Introduction and Objective: Pediatric Type 1 Diabetes (T1DM) management relies on a complex social architecture. While index hospitalizations at diagnosis are nearly universal, subsequent readmissions represent a breakdown of this framework. This 1.5-year longitudinal study quantifies how socio-environmental instability predicts hospital utilization, a proxy for long-term glycemic risk. Methods: An 18-month longitudinal analysis was conducted on 214 unique patients. Social Determinants of Health (SDOH) were captured via parent surveys; clinical metrics and engagement (phone notes, clinic visits) were extracted via chart review. Additional variables were obtained via chart review when not available in structured EMR fields. A mixed-effects ordinal logistic regression model, adjusted for patient clustering, was utilized to identify predictors of increasing admission frequency. Results: Socio-environmental barriers were the primary drivers of readmission frequency. Lack of mental health care was a significant predictor, associated with a 4.14-fold increase in odds of higher-tier hospital utilization (p=0.001). Notably, clinical factors such as pump malfunction (p=0.88) and intercurrent illness (p=0.13) were not significant predictors of readmission. Care escalation occurred: patients with frequent admissions received significantly more medical touchpoints, including more phone notes (median 6.0 vs. 2.0; p0.001) and more frequent clinic visits (median 3.0 vs. 2.0; p=0.004). However, intensified clinical interventions did not mitigate risk from social instability. Frequent admissions were associated with a median A1c of 10.4%. Conclusion: These data suggest recurrent T1DM hospitalizations are largely social rather than purely medical. Readmissions reflect socio-environmental instability. Increasing the frequency of traditional clinical care (visits and calls) is insufficient to overcome structural instability. To improve long-term outcomes, health systems must transition from a reactive medical model to an integrated social and emotional infrastructure. Disclosure L. Higham: None. J. Lyon: None. R. Gonsalves: None. R. Chelala: None. E.A. Davis: None. V. Vaidya: None. P. Kang: None. C.R. Kapadia: Consultant; Ended; Novo Nordisk, MannKind Corporation, Rezolute.
HIGHAM et al. (Fri,) studied this question.