By Synneva Hagen-Lillevik, PhD, MS, RD, 2023 ADCES/CBDCE Postdoctoral Fellow, University of Utah College of Nursing| University of Utah, and

Joy Doll, OTD, OTR/L, Program Director, Health Informatics, Associate Professor, Creighton University

Recently, several policy initiatives have recognized the importance of social determinants of health (SDOH) including the U.S. Playbook to Address Social Determinants of Health which supports “flexible funding to address social needs.” In addition, the Centers for Medicare and Medicaid and the Joint Commission are adding requirements for health-related social needs screening and even reimbursable codes for navigation to connect patients to resources to address their social needs. Gottlieb, et al. (2024), shared recent studies that have demonstrated evidence for a wide variety of programs that impact SDOH in a variety of ways. Food insecurity has been recognized as a priority health-related social need leading to programs including in the White House Challenge to End Hunger and Build Healthy Communities, which focuses on partners to promote healthy food and food delivery. In some cases, these include procured food boxes catered to the unique health needs of patients which has been coined “food as medicine.” The opportunity to combine efforts to address health-related social needs and chronic disease management by providing food catered to either prevent or manage chronic disease is now. However, how do we do that? What works? New and emerging evidence is evolving. 

With the prevalence of chronic diseases that are either prevented or treated with dietary interventions, “food-as-medicine” programs are a hot topic — especially in the context of cardiometabolic diseases like Type 2 diabetes. Doyle and colleagues performed a randomized clinical trial to address how effective one food-as-medicine program was in lowering hemoglobin A1C levels between treatment and control groups in a large group of individuals with Type 2 diabetes and self-reported food insecurity. Participants in the treatment group were invited to a clinic weekly to pick up fresh foods to cook at home, along with the opportunity to meet with a community health worker, a registered dietitian, and a nurse who specializes in diabetes care.

The authors conclusions were less than promising as they found no significant differences in A1C levels between the treatment and control groups after six months; surprisingly, both groups did show an overall decrease in A1C. However, it may not be time to write off food-as-medicine approaches as there were possible confounders in the study design that contributed to the results. This study was not unique in that measuring dietary interventions is inherently challenging and requires systematic approaches. A few pitfalls here are that food insecurity eligibility was limited to two questions on a questionnaire and not a more robust analysis of SDOH, male and female participants were not separated in the analysis (ignoring sex as a biological variable), and it is unknown if the food was consumed by the participant after it left the clinic. Targeted research methods that address SDOH and measure barriers/facilitators to participant engagement in the intervention are needed to truly evaluate the efficacy of food-as-medicine programs.