Effectiveness of Recipe4Health on Obesity and Cardiometabolic Health: A natural experiment

PROJECT SUMMARY The landmark White House Conference on Hunger, Nutrition, and Health in September 2022 underscored the urgency of improving food and nutrition security in the US as a fundamental strategy to promoting health equity. Food insecurity – the lack of consistent access to sufficient quantities of healthy food - disproportionately impacts communities of color and those with limited resources, deepening existing inequities with nutrition-sensitive chronic conditions like obesity, diabetes, and heart disease. To address the interrelated challenges of food and nutrition insecurity and chronic disease, healthcare organizations are increasingly integrating “Food as Medicine” nutrition interventions into healthcare to treat disease. A recent policy change in California provides a powerful, time-sensitive opportunity for a natural experiment to evaluate the impact of “Food as Medicine” in participants compared to controls from the same highly diverse patient population. In 2022, the California Advancing and Innovating Medi-Cal (CalAIM) Medicaid section 1115/1915(b) included a provision for medically supportive food for Medi-Cal (Medicaid in California) beneficiaries. It is extremely critical to start the data collection as soon as possible to be able to complete follow-up data collection before the policy expires. Alameda County, California, home to a diverse community of 1.6 million, has leveraged this policy change to offer Recipe4Health to Medi-Cal beneficiaries. Recipe4Health is a leading Produce Prescription model in the US that includes: 1) Produce Prescription (Food Farmacy); 2) Group and individual health coaching (Behavioral Pharmacy); and 3) Technical assistance to providers for clinical integration of Recipe4Health. We will use a quasi-experimental design to evaluate the impact of Recipe4Health on key markers of cardiometabolic health among patients with a BMI30 (e.g. glycemic control and cholesterol) using two strategies. Strategy 1 will take advantage of BMI and laboratory data available in the electronic health record (EHR),using our experience with retrospective propensity score matching. Strategy 2 will include a prospective random subsample (240 participants+ 240 propensity-score matched controls), with assessments at 0, 3, and 6 months. With this subsample, point-of-care testing at patient homes will support timely collection of key laboratory biomarkers (HbA1c and cholesterol), and telephone surveys will allow for assessment of food and nutrition security and high-quality dietary intake data (24-hour recall). We are uniquely positioned to conduct this study as we have already built bi-directional community partnerships, infrastructure to share EHR data, and the infrastructure to conduct point-of-care testing. Successful completion of these aims will represent a large, generalizable, and rigorous assessment of Food as Medicine in a highly diverse low- income patient population (~50% Latinx, 20% non-Hispanic Black, 10% Asian/Pacific Islander), with direct relevance to patients, the healthcare system, and policy makers.