Food Swaps to Improve the Healthfulness and Reduce the Carbon Footprint of Grocery Purchases

A Randomized Trial of Food Swaps to Improve the Healthfulness and Reduce the Carbon Footprint of Grocery Purchases

The study aims to determine whether viewing health or climate labels (or both) and receiving recommendations for healthier or more climate-friendly swaps (or both) in an online grocery store environment improves the healthfulness and reduces the carbon footprint of consumers' food and beverage purchases compared to shopping as usual without swap recommendations. The online store will record participants' food selections. Participants will also be asked to complete survey measures.

Study Overview

• Status: Completed
• Conditions: Dietary Habits; Food Preferences; Healthy Diet; Food Selection; Nutrition; Sustainability
• Intervention / Treatment: Behavioral: Health swaps; Behavioral: Control; Behavioral: Climate swaps; Behavioral: Combined health and climate swaps

Detailed Description

Participants will complete an online between-subjects randomized experiment. The experiment will involve three study visits.

In the first study visit, participants will complete an online grocery store shopping task without intervention. Participants will be given a budget of $50 and asked to shop as they normally would. The store will record participants' food selections. After completing the shopping task, participants will complete an online survey.

Approximately one week later, participants will complete a second study visit with the same instructions. They will be randomized to 1 of 4 conditions (1) health only, 2) climate only, 3) combined health and climate, 4) control. In the health only condition, participants will view labels with nutrition grades and may be directed to swap their selections for healthier items. In the climate only condition, participants will view labels with climate grades and may be directed to swap their selections for more climate-friendly (lower-carbon-footprint) items. In the combined health and climate condition, participants will view labels with health and climate grades and may be directed to swap their selections for healthier or more climate-friendly items. In the control group, participants will not view any labels or swaps. The store will record participants' selections. After completing the shopping task, participants will complete an online survey.

In the third study visit, the participants will maintain their assigned group and will be asked to repeat the same tasks as in the second visit.

• Study Type: Interventional
• Enrollment (Actual): 1201
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • California
    • Palo Alto, California, United States, 94304
      • Stanford School of Medicine

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

• Aged 18 years or older
• Reside in the United States
• Able to complete a survey in English
• Have internet access to complete the 3 online study visits

Exclusion Criteria:

• Under the age of 18
• Reside outside of the United States
• Unable to complete a survey in English
• Do not have internet access to complete the 3 online study visits

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Prevention
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

4

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Active Comparator: Control; No labels or swaps.	Behavioral: Control; Participants will not view any extra labels or be offered any swaps in the online grocery store.
Experimental: Health swaps; Labels will be "nutrition grades" for products and replacement swap suggestions will be for items with a better nutrition grade.	Behavioral: Health swaps; Participants will view "health grade" labels on all products in the online grocery store indicating their healthfulness as estimated by United Kingdom Ofcom Nutrient Profiling Model scores. The health score labels will mimic Nutri-Score labels, a labeling system used in some European countries, showing a color-coded grade of "A" (green) through "F" (red) on each product. Products with "A" and "B" labels will meet the United Kingdom's cutoff for products that can be marketed to children and "C", "D" and "F" labels products are less healthy than this cutoff (based on tertiles of Ofcom scores within each food group). When participants attempt to add a less healthy product to their cart (e.g., "C," "D" or "F" health label), the store will automatically suggest a healthier product from the same category (e.g., with a "A" or "B" health label).
Experimental: Climate swaps; Labels will be "climate grades" for products and replacement swap suggestions will be for items with a better climate grade.	Behavioral: Climate swaps; Participants will view "climate grade" labels on all products indicating their climate impact. The climate impact is calculated as the greenhouse gas emissions associated with producing the product in carbon dioxide equivalents (CO2-eq) per 100g (i.e., "carbon footprint"). Labels will be applied based on quintiles of carbon footprints in each food group. When participants attempt to add a high-climate-impact product to their cart (e.g., with a "C," "D," or "F" climate label), the store will automatically offer them swaps to more climate-friendly products (e.g., with a "A" or "B" climate label).
Experimental: Combined health and climate swaps; Labels will be "nutrition grades" and "climate grades" for products and replacement swap suggestions will be for items that are better than originally selected products on at least one dimension (nutrition grade or climate grade) and not worse than originally selected products on either dimension.	Behavioral: Combined health and climate swaps; Participants will view both the health and climate grade labels on all products in the online grocery store. When participants attempt to select a product with a "C," "D," or "F" label on either dimension to their cart, the store will automatically offer them swaps to products that offer improvement over the original food on at least 1 dimension (health or climate-friendliness) and were at least as good or better on the other dimension, with the additional guardrail that the store never suggests products with a "C," "D" or "F" label on either dimension.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Healthfulness of Product Selection	The study will asses healthfulness of participants' selections as the average Ofcom Nutrient Profiling Model score of the products the participants select in the shopping task. The score ranges from 0 to 100, where higher scores are healthier. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Carbon Footprint of Participants' Grocery Store Selections	The study will assess the carbon footprint of participants' selections, operationalized as the average carbon footprint (in CO2-equivalents) of the products the participants select in the shopping task. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Calorie Density of Participants' Grocery Selections	The study will assess the average calorie density of participants' grocery selections, calculated as the average kcal per 100g of the selected products. Higher values will indicate higher calorie density. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Sugar Density of Participants' Grocery Selections	The study will assess the average sugar density of participants' grocery selections, calculated as the average grams of sugar per 100g of the selected products. Higher values will indicate higher sugar density. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Sodium Density of Participants' Grocery Selections	The study will assess the average sodium density of participants' grocery selections, calculated as the average milligrams of sodium per 100g of the selected products. Higher values will indicate higher sodium density. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Saturated Fat Density of Participants' Grocery Selections	The study will assess the average saturated fat density of participants' grocery selections, calculated as the average grams of saturated fat per 100g of the selected products. Higher values will indicate higher saturated fat density. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Fiber Density of Participants' Grocery Selections	The study will assess the average fiber density of participants' grocery selections, calculated as the average grams of fiber per 100g of the selected products. Higher values will indicate higher fiber density. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Protein Density of Participants' Grocery Selections	The study will assess the average protein density of participants' grocery selections, calculated as the average grams of protein per 100g of the selected products. Higher values will indicate higher protein density. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Total Spending on Participants' Grocery Selections	The study will assess the total amount participants spent on their grocery selections in United States dollars. Higher values will indicate greater spending. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Thinking About Health	The study will assess thinking about health using 1 item: "We are interested in how you selected products while you were grocery shopping. When you selected foods, how much did you think about each of the following characteristics? ... health." Response options range from "Not at all" (1) to "A great deal" (5). Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Thinking About Climate Impact	The study will assess thinking about climate impact using 1 item: "We are interested in how you selected products while you were grocery shopping. When you selected foods, how much did you think about each of the following characteristics? ... climate impact." Response options range from "Not at all" (1) to "A great deal" (5). Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Thinking About Taste	The study will assess thinking about taste using 1 item: "We are interested in how you selected products while you were grocery shopping. When you selected foods, how much did you think about each of the following characteristics? ... taste." Response options range from "Not at all" (1) to "A great deal" (5). Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Negative Emotional Reactions While Shopping	The study will assess negative emotional reactions using 3 items: "While you were shopping, how much did you feel…" "worried", "guilty," "ashamed"? Response options range from "Not at all" (1) to "Extremely" (5). The investigators will average the 3 items; higher average scores indicate stronger negative emotional reactions. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Positive Emotional Reactions While Shopping	The study will assess positive emotional reactions using 3 items: "While you were shopping, how much did you feel…" "inspired", "proud," "reassured"? Response options range from "Not at all" (1) to "Extremely" (5). The investigators will average the 3 items; higher average scores indicate stronger positive emotional reactions. Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Injunctive Norms to Buy Healthy Foods	The study will assess injunctive norms to buy healthy foods with 1 item: "People who are important to me think I should buy healthy foods." Response options range from "Strongly disagree" (1) to "Strongly agree" (5). Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Descriptive Norms to Buy Healthy Foods	The study will assess descriptive norms to buy healthy foods with 1 item: "Most shoppers buy healthy foods when they are shopping for groceries." Response options range from "Strongly disagree" (1) to "Strongly agree" (5). Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Injunctive Norms to Buy Foods With Low Climate Impact	The study will assess injunctive norms to buy foods with low climate impact with 1 item: "People who are important to me think I should buy foods with low climate impact." Response options range from "Strongly disagree" (1) to "Strongly agree" (5). Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)
Descriptive Norms to Buy Foods With Low Climate Impact	The study will assess descriptive norms to buy foods with lower climate impact with 1 item: "Most shoppers buy foods with low climate impact when they are shopping for groceries." Response options range from "Strongly disagree" (1) to "Strongly agree" (5). Outcome values were calculated as each participant's mean across visits 2 and 3.	1-week follow-up (visit 2), and 2-week follow-up (visit 3)

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Perceived Helpfulness of Nutrition Labels	The study will assess perceived helpfulness of nutrition labels with 1 item: "These labels would help me choose healthier foods." Response options range from "Not at all" (1) to "A great deal" (5).	Assessed at 2-week follow-up (visit 3)
Perceived Helpfulness of Climate Labels	The study will assess perceived helpfulness of climate labels with 1 item: "These labels would help me choose foods with lower climate impact." Response options range from "Not at all" (1) to "A great deal" (5).	Assessed at 2-week follow-up (visit 3)
Perceived Helpfulness of Health Swap Recommendations	The study will assess perceived helpfulness of health swap recommendations with 1 item: "This type of recommendation would help me choose healthier foods." Response options range from "Not at all" (1) to "A great deal" (5).	Assessed at 2-week follow-up (visit 3)
Perceived Helpfulness of Climate Swap Recommendations	The study will assess perceived helpfulness of climate swap recommendations with 1 item: "This type of recommendation would help me choose more foods with lower climate impact." Response options range from "Not at all" (1) to "A great deal" (5).	Assessed at 2-week follow-up (visit 3)
Liking of Nutrition Labels	The study will assess liking of nutrition labels with 1 item: "I like these labels." Response options range from "Not at all" (1) to "A great deal" (5).	Assessed at 2-week follow-up (visit 3)
Liking of Climate Labels	The study will assess liking of climate labels with 1 item: "I like these labels." Response options range from "Not at all" (1) to "A great deal" (5).	Assessed at 2-week follow-up (visit 3)
Liking of Health Swap Recommendations	The study will assess liking health swap recommendations with 1 item: "I like these recommendations." Response options range from "Not at all" (1) to "A great deal" (5).	Assessed at 2-week follow-up (visit 3)
Liking of Climate Swap Recommendations	The study will assess liking climate swap recommendations with 1 item: "I like these recommendations." Response options range from "Not at all" (1) to "A great deal" (5).	Assessed at 2-week follow-up (visit 3)
Acceptability of Nutrition Labels	The study will assess acceptability of the nutrition labels with 1 item: "How much would you approve or disapprove of these labels being displayed on foods sold in grocery stores?" Response options range from "Strongly disapprove" (1) to "Strongly approve" (5).	Assessed at 2-week follow-up (visit 3)
Acceptability of Climate Labels	The study will assess acceptability of the climate labels with 1 item: "How much would you approve or disapprove of these labels being displayed on foods sold in grocery stores?" Response options range from "Strongly disapprove" (1) to "Strongly approve" (5).	Assessed at 2-week follow-up (visit 3)
Acceptability of Health Swap Recommendations	The study will assess acceptability of the health swap recommendations with 1 item: "How much would you approve or disapprove of these recommendations being offered by grocery stores?" Response options range from "Strongly disapprove" (1) to "Strongly approve" (5).	Assessed at 2-week follow-up (visit 3)
Acceptability of Climate Swap Recommendations	The study will assess acceptability of the climate swap recommendations 1 item: "How much would you approve or disapprove of these recommendations being offered by grocery stores?" Response options range from "Strongly disapprove" (1) to "Strongly approve" (5).	Assessed at 2-week follow-up (visit 3)

Collaborators and Investigators

• Sponsor: Stanford University
• Collaborators: National Heart, Lung, and Blood Institute (NHLBI)
• Investigators: Principal Investigator: Anna H. Grummon, PhD, Stanford School of Medicine

Study record dates

Study Major Dates

• Study Start (Actual): May 7, 2025
• Primary Completion (Actual): June 25, 2025
• Study Completion (Actual): June 25, 2025

Study Registration Dates

• First Submitted: October 16, 2024
• First Submitted That Met QC Criteria: October 16, 2024
• First Posted (Actual): October 18, 2024

Study Record Updates

• Last Update Posted (Actual): June 1, 2026
• Last Update Submitted That Met QC Criteria: May 4, 2026
• Last Verified: May 1, 2026

More Information

Terms related to this study

• Keywords: diet; food choices; dietary substitutions; climate
• Additional Relevant MeSH Terms: Behavior; Behavior, Animal; Feeding Behavior; Food Preferences
• Other Study ID Numbers: 76925; K01HL158608 (U.S. NIH Grant/Contract)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: The investigators will post the data to a public repository.
• IPD Sharing Time Frame: The analytic code will be posted to a public repository upon publication of study findings.
• IPD Sharing Access Criteria: Publicly available without restrictions.
• IPD Sharing Supporting Information Type: ANALYTIC_CODE

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No