The project first came together in 2010 after a group of 16 major food companies pledged, as part of first lady Michelle Obama's campaign to combat obesity, to reduce the calories they sell to the public by 1.5 trillion. The Robert Wood Johnson Foundation agreed to fund a study to hold the companies accountable, eventually turning to UNC with grants totaling $6.7 million.
Aided by supercomputers on campus, Popkin and his team have taken existing commercial databases of food items in stores and people's homes, including the store-based scanner data of 600,000 different foods, and matched that information with the nutrition facts panels on the back of packages and government data on individuals' dietary intake.
The result is an enormous database that has taken almost three years so far to construct and includes more detail than researchers have ever had on grocery store items — their individual nutritional content, who is buying them and their part in consumers' diets.
The study will fill gaps in current data about the choices available to consumers and whether they are healthy, says Susan Krebs-Smith, who researches diet and other risk factors related to cancer at the National Cancer Institute.
Government data, long the only source of information about American eating habits, can have a lag of several years and neglect entire categories of new types of products — Greek yogurt or energy drinks, for example.
With those significant gaps, the government information fails to account for the rapid change now seen in the marketplace. Now more than ever, companies are reformulating products on the fly as they try to make them healthier or better tasting.
While consumers may not notice changes in the ingredient panel on the back of the package, the UNC study will pick up small variations in individual items and also begin to be able to tell how much the marketplace as a whole is evolving.