A team of researchers from the University of Chicago Medicine has developed a new risk score that aims to improve the allocation of donor hearts to the patients who need them most. The current system for prioritizing heart transplant candidates is based on an algorithm that takes into account medical urgency, geography, and pediatric status. However, with deceased donor organs being scarce in the United States, many patients are not even placed on waitlists because the chances of receiving a heart transplant are too low.
The research team, led by experts at the University of Chicago Medicine, has created the U.S. Candidate Risk Score (US-CRS), which is designed to predict the likelihood of a patient dying without a heart transplant. This innovative risk score aims to address the limitations of the current therapy-based system and provide a more precise and fair approach to prioritizing candidates based on medical urgency. The results of the study, detailing the development and initial validation of the US-CRS, have been published in JAMA.
According to Dr. William F. Parker, Assistant Professor of Medicine and Public Health at UChicago Medicine and senior author of the paper, the goal is to identify the sickest patients. While all candidates on the waitlist require a heart transplant, some can wait marginally longer than others. The US-CRS improves upon the current system by not solely relying on the treatment decisions made by individual physicians.
The research team took into account various clinical and laboratory measurements associated with end-stage heart failure, such as levels of molecules in the blood linked to liver and kidney failure. They incorporated these variables into the US-CRS, which was then analyzed using data from over 16,900 adult heart transplant candidates in the U.S. allocation system. The researchers found that the US-CRS was highly accurate in predicting mortality within 6 weeks of being placed on the transplant waitlist.
Surprisingly, the standard regression model used in the study was found to be more accurate than advanced deep learning and machine learning models. This suggests that the variables chosen for the US-CRS were well-suited for predicting mortality. The resulting model also has the advantage of being easy to understand.
Compared to the current U.S. heart allocation system, the US-CRS demonstrated a much higher accuracy in predicting mortality in patients who did not receive a heart transplant within 6 weeks.
While the US-CRS will undergo further validation and review processes, it represents an important step towards improving heart transplantation allocation. The researchers are already working on follow-up research projects aimed at refining the heart allocation system as a whole and addressing health inequities.
In the long term, the team believes that their work may serve as a foundation for using algorithms to distribute scarce healthcare resources more fairly and effectively. By tackling the issue of organ transplantation, they hope to uncover lessons that can be applied to broader healthcare resource allocation challenges.
*Note:
1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it
