Evaluation of a State-Level Incentive Program to Improve Diet

TL;DR

A state-level program offering SNAP participants a 50% discount on fruits and vegetables did not significantly improve overall dietary intake but helped those already eating healthily.

Contribution

The study evaluates the effectiveness of a state-level SNAP incentive program and identifies differential impacts based on baseline dietary habits.

Findings

The program was not associated with significant overall increases in fruit and vegetable intake.

Participants with higher baseline fruit and vegetable consumption saw a significant increase.

Program awareness and usage were reported by 36.8% and 26.4% of participants, respectively.

Abstract

Is a state-level Supplemental Nutrition Assistance Program (SNAP) fruit and vegetable incentive program with automatic enrollment associated with improved dietary intake among adults with low income?

In this pre-post cohort study with a comparison site including 725 SNAP participants, the 50% incentive program resulted in a modest, nonsignificant increase in fruit and vegetable intake (0.12 cup equivalents per 1000 kcal). Participants with higher baseline consumption of fruits and vegetables showed significant increases (0.29 cup equivalents per 1000 kcal), while consumers with a lower baseline level showed no change.

This study suggests that financial incentives may be associated with improved diet among individuals already inclined toward healthy eating, indicating that targeted implementation strategies are needed.

Fruit and vegetable incentive programs show promise for improving dietary quality among Supplemental Nutrition Assistance Program (SNAP) participants, but few programs have been implemented at scale. Understanding state-level program effectiveness is critical for informing nutrition assistance policy.

To evaluate the association of Eat Well, Be Well (EWBW), the first state-level SNAP fruit and vegetable incentive program launched in Rhode Island, with dietary intake and program awareness during initial implementation.

In this pre-post cohort study with a comparison site (Connecticut) using difference-in-differences analysis, baseline data were collected from May 15 to September 30, 2023, with follow-up from June 12 to October 6, 2024 (5-8 months after implementation). Community-based recruitment took place across Rhode Island (intervention) and Connecticut (comparison), including food pantries and health fairs, as well as via text messages sent to participants from the Special Supplemental Nutrition Program for Women, Infants, and Children. Participants were current SNAP recipients, aged 18 years or older, English or Spanish speaking and had email or text access.

Automatic enrollment providing a $0.50 credit per $1 spent on fresh fruits and vegetables (up to $25/mo) at 2 grocery chains, with credits deposited on Electronic Benefits Transfer cards.

The primary outcome was change in total fruit and vegetable intake (measured as cup equivalents per 1000 kcal) via validated food frequency questionnaire. Secondary outcomes included diet quality (measured by the Healthy Eating Index–2015), program awareness, and use of the program.

Of the 725 participants (mean [SD] baseline age, 34.5 [10.1] years; 364 in Rhode Island and 361 in Connecticut), 690 were women (95.2%), and the mean (SD) household size was 3.9 (1.6) at baseline and follow-up. Baseline mean (SD) fruit and vegetable intake was 2.05 (1.19) cup equivalents per 1000 kcal in Rhode Island and 2.12 (1.25) cup equivalents per 1000 kcal in Connecticut. There was no significant change in fruit and vegetable intake in Rhode Island relative to Connecticut (0.12 [95% CI, −0.04 to 0.28] cup equivalents per 1000 kcal). Stratified analyses among those with higher baseline consumption of fruits and vegetables (>1.79 cup equivalents/1000 kcal) showed that fruit and vegetable intake significantly increased in Rhode Island relative to Connecticut (0.29 [95% CI, 0.03-0.54] cup equivalents per 1000 kcal), while those with lower baseline consumption of fruits and vegetables showed no change (0.00 [95% CI, −0.17 to 0.18] cup equivalents per 1000 kcal). A total of 134 of 364 Rhode Island respondents (36.8%) correctly identified the program’s purpose and 96 of 364 (26.4%) reported using discounts available through the EWBW.

In this cohort study of SNAP participants, the 50% incentive, automatic enrollment EWBW program was not associated with significant relative changes in fruit and vegetable intake, but was associated with benefits among participants already consuming more fruits and vegetables. Enhanced implementation, including broader retail partnerships, improved outreach, and targeted interventions for populations with lower intake of fruits and vegetables, may be needed to maximize public health outcomes.

This pre-post cohort study evaluates the association of a state-level Supplemental Nutrition Assistance Program fruit and vegetable incentive program with improved dietary intake and program awareness during initial implementation.

Figures