Waiting Time or Wasted Time? The Case for Using Time on Dialysis to Determine Waiting Time in the Allocation of Cadaveric Kidneys
The international transplant community confronts a formidable challenge as it struggles valiantly to distribute the scarce resource of cadaveric kidneys in a manner that is both biologically rational and societally equitable (1). This struggle, however, has only been partially successful; unanticipated and unintended inequities have been reported repeatedly for certain minority ethnic groups, for women, and for the financially or educationally disadvantaged (1–7). Barriers that impact access to transplantation for these groups of patients have been reported at various steps in the process of receiving a cadaveric transplant (8). In March of 2002, a consensus conference held in Philadelphia to analyze the waiting-list focused on these barriers to transplantation access in the USA. Various proposals have been made to address the problem; most of them emphasize patient and professional education (1,9). We contend that though well meaning, these proposals are unlikely to be effective, at least, in the short-run. We wish to propose a relatively simple expedient: changing the way waiting time is assessed in the organ-allocation algorithm – we believe this will go far to addressing the frustrating problem of equalizing access to cadaveric kidneys for all potential candidates.
The Cadaveric Transplant Process
In order to receive a cadaveric kidney transplant, a patient with end-stage kidney disease is first placed on the cadaveric kidney waiting-list. In the USA, this list is currently maintained by the United Network for Organ Sharing (UNOS). In the fullness of time, a kidney is allocated to the patient. The process of placement on the list typically has multiple stages (8). In brief, an interested patient first requests to be referred to a transplant program or is referred to a program by his or her nephrologist. This referral often needs to be approved by the patient's insurance carrier. Once an appointment is made, the patient is evaluated by the transplant program according to more or less standard guidelines and, if necessary, a work-up of varying complexity is recommended to ensure the patient is an acceptable candidate (10). Once the work-up is successfully completed and its results have been reported to the transplant center, the patient is ‘listed’ at UNOS and the waiting-time ‘clock’ starts. If, while the patient is on the active transplant list, an intercurrent illness occurs that contraindicates transplantation, transplant programs are exhorted to report this to UNOS and the patient is placed ‘on-hold’ (status 7). The waiting-time clock stops until the illness resolves; its resolution is then reported to UNOS, and the clock starts again.
In the USA, kidneys are allocated based on a standard UNOS algorithm. This algorithm currently uses two major components: human leukocyte antigen (HLA) matching and waiting time (with the exception of six-antigen matched or zero-mismatched kidneys that are allocated based on HLA only). In some regions of the country, an exception or ‘variance’ is in place whereby waiting time is the only factor used. In the policy approved by UNOS in 2001 to facilitate the rapid placement of marginal kidneys (now called ECD's- extended criteria donors), waiting time is the prime determinant of allocation (11). This policy is scheduled to go into effect in late 2002. Whatever algorithm system is in place, waiting time is always a critical determinant of organ allocation.
It is not difficult to conceive why such a process is prone to inequity. Patients living in rural areas or at a distance from the transplant program may find it more difficult to reach the program for logistic reasons. If these patients have complex medical problems, their transplant work-up may be prolonged or incomplete. Less well-educated patients, who are more likely to belong to minority ethnic groups or to be financially disadvantaged, may be less aware of their therapeutic options and be less aggressive in seeking out transplant centers. Women may have disproportionate child-care responsibilities. Perverse incentives may also be at work, as suggested by the evidence that patients treated at privately owned dialysis centers are less likely to be referred for transplantation than those treated at university-based centers (5). All of these factors, and no doubt others, may delay placement on the list. As long as waiting time is determined by the time the UNOS clock starts, any delay in placement on the list ‘wastes time’ and will delay access to kidneys. It is, therefore, not difficult to anticipate which groups of patients are likely to be disadvantaged.
Waiting Time as Dialysis Time
Starting the waiting-time clock at, or close to, the time a patient starts dialysis is a relatively simple expedient, with the potential to address many of the current inequities in access to cadaveric organs. It is a proposal that is intrinsically egalitarian in that it automatically negates the impact of many of the obstacles that disadvantaged groups currently suffer. It is anti-bureaucratic, in that it is not dependent on the process of referral, evaluation and work-up. It is intuitively fair and easy to understand. In the USA, dialysis time could be determined from the time that Medicare benefits are activated, and would be easy to verify and difficult to manipulate. We also propose that waiting time should not be ‘lost’ when a patient suffers a potentially reversible medical transplant contraindication. This change would encourage what we call ‘truth in listing’, since there would be no disincentive for the transplant program to accurately report the patient's condition to UNOS. Hence the national list would more accurately reflect the medical status of candidates, and the allocation of kidneys to those who are not ready to receive them would be avoided.
The proposal does have some potential drawbacks. Since waiting time is determined independently of the behavior of the patient, there may be a disincentive for some to initiate and complete the evaluation process expeditiously. We suspect that these patients are more likely to belong to one of the currently disadvantaged groups. It would need to made clear to all patients that they could not be allocated a kidney until their work-up had been completed and they were placed on the list, and that delays in listing might lead to them missing an opportunity to receive a six-antigen-matched kidney. Currently, in the USA, pre-dialysis patients may be placed on the list and accrue waiting time once their estimated GFR is less than 20 mL/min. If waiting time is assessed as dialysis time, this accrual would not be permitted. In practice, however, pre-dialysis patients are only likely to be offered highly matched kidneys since their accrued waiting time is limited. They would still be offered six-antigen matched or zero-mismatched kidneys under our proposed modification. Children, in a critical growth phase, could be potentially disadvantaged by not permitting pre-dialysis accrual of waiting time, and it may be wise to exclude them from the proposed modification.
Would patients currently on the list who had expeditiously completed their evaluation process or who had accrued a long waiting time feel it was unfair if their position on the list was lowered by the proposed change? Since the time spent on dialysis has been shown to be a factor affecting long-term patient and graft survival (12), is it possible that the proposed change could have a negative impact on transplant success rates? Experience from Eurotransplant, adduced below, suggests that the impact of the proposed change on both these concerns is likely to be small. It could be argued that, since time on dialysis is a negative factor in long-term success, waiting time should not be given such importance in organ distribution. Such a utilitarian argument may be a rational one, but it is not directly relevant to our proposal. It would also require a radical rethinking of allocation priorities that is beyond the purview of this discussion.
The Eurotransplant Experience
Eurotransplant (http://www.eurotransplant.org) is an organ-exchange organization of six countries, with a combined population of 119 million, approximately 12 000 patients wait-listed for kidney transplants and an annual kidney transplant rate of 3000. As of April 2000, for reasons similar to those we have discussed, Eurotransplant elected to define waiting time in its allocation algorithm as the time a patient had been dialyzed in an uninterrupted fashion. The 2 years that have elapsed since this change permit us to draw some preliminary conclusions, particularly regarding the introduction of the program. Following the redefinition of waiting time, 58% of the patients on the list were given extra waiting time, with a median gain of 6 weeks. Only 1% of the candidates, all of whom had been preemptively listed, suffered a median loss of waiting time of 12 weeks. In USA, this percentage is likely to be higher since preemptive listing is more common than in Europe. Under the new definition of waiting time, there was a 3% increase in the number of long-waiting (more than 5 years) patients transplanted. Though it is too early to reliably detect any change in overall graft survival since the changed definition of waiting time, no such trend has been detected. Eurotransplant intends to continue with the new definition.
The use of dialysis time to assess waiting time is certainly no panacea for the problems facing transplant candidates. It does not increase the availability of organs or reduce the interminable wait. It does, however, offer a practical and easily instituted mechanism whereby access to kidney transplants can be made more equitable. Such a change would represent a meaningful response to the challenge represented by repeated reports of inequity of access. We exhort the transplant community to give the proposal critical consideration.