An Eye to Quality

Abstract

Over the last 20 years, there has been remarkable progress in patient and graft survival following transplantation. Results now achieved after solid organ transplantation would have seemed Olympian by the pioneers of transplantation. With this success, transplantation has been gradually transformed from an experimental therapy to the routine. This success has also brought raised expectations. Transplantation, while still challenging for the physician and patient, is now viewed by the same critical eye that is examining outcomes in all of medicine. Unfortunately, this critical eye has poor vision in many fields of medicine other than transplantation. There have been few databases outside of transplantation that allow for identification of specific center outcomes and even fewer that allow for risk adjustment. Risk adjustment, based upon factors that are accepted confounders of outcome such as patient age, or comorbid conditions, has been identified as being crucial to allow a fair comparison between outcomes. Many of the studies examining outcomes have been carried out on ‘administrative’ databases, such as Medicare financial databases, where the clinical information allowing risk adjustment is lacking. Because of this lack of clinical source data, the scrutiny is frequently based on surrogate measures of quality. Center volume has been shown to be related to outcome in many different, primarily procedural based, areas (1). With the identification of this surrogate of quality, some have advocated a move to monitor center volume and to insist that centers meet certain volume criteria in order to be considered ‘centers of excellence’. It now appears that center volume may actually be a relatively poor surrogate of center outcome. A recent publication pointed out that surgeon rather than center volume is an important contributor to patient outcome (2). This paper demonstrates that high-volume surgeons within low-volume centers can have outcomes that may exceed those of a low-volume surgeon in a high-volume center. This recent information helps to explain what has been relatively obvious; that is, there are low-volume centers that have very good outcomes, possibly because of the work of a high-volume surgeon. It might be expected that there maybe other confounders for the use of volume as a surrogate of quality and over time these will become clear (3). In any case, it appears that there is a substantial overlap of center results using volume as a surrogate. The paper by Axelrod in this edition of the American Journal of Transplantation demonstrates the effect of center volume on the outcomes following transplantation (4). It extends previous work on this topic in the field, and demonstrates the center–volume effect on graft outcomes in kidney transplantation and patient outcomes in liver transplantation. The authors demonstrate significant, albeit relatively small, differences in outcomes following transplantation. The authors remind us that there is substantial overlap in the results among the different volume classes. The major question is where do we go from here? The direction that we should not go is to use volume for a surrogate for transplant center quality. The primary reason for this is that transplantation is far better prepared than almost any other field of medicine to look directly at quality in outcomes. The clarity of our vision can be surmised from Axelrod's paper. The authors examine tens of thousands of transplants by using data reported primarily from the transplant centers themselves. This could not be carried out without the presence of outstanding center reporting and a fantastic database. This type of data allows the United States to demonstrate transplant center quality directly, with appropriate risk stratification, and not to surmise quality using volume as a surrogate. The information on center outcomes is currently being reported by the publication of center-specific outcome data available to the public (5). Transplant centers identified as being below standard have efforts at remediation effected by the OPTN. While there is an open question about the best way from a statistical basis to identify centers that are failing to meet the standard, the overall picture is clear without using volume as a surrogate of quality.