Evaluation of a Distribution Model to Increase Access to Affordable Fruits and Vegetables.

Abstract

Identifying effective, sustainable strategies to increase fruit and vegetable consumption is critical to addressing chronic disease risk. Models that provide incentives for produce purchases through reduced-cost or no-cost produce shares are promising. The purpose of our study was to examine the impact on fruit and vegetable intake of Good Food for All, a community-based program to distribute no-cost produce boxes to participants with low incomes. We also assessed program satisfaction and future interest in purchasing an affordable produce box. The Good Food for All program was implemented in 22 US cities. Surveys were administered at baseline and postintervention. An online research panel was used as a comparison group and weighted to be demographically comparable to the intervention group. Descriptive statistics and adjusted difference-in-difference (ADID) models were used to examine differences in outcomes between groups. Respondents (intervention n = 632; comparison n = 1,153) were primarily White, non-Hispanic, and female. Intervention participants had a greater increase in total fruit consumption frequency (ADID: 0.43 times/d; 95% CI, 0.21-0.64; P < .001), total vegetable consumption frequency (ADID: 0.52 times/d; 95% CI, 0.12-0.91; P = .01), and total fruit and vegetable consumption frequency (ADID: 1.03 times/d; 95% CI, 0.49-1.56; P < .001) than comparison respondents. Most intervention participants reported boxes contained the right amount of food (71.9%) and high-quality produce (68.4%) and were willing to purchase a future produce box (85.0%). Findings indicate that a produce box distribution model increased fruit and vegetable consumption, had high satisfaction among participants, and generated interest in purchasing affordable produce boxes. Future studies should explore feasibility of offering low-cost produce boxes at grocery stores and determine appropriate pricing models to enhance access and sustainability.