Disordered Eating Behavior, Diabetes-Related Distress, and HbA1c among Adolescents and Young Adults with Type 1 Diabetes

Abstract

Background: Adolescents and young adults (AYA) with type 1 diabetes (T1D) face elevated risk for poor glycemic control due to both psychosocial and social determinants related factors. Disordered eating behaviors (DEBs) and diabetes distress (DD) are two key psychosocial factors linked to worse outcomes. While DEBs and DD are well-documented contributors to poor glycemic outcomes in AYA with T1D, less is known about how these factors may be modified by unmet social needs. Objective: To examine (1) associations of DEBs and DD with Hemoglobin A1c (HbA1c), and (2) whether unmet social needs modify these relationships, in a cohort of AYA with T1D. Methods: This cross-sectional study used baseline data from 530 AYA (ages 15–23) enrolled in the Achieving Health in Emerging Adults with Diabetes (AHEAD) program at Seattle Children’s and UW Medicine. DEBs and DD were assessed using the Diabetes Eating Problem Survey – Revised (DEPS-R) and Problem Areas in Diabetes – Teen Version (PAID-T) tools, respectively. Unmet social needs were characterized using a 9-item screening tool. HbA1C and demographic measures were obtained from medical records. Linear regression models adjusted for age, sex, and diabetes duration were used to assess associations of DEBs and DD with HbA1c. Potential effect modification of associations by unmet social needs was examined using stratified analyses and models with interaction terms. Results: Of 530 participants (mean age 19 ± 2, 46% female, mean HbA1c 8.7% ± 2.2%, 73% non-hispanic white, 73% private insurance), 19% screened positive for DEBs and 19% reported DD. Both DEBs and DD were statistically significantly associated with higher HbA1c levels (DEBs: β = 1.39, 95% CI: 0.86–1.91, p-value: <0.0005; DD: β = 1.94, 95% CI: 1.42–2.46, p-value: <0.001). Stratified models showed that DD was associated with HbA1c among those without unmet social needs (β = 3.02, CI: 1.12-4.91, p = 0.002), but not among those reporting one or more needs (β = -0.46, CI: -3.99-3.06. p = 0.78) (interaction p-value=0.86). Differences in DEB-HbA1c or DD-HbA1c associations were not observed among groups defined by social needs. Conclusion: DEBs and DD are positively associated with worse glycemic control in AYA with T1D. The DD-HbA1c association was observed among youth without reported social needs, although the effect modification was not statistically significant. Integrating psychosocial and social risk screening into diabetes care may support improved outcomes.