P78 Low-Income Adults Enrolled in a Cost-Offset, Community-Supported Agriculture Intervention are not Nationally Representative

Abstract

<h3>Background</h3> Cost-offset, Community-supported Agriculture (CO-CSA) is a model in which low-income households receive subsidized shares of a local farm's harvest. The goal is to address diet and health disparities among a population with historically poor access to and consumption of healthy foods. The appeal of such an intervention, and thus its reach, is limited to a self-selected sub-sample of the low-income population that may be non-representative. <h3>Objective</h3> To compare characteristics of CO-CSA enrolled adults from low-income households with those of a nationally-representative sample of low-income adults. We hypothesized that CO-CSA enrollees would be less demographically diverse, have greater nutrition-related knowledge, and report better health than the national sample. <h3>Study Design, Setting, Participants</h3> CO-CSA enrollee data were drawn from the baseline survey of a multi-state randomized controlled trial testing the CO-CSA model in conjunction with nutrition education among low-income households with children in Vermont, New York, North Carolina, and Washington (n=305). Those data were compared to a nationally-representative sample of low-income adults from households with children evaluated via the National Health and Nutrition Examination Survey (2007-2010). <h3>Measurable Outcome/Analysis</h3> Descriptive statistics for demographic, food security, nutrition knowledge, and self-reported health status data were generated using SPSS for both samples. <h3>Results</h3> Compared to the national sample, a greater proportion of CO-CSA enrollees were female (97% vs 57%), white (76% vs 45%), highly educated (49% vs 7% college graduates), and food insecure (57% vs 49%); a smaller proportion were Hispanic (6% vs 34%). A greater proportion of CO-CSA enrollees knew the daily recommendation for fruits and vegetables (3-5 cups; 61% vs 16%) and reported being in good or excellent health (34% vs 25%). <h3>Conclusions</h3> Differences between the CO-CSA and NHANES samples may be related to the CO-CSA study sampling frame (English-speaking households in certain regions of the U.S.). Results support continued exploration into whether CO-CSA and other local food system interventions have the potential to reach beyond female, white, educated, and nutritionally-knowledgeable segments of the target population.