(P20-105-24) The Association Between Program Utilization and Enrollee Characteristics in Seattle’s Fresh Bucks Fruit and Vegetable Incentive Program

Objectives: Financial incentives to support the consumption of fruits and vegetables and to address food insecurity are becoming increasingly common. The Fresh Bucks program is a fruit and vegetable (F/V) incentive program currently serving approximately 12,000 lower-income households in Seattle. It provides low-income families with $40 per month on an electronic benefits card for the purchase of F/V at retailers throughout Seattle. We assessed differences in utilization of program benefit according to language spoken, race/ethnicity and method of enrollment (via a community-based organization (CBO) or public application) to inform whether the program is equitably meeting participant needs.

Methods: We used an administrative dataset of all Fresh Bucks transactions from 10/2021-12/2022 maintained by the City of Seattle that includes a unique customer identifier, demographics, transaction date, and amount of benefit spent via the electronic benefit card. We utilized a two-part generalized estimating equations (GEE) model, first examining the odds of any program utilization with a logistic GEE model, then examining the percent of benefit redeemed among those with any use in a given month in a linear GEE model.

Results: Our study sample included 18,033 participants with 192,826 total monthly observations. We find that, compared to participants with a primary language of English, participants whose primary language was Chinese had 2.6 times the odds of utilizing the Fresh Bucks program (95%CI: 2.2, 3.1), and used 4.7 percentage points more of their allocated benefit per month (95%CI: 3.7 pp, 5.7 pp). Black participants had 20% lower odds of any program utilization (OR: 0.8; 95%CI: 0.7, 0.9), while Asian participants had 80% higher odds of program utilization (OR: 1.8; 95%CI 1.5, 2.0), compared to White participants. People who enrolled via a CBO (vs public application) had 8% lower odds of any program utilization (OR: 0.92; 95%CI: 0.86, 0.99).

Conclusions: We find evidence of differences in program utilization by primary language spoken, race/ethnicity and program enrollment path. Future work should identify specific barriers to program utilization and potential impacts on food security and F/V consumption to ensure equitable outcomes.

Funding Sources: This work is funded by the National Institutes of Health, National Heart, Lung, and Blood Institute.