Posted on January 30, 2020 at 1:54 PM
by Hannah Giunta D.O., Ph.D. & Richard Sharp, Ph.D.
In their article, MacKay and Saylor analyze the issue of fair subject selection in clinical research and suggest that this overarching principle is best understood as a collection of four sub-principles, namely fair inclusion, fair opportunity, fair burden sharing, and fair distribution of third-party risks. After describing these principles, the authors suggest several strategies for managing potential tensions between these four sub-imperatives, as conflicts are likely in practice. The authors’ strategies for negotiating the multiple ethical perspectives highlighted by these sub-imperatives are crucial because they support a robust concept of fairness that acknowledges the need to protect research volunteers without denying them access to potentially beneficial research. In this regard, the analysis offered by MacKay and Saylor provides a useful way of revealing several salient ethical considerations that are relevant to a claim to scarce healthcare resources.
It is our belief that the authors’ insights regarding principles of fair subject selection in clinical research may be useful in other settings, particularly in clinical settings in which there is high demand for resource-intensive therapies. In such settings, decisions to offer therapy to one patient often result in other patients no longer having that option, thereby introducing ethical questions about the fairness of the decision-making process or its outcome. In what follows, we use the framework developed by MacKay and Saylor to illustrate the potential value of more explicitly delineating relevant sub-principles of fair allocation of resource-intensive therapies.
Extracorporeal membrane oxygenation (ECMO) is an example of a resource-intensive therapy that raises multiple questions about its availability to patients in need. ECMO is an advanced, highly expensive medical technology that requires significant levels of coordination across healthcare specialists. Although ECMO is available at many tertiary care hospitals, demand greatly exceeds capacity across the U.S. As a result, decisions about who should be offered this technology have generated much controversy, largely centered on how to distribute this life-preserving technology fairly.
A widely accepted paradigm for distributing ECMO therapy is based on the likelihood of eventual recovery following definitive therapy. Under this approach, ECMO is typically not offered if a patient’s prognosis is poor or if the patient is not a candidate for future therapies such as transplantation. For instance, ECMO is often used as a bridge to lung transplantation. ECMO may be withdrawn, however, if a patient becomes ineligible for transplant. Given the limited availability of ECMO therapy, it is often considered a temporary treatment strategy for patients who are likely to recover or as a time-limited trial while options for future intervention are explored.
These considerations highlight how decisions about offering ECMO therapy are similar to many types of clinical research, particularly studies in which a promising new therapy may only be available through participation in a clinical trial. In both settings, there is no guarantee of benefit and an overarching presumption that each new case will be instructive in guiding future clinical applications. In addition, in both settings, there are significant practical limits on the number of individuals who are able to receive therapy, which introduces a number of ethical considerations related to determining eligibility for these therapies. In the remainder of this article, we lay out how each of the four sub-principles introduced by MacKay and Saylor might be extended to inform decisions about the provision of ECMO therapy.
The first sub-imperative introduced by MacKay and Saylor is fair inclusion. It is essential that a wide variety of participants with conditions relevant to a research question be included in a study if it is to yield generalizable knowledge. Similarly, the need for fair inclusion is apparent when distributing ECMO resources since it would be wrong to restrict ECMO therapy to a small number of individuals (e.g. to patients with congenital heart disease but not septic shock) given its life-preserving potential. In addition, the resource-intensive nature of ECMO may suggest that health professionals prioritize those patients who are most likely to receive long-term benefits from the technology. To balance these competing priorities, physicians and other health professionals may begin a patient on ECMO therapy but elect to reevaluate its continuation as the patient’s prognosis evolves. Unfortunately, health care providers may not agree on which circumstances warrant offering ECMO therapy. In a recent survey, for example, physicians held divergent views on which groups of children should be offered ECMO therapy, often citing the same supporting reasons (i.e. quality of life, benefit vs. burden, etc.) but reaching different prioritization schemes. This scenario is not dissimilar to disagreements that may occur among members of institutional review boards (IRBs) reviewing eligibility criteria for clinical trials.
MacKay and Saylor also point out that investigators may have a duty to carry out research that fairly benefits people generally via health priority setting. Physicians likewise have a duty not only to treat individual patients fairly but also to promote the health of the public at large. Here, ECMO therapy is less similar to other clinical interventions, since most people will never need to be placed on ECMO—while the few individuals who require this therapy will also need a much larger than typical proportion of intensive care resources. In this way, ECMO therapy is not dissimilar to clinical research on extremely rare or recalcitrant diseases.
The second sub-imperative described by MacKay and Saylor is fair opportunity. It makes sense that patients who might receive clinical benefit from a therapeutic option offered through a research trial should have a fair opportunity to enroll in that trial. This raises interesting challenges for both clinical trials and for allocating extraordinary resources like ECMO. When it comes to allocation, it is very difficult to make conditions objectively fair. MacKay and Saylor highlight the fact that there is a tension between including under-represented groups in research and ensuring, despite relative disadvantages, that they do not bear a disproportionate amount of the burdens of research participation. While it is patently unfair to ask a subgroup who is already disadvantaged to accept excessive burdens as a result of participation in clinical research, it would be just as unfair to fail to include those individuals in potentially beneficial research projects. MacKay and Saylor acknowledge but fail to explore the ethical salience of the fact that each research participant comes with a rich personal history that shapes their experience of research, including their assessment of the fairness of study eligibility. If we recognize this diversity of personal experiences, then factors unique to each individual patient who may benefit from a clinical trial matter. Personal stories not only touch the hearts of clinicians, investigators, and ethicists but are highly relevant to decisions about fair opportunities to receive advanced therapies or participate in clinical research.
In ECMO allocation, a similar issue arises when considering the right to try an innovative technology. ECMO therapy is only available at high volume, well-resourced medical centers. The need for ECMO can present itself in a variety of settings. For instance, a person who drowns in very cold water may have a good chance for meaningful recovery if ECMO cannulation occurs promptly. However, duration of CPR while traveling to the nearest ECMO center impacts the chance for neurologically intact survival. A person who is closer to an ECMO center not only has a better chance of receiving ECMO but has a better chance that such treatment will be successful. There are no doubt a number of these sorts of factors that impact if a person is offered ECMO therapy, including proximity to medical attention, local availability of a well functioning EMS system, prompt recognition of ECMO candidacy, and provider familiarity will regional ECMO facilities. Of course, some factors impact a patient’s clinical condition, and individual providers may have to turn away a patient who is unlikely to recover even with ECMO therapy. Nonetheless, there does seem some obligation to mitigate system-level factors that impact ECMO eligibility. Given the longitudinal impact of the intervention itself, there is also an obligation to develop transparent criteria for ECMO cannulation rather than deferring entirely to subjective assessments by clinical staff.
The third sub-imperative presented by MacKay and Saylor is fair burden sharing. Research participants ought not to be asked to bear excessive burdens in the name of research that may benefit future patients. Although highly relevant to decisions about research eligibility, bedside assessments of benefit and burden are somewhat different. The magnitude and likelihood of direct benefit is typically greater, and the assessment of potential burden is often more precise. It is easier to assess potential burdens at the bedside because providers have an ongoing relationship with the patient and can monitor burdens that may be associated with ongoing care. By contrast, reviewing a research protocol requires an a priori assessment of its potential burdens, with limited opportunities to assess the lived experiences of research volunteers. However, the personal relationship that makes it easier to assess actual benefits and burdens in clinical settings can lead providers to dismiss the therapeutic burdens of invasive technologies, including ECMO, in hopes for a good outcome.
The final sub-imperative offered by MacKay and Saylor is fairness in the distribution of third-party risks. In research, third-party risks are downstream impacts of a research project (e.g., potential infection with a study agent, etc.). Because research may present risks to unsuspecting community members, there must be a careful analysis of any risks and adequate mitigation. By the very fact that people requiring ECMO therapy are seriously ill, and their loved ones often desperate for a cure, family members and friends of ECMO patients may be least capable of bearing the substantial emotional and psychological burdens associated with ECMO therapy. Undoubtedly, healthcare providers do their best to provide support for patients and their families but there may be an additional duty to insure third parties are not overly burdened when deciding whether to pursue or continue ECMO therapy. The less likely ECMO therapy is to provide a path toward either curative or palliative good, the more burdens to third parties should be considered.
In their final analysis, MacKay and Saylor argue for the primacy of inclusion, so long as prospective participants do not face unacceptably high burdens and there is not a patently unfair distribution of third-party risks. Researchers have a responsibility to generate new knowledge, and overly restrictive inclusion criteria may jeopardize this objective. In a similar way, decisions to offer ECMO therapy should promote inclusion, so long as the anticipated burdens to the patient and to others are not unacceptably high. Those individuals and programs that offer ECMO therapy should analyze the potential burdens and establish objective exclusion criteria, both within and across institutions. At the same time, clinical settings allow for more personalized assessment of individual benefits and burdens as a patient’s ECMO course continues and new information becomes available. As the potential long-term benefit from ECMO diminishes, mitigating risk to the patient and to third-parties – including other patients who might benefit as a result of receiving ECMO therapy – become higher priorities.
Our goal was to consider how the sub-principles offered by MacKay and Saylor might be applied in a clinical setting; however, we recognize that clinical settings and local contexts vary considerably. Additionally, although we believe the framework introduced by MacKay and Saylor could be usefully adapted to serve as a guide in determining patient eligibility for resource-intensive therapies such as ECMO, it is important to acknowledge that clinical settings and research settings differ in significant ways. While clinical care is designed to benefit individual patients, the primary objective of research is the discovery of new knowledge, though incidental clinical benefits may accrue to research participants. Research projects require approval by a local institutional review board (IRB) and compliance with additional federal regulations. Since IRBs are local institutions, they often vary in their interpretation of those regulations; however, there are shared principles and non-negotiable elements recognized by all IRBs. By contrast, the overarching rules that govern the delivery of innovative forms of healthcare such as ECMO therapy are rarely as well defined or agreed upon across institutions.
Despite these caveats, the value of a more systematic delineation of principles of fair distribution of scarce healthcare resources in either research or patient-care settings is critical. Individual physicians and institutions attempt to do the best they can for patients, but in the absence of a clear professional consensus or practice guidelines for allocating healthcare resources, these individuals are left to make these decisions on their own. Clinicians may decide to use new technologies for their patients, with incomplete knowledge of their potential benefits and harms, amid a backdrop of ethical controversy about whether the technology ought to be available at all. Moreover, clinical innovation is a non-linear process, with many unknowns even after widespread adoption. A more nuanced analysis of the principles of fair patient eligibility for resource-intensive therapies can help by filling a critical void along the path toward such professional consensus, giving structure to bedside discussions of patient eligibility and revealing latent concerns about fairness that might not otherwise be considered.