New Computer Studies Identify Transplant Allocation Policies That Save More Lives than Current System
PITTSBURGH, October 7, 1998 — As the transplant community debates the possible consequences of a federal regulation calling for new organ allocation policies that minimize the geographic differences in patients’ waiting times and give increased priority to the most medically urgent patients awaiting transplantation, preliminary results of new computer simulation projections released today demonstrate that more lives can be saved with liver allocation policies with broader geographic sharing of donated organs.
The current system of liver allocation is based primarily on geography, whereby a donated liver is placed locally to candidates in decreasing order of medical urgency before being offered to patients regionally or on a national basis. The current system is plagued by large geographic disparities in waiting times and patient deaths on the waiting list.
"Several alternative liver allocation policies were determined to meet the performance standards of the U.S. Department of Health and Human Services (HHS) regulation and result in a significant increase in patient lives saved," said Mark A. Joensen, Ph.D., a health policy analyst and vice president of CONSAD Research Corporation, who consults for the University of Pittsburgh Medical Center.
CONSAD’s just-released preliminary results of a new computer modeling analysis of alternative national liver allocation policies are based on results of modeling that incorporate newly acquired UNOS data describing liver transplant patients during 1996 and 1997.
Dr. Joensen said that a system that allocates donated livers through a single national waiting list will save patient lives and equalize waiting times. The CONSAD modeling indicates that during the first years after implementation of a single national list, approximately 275 patient lives would be saved per year compared to the current system. Over the six-year simulation period of the computer analysis, a total of 1,000 patient lives would be saved by the single national waiting list policy. In addition, the computer model results demonstrate that patient life-years would be increased, disparities in patient waiting times at different centers would be diminished, and disparities among centers in the percentage of patients who die pre-transplantation would be reduced under a single national list.
In addition to analyzing a single national waiting list, CONSAD also projected the impacts of other alternative allocation policies that would result in broader geographic sharing of donated livers. Although these other policies would not achieve the same magnitude of patient lives saved and waiting time equity as the single national list, they would represent great improvements over the current system.
"To achieve equity for patients and comply with the HHS rule, a new liver allocation policy must assure that the movement of donated organs is not limited to restrictive and unequal geographic boundaries. These modeling results show that a number of alternative liver allocation policies exist with broad geographic sharing of donated organs which will save patient lives and have more equitable waiting times. The transplant community can consider a number of policies that meet the performance guidelines of the HHS rule," said Dr. Joensen.
The HHS regulation does not specify any one allocation policy for donated livers, but specifies performance goals that allocation policies must meet, including minimizing waiting time differences among different transplant programs. The rule calls for the private contractor that operates the national Organ Procurement and Transplantation Network (OPTN), the United Network for Organ Sharing (UNOS), to propose new organ allocation policies. The OPTN has 60 days from the effective date of the HHS rule to identify a new liver allocation policy and one year to propose new policies for all other organs.
But UNOS has vehemently opposed the rule based on its interpretations that it calls for a "single national list, sickest patient first" allocation system. Based on computer projections performed by the Pritsker Corporation, UNOS has claimed that such a policy would have a number of harsh consequences, including an increased number of patient deaths.
CONSAD plans to make the results of its recent computer models available to members of UNOS’s Liver and Intestine Transplantation Committee, which is responsible for making recommendations about liver allocation policy to the UNOS board.
Key differences between the CONSAD and Pritsker modeling of future impacts of alternative allocation policies are the assumptions about future trends in liver transplantation.
"Clearly, any health policy analysis that makes projections into the future must correctly account for changes in transplantation that are expected to occur in the future. This is where the two models are different. As an example, the Pritsker model assumes that there will be only 4,000 liver transplants annually between now and the year 2003. UNOS data shows that there were 4,065 liver transplants in 1996, 4,167 in 1997, and so far this year we have seen a 5 percent increase over last year. The CONSAD model accounts for the likely increase in the annual number of liver transplants based on historical trends, whereas the Pritsker model does not," explained Dr. Joensen.
"More importantly, the Pritsker model assumes that transplant graft survival rates will remain constant into the future at the same rate experienced by patients entering in 1991 and 1992. Separate UNOS studies from 1994 and 1998 indicate that risk-adjusted graft survival rates have increased each year since 1989. The CONSAD model uses the statistical results of UNOS’s graft survival studies to predict likely changes in survival rates between now and the year 2003. It is differences like these in how the two models predict changes in transplantation in the future that account for the differences in the two models’ results. I believe we are doing a much better job of representing in our computer model the changes in transplantation that will occur between now and the year 2003," Dr. Joensen added.
Since 1993, Pittsburgh-based CONSAD has studied the impacts of alternative national systems for liver allocation on both the medical outcomes of patients and patient equity. Working as a consultant to the University of Pittsburgh Medical Center, CONSAD has constructed computer models based on UNOS data, providing its results to UNOS, HHS and members of the general public.