Social needs screening and uptake patterns of a produce prescription program in an urban pediatric primary care clinic

OBJECTIVE: The WE CARE social needs screener (WC screener) is a self-report instrument for 8 social needs. Our objective was to determine whether screening positive for wanting help with food is associated with increased odds of enrollment in Fresh Connect (FC), a produce prescription program. 

METHODS: Families who completed the WC screener at a well-child visit (n=1049) and spoke English, Spanish, or Portuguese were called within two weeks and offered FC regardless of screener results. Child age, race and ethnicity, caregiver's preferred language, and insurance were extracted from medical records. A multivariable adjusted logistic regression model was utilized to describe the association between WC responses and likelihood of enrolling in FC. 

RESULTS: Overall, 38% of the children were of non-Hispanic White race/ethnicity; 97% of caregivers spoke English, and 52% utilized public insurance. Among caregivers who completed the WC screener, 6% requested help with food. Overall, 20% enrolled in FC. Compared to those who said "no", saying "yes" to wanting help with food did not increase odds of FC enrollment after adjusting for other variables (aOR 0.93, 0.44-1.97), yet saying "maybe later" did (aOR 3.2, 95% CI 1.2-8.3). Additionally, combining "yes" and "maybe later" also did not have significantly higher odds of enrollment compared to those who said no or left the question blank (aOR 1.46, 95% CI 0.86-2.46). 

CONCLUSIONS: This pilot study is the first to explore concordance between screening responses and produce prescription enrollment. Responding "yes" to wanting help with food on the WC screener was not predictive of FC enrollment, which suggests that future primary care-based interventions should evaluate the role of social needs screeners in the referral pathway and consider, when possible, offering food-related assistance to all families regardless of screening responses.