Automated Insulin Delivery Use Effective During Hospitalization

Why It Matters

The findings demonstrate that closed‑loop insulin delivery can close a long‑standing gap in inpatient diabetes care, offering safer, more effective glucose management and potentially reducing complications and costs for hospitals.

Key Takeaways

• Automated insulin delivery raised time-in-range to 67.7% vs 40.8% control.
• No increase in hypoglycemia despite higher basal insulin doses.
• Severe hyperglycemic events fell from 7.7% control to 0% automated.
• Implementation needs extensive nurse training but proved feasible in trial.
• Trial enrolled 130 adults with type 1 and type 2 diabetes.

Pulse Analysis

Inpatient diabetes management has traditionally relied on basal‑bolus insulin regimens, which often fall short of achieving target glucose ranges, especially in acutely ill patients. The recent AIDING trial, presented at the American Diabetes Association Scientific Sessions, introduced a closed‑loop system—Omnipod 5 paired with Dexcom G7 CGM—into three U.S. hospitals. By enrolling a balanced cohort of type 1 and type 2 patients across diverse racial groups, the study provided a realistic snapshot of how automated insulin delivery performs under real‑world hospital conditions, moving beyond earlier outpatient‑focused research.

The trial’s outcomes were striking: patients on the automated system spent nearly 68% of their time within the 70‑180 mg/dL range, a 27‑percentage‑point jump over standard care. Importantly, this improvement did not come at the expense of safety; rates of hypoglycemia (<54 mg/dL) were unchanged, and severe hyperglycemic excursions (>400 mg/dL) vanished in the intervention arm. While total daily insulin increased modestly, the basal component rose more than the bolus, suggesting the algorithm fine‑tuned background insulin delivery to match fluctuating metabolic demands. These data signal that closed‑loop technology can deliver tighter glucose control without adding hypoglycemia risk, a balance that has eluded clinicians for decades.

Translating these results into routine practice, however, hinges on workflow integration and staff education. Hospital nurses expressed initial hesitancy, but hands‑on training and demonstrable patient benefits shifted attitudes, underscoring the importance of comprehensive onboarding programs. As health systems grapple with rising diabetes‑related costs, adopting automated insulin delivery could lower complication rates, shorten lengths of stay, and improve overall quality metrics. Continued research will be essential to refine protocols, assess cost‑effectiveness, and scale the technology across diverse care settings, positioning it as a potential new standard for inpatient glycemic management.