Participant Characteristics Contributed to Variable Responsiveness to Personalized Healthy Food Incentives—A Secondary Analysis of a Randomized Crossover Trial

Abstract

ObjectivesThe recently completed Smart Cart Study, designed to promote healthier food purchases, found that personalized healthy food incentives modestly improved mean grocery purchase quality, with considerable variation in participant responsiveness. The present study explored participant characteristics associated with variable responsiveness to this healthy food incentives intervention. MethodsA secondary analysis examined the Smart Cart Study, where participants received personalized weekly coupons during in the intervention and occasional generic coupons during the control period (n = 209). The outcome variable was 3-month changes in Grocery Purchase Quality Index (GPQI), a validated score calculated from purchasing data that compares the % spending relative to recommended spending to reflect the dietary quality of grocery purchases, in the intervention minus control periods. Potential exposure variables were selected a priori based on a literature review included age, sex, race/ethnicity, education, income, general health status, food literacy, food neophobia, nutritional self-efficacy and eating perception. Multivariable linear regressions evaluated characteristics that predict changes in GPQI. Stepwise model selection guided variable retention with specification of pr = 0.2. Analyses were conducted in Stata 16.1. ResultsParticipants were predominately female (89.5%), Non-Hispanic white (94.1%), higher income (50.3% ≥$100K/year), higher education (48.4% >bachelor’s degree), with a mean age of 55.3. In the final model, changes in % spending on couponed foods (β = 15.58, P < 0.001) and food neophobia (β = –1.77, P = 0.047) predicted changes in GPQI, while other predictors were not significant. The final model explained 13% of the variation in GPQI changes. ConclusionsGreater % spending on couponed foods was associated with higher GPQI while food neophobia was associated with lower GPQI. Given that food neophobia may prevent participants from increasing their spending on new couponed healthy foods, future personalized couponing interventions should explore strategies to address food neophobia, potentially through free samples of new foods or providing coupons for familiar healthy foods, especially in diverse populations. Funding SourcesThe Smart Cart Study was funded by the Foundation for Food and Agricultural Research.