Household food insecurity (HFI) continues to be a significant public health challenge in the United States (US) with more than 1 in 10 households and 1 in 8 households with children identified to be food insecure in 2021.¹ The SEARCH for Diabetes in Youth (SEARCH) study recently reported that youth and young adults (YYA) with type 1 diabetes are even more disadvantaged, as almost 1 in 5 (18%) YYA with type 1 diabetes are food insecure.² Given that diabetes management requires careful alignment of medication and dietary intake, YYA with type 1 diabetes face additional health challenges secondary to uncertainty in the availability of food stemming from HFI.

Defined by the American Diabetes Association as the unreliable availability of nutritious food and the inability to consistently obtain food without resorting to socially unacceptable practices,^3,4 HFI can negatively influence glycemic control through its potential impact on nutrition therapy, physical activity, and diabetes self-management. Thus, HFI represents a potentially modifiable risk factor for adverse diabetes outcomes among socially disadvantaged populations. Pilot work carried out through the SEARCH Food Security study demonstrated that YYA with type 1 diabetes who lived in food insecure households had 2.4 higher adjusted odds of having a hemoglobin A_1c (HbA_1c) >9.0% (>75 mmol/mol) compared to those living in a food secure household.⁵ Furthermore, persons with marginal food security, who technically are classified as food secure on the U.S. Household Food Security Survey Module (HFSSM), had an elevated risk of higher HbA_1c. This pilot study, however, enrolled participants in Washington and South Carolina only, which limits its generalizability. Thus, these findings require confirmation in a larger and more diverse sample, as these findings would have strong implications for practice guidelines, including more nuanced recommendations for diabetes care depending on level of HFI and HbA_1c.

Studies examining the impact of HFI in adults with chronic disease have found that food-insecure adults with chronic disease must often choose between paying for health care and for food, leading to suboptimal adherence to medications, prolonged periods of fasting, and increased health care utilization.^6,7 Similarly, in older adults with type 2 diabetes, HFI has been found to be associated with more episodes of severe hypoglycemia and emergency department visits and hospitalizations.^8–11 In YYA with type 2 diabetes, food insecure YYA have been reported to have higher odds of diabetic ketoacidosis (DKA) compared to YYA living in food secure households.¹² To our knowledge, no population-based studies have examined the association between HFI and rates of DKA and severe hypoglycemia in YYA with type 1 diabetes. We sought to examine the association of HFI with 1) glycemic control, 2) episodes of DKA, and 3) episodes of severe hypoglycemia among a large cohort of YYA with type 1 diabetes from the SEARCH for Diabetes in Youth study.

The overall prevalence of HFI in YYA with type 1 diabetes was 16.4%, with 8.6% reporting living in households with low food security and 7.8% reporting very low food security. Additionally, 11.7% were identified as marginally food secure. The mean age for the sample was 20.8 ± 5.0 years (Table 1A). Most participants were female (52.1%), Non-Hispanic White (70.0%), and had private insurance (75.4%). Approximately 18% of the sample reported a household income less than $25,000 per year with an additional 21% reporting a household income $25,000 to $50,000.

The mean HbA_1c for YYA in this study was 9.2% ± 2.0% SD (77 mmol/mol ± 22 mmol/mol SD; Table 1B). In the past 12 months, 18.2% reported experiencing DKA and 9.9% reported severe hypoglycemia. Of participants reporting an acute diabetes-related complication, 2.4% experienced both a DKA and severe hypoglycemia episode in the past 12 months.

YYA who were food insecure had a HbA_1c that was 0.33% (95% CI: 0.003–0.657) higher than those who were food secure when adjusting for age, sex, race, participant education, parental education, income, health insurance, study site, insulin pump use, CGM use, and diabetes duration. When evaluated as a continuous variable, we observed a positive linear relationship between HFSSM scaled score and HbA_1c (Figure 1). In the linear regression analysis, a 1-point higher HFSSM scaled score was associated with a 0.12% (95% CI: 0.05–0.18) higher HbA_1c.

For YYA that had HFI, the odds of experiencing DKA was 1.58 (95% CI: 1.13–2.21) times the odds of experiencing DKA for those without HFI after adjusting for all covariates (Table 2). When using 4 categories of HFI, the adjusted odds of experiencing DKA was the highest for YYA reporting very low food security (OR 1.87, 95% CI: 1.19–2.92), similarly elevated but not statistically significant for the low food security group (OR 1.51, 95% CI: 0.96–2.40), but much weaker in the marginal food security group (OR 1.19, 95% CI: 0.78–1.81). With respect to severe hypoglycemia, while the odds of experiencing severe hypoglycemia among the food insecure YYA compared to those who were food secure were higher, this finding did not reach statistical significance after adjusting for covariates (OR 1.53, 95% CI: 0.99–2.37).

Data from this large multicenter study demonstrates that HFI is associated with higher HbA_1c levels and higher likelihood of episodes of DKA in US YYA with type 1 diabetes. That HFI is associated with higher HbA_1c levels is consistent with two smaller studies examining food insecurity and its association with glucose levels in U.S. and Canadian YYA with type 1 diabetes.^5,14 Unlike the Mendoza et al. study, however, we did not identify an inverted U-shaped non-linear relationship of HFI with HbA_1c.⁵ Instead, in our larger sample of YYA with type 1 diabetes, we found a positive linear relationship between the two variables such that as higher food insecurity was associated with higher HbA_1c. In addition, the odds of DKA were higher for food insecure YYA as compared to YYA from food secure households.

These findings extend our understanding of food insecurity among YYA with type 1 diabetes, which is an understudied population with respect to HFI. Several mechanisms have been proposed through which HFI may influence glucose levels, including through behavioral, mental health, and nutrition paths.¹⁵ In type 1 diabetes, a condition that is dependent on insulin administration and glucose monitoring throughout the day to meet glycemic targets, perhaps the most influential mechanism is on diabetes self-management behaviors. Our finding that YYA who are food insecure have a HbA_1c that was 0.33% higher than those who were food secure provides additional support to the hypothesis that HFI affects health in YYA with type 1 diabetes and may divert resources away from reliable insulin administration and glucose monitoring. This study’s findings are in contrast to a recently published study examining HFI in YYA with type 2 diabetes.¹² In a smaller sample from the SEARCH study, Reid et al. did not observe an association between HFI and glycemic control in YYA with type 2 diabetes. Our results are, however, consistent with what cross-sectional studies in the adult literature have reported for type 1 and type 2 diabetes,^8,11 including a longitudinal study in adults with diabetes that demonstrated food insecurity was associated with a 0.6% higher HbA_1c.¹⁶

We found that every 1-point higher scaled HFSSM score was associated with a 0.12% higher HbA_1c. YYA with type 1 diabetes experiencing increasing severity of HFI may be forced to balance the costs associated with obtaining diabetes medications and supplies with maintaining adequate and nutritious food. As a result, food insecure individuals and their families may compromise on one or both aspects to reduce expenses.^17,18 These compromises also often necessitate a shift in dietary intake in YYA as poorer nutritional intake has been shown to be associated with food insecurity.¹⁹ Changes in intake from relatively expensive fruits and vegetables and toward foods that are relatively less expensive but lower nutritional value, is also likely to contribute to higher HbA_1c levels.²⁰

The demands for using finite household resources to meet competing needs are only magnified in an era of rapidly rising insulin prices²¹ and when cost of living, including food-related costs, are also at record levels. At its most extreme, food insecure YYA may omit insulin administration altogether. We found that the odds of experiencing DKA was 1.64 times the odds of experiencing DKA for those without HFI after adjusting for all covariates, which could be explained by insulin omission due to finite household resources. Reid et al. similarly reported that food insecure YYA with type 2 diabetes had 3.08 times the odds of experiencing DKA than YYA who lived in food secure households.¹² The relationship between HFI and underutilization of medical care is well documented,^6,7,17 with evidence that adults experiencing food insecurity are more likely than their food-secure counterparts to delay or forgo medical care because of cost concerns.²² The inability to afford to see a diabetes provider or fill prescriptions presents another potential challenge for food insecure YYA given the requirement to have active insulin prescriptions to support reliable daily insulin administration, increasing the risk for insulin omission.

Previous studies in adults with diabetes have reported food insecurity is a risk factor for clinically significant hypoglycemia.^10,23 We also observed higher odds of severe hypoglycemia in food insecure YYA, albeit this did not reach the threshold of statistical significance (1.53, 95% CI: 0.99–2.37) in our sample. The American Diabetes Association recommends that glycemic targets be tailored to the individual patients to keep them safe and to avoid significant hypoglycemia.²⁴ Thus, diabetes providers may relax glycemic targets when patients encounter severe hypoglycemic episodes, which may represent an additional mechanism by which HFI leads to higher average glucose levels. Food insecurity is frequently a transient and cyclic phenomenon and thus, the co-existence of significant hypoglycemia in the setting of hyperglycemia in food insecure YYA can be explained by periods of adequate food supplies alternating with inadequate supplies during the year. Given that the HFSSM queries whether the food security-related experience or behavior ever occurred in the past 12 months, more work should be done to characterize the frequency and duration of the food insecurity experience for YYA with type 1 diabetes within that 12-month window.

This study has some limitations. First, given the cross-sectional nature of the analysis, the direction of the proposed mechanistic paths cannot be inferred. Second, the time frame of the food insecurity questions is wider than that of the HbA_1c measurement. Thus, it is possible that HbA_1c measurements do not reflect concurrent food insecurity experiences. Third, our study may be impacted by recall bias since HFI, DKA episodes, and severe hypoglycemia events were self-reported. Finally, we were unable to assess emotional distress related to diabetes as a potential mechanism for reduced capability for diabetes self-management, as this may be worsened due to HFI and interfere with a patient’s ability to manage their diabetes. These limitations, however, are balanced by several strengths. Strengths of the present study include its large sample size, a study population drawn from what is the largest multiethnic population-based registry of individuals with youth-onset diabetes in the U.S. with demographic and clinical characteristics similar to those of the overall U.S. population, and the use of the full 18-item HFSSM, the U.S. reference measure for HFI.

This study underscores the importance of assessing HFI in clinical settings, so that diabetes care teams can identify patients who may have challenges with following certain prescribed treatments due to the competing demands exacted by HFI and who are at higher risk for acute diabetes-related complications.³ The Hunger Vital Sign is a validated two-item food insecurity screening instrument that has been shown to have 97% sensitivity when compared to the reference HFSSM,²⁵ and thus can be more easily implemented in the clinical setting. A global tool that can be used to evaluate the severity of food insecurity at the household or individual level and allows for comparison between countries is the Food Insecurity Experience Scale.²⁶ Regardless of the prevalence of food insecurity, local or national programs that identify food-insecure patients should be implemented in the context of institutional support for linking patients to more reliable food assistance programs and resources.²⁷

Our work has important public health implications because of the substantial morbidity and mortality associated with HFI, as well as suboptimal glycemic control in diabetes. The high prevalence of HFI in the U.S. documented in this study, and its association with worse glycemic control and potentially preventable episodes of DKA, highlight the need for further investigation on the most effective ways of eliminating food insecurity among YYA with type 1 diabetes. This is particularly relevant for low and middle-income countries where food insecurity prevalence is higher than in the U.S.²⁸ Altogether, our findings suggest that policies and programs to identify and alleviate HFI among YYA with type 1 diabetes are urgently needed and may improve their glucose levels and reduce health care utilization for potentially preventable acute diabetes-related complications.