Managing Expectations: Delivering the Worst News in the Best Way?

This post also appears in the January 2018 issue of the American Journal of Bioethics

by Alyssa M. Burgart & David Magnus

In this issue, Weiss and Fiester’s (2018) “From ‘Longshot’ to ‘Fantasy’: Obligations to Patients and Families When Last-Ditch Medical Efforts Fail” calls attention to the weight of clinician word choice when discussing interventions in the pediatric population. Their work focuses on communication in a highly narrow slice of intervention options, from unlikely to work therapies to impossible ones. Regardless of a therapy’s low probability of success, physicians and parents suffer from forms of misconception: physicians tend to be overly optimistic in both their prognostic estimates and in their disclosure of illness severity, and parents tend to be highly likely to believe that their child is the one of many who will benefit from therapy. For Weiss and Fiester’s proposal to work, clinicians must realize and acknowledge early on that a “longshot” therapy is unlikely to work. This insight allows the team to share this understanding with the family and for the family to then, the authors assume, process the information. Adequate communication is a core clinical skill, yet not one innate to the best clinicians and few evidence based resources exist to support clinicians in achieving mastery in this arena. (One notable exception is Vital Talk [www.vitaltalk.org], however, the program is not specifically geared towards pediatric providers.)

The authors advocate for expectation management at its finest: clear, highly specific communication that empowers parents to understand their child’s illness better and to come to terms with narrowing treatment options. They propose a four-step process in which clinicians may build a series of signposts for parents, connecting an already unlikely cure to an unrealistic one. As the clinical picture evolves, parents are aware that forks in the road lie ahead. Presumably, this helps parents and clinicians build a stronger relationship while working in tandem to manage expectations of success and failure.

We have some concerns about whether this proposal would be workable or even desirable in practice. There are several ambiguities and assumptions that need to be explored. First, Weiss and Fiester make no distinction between highly innovative “longshot” therapies and routine “longshot” intensive care. Commonly, tertiary care children’s hospitals accept referrals of patients previously declined by other centers. Often times, there may only be one center that offers a specific therapy, and such centers are willing to accept a high failure risk for a slim possibility of reward for patients and families. For example, there may be few surgeons who offer an innovative procedure and one that has evidence that her outcomes are better than other surgeons in the field. This may present a child’s best and only hope for disease modifying treatment, even if the chances of cure are low.

This is significantly different from a patient who has languished in the intensive care unit, failing to improve despite high quality standard therapy (ventilator, antibiotics, pressors, etc.), who slips from care that’s likely to work (a point upstream from the 4 stages presented) to care highly unlikely to work. In the case of a highly innovative therapy, the lines may be easier to draw, when families have been specifically referred to a tertiary care pediatric hospital for a “longshot” therapy, providing a starting point to clearly lay out the path(s) that may lie ahead. It is much trickier to do so for patients who have crept into the realm of “fantasy” care during their hospitalization. As a result, we are skeptical that the approach will work in these types of cases.

Weiss and Fiester’s plan hinges on eventual “universal clinical agreement” and a zero probability of success. Critical care occurs along a continuum, and is not easily dissected into artificially and nebulously defined stages. Two clinicians may believe they are in different stages at the same time, making it impossible to achieve the consensus the authors expect. Therefore, clinicians never make it to step four, no matter how poor the outcome. At any point along this continuum, clinicians might advise parents against a longshot therapy, or alternatively, parents might decide to withdraw treatment because the risks of therapy do not outweigh the suffering caused. The idea that zero probability of success is the bright line at which clinicians should argue for redirection of treatment is deeply troubling. It is not clear that a true zero probability is ever reached. Almost every pediatrician has at least one story of a patient for whom they felt there was no hope for survival, who persisted and survived none the less. With a bar set so high, most critically ill children will experience unnecessary suffering for no benefit. In fact, by the time clinicians find themselves in widespread agreement about clinical “fantasy,” innumerable communication opportunities that focus on whether continued critical care interventions make sense will have been missed.

Finally, based on our extensive experience talking to families dealing with serious illness, we are skeptical that the words “longshot” and “fantasy” are uniquely suited to convey precise meaning, and are worried families may find the term “fantasy” offensive and confusing. Certainly, anyone who has been involved in clinical ethics has had families hope for or even demand miracles. As many of the OPC’s point out, the word “fantasy” will likely not save us from the pitfalls of “futility,” and any individual word risks suffering the same lack of uniform interpretation. We should ask the empirical question: what language works with what families? Communication is complicated, even between two people who are already on the same page.

Weiss and Fiester raise questions vital to quality pediatric care, as clinicians struggle to determine how to communicate complex information to parents in a palatable manner. James Tulsky and colleagues (2017) have called for research into communication practices and proposed an agenda to include formal evaluation of both verbal and non-verbal communication. Studies in adults are helpful, but we need to know more about parents and children. What is the best way to deliver the worst news to parents and children? Are specific words particularly helpful when explaining small to miniscule probabilities of success? Are words or phrases equally effective for communication between each clinician, child, and parent? How can clinicians move a family towards understanding a poor prognosis in a compassionate and honest way? Unfortunately, neither the authors’ framework or the choice of language is based in empirical evidence. The development of a four-stage approach to communication is intriguing, although untested and atheoretical. Such a proposal may provide clinicians with an organizational framework for their thoughts on a case, but without methodical exploration, we won’t know if it actually helps clinicians and families.