Affiliation: Gillings School of Global Public Health, Department of Nutrition
Many low-income populations have limited access to healthful foods, which is associated with poor health outcomes. Federal nutrition education programs often do not address access to healthier foods. Integrating an affordable Community Supported Agriculture (CSA) program within these nutrition education programs may increase program impact. There is a need to understand how this program should be designed to meet customer needs and be successfully delivered through existing federal nutrition education programming. As part of a larger randomized trial testing a CSA program delivered through an Extension model in four US states, this mixed-methods formative evaluation included: 1) survey of low income adults (n=262) participating in federal nutrition programs across North Carolina, 2) choice experiment to determine program preferences among a low income population (n= 42) in four US states, and 3) in-depth interviews of Extension community nutrition educators (n=20) in four states followed by focus groups to understand perceived barriers and facilitators to implementing the proposed program. Nutrition Educators showed strong interest in the idea of a CSA plus education program. Making the program convenient, educational, and involving children was important, with staff time and program logistics being potential issues. The choice experiment indicated that the ideal CSA program would be a large mixed variety box, distributed 2 times per month, less than $15, no more than 10 minutes further than the supermarket from their home, and less expensive but no more than 20% more expensive than supermarket prices. There were statistically significant differences in willingness to participate given certain program scenarios across race and household size. The survey indicated high overall interest (85%), and more interest in the nutrition education program if there was a CSA (84%). There were statistically significant differences in willingness to participate by Race (p=.03), but not by Age (p=.70) or BMI (p=.057). Adaptations of the typical CSA disbursement frequency and price points may be needed to be attractive to low income populations. Results of this research were used to inform the larger randomized trial of CSA program impact on dietary intake and economic opportunity for farmers.