OBJECTIVES: Parkinson's disease (PD) affects approximately 8.5 million people globally, with projections suggesting an increase to over 12 million by 2040. Deep Brain Stimulation (DBS) offers substantial relief for advanced PD (aPD), yet timely identification of aPD remains challenging due to individual disease progression and the absence of reliable biomarkers. This study assesses the early cost-effectiveness of integrating Artificial Intelligence (AI) with Remote Monitoring Solutions (RMS) for the timely detection of aPD, aiming to enable earlier DBS interventions and enhance patient outcomes.

METHODS: We employed a Markov model reflecting a 16-year horizon from a Finnish healthcare perspective, incorporating data from controlled trials and observational studies. Costs and health effects were discounted at 3%. The model presumes all persons with PD (PwPD) develop aPD within ten years post-diagnosis. AI-enhanced RMS is posited to timely diagnose aPD with 100% accuracy, contrary to conventional care (CC) which experiences a one-year diagnostic delay. Under this model, 10% of aPD diagnosed patients are eligible and receive DBS.

RESULTS: The incremental cost-effectiveness ratio (ICER) for AI-enhanced RMS stood at €83,077 per quality-adjusted life year (QALY) gained, surpassing Finland's typical willingness-to-pay threshold. Scenario analyses suggest it becomes cost-effective if earlier DBS intervention delays nursing home admissions by one year, or if RMS alone boosts quality-of-life by 10%.

CONCLUSIONS: While AI-enhanced RMS does not appear to be cost-effective for timely aPD detection, its feasibility might improve with new evidence supporting the benefits of earlier DBS interventions. Furthermore, beyond traditional metrics, digital technologies may provide additional patient-centric benefits, including enhanced satisfaction, peace of mind from reduced diagnostic uncertainty, and the intrinsic value of hope, which extends to family spillover effects. These considerations suggest that Health Technology Assessment (HTA) methodologies should incorporate these broader impacts in the analysis of digital health technologies, thereby capturing the full value of innovations like AI-enhanced RMS.