A Standardized Guide to Developing an Online Grocery Store for Testing Nutrition-Related Policies and Interventions in an Online Setting

Feasibility and acceptability of a personalized, pre-filled online grocery cart to improve benefit redemption and diet quality of grocery purchases among participants of the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC)

Most participants in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) do not fully redeem their benefits due to barriers like transportation, confusing product eligibility, and unclear labeling. Online food shopping enables choice architecture to promote benefit redemption and maximize diet quality. We conducted a mixed-methods pilot randomized-controlled trial to assess the feasibility and acceptability of a pre-filled online grocery shopping cart to improve WIC benefit redemption and diet quality of grocery purchases. Rhode Island WIC participants (n = 24, mean age 29.4 ± 1.1 years, 75% Hispanic, 54% had never grocery shopped online) completed a baseline questionnaire and a simulated shopping episode (SSE), buying WIC and non-WIC items. After a week, we randomized participants into the intervention (personalized, modifiable carts pre-filled with 100% of the 2022 proposed WIC packages) or control (selected their items individually) groups before the second SSE. Both groups had WIC labels. We assessed feasibility using process data and percent agreement to feasibility questions, and acceptability via percent agreement to acceptability questions and post-intervention qualitative interviews. We conducted exploratory analyses to examine differences within and between groups at each timepoint for percent WIC benefit redemption and diet quality of grocery purchases, evaluated using the Grocery Purchase Quality Index-2016 (GPQI-2016) scores. Quantitative study measures suggest that the intervention was feasible and that the personalized, modifiable pre-filled cart was acceptable. These findings were supported during qualitative interviews, where participants highlighted time-savings, flexibility, and WIC labels as facilitators of WIC online shopping. Exploratory results showed significant increases in mean percent redemption of most WIC food categories and non-significant moderate increases in most GPQI-2016 scores. These measures are vital for the future adaptation of a full-scale efficacy trial in real-life settings.

Online retail nudges to help parents with lower-income choose healthy beverages for their children: A randomized clinical trial

BACKGROUND

Nudges offer a promising tool to reduce sugary drink intake among children who are most at risk for diet-related disease.

OBJECTIVE

To examine the impact of online store nudges on purchases of sugary drinks for children in lower-income households.

METHODS

Caregivers with lower-income were recruited to an online shopping experiment and instructed to spend $10-$30 on three beverages for their child aged 1-5 years. Participants were randomized to navigate an online supermarket in its standard version (n = 1106) or a version with nudges (n = 1135), including a product placement nudge (i.e. placing healthy beverages in prominent positions) and a swap nudge (i.e. offering a swap of water, plain milk and/or 100% fruit juice upon selection of sugary drinks).

RESULTS

On average, participants purchased 1887 (SD = 2113) and 620 (SD = 1528) calories from sugary drinks per basket in the control and experimental conditions, respectively. Model-based results indicate that those in the experimental condition purchased 1267 (95% CI: 1419, 1114) fewer calories from sugary drinks, and fewer grams of total sugar (β = -253.5 g (95% CI: -286.3, -220.6)) and added sugar (β = -287.8 g (95% CI: -323.1, -252.5)) purchased from sugary drinks.

CONCLUSION

Nudges may be an effective, acceptable, scalable strategy for leading caregivers in lower-income households to purchase fewer sugary drinks for their children.

Effects of red meat taxes and warning labels on food groups selected in a randomized controlled trial

BACKGROUND

High consumption of red and processed meat contributes to both health and environmental harms. Warning labels and taxes for red meat reduce selection of red meat overall, but little is known about how these potential policies affect purchases of subcategories of red meat (e.g., processed versus unprocessed) or of non-red-meat foods (e.g., cheese, pulses) relevant to health and environmental outcomes. This study examined consumer responses to warning labels and taxes for red meat in a randomized controlled trial.

METHODS

In October 2021, we recruited 3,518 US adults to complete a shopping task in a naturalistic online grocery store. Participants were randomly assigned to one of four arms: control (no warning labels or tax), warning labels only (health and environmental warning labels appeared next to products containing red meat), tax only (prices of products containing red meat were increased 30%) or combined warning labels + tax. Participants selected items to hypothetically purchase, which we categorized into food groups based on the presence of animal- and plant-source ingredients (e.g., beef, eggs, pulses), meat processing level (e.g., processed pork versus unprocessed pork), and meat species (e.g., beef versus pork). We assessed the effects of the warning labels and tax on selections from each food group.

RESULTS

Compared to control, all three interventions led participants to select fewer items with processed meat (driven by reductions in processed pork) and (for the tax and warning labels + tax interventions only) fewer items with unprocessed meat (driven by reductions in unprocessed beef). All three interventions also led participants to select more items containing cheese, while only the combined warning labels + tax intervention led participants to select more items containing processed poultry. Except for an increase in selection of pulses in the tax arm, the interventions did not affect selections of fish or seafood (processed or unprocessed), eggs, or plant-based items (pulses, nuts & seeds, tofu, meat mimics, grains & potatoes, vegetables).

CONCLUSIONS

Policies to reduce red meat consumption are also likely to affect consumption of other types of foods that are relevant to both health and environmental outcomes.

TRIAL REGISTRATION

NCT04716010 on www.

CLINICALTRIALS

gov .

Impact of taxes and warning labels on red meat purchases among US consumers: A randomized controlled trial

BACKGROUND

Policies to reduce red meat intake are important for mitigating climate change and improving public health. We tested the impact of taxes and warning labels on red meat purchases in the United States. The main study question was, will taxes and warning labels reduce red meat purchases?

METHODS AND FINDINGS

We recruited 3,518 US adults to participate in a shopping task in a naturalistic online grocery store from October 18, 2021 to October 28, 2021. Participants were randomized to one of 4 conditions: control (no tax or warning labels, n = 887), warning labels (health and environmental warning labels appeared next to products containing red meat, n = 891), tax (products containing red meat were subject to a 30% price increase, n = 874), or combined warning labels + tax (n = 866). We used fractional probit and Poisson regression models to assess the co-primary outcomes, percent, and count of red meat purchases, and linear regression to assess the secondary outcomes of nutrients purchased. Most participants identified as women, consumed red meat 2 or more times per week, and reported doing all of their household's grocery shopping. The warning, tax, and combined conditions led to lower percent of red meat-containing items purchased, with 39% (95% confidence interval (CI) [38%, 40%]) of control participants' purchases containing red meat, compared to 36% (95% CI [35%, 37%], p = 0.001) of warning participants, 34% (95% CI [33%, 35%], p < 0.001) of tax participants, and 31% (95% CI [30%, 32%], p < 0.001) of combined participants. A similar pattern was observed for count of red meat items. Compared to the control, the combined condition reduced calories purchased (-312.0 kcals, 95% CI [-590.3 kcals, -33.6 kcals], p = 0.027), while the tax (-10.4 g, 95% CI [-18.2 g, -2.5 g], p = 0.01) and combined (-12.8 g, 95% CI [-20.7 g, -4.9 g], p = 0.001) conditions reduced saturated fat purchases; no condition affected sodium purchases. Warning labels decreased the perceived healthfulness and environmental sustainability of red meat, while taxes increased perceived cost. The main limitations were that the study differed in sociodemographic characteristics from the US population, and only about 30% to 40% of the US population shops for groceries online.

CONCLUSIONS

Warning labels and taxes reduced red meat purchases in a naturalistic online grocery store. Trial Registration: http://www.clinicaltrials.gov/ NCT04716010.

Effect of Financial Incentives and Default Options on Food Choices of Adults With Low Income in Online Retail Settings: A Randomized Clinical Trial

Importance

Despite recent growth in online redemption of Supplemental Nutrition Assistance Program (SNAP) benefits, no previous work has tested the impact of economic and behavioral economic strategies on food purchasing behaviors in an online grocery retail setting among adults with low income.

Objective

To examine the extent to which financial incentives and default shopping cart options influence fruit and vegetable purchases.

Design, Setting, and Participants

This randomized clinical trial used an experimental online grocery store for adults who currently or have ever received SNAP benefits. From October 7 to December 2, 2021, participants were instructed to shop for a week's worth of groceries for their household, with a budget tailored to household size; no payment was taken.

Interventions

Random assignment to 1 of 4 conditions: no intervention, 50% discount on eligible fruits and vegetables, prefilled shopping carts with tailored fruit and vegetable items (ie, default options), or a combination of the discount and default options.

Main Outcomes and Measures

The primary outcome was the percentage of nondiscounted dollars spent on eligible fruit and vegetables per basket.

Results

Of 2744 participants, mean (SD) age was 46.7 (16.0) years, and 1447 (52.7%) identified as women. A total of 1842 participants (67.1%) reported currently receiving SNAP benefits and 1492 (54.4%) reported shopping online for groceries in the previous 12 months. Participants spent a mean (SD) 20.5% (23.5%) of total dollars on eligible fruits and vegetables. Compared with no intervention, those in the discount condition spent 4.7% (98.3% CI, 1.7%-7.7%) of more total dollars on eligible fruits and vegetables; those in the default condition, 7.8% (98.3% CI, 4.8%-10.7%) more; and those in the combination condition, 13.0% (98.3% CI, 10.0%-16.0%) more (P < .001 for all). There was no difference between the discount and the default conditions (P = .06), but the effect in the combination condition was significantly larger than both discount and default conditions (P < .001). Default shopping cart items were purchased by 679 participants (93.4%) in the default condition and 655 (95.5%) in the combination condition, whereas 297 (45.8%) in the control and 361 (52.9%) in the discount conditions purchased those items (P < .001). No variation was observed by age, sex, or race and ethnicity, and results were similar when those who reported never shopping online for groceries were excluded.

Conclusions and Relevance

In this randomized clinical trial, financial incentives for fruits and vegetables and default options, especially in combination, led to meaningful increases in online fruit and vegetable purchases among adults with low income.

Trial Registration

ClinicalTrials.gov Identifier: NCT04766034.

Consumer behaviour and experiences in a naturalistic online grocery store: implications for nutrition research

Naturalistic online grocery stores could provide a novel setting for evaluating nutrition interventions. In 2021–2022, we recruited US adults (n 144, 59% low-income) to complete two weekly study visits: one in a naturalistic (‘mock’) online grocery store developed for research and one in a real online grocery store. Participants selected groceries and responded to survey questions. Analyses examined survey responses and expenditures on fifteen food categories (e.g., bread, sugar-sweetened beverages). Nearly all enrolled participants completed both visits (98% retention). Moreover, nearly participants all reported that their selections in the naturalistic store were similar to their usual purchases (95%) and that the naturalistic store felt like a real store (92%). Participants’ spending on food categories in the naturalistic store were moderately-to-strongly correlated with their spending in the real store (range of correlation coefficients: 0⋅36–0⋅67, all P-values < 0⋅001). Naturalistic online grocery stores may offer a promising platform for conducting nutrition research.

Using a Naturalistic Store Laboratory for Clinical Trials of Point-of-Sale Nutrition Policies and Interventions: A Feasibility and Validation Study

Point-of-sale policies such as warnings and taxes are promising tools for improving the nutritional quality of food purchases. Research studies conducted in naturalistic store laboratories could improve the quality of evidence about point-of-sale interventions by allowing for realistic exposure in a controlled setting. This study aimed to assess whether purchasing behavior in a naturalistic store laboratory setting was similar to real-life purchasing behavior and to evaluate participants’ perceptions of store realism and the acceptability of research study protocols in this setting. In a longitudinal observational study in 2019, Latinx parents in North Carolina (n = 61) attended five weekly visits at the UNC Mini Mart, a naturalistic store laboratory that resembled a small convenience store. At each visit, participants purchased a week’s supply of beverages. Purchases of beverages in the Mini Mart were compared to participants’ purchases from receipts submitted the week prior to the study. Analyses compared the percentage of participants buying sugary drinks and non-sugary drinks in the Mini Mart vs. in real stores using Chi-Square tests with Fisher’s p. The percentage of parents who purchased sugary drinks in the Mini Mart (93%) was not significantly different from the percentage who purchased sugary drinks during the week before the study (74%, p = 0.28). The percentage purchasing non-sugary drinks was similar in the two settings (85% in the Mini Mart vs. 85% from receipts, p = 0.33). Nearly all participants reported that their Mini Mart purchases were similar to real-life purchases (96%); the Mini Mart felt like a real store (94%); they could find all the beverages they were looking for (92%); and they could imagine doing their real-life beverage shopping in the Mini Mart (92%). Moreover, retention was high, with 97% of participants attending the final study visit. These results indicate that naturalistic store laboratories are a promising method for increasing the ecological validity of trials to evaluate point-of-sale interventions.