Testing Non-Nutrition Menu Labels on Food Selections

The primary objective of this study is to test the relative effects of environmental harm menu label designs on the healthfulness of consumers' fast-food meal choices. Participants will complete hypothetical online meal ordering tasks using a survey which emulates the online menus of two types of fast-food chain restaurants: a burger restaurant and a sandwich restaurant. Participants will be randomized the view both menus, presented in random order, with one of five labeling conditions applied. Secondary objectives include energy and nutrient content of meals ordered, prices of meals ordered, and, through a post-order survey, noticeability of the labels and perceptions of labels between the conditions.

This online randomized trial will test the relative effects of environmental harm menu label designs on the healthfulness of consumers' fast-food meal choices and perceptions of menu items. The survey research firm Verasight will recruit a nationally representative sample of 7,000 adults aged 18 years and older who speak English and reside in the United States from their survey panel. Participants will complete a between-subjects online experiment in which the participants will be randomized to view two fast-food online ordering menus (in randomized order) with one of five labeling schemes applied: (1) a QR code on all items (control); (2) a low environmental harm label on all low harm items (3) a high environmental harm label on high harm items; (4) traffic light labels on all items by level of environmental harm; and (5) environmental grade labels on all items by level of environmental harm. Each restaurant menu will be presented across 2 pages; page 1 will display all main items, and page 2 will display categories for side, dessert, and drink items. Labels will only appear alongside main menu items and will be assigned using thresholds for environmental harm set a priori which factor in estimated cropland, pastureland, and water usage associated with each food item. Each hypothetical order must include between 1 and 4 items. Participants will be required to select 1 main item and, optionally, up to 3 additional items (drinks, sides, desserts). After completing the meal ordering task, participants will be prompted to answer questions about whether they noticed the correct label for their condition, whether they used the label when making their meal selections, perceptions of the labeled menu items and the labeling condition which the participants were randomized to view. Additionally, participants will be asked about their level of food security, participation in the Supplemental Nutrition Assistance Program (SNAP), support of governmental policies and practices regarding food and nutrition, use of GLP-1 agonists, personal values, dietary patterns, and other demographic information. The primary outcome will be the healthfulness of meal orders selected from the online menu, indicated by a modified Nutrient Profile Index meal score for all food items ordered. Secondary outcomes will include total calories ordered, whether a sugar-sweetened beverage was ordered, and perceived message effectiveness of the environmental harm labels. Analyses will be conducted overall (summing the meal selections across both restaurants) and stratified by restaurant.

Inclusion Criteria: * 18 years of age or older * Member of the Verasight Panel * Residing in the United States Exclusion Criteria: * \<18 years of age * Not residing in the United States * Completed the survey in less than one-third of the median completion duration of all participants * Respondents with high refusal rates (skipped or refused more than 50% of questions) * Respondents who 'straight-line' all question grids with 6 or more items, where responses are not internally consistent