Food Selection
Conditions
Keywords
Sustainability, Climate change, Diet, Food choices, Food labeling
Brief summary
The aim of this study is to compare responses to 6 different types of labels for restaurant menus: 1) a QR code on all items (control); 2) High Climate Impact label; 3) High Climate Impact Warning label; 4) Climate Grade label; 5) Climate Grade label also displaying full range of possible grades; 6) Estimated Environmental Cost label. Participants will be randomized to 1 of these 6 labeling arms. Each participant will view a menu based on a real-world restaurant with one of the 6 labels shown on applicable menu items, select the menu item they would like, and then respond to survey questions about each label.
Detailed description
This is an online randomized controlled trial in which researchers will use a survey company to recruit participants to an online survey. In the study, participants will be randomized to 1 of 6 labeling arms: 1) a QR code on all items (control); 2) High Climate Impact label; 3) High Climate Impact Warning label; 4) Climate Grade label; 5) Climate Grade label also displaying full range of possible grades; 6) Estimated Environmental Cost label. Labels will only appear alongside main menu items and will be assigned using thresholds for greenhouse gas emissions set a priori in kilograms of CO2 equivalent per kilogram of food. Each participant will view a menu based on a real-world restaurant with one of the 6 labels shown on applicable menu items, select the menu item they would like, then respond to survey questions about each label. For dichotomous outcomes, the investigators will directly estimate the probability ratio using Poisson regression with a robust error variance, regressing the outcome on indicators for experimental condition. For continuous outcomes, the investigators will use linear regression models, regressing the outcome on indicators for experimental condition. A critical alpha 0.05 will be used, and statistical tests will be two-tailed. Investigators will compare all experimental label conditions to the control label, then compare each experimental label to each other using pairwise comparisons.
Interventions
BEHAVIORALMenu label
Participants will view a sample of items from a restaurant menu, displayed with labels
Sponsors
University of California, Davis
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
PREVENTION
Masking
SINGLE (Investigator)
Intervention model description
Participants will be randomized to view a menu based on a real-world restaurant with one of six labeling schemes applied: 1) a QR code on all items (control); 2) High Climate Impact label; 3) High Climate Impact Warning label; 4) Climate Grade label; 5) Climate Grade label also displaying full range of possible grades; 6) Estimated Environmental Cost label. Menus will include main items only. Labels will only appear alongside main menu items and will be assigned using thresholds for greenhouse gas emissions set a priori in kilograms of CO2 equivalent per kilogram of food. Each participant will view a menu based on a real-world restaurant with one of the 6 labels shown on applicable menu items, select the menu item(s) they would like, then respond to survey questions about each label.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* 18 years and older * English-speaking * U.S. residents * Have eaten restaurant food at least once in the last month * Participants will reflect the U.S. Census Bureau's American Community Survey estimates for sex, race/ethnicity, educational attainment, and age
Exclusion criteria
* Non-US resident * Younger than 18 years old * Failing the attention check question * Completing the survey in less than 33% of the median completion time
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| High climate impact menu item selected - dichotomous variable | Up to approx. 1 minutes | Measured as a dichotomous variable (yes/no) indicating whether participants' hypothetical meal orders include a high climate impact menu item. |
| Total greenhouse gas emissions - continuous variable | Up to approx. 1 minute | Cumulative total kilograms of carbon equivalent (kgCO2e) per kilogram of food for all meal items. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Total energy (kcal) selected - continuous variable | Up to approx. 1 minute | The total number of calories in participants' hypothetical meal orders, calculated as the sum of calories across all menu items selected. Higher values indicate more calories selected. |
| Modified Nutrient Profile Index (NPI) score - continuous variable | Up to approx. 1 minute | NPI scores are based on the UK Ofcom Nutrient Profiling Model, which is used to score individual foods in the U.K. to determine which ones can be marketed to children. The NPI generates a 0 to100-point score for foods and beverages where scores \>=64 are considered healthy. NPI scores are based on points awarded from nutrients to encourage (e.g., percent fruits and vegetables, grams of fiber and protein) and nutrients of concern (e.g., sodium, sugar) per 100 grams. For this study, we will generate a modified NPI score to evaluate the healthfulness of full meals comprised of multiple food items. This measure will be determined by calculating the weighted mean NPI score of all food items selected per meal, with each item's NPI score weighted by its proportionate contribution of mass in grams to the total mass of the meal. Reference: Rayner, M., Scarborough, P., Boxer, A., & Stockley, L. (2005). Nutrient profiles: development of final model. London: Food Standards Agency. |
| Total saturated fat (g) selected - continuous variable | Up to approx. 1 minute | The total grams of saturated fat in participants' hypothetical meal orders, calculated as the sum of grams of saturated fat across all menu items selected. Higher values indicate more saturated fat selected. |
| Total sodium (mg) selected - continuous variable | Up to approx. 1 minute | The total milligrams of sodium in participants' hypothetical meal orders, calculated as the sum of milligrams of sodium across all menu items selected. Higher values indicate more sodium selected. |
Other
| Measure | Time frame | Description |
|---|---|---|
| Support - categorical variable | Up to approx. 1 minute | Measured with a single item: Imagine the city you live in is considering a policy that would add the label(s) shown below to relevant menu items in chain restaurants. Would you…? With an image of the assigned label condition the participant was randomized to and the response options: Strongly oppose this policy, Somewhat oppose this policy, No opinion, Somewhat support this policy, Strongly support this policy. There is no formal name for this measure. It was developed for this study. |
| Noticeability - dichotomous variable | Up to approx. 1 minute | Measured as a dichotomous variable (yes/no) indicating whether participants noticed and can correctly recall the climate, environmental, or control label with the question: Did you notice any labels (other than calories) next to or below the menu items?. With response options: Yes, a climate label (or Yes, an environmental label in the environmental cost condition), Yes, a heart healthy label, Yes, a vegan label, Yes, an organic label, Yes, a local label, Yes, a QR code label, No, I did not see a label, Not sure. There is no formal name for this measure. It was developed for this study. |
| Effectiveness - categorical variable | Up to approx. 1 minute | A single item adapted from the UNC-PME scale to measure how much the climate labels discouraged wanting to purchase an item that has a high climate impact with the question: How much does this label discourage you from wanting to purchase an item that has a high climate impact? With response options: Not at all, A little bit, Somewhat, Quite a bit, and A great deal. Reference: Baig, S. A., Noar, S. M., Gottfredson, N. C., Boynton, M. H., Ribisl, K. M., & Brewer, N. T. (2019). UNC perceived message effectiveness: validation of a brief scale. Annals of Behavioral Medicine, 53(8), 732-742. |
Countries
United States
Outcome results
None listed