The public discourse surrounding healthcare reform is replete with references to “accountable care” and “population health” . It is widely asserted that shifting the focus of the US healthcare delivery system from services rendered to health outcomes achieved will result in reduced spending and improved patient outcomes. At the core of this argument is the belief that better public reporting of the health status of populations cared for by specific providers will lead to better decision making at the bedside. Unfortunately, the act of measurement itself is not neutral and systems of care will evolve to maximize the prioritized outcomes, potentially at the expense other considerations.
The enhanced regulatory attention on posttransplant patient and allograft survival provides an important case study of the potential pitfalls of the “accountable healthcare” paradigm. The decision to list a patient for transplantation and subsequently accept a deceased donor organ rests at the intersection of many competing priorities. As elegantly described by Schnier et al. in this issue of AJT, the transplant physician must weigh patient demographics, donor quality, financial realities and regulatory consequences when offered a potential donor organ . Ideally, the decision to accept the organ would focus on the straightforward question, “will this patient benefit from this organ”? However, as described by Schnier, medical professionals are, by nature, risk adverse despite our patient-centered focus. Consequently, the physician is more likely considering whether “the likelihood of success of this transplant is sufficiently high to risk the program's outcome results if we take this organ”? The difficulty of this perspective is compounded by our limited ability to accurately assess risk either clinically or through the statistical models. Consequently, it is likely that this shift in perspective will limit access to transplantation, particularly for patients with risk factors that are poorly captured in the current risk adjustment methodology. This has been demonstrated by an overall reduction in the proportion of high cardiac risk patients reaching kidney and liver transplant  and marked disparities in the use of high donor risk organs among centers .
The adoption of a risk avoidance strategy is predictable given the significant impact on a transplant center's viability if it is identified as having poor posttransplant patient or graft survival. In their report in this issue of AJT, Schold et al. consider the impact of public citations for poor outcomes on kidney transplant center volume . Among the 23% of centers identified as low-performing centers, the average volume of transplants performed declined by 22 cases, whereas programs with better statistics saw average volumes increase by 7.8 cases. Furthermore, centers with low-performance signals experienced a decline in the proportion of patients with private health insurance undermining the financial underpinnings of these programs, likely as a result of loss of contracts through center of excellent networks. Clinically, low-performing centers reduced utilization of marginal donor kidneys that clearly benefit appropriate patients but are perceived to increase the risk of regulatory scrutiny.
While Schold's observations are not conclusive evidence that enhanced regulatory review led to these changes, his conclusions are consistent with concerns expressed by other transplant professionals . In assessing the impact of regulatory changes on practice, three important factors must be considered: inadequate risk adjustment due to uncaptured comorbidity in our current evaluation system, true failures of center performance and processes and, most importantly, the impact of the choice of outcome metric. There have been extensive and significant discussions about the limitations inherent in the current risk adjustment models. Socioeconomic status, cardiovascular disease burden and donation service area characteristics are all missing from the current risk models, despite demonstrated impact on patient and graft survival [7, 8]. Absent changes in the models and given natural risk aversion, patients with adverse characteristics will be less likely to be listed and transplanted.
It is also clear that among poor performing centers there are significant deficiencies in care processes that contribute to clinically important differences in outcome. As reported by Tom Hamilton, Director of CMS Survey and Certification Group, centers receiving outcomes citations were also more likely to be cited for poor patient care practices (45% vs. 24%), lack of multidisciplinary planning (49% vs. 21%) and inadequate quality assurance/process improvement programs (27% vs. 7%) . In reality, only a minority of programs have been cited and many centers transplant complex patients successfully with excellent outcomes. Clearly, the transplant community needs to adopt and maintain consistently excellent standards for care delivery to meet our core responsibilities to our patients.
Finally, the transplant community needs to embrace the inevitable shift to an “accountable care”-based healthcare delivery system. However, it is crucial that we accurately define both the population we are accountable for and the metrics to which transplant centers should be held. Transplant should be, and largely is, the best model for accountable care organizations as we already struggle to attain the maximum benefit possible within a resource-constrained environment. Currently, we are judged by the outcome of those patients who actually receive a transplant independent of how we care for listed patients with end organ failure. This focus leads to risk aversion and limitation of access. If we shift accountability to the population of patients with end-stage organ failure who we agree to list for transplant, the incentives will change. Centers will then be recognized for their efforts to maximize access to transplant, utilize marginal organs and encourage innovation even at the expense of a limited reduction in posttransplant outcome. This change should help to align the incentives of the transplant physician with the needs of the potential recipients. While there will still be limitations of risk adjustment and the financial jeopardy inherent in caring for complex patients, by choosing the appropriate metric to judge success, we can put the patient first and transplant them with less conflict with our natural desire to avoid risk.