The Day One Talk, 20 Years Later: What Have We Learned?

Nearly 20 years ago, Holcombe Grier, MD, and I wrote The Day One Talk,¹ an article describing our process for disclosing a new cancer diagnosis to children and their families. Whenever possible, we built on existing evidence for best communication practice, but the evidence in pediatric oncology was limited at the time. As a result, much of the article was based on our experience. Holcombe in particular had years of experience talking with children and families, and in turn, he taught his process to most of the fellows and faculty at our institution. The article was our effort to share this human and humanizing aspect of medicine with a wider set of colleagues and to bring this part of clinical practice into the academic conversation.

In the years since its publication, a growing group of investigators has created an evidence base for communication practice in pediatric oncology, including answering basic questions about what children and parents want to know, what they take away from conversations, how conversations affect care, and how conversations affect the individuals who are part of them. Although I believe we would still propose largely the same Day One Talk today, some of the things learned through research in the intervening years surprised us. In particular, contrary to the fears of many clinicians, we found that providing honest prognostic information seems to support hope and peace of mind for both parents and adolescents.^2-4 Difficult news can be supportive and even healing for those who are facing the unknown.

A welcome new paper by Kathryn Six et al⁵ examines current practice in new diagnosis discussions in pediatric oncology. This work gives us a chance to reflect on the state of pediatric oncology communication and look forward toward areas where we need to grow. This group assessed the state of the field in two phases, beginning with interviews with 20 pediatric oncologists at a single center and then surveying fellowship program directors and fellows about their own practice. A majority of participants, by design, were fellows, who lack the benefit of years of experience but who are often clinically immersed and who represent the future of the field.

One of the notable findings of the manuscript is the extent to which conversations are individualized in hopes of meeting patient and family needs. Identified areas of particular individualization include decisions about whether the child should be part of the conversation and whether and how to discuss prognosis. Both decisions can be quite challenging; preferences for each are known to vary and are subject to complex influences.^4,6-7 The decision to tailor conversations to personal needs represents a patient- and family-centered approach. However, whenever care is highly individualized, threats to quality and equity can arise if practitioners rely on instinct, which can introduce bias.

Existing evidence does in fact support a potential role of bias in communication about prognosis in pediatric oncology. In a previous study of 357 parents of children with cancer,⁸ most parents (87%) said that they wanted as much information as possible about their child's prognosis, with no differences by race or ethnicity. Yet, oncologist perceptions of parents' information needs did differ by parent race and ethnicity; oncologists identified Black and Hispanic parents as less interested in prognostic information than White parents. They also tended to believe that Black, Hispanic, and Asian parents had poorer comprehension of medical information than White parents. These biases have potential clinical consequences for minority parents and their children, including lower prognostic understanding,^8,9 higher rates of invasive measures at the end of life for children with incurable cancer,¹⁰ and higher rates of decisional regret.¹¹ Although not every parent wants prognostic information, most do, and we need to meet those needs to provide the best care.

In our previous work, we have proposed a simple solution to the difficulty of ascertaining whether an individual parent or child might want prognostic information: just ask. This is reflected by the practice reported by one of the participants in the study by Six⁵: “I usually give them an opportunity to say are you a numbers person? Do you want to hear a specific number based on our past research, what we think? … Or would you rather just hear, more in general terms, such as it may be very difficult to cure this or we have a good chance of curing this? And so I give them the opportunity to say which one they are.” Notably, other oncologists said they only give information if families ask. However, parents may not ask questions for many reasons—they do not know what to ask, they worry about seeming foolish, or they trust the oncologist to give them all important information—and not only because they do not want such information.¹² Six et al⁵ also found that the decision to provide numeric versus qualitative prognostic information was determined by parent preference only 36% of the time when prognosis was favorable and 38% of the time when prognosis was poor. Otherwise, the oncologist made the decision about which type of information to provide. In other words, while there is some degree of individualization, for most parents, a lack of standardization is not based on their own personal preferences but rather the decision of the oncologist.

As a result, the study by Six⁵ raises the concern that practice surrounding new diagnosis discussions in pediatric oncology does not always follow existing evidence or fully reflect parent preferences. Pediatric oncologists are accustomed to using evidence-based medicine when it comes to cancer therapy and management or prevention of complications. Yet, when it comes to communication, oncologists often seem to rely on a personal sense of what will work best for a particular patient or family.

This raises the question of how findings from communication research can best be integrated into practice. Previous work has proposed that successful integration of research findings into practice depends on three elements—the nature of the evidence, the practice context or setting, and the integration process.¹³ When it comes to prognosis communication and other elements of the Day One Talk, a particular challenge is the nature of the evidence. Limitations in the current literature—namely that studies tend to be observational and describe associations rather than establishing cause and effect—are important. But even more important is the evidence supplied by personal experience, which may be contrary to the evidence in the medical literature and yet weighed heavily by clinicians. In other words, talking about prognosis may feel wrong even if oncologists are aware that the evidence supports it. Especially given evidence that not every parent or child will want prognostic information, a bias toward nondisclosure may feel like the best way to avoid harm and support hope. As one oncologist in the study said, “I think you have to leave them with some hope that is reasonable. They will get the realities and things in time, at first they just need to be able to get up off the floor.” This instinct is so strong that, even in my own practice, I often have to remind myself to ask parents and children what they want to know.

Even if these challenges are overcome, successful changes in practice require attention to the context of implementation,¹³ including a culture that values learning from evidence, the organizational structure and leadership, and use of measures that promote attention to the relevant issue, as well as attention to the implementation process,¹³ including communication training. In the study by Six,⁵ most fellows and program directors reported that some instruction in new diagnosis discussions took place but often this took the form of shadowing rather than formal didactic methods or simulation. Fellows may have learned about topics such as prognosis communication simply by following the example of a more senior oncologist. We know from previous literature that even intensive communication training programs often have mixed results. Taken together, all of these factors, including data that go against personal experience, environments that do not always prioritize communication quality, and a lack of effective training, all create threats to the best communication experience of patients and families.

The article by Six et al⁵ makes clear that pediatric oncologists put tremendous care into new diagnosis discussions and yet their efforts to meet patient and family needs can sometimes lead them to bypass the evidence. I suspect that people who cared less might be more consistently direct, and as we come up with solutions to this challenge, we need to draw on these caring instincts. Pediatric oncologists need continued support to see direct communication about difficult topics as an act of compassion, not a repudiation of it. As we wrote in the original Day One Talk, “conversation is a major, and sometimes the only, way for physicians to alleviate suffering.”

AUTHOR'S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The Day One Talk, 20 Years Later: What Have We Learned?

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/op/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

No potential conflicts of interest were reported.