Forecasting Lung Transplant Benefit: A Dynamic Risk Modeling Approach

点击翻译按钮获取中文摘要

Project Summary Increasing clinical demand for lung transplants has exacerbated the problem of rationing this limited yet life- saving societal resource. The Lung Allocation Score (LAS) system was developed to improve overall survival by identifying patients who would likely benefit the most from transplant. Despite this effort, there have been increasing rates of waiting list mortality, declines in long-term survival after transplant and dramatic increases in healthcare costs and utilization among transplant patients. Our project focuses on improving the LAS system by: 1) designing better methodologies to more accurately identify the progression of illness in a patient who is awaiting transplant, 2) predicting ideal timing of transplant to maximize the number of years gained from a transplant, and 3) evaluating different allocation strategies and their impact on individual and population level survival. We will achieve this by carrying out the following aims: Aim 1: Update the lung allocation score (LAS) underlying risk models to better accommodate subpopulation-level differences over time among lung transplant candidates. Aim 2: Develop and validate a forecasting model for lung transplant candidates’ dynamic health state and likelihood of transplantation over time using a systems-based microsimulation modeling approach. Aim 3: Evaluate the impact of lung allocation strategies that optimize patient- and population-level functional and survival outcomes. The results of this work will provide the foundation for improving lung allocation in the United States. We will optimize timing of lung transplantation to maximize transplant benefit at the individual patient and population levels. The methods identified in this project can be utilized in other scenarios where limited life saving resources must be rationed.