Tracking What America Eats: The FoodAPS-2 Field Test

Imagine if every time you bought food, you scanned it with your phone so researchers could learn more about what people are really eating. That’s exactly what happened in the FoodAPS-2 field test.

For one week, participants used a smartphone app to record everything they acquired—by scanning barcodes, typing in descriptions, or entering special produce codes (PLUs). This gave researchers a detailed look at grocery store purchases, restaurant meals, and other food acquisitions.

To make sense of all this data, researchers connected the barcodes and codes to several big food databases, including those from the USDA, Circana, and Nutritionix. They used a special tool called the Purchase to Plate Crosswalk (PPC) to match foods with the USDA’s nutrient database. When the computer couldn’t find a match, Nutritionists stepped in to review and assign the right food codes.

Once the data was cleaned up, analysts added even more details—like restaurant characteristics, store information, portion sizes, package sizes, ingredients, and even label claims.

Why it matters: This field test shows how technology and databases can work together to give us a clearer picture of America’s food habits. By linking purchases to nutrition information, researchers can study not just what people buy, but how healthy those choices are, how much they cost, and how they fit into overall eating patterns.

In short: The FoodAPS-2 field test is powered by smartphones and science, helping us understand the connection between what we buy and what we eat.

Read more in the Journal of Food Composition and Analysis.

Written by Amber Brown McFadden, Deirdre Douglass, and Thea Palmer Zimmerman.

Utilization of multiple food and nutrient databases to code food logs in the FoodAPS-2 field test,

Journal of Food Composition and Analysis, Volume 145, 2025, 107851, ISSN 0889-1575,

https://doi.org/10.1016/j.jfca.2025.107851.

(https://www.sciencedirect.com/science/article/pii/S0889157525006660)

Abstract: Westat executed the field test for the second National Household Food Acquisition and Purchase Survey (FoodAPS-2) in 2022. Participants recorded acquisitions with a smartphone app by scanning barcodes and entering descriptions and price look-up (PLU) codes for 7 days. Three processes linked the food and beverage data to food codes. Barcode data were linked to the Circana, U.S. Department of Agriculture (USDA), and Nutritionix databases to identify foods. The Purchase to Plate Crosswalk (PPC) linked Universal Product Codes (UPCs) to the 2017–2018 version of the Food and Nutrient Database for Dietary Studies (FNDDS). PLU codes were linked to FNDDS with a crosswalk. During data processing, coders reviewed automated matches and coded unmatched items with food codes from USDA’s FoodData Central while coding additional attributes. After the coding phase, SAS analysts linked USDA Economic Research Service (ERS) Food Purchase Groups, NPD Group restaurant characteristics, and store information. The FoodAPS-2 field test utilized multiple food and nutrient databases to describe food and beverage acquisitions. This combination of databases yielded information to identify food and beverage nutrients, food patterns equivalents, ingredients, portion size, package size, PPC form, PPC refuse, label claims, restaurant data, store data, ERS Food Purchase Groups, and brand names.

Keywords: Database linking; Nutrient analysis; Nutrient databases; Food coding; Food surveys; Household food scanner data; Household food acquisition