Letter: Comments on: Real-World Total Daily Insulin Dose in over 14,000 Young Individuals with Type 1 Diabetes in the United States

To the Editor,

Molinsky et al. ¹ should be congratulated for delivering one of the largest and most clinically relevant contemporary analyses of total daily insulin dose (TDD) in pediatric and young-adult type 1 diabetes. At a time when diabetes technology is rapidly advancing from glucose monitoring toward intelligent therapeutic ecosystems, their evaluation of 14,358 individuals provides far more than a dosing benchmark—it offers a developmental map of real-world insulin physiology across childhood, puberty, and young adulthood. Particularly compelling is the demonstration that insulin requirements evolve differently according to age, sex, body mass index, glycemic status, and technology modality, thereby underscoring that TDD is not a static pharmacologic variable but a dynamic systems-level biomarker reflecting endocrine maturation, behavioral adaptation, and therapeutic responsiveness. ¹ By translating large-scale real-world data into clinically interpretable trajectories, this study advances the field beyond conventional dose reporting and opens an important pathway toward precision pediatric diabetes management in the era of continuous glucose monitoring, insulin pumps, and automated insulin delivery.

The most promising implication is not merely that TDD changes with age, sex, and body mass index (BMI), but that TDD may function as an early “dose-intelligence” marker. The observation that mean HbA1c remained 8.4%, and that 80.2% of participants had HbA1c ≥ 7%, despite 80% continuous glucose monitoring use and 70% pump use, ¹ suggests that technology access alone does not guarantee metabolic success. This finding is consistent with current standards emphasizing diabetes technology, ² but it also highlights the need for adaptive workflows that combine device data with pubertal timing, insulin resistance, missed boluses, psychosocial context, and clinical inertia.

A future validation pathway should therefore move from descriptive dose benchmarking toward a pediatric dose-intelligence framework. Such a framework could integrate longitudinal TDD trajectories with CGM-derived time in range, BMI velocity, pubertal stage, insulin-delivery mode, socioeconomic risk, and acute-event history to identify children at risk of underinsulinization, rising insulin resistance, or technology-workflow mismatch. Recent evidence indicates that automated insulin delivery improves glucose management in children and adolescents, ³ yet global implementation must account for unequal access to insulin, CGM, pumps, and reimbursement. ⁴ Thus, TDD algorithms derived from high-resource settings should be externally validated across diverse health systems before being generalized.

Future research should move beyond static insulin-dose estimation toward globally connected, equity-sensitive pediatric dose-intelligence systems capable of anticipating metabolic vulnerability before deterioration occurs. International multicenter registries integrating TDD trajectories with continuous glucose monitoring metrics, pubertal staging, behavioral adherence, insulin-delivery modality, and social determinants of health could establish the first universally adaptable reference architecture for precision insulin management in children and adolescents with type 1 diabetes. Such frameworks may enable earlier recognition of insulin resistance transitions, therapeutic inertia, technology-workflow mismatch, and impending diabetic ketoacidosis risk while simultaneously improving personalization of automated insulin-delivery systems. Importantly, the next era of diabetes technology should not be defined solely by device sophistication but by its capacity to deliver equitable clinical intelligence across diverse populations and healthcare settings worldwide. ⁵ In conclusion, Molinsky et al. ¹ have provided an exceptionally valuable real-world foundation upon which the field can build a new generation of predictive, developmentally adaptive, and globally scalable diabetes care models. This letter aims to support that translational evolution—from descriptive insulin dosing toward precision pediatric dose intelligence capable of advancing safer, smarter, and more equitable diabetes care worldwide.