Presentation (CTI)

OBJECTIVES: Type 2 diabetes is one of the costliest chronic conditions in the United States, resulting in higher medical utilization, prescription costs, and lost productivity for employers. Twin Health’s AI-enabled digital twin technology, which creates personalized metabolic models paired with clinical support, is a digital health intervention aimed at improving metabolic outcomes. This study evaluated the total cost of care associated with participation in Twin Health’s digital twin program among employees with type 2 diabetes.
METHODS: This study analyzed claims data for 228 employees with type 2 diabetes from a large employer with continuous eligibility. Claims from 2022-2025 were analyzed, covering medical (inpatient, emergency room, outpatient) and pharmacy services. Costs were compared between program participants and a propensity-score matched control group using a 12-month pre- and post-enrollment difference-in-differences approach.
RESULTS: Average savings per member were $9,047 over the 12-month study period compared to controls. Medical cost savings totaled $4,357 (48% of total), driven primarily by reduced hospitalizations and emergency room visits. Pharmacy savings totaled $4,690 (52% of total), with 66% attributed to reduced antidiabetic medication use, particularly GLP-1s and SGLT-2 inhibitors. Participants who tapered off GLP-1s achieved the largest prescription cost reductions.
CONCLUSIONS: Participation in Twin Health’s AI-enabled digital twin program was associated with meaningful reductions in total cost of care for employees with type 2 diabetes. Cost savings were driven by both reduced acute care utilization and lower prescription drug spending, including de-escalation of glucose-lowering therapies. These findings suggest that personalized, data-driven metabolic health interventions may offer a scalable strategy for cost containment in diabetes care. Estimated savings are conservative, as indirect costs such as absenteeism and productivity loss were not included.