06/10/2020
Breathing New Life into Lung Transplant Allocation: Policies Require Dynamic Forecast Modeling
A multi-disciplinary team of researchers has received a four-year, $3 million grant from the National Heart, Lung, and Blood Institute to develop an improved risk modeling approach to help prioritize patients with advanced lung diseases who need a transplant, standing to make a significant impact on lung allocation policy in the United States.
The current system to prioritize U.S. patients waitlisted for a lung transplant is called the Lung Allocation Score (LAS) system. This system takes into account predicted waitlist mortality (the likelihood a patient will die while waiting for transplant) and post-transplant survival (the extent to which receiving a transplant is likely to extend a patient’s life) when determining which patients are in greatest and most urgent need.
However, many researchers, clinicians and public health experts—including Jarrod Dalton, PhD, a data scientist in the Department of Quantitative Health Sciences and one of the grant’s principal investigators—believe the LAS leaves room for improvement.
“It’s important that the field adopts a model that goes beyond just one-dimensional mortality risk predictions, focusing instead on developing a more dynamic system that incorporates day to day changes in patients’ clinical status that undoubtedly impact their survival before and after transplant,” he explained.
The team’s model will account for interactions between sex, race/ethnicity and neighborhood socioeconomic factors. It will also track multiple disease indicators over time—including more than 10 measures of pulmonary, cardiac, and renal function—and how the interaction of one or more of these indicators dynamically affect a candidate’s disease state and, as a result, mortality risk. According to Dr. Dalton, this type of model is termed a “microsimulation model.”
“The survival risk prediction models that underlie the current LAS assume that the association between clinical risk factors and mortality risk are fixed across all transplant candidates in the end stages of lung disease,” said Maryam Valapour, MD, MPP, Director of Lung Transplant Outcomes, Respiratory Institute and co-principal investigator. “But in the 15 years since the LAS system’s development, and as we have moved towards a more precision medicine-based approach to care, we have learned that there is significant variance in disease burden and trajectory among patient subgroups. This is the gap we are working to fill.”
This new award will not only help support Drs. Dalton and Valapour in the model’s development, but also to evaluate its impact on both patient- and population-level outcomes, as well as how it measures in metrics of fairness and utility compares to other allocation strategies. Using electronic medical record data from 12-15 thousand transplant candidates on the waitlist between 2015 and 2019, the researchers will run simulations of their model compared to other allocation strategies.
“We are eager to see how our microsimulation model compares in terms of functional and survival outcomes,” Dr. Dalton said of the team, which spans Cleveland Clinic, Hennepin Healthcare and the University of Minnesota. Hennepin Healthcare serves as the data coordinating center for the U.S. Scientific Registry for Transplant Recipients (SRTR), an initiative of the Health Resources and Services Administration and source for patient data that will be used in the team’s simulations.
“Ultimately, our hope is that our research will inform meaningful change in lung transplant allocation policy,” said Dr. Valapour, who also serves as Senior Lung Transplant Investigator of SRTR and as scientific advisor to U.S. lung transplant policymakers. “When LAS was developed, it was a significant improvement over the preexisting system. Now that we know more and have improved methodologies available to us, it is time our policies reflect these.”
Featured Experts
Jarrod Dalton, PhD
Associate Staff
Director, Center for Populations Health Research
Lab Profile
