COST SAVINGS OF AN AI-ENABLED DIGITAL TWIN PROGRAM IN PREDIABETES AND HEALTHY WEIGHT MANAGEMENT
Author(s)
Tyler Koep, PhD, Caleb Brooks, BA, Aleksandra V. Schifman, BS;
Twin Health, New York, NY, USA
Twin Health, New York, NY, USA
OBJECTIVES: Early-stage metabolic interventions can prevent progression to type 2 diabetes (T2D) and reduce associated healthcare costs. Preventing progression at this stage is critical to reducing long-term diabetes-related complications and healthcare expenditures. Twin Health’s AI-enabled digital twin technology pairs personalized metabolic models with clinical support to guide lifestyle changes and medication optimization. This study assessed clinical outcomes and cost savings over 24 months for participants in a prediabetes and healthy weight management program, including GLP-1 medication de-escalation.
METHODS: Data from 302 participants were collected continuously over 24 months, including biometric sensor measurements and pharmacy records. Outcomes used to inform the economic model included weight change, GLP-1 utilization, and T2D incidence. Cost estimates were derived by comparing pre- and post-program medication use and health improvements. Economic modeling incorporated both observed clinical outcomes and published evidence on weight-loss-related cost reductions to estimate program value.
RESULTS: At 24 months, participants lost an average of 16 lbs, and 95% remained free from progression to T2D. Medication reduction was strongly associated with savings. Among GLP-1 users, 67% reduced or discontinued therapy by the end of Year 1. Projected gross savings were $2,756 per participant in Year 1, reaching total savings of $7,532 over two years as weight loss was maintained and medication use declined.
CONCLUSIONS: Participation in Twin Health’s AI-enabled digital twin program supported sustained weight management, reduced dependence on costly GLP-1 therapy, and lowered T2D progression risk, delivering substantial economic value. These estimated savings are conservative, as they exclude indirect costs and the long-term benefits of disease prevention. These findings demonstrate that precision health approaches can meaningfully reduce healthcare costs while preventing metabolic disease progression.
METHODS: Data from 302 participants were collected continuously over 24 months, including biometric sensor measurements and pharmacy records. Outcomes used to inform the economic model included weight change, GLP-1 utilization, and T2D incidence. Cost estimates were derived by comparing pre- and post-program medication use and health improvements. Economic modeling incorporated both observed clinical outcomes and published evidence on weight-loss-related cost reductions to estimate program value.
RESULTS: At 24 months, participants lost an average of 16 lbs, and 95% remained free from progression to T2D. Medication reduction was strongly associated with savings. Among GLP-1 users, 67% reduced or discontinued therapy by the end of Year 1. Projected gross savings were $2,756 per participant in Year 1, reaching total savings of $7,532 over two years as weight loss was maintained and medication use declined.
CONCLUSIONS: Participation in Twin Health’s AI-enabled digital twin program supported sustained weight management, reduced dependence on costly GLP-1 therapy, and lowered T2D progression risk, delivering substantial economic value. These estimated savings are conservative, as they exclude indirect costs and the long-term benefits of disease prevention. These findings demonstrate that precision health approaches can meaningfully reduce healthcare costs while preventing metabolic disease progression.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
EE322
Topic
Economic Evaluation
Topic Subcategory
Trial-Based Economic Evaluation
Disease
SDC: Diabetes/Endocrine/Metabolic Disorders (including obesity)