Around the world, organ transplant programs face the question of how to maximize the number of organs available for transplantation without compromising recipient safety. The fundamental problem is that not all donor organs are equal—some will function better for longer, others carry a higher risk of disease transmission. The challenge for allocation systems is to distribute organs to the waiting list so that all usable organs are ultimately transplanted into an appropriate recipient while maintaining ethical standards and preventing adverse events.
In this issue of Transplantation, several articles evaluate policies introduced by the United States Organ Procurement and Transplantation Network (OPTN) and the UK National Health Service Blood and Transplant (NHSBT) to address the problem above. Their findings illustrate how well-intentioned policies can have unintended consequences when failing to take into account human biases in decision-making.
In Kahneman and Tversky's1 seminal paper in the field of behavioral economics, the authors describe how, when making decisions between prospects and gambles, people tend to make choices that would not be considered rational by traditional economic theory. Instead, people overweight risks of low probability but large potential loss, and prefer a smaller but certain gain to a larger gain that lacks certainty. That is, when seeking gains, we are biased toward risk aversion. Further, perceptions of risk are acutely sensitive to how choices are described. In the context of transplantation, prospect theory predicts that the risk of an adverse outcome will weigh disproportionately heavily in organ acceptance decisions. Physicians and patients will be biased toward waiting longer for a “good” organ instead of accepting an organ described as “increased-risk.” Decisions are made in terms of near-term gains and losses, relative to a patient’s current state of health, rather than in terms of the aggregate lifetime benefit to be gained from a given organ.
The impact of labeling organs as “good” or “bad” has previously been observed with the introduction of the standard criteria/expanded criteria donor (SCD/ECD) classification for kidneys by OPTN in 2002. The ECD classification was introduced to expedite transplantation for patients unable to tolerate a lengthy wait by promoting the retrieval and utilization of “increased-risk” kidneys. However, although retrievals increased, the acceptance rate for kidneys meeting ECD criteria fell and rates of discard remained high.2 Given this result and the limitations of a binary classification system, the OPTN introduced the Kidney Donor Profile Index (KDPI) 2012, and in 2014 implemented a new kidney allocation scheme (KAS) under which kidney offers are made according to 4 tiers of KDPI. The main goals of the KAS were to improve the utility of the existing donor pool by matching KDPI with the estimated posttransplant survival of the recipient, and to expedite transplantation of “increased-risk” kidneys (now defined as KDPI >85%) to recipients who would benefit from them, thereby reducing discards and improving outcomes overall.
Contrary to intended outcomes, however, evaluation of the first 18 months of the KAS shows an increase in the discard rate from 18% to 20%. In the present issue, Heilman et al3 apply prospect theory to eloquently explain this observation. Not only has the introduction of the KDPI not corrected for the labeling effect of the SCD/ECD era (and possibly it has made it worse), the regulatory context in which transplantation programs operate nudges decision making toward loss aversion. Current allocation policies and regulations, they argue, bias physicians and patients away from accepting high KDPI kidneys and make organ discard more likely.
Also in this issue, articles by Volk et al, and Pruett et al evaluate the OPTN policy introduced in 2008 requiring transplant programs to obtain informed consent before accepting organs that meet the United States Public Health Service criteria for high risk of human immunodeficiency virus, hepatitis B, and hepatitis C.6 Volk and Pruett independently show that adjusted utilization rates for organs meeting these criteria (labeled PHS-IR) are significantly lower than for non–PHS-IR organs (with the exception of livers), despite the risk of disease transmission being extremely low and posttransplant survival being equivalent to non–PHS-IR organs.4,5 Both articles conclude that if the PHS-IR label did not exist, several hundred more transplants might be performed each year.
Volk et al4 propose patient and provider education as a solution to unnecessary discards of organs bearing a PHS-IR label. As the authors acknowledge, however, even if patients fully comprehended the low risk of disease transmission, they may still decline an organ labeled PHS-IR. Indeed prospect theory would predict exactly that outcome: people will still overweight low risks of serious adverse events and choose certainty over uncertainty, regardless of the precision with which they comprehend the risks involved. Even with education, a label of PHS-IR or KDPI greater than 85% will provoke stereotyping, overestimation of the differences between organs, and decisions based on heuristics rather than probable utility.
As rational and utilitarian as the intended outcomes of transplantation policies might be, the ultimate decision to accept an organ is made by a human who is subject to all of the standard biases predicted by behavioral economics. As Heilman et al3 argue, if transplantation systems are to achieve utilitarian goals then it is necessary to avoid setting policies that are defined in terms of risks and losses, and to instead frame utilization in terms of opportunities.
The UK's strategy for maximizing kidney utilization provides one example of how this might be done. In 2012, the NHSBT introduced the UK Kidney Fast-Track Scheme (KFTS) whereby kidneys that meet prespecified criteria identifying them as being at risk of discard are simultaneously offered to centers willing to transplant them. Importantly, no explicit “increased-risk” label is applied to kidneys entering the KFTS. In this issue, Mittal et al7 show that—although discard of useable organs still occurs and there is room for improvement—the KFTS has been successful in identifying kidneys at risk of discard and offering them appropriately. The discard rate in the UK in 2015 was 11%.8
Transplantation regulatory authorities can learn from these experiences to design more effective strategies for maximizing organ utilization and minimizing discards. Adapting knowledge from the field of behavioral economics has the potential to bring systems closer to the overall goal of optimal outcomes for the largest number of patients.
1. Kahneman D, Tversky A. Prospect theory: an analysis of decision under risk. Econometrica
2. Hirth RA, Pan Q, Schaubel DE, et al. Efficient utilization of the expanded criteria donor (ECD) deceased donor kidney pool: an analysis of the effect of labeling. Am J Transplant
3. Heilman R, Kaplan B, Green E, et al. Potential impact of risk and loss aversion on the process of accepting kidneys for transplantation. Transplantation
4. Volk ML, Wilk AR, Wolfe C, et al. The "PHS Increased Risk" label is associated with nonutilization of hundreds of organs per year. Transplantation
5. Pruett T, Clark M, Taranto S. Deceased organ donors and PHS risk identification: impact upon organ usage and outcomes. Transplantation
6. Seem DL, Lee I, Umscheid CA, et al. PHS guideline for reducing human immunodeficiency virus, hepatitis B virus, and hepatitis C virus transmission through organ transplantation. Public Health Rep
7. Mittal S, Adamusiak A, Horsfield C, et al. A re-evaluation of discarded deceased donor kidneys in the UK: are usable organs still being discarded? Transplantation