AI-ENABLED VIRTUAL REALITY REHABILITATION: A HEALTH ECONOMIC FRAMEWORK FOR NEURO-ONCOLOGY
Author(s)
Farah Farahati, PhD1, Kevan Farahati, MD1, Evelyn Rizzo, MSc2.
1Global Wellness Advisors, Westlake, OH, USA, 2Mobility HEOR, Akron, OH, USA.
1Global Wellness Advisors, Westlake, OH, USA, 2Mobility HEOR, Akron, OH, USA.
OBJECTIVES: To develop a transparent, scenario‑based economic framework integrating SEER survivorship data, VR rehabilitation effectiveness evidence, and AI‑enabled measurement enhancements to estimate cost‑effectiveness, budget impact, and return on investment (ROI) for Medicare/public payers.
METHODS: A Markov model was considered but not used due to insufficient transition probability data. Instead, a scenario‑based decision‑analytic model was developed in alignment with ISPOR recommendations for early‑stage digital health technologies. The model adopted a Medicare/public payer perspective over a 5‑year horizon with 3% discounting. SEER data informed population size (~80,000 survivors) and baseline hospitalization rates. VR effectiveness parameters—including functional improvement, fall reduction, hospitalization avoidance, and QALY gains (0.05-0.15)—were synthesized from VR neurorehabilitation and cancer survivorship studies. AI‑enabled analytics were incorporated conceptually to support improved measurement precision, adherence monitoring, and scalable deployment. Scenario analyses varied uptake (25-50%), QALY gain (0.05-0.15), and program cost (±20%). Outputs included ICERs, budget impact, and ROI.
RESULTS: Across modeled scenarios, VR rehabilitation demonstrated favorable value profiles. ICERs ranged from $22,000-$48,000 per QALY depending on QALY gain and program cost assumptions. Five‑year budget impact estimates ranged from $40M to $250M in net savings, and ROI ranged from 0.9 to 1.8. Hospitalization avoidance—derived from VR‑related functional gains applied to SEER hospitalization rates—was the dominant driver of economic value. AI‑enabled enhancements strengthened implementation feasibility and supported higher‑uptake scenarios.
CONCLUSIONS: This study introduces a transparent, AI‑enabled scenario‑based economic framework for evaluating VR rehabilitation in neuro‑oncology survivorship. The approach demonstrates how early‑stage digital health interventions can be assessed using HTA‑aligned methods even when evidence is heterogeneous. Findings support the potential for VR rehabilitation to be incorporated into value frameworks, reimbursement deliberations, and future real‑world evidence initiatives.
METHODS: A Markov model was considered but not used due to insufficient transition probability data. Instead, a scenario‑based decision‑analytic model was developed in alignment with ISPOR recommendations for early‑stage digital health technologies. The model adopted a Medicare/public payer perspective over a 5‑year horizon with 3% discounting. SEER data informed population size (~80,000 survivors) and baseline hospitalization rates. VR effectiveness parameters—including functional improvement, fall reduction, hospitalization avoidance, and QALY gains (0.05-0.15)—were synthesized from VR neurorehabilitation and cancer survivorship studies. AI‑enabled analytics were incorporated conceptually to support improved measurement precision, adherence monitoring, and scalable deployment. Scenario analyses varied uptake (25-50%), QALY gain (0.05-0.15), and program cost (±20%). Outputs included ICERs, budget impact, and ROI.
RESULTS: Across modeled scenarios, VR rehabilitation demonstrated favorable value profiles. ICERs ranged from $22,000-$48,000 per QALY depending on QALY gain and program cost assumptions. Five‑year budget impact estimates ranged from $40M to $250M in net savings, and ROI ranged from 0.9 to 1.8. Hospitalization avoidance—derived from VR‑related functional gains applied to SEER hospitalization rates—was the dominant driver of economic value. AI‑enabled enhancements strengthened implementation feasibility and supported higher‑uptake scenarios.
CONCLUSIONS: This study introduces a transparent, AI‑enabled scenario‑based economic framework for evaluating VR rehabilitation in neuro‑oncology survivorship. The approach demonstrates how early‑stage digital health interventions can be assessed using HTA‑aligned methods even when evidence is heterogeneous. Findings support the potential for VR rehabilitation to be incorporated into value frameworks, reimbursement deliberations, and future real‑world evidence initiatives.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
EE254
Topic
Economic Evaluation
Disease
SDC: Neurological Disorders, SDC: Oncology, STA: Multiple/Other Specialized Treatments