Polytechnic State University of 50 years of stories in the New York Times, and 17 years of stories in the London Times, found that when vegetables were more frequently named in news stories in a particular year, obesity rates declined three years later. Similarly, when sweet snacks were mentioned more frequently in a year, obesity rates rose three years later.

The findings, published in BMC Public Health, were consistent across all years for both newspapers and could provide public health officials with a highly accurate predictor to help them tailor the types of programs they fund to fight obesity based on trends.

“Billions of dollars of taxpayer money, company development and marketing money is spent on future forecasts of obesity levels, and we know in recent years that these forecasts have not been accurate,” said study co-author Brian Wansink, professor and director of the Cornell Food and Brand Lab.

“Being able to predict accurately the obesity levels in three to five years could profoundly affect taxpayer money, change the types of programming used to fight obesity and improve marketing methods to eat lower-calorie food,” said Wansink, author of the new book, “Slim by Design: Mindless Eating Solutions for Everyday Life.”

Brennan Davis, associate professor of marketing at the Orfalea College of Business at California Polytechnic State University, San Luis Obispo, is the paper’s first author.

When designing the study, the researchers recognized that journalists’ writing reflected societal movements and that made them “good predictors of trends,” Wansink said. 

Wansink and Brennan examined words in the press such as “heart attack,” “blood pressure” or “diabetes” related to obesity health factors. They also tracked words related to the top five fruits and vegetables and less healthy salty and sweet snacks. For the U.S., words were tracked between 1960 (the first year that both U.S. obesity prevalence data and New York Times media mentions were available) and 2010, and in the United Kingdom between 1993 and 2010.

While salty snacks and obesity-related words like “heart attack” and “diabetes” did not predict future obesity rates, and mention of fruits was not highly correlated, the frequency of words for the top five vegetables and sweet snacks were strongly correlated to obesity rates three years later.

A research assistant in Wansink’s lab entered key words such as “broccoli” or “ice cream” and tracked every mention, no matter the context.

“This is a way to not be caught off guard,” said Wansink. “En masse, journalists’ trend-spotting abilities seem to do better than what the experts predict using scientific measures.” Adding in the number of mentions of vegetables and sweets in the news to other measures will dramatically improve the accuracy of big economic models used to predict future levels of obesity, he said.

The authors used no external funding for the study. 

Media Contact

Melissa Osgood