A comprehensive simulation framework for evaluating U.S. lung allocation policies.

Abstract

The rapidly evolving landscape of lung allocation policy, including the recent adoption of the Continuous Distribution framework, demands simulation tools that are both flexible and computationally efficient. Existing lung allocation simulators are often difficult to adapt to new policy designs and can be computationally burdensome when large numbers of replications are needed for uncertainty quantification and optimization. This paper introduces the Modular Lung Allocation Policy Simulation Model (MoLAPS), a high-fidelity, discrete-event simulation framework designed to address these limitations through a modular policy architecture and parallel execution. MoLAPS integrates detailed patient and donor data, dynamic medical status updates, and a predictive organ-offer acceptance model. It is implemented with parallel computing support, substantially reducing runtimes for multi-replication and longer-horizon experiments and enabling the exploration of large-scale policy design spaces. The model features a modular architecture that isolates policy components, including allocation, acceptance, and prioritization rules, allowing policymakers to easily evaluate existing and proposed allocation policies, or hypothetical alternatives. A surrogate-based calibration procedure aligns simulation outputs with national organ transplant data, improving empirical validity. MoLAPS is rigorously validated using prediction intervals, quantile-quantile plots, and Chi-squared tests. Simulation results closely align with real-world data across key performance indicators, including transplant rates, waitlist mortality, and organ nonuse. MoLAPS provides clinicians, policymakers, and researchers with a timely, open, and extensible platform for policy evaluation. By enabling realistic, data-driven assessments of lung allocation strategies, it supports more equitable, efficient, and evidence-based transplant policy decisions.