Deceased organ donation by Ashlagi and Roth in the Annual Review of Economics (with open questions)

 Here's a review paper concentrating primarily on transplantation of  organs (primarily kidneys) from deceased donors. It closes with a set of open research questions.

Organ Allocation and Transplantation  by Itai Ashlagi and Alvin E. Roth,  Annual Review of Economics, Volume 18, 2026, Vol. 18:613-645, August 2026, https://doi.org/10.1146/annurev-economics-092425-123425     
 
ABSTRACT: There is a large shortage of solid organs for transplants. This survey reviews the allocation of organs (particularly kidneys), with an emphasis on how deceased donor organs are obtained and allocated in the United States but with pointers to related issues involving living donors and transplantation around the world. We review some of the key institutional details and theoretical and empirical studies and describe some open questions that we hope will continue to attract attention from researchers interested in the economic and operational aspects of organ allocation.

 
1. INTRODUCTION
 "Since at least the turn of the century, organ transplants have attracted considerable interest from economists. This is due both to their importance in modern medicine and to the fact that organ transplantation deals directly with some of the most basic questions of economics—namely, how to efficiently allocate scarce resources and how to make them less scarce. "

...

OPEN QUESTIONS AND RESEARCH DIRECTIONS
 
This section highlights research opportunities, some of which have been raised above.

1.  Can we design allocation rules free of justified envy that place hard-to-match organs quickly enough so that they remain transplantable? A large number of recovered organs that might be transplanted are discarded. How can we increase utilization of marginal-quality organs and get them accepted after fewer offers and less CIT? How can we do this in a transparent and equitable way? How can we use the information about organ quality that arrives during the offering process? 91 How can we use offer refusals in real time? How should we design allocation rules in the presence of social learning and potential information cascades (Zhang 2010, Doval et al. 2024)? How should the variability in centers’ willingness to accept marginal organs be accounted for?
2.  How to design priorities to balance efficiency and equity? How can priorities be designed to account for both longevity and urgency (Zhang et al. 2024, Lazenby et al. 2025)? How can priorities depend on patients’ treatments, taking into account that treatment decisions may in turn depend on priorities (Sweat 2023)? Countries use different priority rules, and there is a need to compare these to better understand trade-offs between these rules. The OPTN is seeking to implement continuous distribution allocation policies for hearts and kidneys, which will assign priority points smoothly across patient and match features. There are not yet studies that offer guidance to the resulting (in)efficiencies. 92
3.  How should organ quality be measured and reported? The KDPI is currently used for both acceptance decisions by surgeons and prioritization in the allocation system. As supply changes over time, percentiles correspond to different qualities. For example, a KDPI of 85% (which is the threshold for requiring patient consent to receive an offer) in 2018 corresponds to (i.e., has the same underlying nonnormalized KDRI risk measure as) a KDPI of 75% in 2024 due to the recovery of more organs and therefore the higher availability of organs of marginal quality. Would the use of the nonnormalized KDRI substantially change behavior and outcomes?
4.  How should surgeons’ and patients’ preferences be elicited? The OPTN allows transplant centers to express preferences via filters that screen organ offers. Despite this, many organs are still declined hundreds and even thousands of times (Agarwal et al. 2021, Guan et al. 2025). How can we better design these filters? Can we use AI to help recommend filters or to predict organ refusals in real time? More broadly, how should waitlist mechanisms be designed given incomplete patient and center preferences?
5.  How should transplant centers’ performance be measured? Should transplant centers be incentivized to accept marginal organs, for example, for patients facing many years of dialysis? How should we regulate transplant centers? Are new incentive schemes useful in increasing utilization? For example, how will the acceptance rate and pre-transplant mortality rate metrics impact centers’ behavior? The CMS has also launched the Increasing Organ Transplant Access model, 93 using a long-term field experiment that offers monetary incentives based on the number of transplants performed, graft survival, and percentage of accepted offers (Chan & Roth 2026.) How should we incorporate the huge role that transplant centers play in acceptance/rejection decisions when studying organ allocation mechanisms (Chan & Roth 2024, Ashlagi et al. 2025a)? There is a need for research on designing centers’ performance measures that include outcomes for patients on the waiting list.
6.  How should OPO performance be measured? OPOs vary in performance in terms of organ recovery and utilization, and incentive schemes remain controversial. How effective is the new tier system for monitoring OPOs based on donation and age-adjusted transplant rates? How should OPOs be incentivized to improve performance and transparency and to coordinate with transplant centers (see Chan & Roth 2024, Bae et al. 2026, Skowronski et al. 2026)?
7.  What should we incorporate in transplant centers’ preferences? Most theoretical models treat patients as the decision makers who accept or reject organ offers. In practice, however, transplant centers exercise substantial discretion over acceptance decisions, listing practices, and setting filters. How do equilibrium outcomes reflect centers’ preferences? There is a need for empirical and theoretical studies of how centers’ characteristics and incentives shape outcomes.
8.  Can we design simulation tools that account for equilibrium behavior? Simulations and historical data have been instrumental to the study and design of allocation policies. Existing simulations apply myopic acceptance models that are invariant to the allocation rules. There is a need for simple modern simulation tools that account for the incentives of patients and centers. 94 It would be interesting to create tools that generate not only better equilibrium predictions but also short-term equilibration effects. 95
9.  What kinds of experiments should the OPTN conduct? The OPTN conducts very few controlled experiments, despite facing ongoing allocation policy decisions. While simulation tools are useful to inform policy, field experiments offer another approach. Experiments can generate useful data to evaluate different approaches (e.g., for expediting placements). The size and heterogeneity of the United States present opportunities to test different approaches; but they also raise questions about whether different approaches would work differently in different regions, and about how to overcome the reluctance to conduct experiments.
10.  How can countries with low transplant rates be helped? Many countries have low living and/or deceased donation rates. Understanding the underlying barriers (organization structure, population preferences, etc.) would be fruitful in making progress toward increasing transplant rates (see, e.g., Okubo et al. 2026).
11.  What can be done to increase incentives for organ donation in ways that attract social support? What accounts for the widespread reluctance to increase organ donation by compensating donors? How can economists better explain the value of incentives in addressing scarcity and gather persuasive evidence to engage in a productive dialog with those who fear that this will increase rather than decrease black markets, or who feel there are fundamental religious or ethical objections to compensating organ donation? [For some thoughts on this, readers are referred to Roth (2026b)].