A Collaborative Pilot to Support Patients With Diabetes Through Tailored Food Box Home Delivery

Abstract

We conducted a randomized, controlled prospective pilot study to determine feasibility and impact of food bank and health system collaboration to home-delivered food to adults with type 2 diabetes mellitus experiencing food insecurity. Treatment group received biweekly, ethnically tailored, home-delivered food for 24 weeks. Analysis included intervention feasibility and impact on healthcare utilization, HbA1c, and other health-related measures. Intervention was feasible and successful with high levels of participant satisfaction. At baseline, participants with highest HbA1c reported poorer health, lower medication adherence and self-care, and higher diabetes distress and medicine for food tradeoffs. At 24 weeks, treatment group reported improved food security and health status. There were no differences in HbA1c or healthcare utilization measures between the two groups. It is feasible for a community food bank and nearby hospital to successfully collaborate and provide supplemental food staples to food insecure adults with type 2 diabetes and improve food insecurity and health status. Public policy efforts should utilize and expand this strategy with the goal of improving health and reducing health disparities. Future work could include more comprehensive food support focused on those with poorest glycemic control, and expanded, coordinated interventions directed at other social determinants of health. Future programming and policies should be cocreated with community input to ensure greatest success.

A Systematic Review of Interventions That Address Food Insecurity for Persons With Prediabetes or Diabetes Using the RE-AIM Framework

“…With respect to medically tailored meals or groceries, participants picked up diabetes-appropriate food 3,29,40 or healthy produce 35,36,44 at a food pantry or distribution center. There were also 3 studies in which medically tailored meals or groceries were delivered to the participant's home, 34,43,45 and in 1 study, groceries were picked up by the participant or their surgoate. 30 With respect to produce prescriptions/vouchers, vouchers were provided for fresh produce 32,33,[37][38][39]41,42,46,47 or healthy food 31 at specific locations or farmers markets.…”

“…Mean sample size was 256 participants. There were 10 studies with a sample size less than 100, [30][31][32][33][34][35][36][37][38][39] 5 studies with a sample size between 100 and 500, [40][41][42][43][44] and 5 studies with a sample size greater than 500. 3,29,[45][46][47] The average age of participants ranged from 48.5 to 61.5 years, with a mean age of 55.40 years.…”

“…3,30,40 Self-efficacy was improved in 40% of studies that evaluated this outcome (2/5 studies). 3,31 Improvements in other psychosocial outcomes were promising, with 67% (4/7) of studies that evaluated depressive symptoms/diabetes distress 3,30,31,40 and 100% (5/5) of studies that evaluated quality of life 34,40,[43][44][45] reporting significant improvements over time.…”

“…The target location of interventions included farmers markets/green markets (n = 9), 32,33,[37][38][39]41,42,47 food banks (n = 4), 3,29,40,46 community-based distribution centers (n = 4), 30,31,35,44 or home delivery (n = 3). 34,43,45 There was limited information describing the setting included in study reports. In 2 studies by the same author, established food banks were invited to submit an application and were selected based on a competitive process.…”

“…3,29 The delivery agent (or who provided the intervention) was not specified in many of the studies (n = 8). 30,33,[35][36][37]41,45,47 Other studies identified food pantry staff (n = 3), 3,29,40 health care provider (n = 6), 31,32,38,39,44,46 or guidance by a dietician (n = 5) 3,29,34,42,43 as delivery agents.…”