The Impact of Nutrition Label Usage on Food Selection and Consumption in a University Dining Setting

Background Although adding nutrition labels at the point of purchase is commonly recommended to improve food choice, most studies have relied on self‐reported dietary intake or have inferred usage based on sales data without examining actual food selection and consumption of label users and non‐users. In this study, we combine survey and plate photo data to compare food selection and consumption of label users and non‐users in a university dining hall setting. Methods We surveyed n =606 diners in two university dining halls in the Fall 2014 semester. Food selections were recorded, and pre‐ and post‐meal photos were taken of each diner's plate. Photos were coded to identify the presence, servings, and consumption of MyPlate food categories. We compared the plates of self‐reported label users (those who agreed they used nutrition labels to make their food choices at the meal surveyed) and non‐users. Results Across the entire sample, the most commonly selected categories were protein (86.1% of diners selected protein items), refined grains (79.0%), and vegetables (70.6%); whole grains (5.3%), beans (14.0%), and low‐fat dairy (15.4%) were the least commonly selected categories. Label users were more likely to select vegetables and beans and less likely to select potatoes (all P<0.05) than non‐users. Serving sizes only differed between label users and non‐users for the vegetables, beans, potatoes, and refined grains categories. On average, label users selected 0.53 ± 0.51 cups vegetables versus 0.38 ± 0.40 cups for non‐users (P=0.0001), and 0.18 ± 0.46 ounces beans versus 0.09 ± 0.28 ounces for non‐users (P=0.0055). Users selected an average of 0.22 ± 0.27 cups potatoes while non‐users selected 0.31 ± 0.31 cups (P=0.0024). For refined grains, label users served themselves 1.35 ± 1.2 ounces on average, whereas non‐users selected 1.59 ± 1.31 ounces (P=0.0485). Net consumption, calculated by multiplying serving size by the percentage consumed, showed label users consumed 1/10 cup more vegetables (p=0.0538) and 1/5 ounce fewer refined grains (p=0.0320) compared to non‐users. Conclusions Label users behave differently than label non‐users in selecting and consuming food. Further, since our results suggest both groups must improve significantly to meet current dietary guidelines, educational efforts beyond nutrition labeling may be needed. Support or Funding Information This research was supported by the Agriculture and Food Research Initiative of the USDA National Institute of Food and Agriculture as part of the AFRI Childhood Obesity Prevention Challenge (2011‐67001‐30101) to the Division of Nutritional Sciences at the University of Illinois Funding began on 02/01/2011.