Comparison of Fruit and Vegetable Intake Among Urban Low-Income US Adults Receiving a Produce Voucher in 2 Cities

IMPORTANCE: Fruit and vegetable vouchers have been implemented by cities and counties across the US to increase fruit and vegetable intake and thereby improve overall nutritional quality. OBJECTIVE: To determine whether and why use of fruit and vegetable vouchers are associated with varied nutritional intake across different populations and environments. DESIGN, SETTING, AND PARTICIPANTS: In a population-based pre-post cohort study of 671 adult participants with low income before and during (6 months after initiation) participation in a 6-month program, fruit and vegetable vouchers were distributed for redemption at local San Francisco and Los Angeles neighborhood grocery and corner stores between 2017 and 2019. A transportability analysis was performed to identify factors that may explain variation in voucher use between cities. EXPOSURE: Receipt of $20 per month in produce vouchers for 6 months from 2017 to 2019. MAIN OUTCOMES AND MEASURES: Change in total fruits and vegetables (as defined by the US Department of Agriculture) consumed per person per day (change in cup-equivalents between month 6 and month 0). RESULTS: A total of 671 adults (median age, 54.9 years [interquartile range, 45.0-65.0 years]; 61.7% female; 30.9% Black; 19.7% Hispanic) were enrolled. An increase in fruit and vegetable intake of 0.22 cup-equivalents per day overall (95% CI, 0.14-0.31 cup-equivalents; P < .001) was observed. However, the observed increase was larger in Los Angeles compared with San Francisco (0.64 cup-equivalents per day; 95% CI, 0.41-0.88 cup-equivalents vs 0.10 cup-equivalents per day; 95% CI, 0.01-0.19 cup-equivalents). When the concurrently sampled San Francisco group (n = 157) was weighted in transportability analysis to demographically match the Los Angeles group (n = 155) in observed covariates, the weighted San Francisco group had an estimated increase of 0.53 fruit and vegetable cup-equivalents per day (95% CI, 0.27- 0.79 cup-equivalents, P = .03), with income being the variable needed to allow the 95% confidence intervals to overlap between the weighted San Francisco and unweighted Los Angeles populations. CONCLUSIONS AND RELEVANCE: In this study, the use of fruit and vegetable vouchers appeared to be associated with greater benefit among those with lower incomes, suggesting that further investigation of flat-rate rather than income-scaled benefits is warranted.