A Simulation-Based Discrete Choice Experiment Framework for Alzheimer’s Disease-Modifying Therapy Implementation: Methodology and Interactive Decision Support Tool

Background: Health technology assessment bodies evaluating Alzheimer’s disease-modifying therapies (DMTs) require preference-based evidence to support patient-centred decision-making. However, primary DCE data collection across multiple stakeholder groups and countries is resource-intensive and time-consuming. No published preference evidence currently exists for DMT implementation outside the United States. This paper presents a simulation-based DCE framework with an accompanying interactive decision support tool, designed to anticipate likely preference structures, demonstrate the analytical pipeline, and provide a ready-to-deploy platform for real-world data collection by national working groups. Methods: A complete DCE analytical framework was developed comprising nine attributes across three domains: treatment characteristics (efficacy, ARIA risk, mode of administration, treatment duration), implementation factors (diagnostic testing, safety monitoring, geographic access, waiting time), and cost. Monte Carlo simulation of DCE response data was calibrated to published clinical trial evidence, healthcare system data, and expert elicitation from the Alzheimer’s disease health technology assessment literature. Simulated choice data for four stakeholder groups (patients/caregivers, clinicians, HTA professionals, general public) were analysed using conditional logit, mixed logit, and latent class models. An interactive web-based decision support tool was developed to enable policymakers to explore preference estimates, simulate market scenarios, and evaluate budget impact under different implementation configurations. Results: Under calibrated simulation parameters, slowing of cognitive decline emerged as the most important attribute (relative importance 25.7%), followed by annual cost (17.6%), ARIA risk (15.4%), and travel to treatment centre (13.0%). Implementation-related attributes collectively accounted for 42.5% of total attribute importance, exceeding treatment characteristics alone (35.9%). Simulated willingness-to-pay for a 10-percentage-point improvement in cognitive decline slowing was €12,857 per year. Latent class analysis identified three illustrative preference segments: efficacy-driven (42%), access-prioritising (35%), and risk-averse (23%). Sensitivity analysis across a wide range of plausible parameter values confirmed the robustness of the finding that implementation factors rival treatment characteristics in importance. Conclusions: This simulation-based framework provides a methodologically rigorous, immediately deployable platform for preference elicitation in Alzheimer’s DMT implementation. The interactive dashboard enables national DMT working groups and HTA bodies to adopt the framework for real-world data collection, replacing simulated parameters with empirical estimates while retaining the validated analytical infrastructure. The framework is designed for direct adoption by national DMT working groups and HTA bodies globally.