Supermarket price discounts increase fruit and vegetable intake in adult shoppersSupermarket Discounts Linked to Higher Fruit and Vegetable Intake

BACKGROUND: Despite the well-known health benefits of fruit and vegetable (FV) intake, dietary quality among Americans remains low. Barriers to healthy food intake, such as cost, may contribute to suboptimal FV intake. OBJECTIVE: The aim of this study was to assess the effects of multi-level supermarket price discounts of 30%, 15%, or 0% (control) on fruits, vegetables, and noncaloric beverages (NCB) on FV and NCB intake and health outcomes. DESIGN: In this randomized controlled trial (RCT), Multi-level Supermarket Discounts of Fruits and Vegetables' Impact on Intake and Health, participants underwent an 8-week baseline without price discounts, a 32-week intervention, and a 16-week follow-up period. At week 8, participants were randomized to receive a 30%, 15%, or 0% price discount on FV and NCB. PARTICIPANTS/SETTING: Adult supermarket shoppers (n = 312) were recruited in New York City, starting in September 2018 through August 2021. Participants had body mass indexes (BMIs) of 25 to 50, were aged 18 and 70 years, and were primary household shoppers. Of those recruited, 167 were randomized. The final intention-to-treat analysis was based on 33 participants in the 30% discount group, 38 participants in the 15% discount group, and 36 participants in the 0% discount group. INTERVENTION: At week 8, participants were equally randomized into 1 of 3 price discount intervention groups: 30%, 15%, or 0% (control) discount on qualifying FV and NCB. MAIN OUTCOME MEASURES: The primary outcomes were FV and NCB intake (g/day) and body weight (kg). FV and NCB intake were assessed through unannounced 24-hour dietary recalls conducted on 2 weekdays and 1 weekend day, covering the previous day's consumption. Dietary outcomes were collected a month before weeks 8, 24, 40, and 56, and body weight was obtained at weeks 0, 8, 24, 40, and 56. STATISTICAL ANALYSES: Linear mixed models were used to detect significant group-by-time interactions with planned contrasts to examine the differential change (shown as mean ± standard error [SE]) in primary outcomes between groups over time for the intention-to-treat and complete case samples. A sensitivity analysis was then performed on the intention-to-treat sample by adding key covariates to the model: sex, age at enrollment, season at enrollment, primary supermarket, income level, and COVID-19 pandemic presence. RESULTS: Planned contrasts (shown as mean ± SE) showed that the 15% discount group had greater fruit intake from baseline to the end of the follow-up (+137.4 g/day ± 38.7; P < .001) vs the 0% discount group. Contrasts also revealed the 30% discount group had increased vegetable intake during the mid-intervention period (+88.6 g/day ± 40.1; P = .028) vs the 0% discount group. Additionally, contrasts showed that diet soda intake increased in the 30% discount group (vs the 0% discount group) from baseline to mid-intervention (+67.5 g/day ± 20.3; P = .001), to the end of intervention (+42.5 g/day ± 21.2; P = .046), and follow-up (+56.7 g/day ± 22.4; P = .012). Finally, contrasts showed that the 30% discount group lost more weight from baseline to mid-intervention vs the 0% discount group (-2.4 kg ± 0.918; P = .010). CONCLUSIONS: The results showed that the 15% discount on fruits led to increased intake during the follow-up period. The 30% discount on vegetables and diet soda led to greater intake during the first half of the intervention, and sustained increases in diet soda intake throughout the intervention and follow-up periods. Participants in the 30% discount group also exhibited decreased body weight during the first half of the intervention compared with controls. These findings support the use of supermarket-based economic incentives as an effective strategy to promote healthier food intake.