To the Editor:
The impetus to design a new kidney allocation system in the United States stemmed from the inefficiencies of the current system. Even though the basic principle underlying the current allocation system, to give all candidates equal access to the kidney pool and prioritize by time on the waiting list is intuitively the principle to follow, the resultant possibility of allocation of 20-year-old organs to 70-year-old recipients and vice versa has raised concerns. In fact, it makes sense to have an allocation system that aims to match the expected life of the transplanted organ with the expected life of the recipient, thereby minimizing loss of viable organs through patient death.
However, predicting both donor and recipient life expectancy accurately is problematic, particularly when based on limited data. Based on the data available in large databases, the age of both the donor and the recipient is the strongest predictor of survival for the general population, but predictability at the patient level still remains poor. The addition of other readily available variables only marginally improves the predictability of the models. This is in part due to the fact that the detailed medical information that might allow for better modeling of predicted survival is not currently recorded in databases. As a result, even though it might be possible to improve the predictably of the models by including detailed medical information, the generation of such models is not currently feasible. It is difficult to justify replacing the detailed medical history used by physicians to assess patients with a few variables that happen to be recorded in databases. It is therefore quite clear that beyond the implications of the age of the recipient and the donor, it is at present difficult to model reliable and justifiable indicators of longevity based on existing data.
The UNOS KARS committee has done a tremendous amount of work to generate a workable compromise for a new kidney allocation system.
Unfortunately, the proposed new system appears to fall short of the intent to produce an accurate and transparent system that is based on the right balance between efficiency, justice and equity.
There are fundamental principles that most involved in this discussion probably can agree upon. Waiting time on dialysis is probably the strongest modifiable risk factor for death in our patients, and based on the justice principle should probably be treated the same for all patients. On the other hand, there is some consensus that some matching of donor and recipient age could be reasonable. Basic principles like these should be agreed upon in the transplant community and, after consensus is attained, modeling should start to achieve the predetermined goals. A new system should be rational, transparent and easy to explain to our patient. Transparency in the development process is essential for generating a widely accepted revision of the current system.
While it is clear that a revision of the current allocation system is required to improve the efficiency of the organ allocation process, the new allocation system proposed by UNOS will ultimately have a more profound effect on who gets transplanted rather than the type of organ they receive, a shift the transplant community may not be ready to accept. In the present format, the proposed new allocation scheme does not achieve the acceptable justice and equity standards necessary for this proposed policy change to be embraced by the transplant community.