UCSF

Objective

To examine the association between level of food security and diabetes self-management among food pantry clients, which is largely not possible using clinic-based sampling methods.

Design

Cross-sectional descriptive study.

Setting

Community-based food pantries in California, Ohio and Texas, USA, from March 2012 through March 2014.

Subjects

Convenience sample of adults with diabetes queuing at pantries (n 1237; 83 % response). Sampled adults were stratified as food secure, low food secure or very low food secure. We used point-of-care glycated Hb (HbA1c) testing to determine glycaemic control and captured diabetes self-management using validated survey items.

Results

The sample was 70 % female, 55 % Latino/Hispanic, 25 % white and 10 % black/African American, with a mean age of 56 years. Eighty-four per cent were food insecure, one-half of whom had very low food security. Mean HbA1c was 8·1 % and did not vary significantly by food security status. In adjusted models, very-low-food-secure participants, compared with both low-food-secure and food-secure participants, had poorer diabetes self-efficacy, greater diabetes distress, greater medication non-adherence, higher prevalence of severe hypoglycaemic episodes, higher prevalence of depressive symptoms, more medication affordability challenges, and more food and medicine or health supply trade-offs.

Conclusions

Few studies of the health impact of food security have been able to examine very low food security. In a food pantry sample with high rates of food insecurity, we found that diabetes self-management becomes increasingly difficult as food security worsens. The efficacy of interventions to improve diabetes self-management may increase if food security is simultaneously addressed.