Purpose: To evaluate the effectiveness of online healthy food nudging on healthy food purchasing behaviors in a real-life online supermarket, and to test the potential differential effects of nudging across food groups, and whether these effects were modified by neighborhood socioeconomic position (SEP).

Methods: Over the course of five constructive weeks, online grocery shoppers were randomly assigned to one of four versions of a Dutch online supermarket: Control group: regular online supermarket; Group 1: product labels (three types of symbols) on healthy products; Group 2: unhealthy product swap suggestions and healthy check-out suggestions; Group 4: groups 2 and 3 combined. Sales data from all food and beverage products were retrieved from the supermarket chain. Linear mixed models were used to assess the mean differences between the control group and the intervention groups in the percentage of all healthy food purchases (primary outcome). Nonparametric bootstrapping for hierarchical data was used to investigate differences across all food groups and we tested for effect modification by neighborhood SEP.

Results: Sales data from 11,775 individual shops were analyzed. Among costumers from low SEP neighborhoods, those in group 1 purchased a 2.39% (95%CI=0.82,3.96) higher percentage of healthy products compared to the control group. No significant differences were observed in group 2 (β=-1.29; 95%CI=-2.84,0.27) and group 3 (β=-0.10; 95%CI=-1.69,1.48) compared to the control group. For costumers from high SEP neighborhoods, those in group 1 (β=-1.61; 95%CI=-3.17,-0.06) and group 3 (β=-2.12; 95%CI=-3.64,-0.60) purchased a significant lower percentage of healthy products, whereas this difference was non-significant in group 2 (β=-0.85; 95%CI=-2.39,0.69). Evaluation of differences across food groups revealed that the groups of pasta and rice (Group 3: β=7.63; 95%CI=2.36,12.68) and milk and yogurt (Group 1: β=4.62; 95%CI=0.47,8.81) mostly drove the overall higher percentage of healthy purchased among those with a lower SEP.

Conclusions: Online nudging strategies seem to benefit those living in lower SEP neighborhoods as they purchased a higher percentage of healthy products per shop, particularly from healthier purchases of grains and dairy. Further research is needed to better understand the potential adverse effects on purchasing behaviors for those living in higher SEP neighborhoods. 

Purpose Nutrition labels show potential in increasing healthy beverage purchases, but their effectiveness seems to depend on the type of label, the targeted category and the setting, and evidence on their impact in real-world settings is limited. The aim of this study was to evaluate the effectiveness of on-shelf sugar labeling on the sales of beverages with no, low, medium and high sugar content implemented within a real-world supermarket.

Methods In week 17 of 2019, on-shelf sugar labels were implemented by a Dutch supermarket chain. Non-alcoholic beverages were classified using a nutrient-specific traffic-light labeling system and included the beverage categories “green” for sugar free (<1.24g/250ml), “blue” for low sugar (1.25–6.24g/250ml), “yellow” for medium sugar (6.25–13.5g/250ml) and “amber” for high sugar (>13.5g/250ml). Store-level data on beverage sales and revenue from 41 randomly selected supermarkets for 13 weeks pre-implementation and 21 weeks post-implementation were used for analysis. In total, 30 stores implemented the on-shelf sugar labels by week 17, and the 11 stores that had not were used as controls. Outcome measures were differences in sale changes between the control stores and implementation stores of the four beverage categories and the total beverage revenue. Analyses were conducted using a multiple-group Interrupted Time Series Approach. Results of individual store data were combined using random effect meta-analyses.

Results The change in green (B 3.4, 95%CI -0.3; 7.0) and blue (B 0.0, 95%CI -0.6; 0.7) beverage sales was not statistically significantly different between stores that did or did not implement on-shelf sugar labels. Similar to green and blue beverages, the sales of amber (B 0.9, 95%CI -5.5; 7.3) and yellow beverages (B 1.3, 95%CI -0.9; 3.5) or the total beverage revenue (B 0.8, 95%CI -12.3; 14.0) was not influenced by implementation of the on-shelf sugar labels.

Conclusion The implementation of an on-shelf sugar labeling system did not significantly decrease unhealthy beverage sales and also did not significantly increase sales of beverages labeled as healthier. Nutritional labeling initiatives combined with complementary strategies, such as pricing strategies or other healthy food nudging approaches, should be considered to promote healthier beverage purchases.

Purpose: Most studies on environmental determinants of childhood obesity have analysed the food or physical activity environment in isolation. However, it is likely that different environmental constructs interact in their relationship with health behaviours and health outcomes. In addition, while assumed by many, it is unclear if the geographical availability of an environmental facility (e.g., fast-food outlet) translates to the use of that facility, and would thus reflect actual exposure. We aim to address these research gaps by investigating how combined measures of availability and use of facilities in the food, physical activity, and social environment are related to overweight among Brazilian schoolchildren.

Methods: In this cross-sectional study 2,026 schoolchildren aged 7-14 years were recruited from a probabilistic sample of public or private schools from a southern Brazilian city. Food environment facilities analysed were snack outlets, grocery stores, and restaurants; physical activity facilities were outdoor recreational facilities such as parks and sports courts; social-assistance facilities included public centres for health care and centres for supplementary education. We conducted multilevel logistic regression analyses with participants clustered within schools. Separate food retailers, all three food retailers combined, and social and physical activity facilities were entered in individual and combined models, which were adjusted for relevant individual (e.g., age, sex, maternal educational) and environmental level covariates (population density, and average area income).

Findings: In fully adjusted models, larger availability of restaurants within 400 metres of schoolchildren’s homes was associated with higher odds of overweight (OR=1.40; 95% CI=1.07-1.84) and this association was stronger for those reporting to use restaurants available near the home (OR=1.48; 95% CI=1.15-1.90). Additionally, schoolchildren who had social-assistance facilities surrounding their homes but reported not to use them also showed higher odds of being overweight (OR=1.34; 95% CI=1.01-1.78). The physical activity environment was not found to be associated with obesity.

Conclusions: Availability of food and social environments facilities around the home were significantly associated with about 40% higher levels of overweight in this sample. The use (or not) of facilities was relevant for these associations.