Understanding the Discordance About Prognosis Between Clinicians and Terminally Ill Patients and Their Surrogates

A Conversation With Douglas B. White, MD, MAS

Get Permission

Research shows that about half of adults near the end of life in the United States are too ill to participate in decisions about whether to accept life-prolonging treatment,1 requiring family members and other proxies to serve as surrogate decision-makers for their critically ill loved ones. However, research also shows that surrogates of patients with advanced illness often have optimistic expectations about prognosis, which often lead to the increased use of invasive treatment (including life support) in dying patients and delays in the integration of palliative care.2

Douglas B. White, MD, MAS

Douglas B. White, MD, MAS

Douglas B. White, MD, MAS, and his colleagues, investigated the optimistic expectations about prognosis by surrogate decision-makers and whether those expectations led to greater use of life support at the end of life. In their study, they found that, overall, 45% of surrogates held optimistic expectations about the prognosis of their loved one. And that optimistic expectation, defined by the researchers as a prognostic estimate that was at least 20% more optimistic than the physician’s, was associated with a significantly longer duration of intensive care treatment, 56%, among patients before death.3 According to Dr. White, this difference was associated with a significantly longer time to the withdrawal of life support among dying patients whose surrogates had optimistic prognostic expectations compared with those surrogates who did not have those ­expectations.

“Some of the prognostic discordance between surrogates and clinicians is optimistic bias, which is to say the surrogate heard the clinician say his/her loved one has a 90% chance of dying, but he/she does not believe the patient’s chances are that bad,” said Dr. White. “When caretakers do not believe what the physician is saying, it is difficult to know how to change that misperception, because bombing them over the head with the same facts over and over is not likely to change their belief system.”

Misperceptions about prognosis, however, are not limited to patient surrogates. Research shows that only 5% of patients with advanced cancer with up to 6 months to live have an accurate understanding of their illness, with few reporting having a recent discussion about prognosis with their oncologist.4 According to ASCO’s Patient-Clinician Communication Consensus Guideline, conversations about patients’ end-of-life preferences and their advance care planning should be initiated early in the course of incurable illness and readdressed periodically based on changes in their clinical status or patient preferences.5

The reasons these conversations are not happening are many, including the difficulty many oncologists have in initiating conversations about noncurative options with their patients, the fear such a conversation will lead patients to abandon hope, and the worry that the news will adversely affect patients and lead to worse outcomes. Patients may also be hesitant to broach the subject with their oncologists over concern it may be interpreted as an expression of disappointment in their oncologists and in the care they received.6

The consequences of not understanding prognosis can be profound for both patients and their caregivers. They also can result in overly aggressive therapy that prolongs unnecessary pain and suffering for patients and extended grief and trauma for their loved ones.

In a wide-ranging interview with The ASCO Post, Dr. White discussed how oncologists can help their patients and caregivers have a greater understanding of disease prognosis, and why so many terminally ill patients still receive aggressive treatment at the end of life. Dr White is Vice Chair and Professor of Critical Care Medicine and UPMC Endowed Chair for Ethics in Critical Care Medicine and Director, Program on Ethics and Decision-Making in Critical Illness at the University of Pittsburgh Medical Center.

Getting at the Root of Prognostic Misunderstanding

Part of the work of being a good clinician is getting at the root of misunderstandings about prognosis and why patients believe their prognosis is different from the reality.
— Douglas B. White, MD, MAS

Tweet this quote

A study performed a decade ago found that a large majority of patients with incurable lung or colorectal cancer, 69% and 81%, respectively, believed chemotherapy could cure their disease, even after they were told by their oncologists that chemotherapy was very unlikely to do so.7 Why do so many patients with a terminal illness hold onto a belief that their cancer can be cured?

There are likely a number of issues at play in these circumstances. Part of the work of being a good clinician is getting at the root of inaccurate expectations about prognosis and why patients believe their prognosis is different from the reality. Sometimes, the disconnect is because the information is so new, the patient hasn’t had time to process it. Other times, it is because the oncologist  did not deliver the information clearly or the patient believes, despite the prognosis, he will have a better outcome than his peers.

Still in other cases, patients may have understood the information they were given, and they trust their oncologist, yet they have what we call a different explanatory model of illness. For example, patients may believe their cancer outcome is in God’s hands as opposed to in the knowledge purview of their clinicians.

The bottom line is we cannot begin to address the problem of optimistic expectations about prognosis if we do not know the cause.

Gently Engaging Patients in End-of-Life Discussions

What is the most effective language for physicians to use to help their patients comprehend their prognosis, so they have a realistic understanding of their medical situation?

How we engage patients in these discussions matters a huge amount. I am less keen on the idea that if we use clearer language about prognosis, it will make patients’ unrealistic expectations go away. Occasionally, patients just misunderstand the information. But many times, the information is so personally threatening, no matter how clearly the oncologist explains the situation, the patient may hold a more optimistic belief because there may be something much more powerfully psychological going on for that patient, and that merits further exploration.

The bottom line is we cannot begin to address the problem of optimistic expectations about prognosis if we do not know the cause.
— Douglas B. White, MD, MAS

Tweet this quote

Being an effective communicator has long been identified as an important physician competency. However, among the basic principles of transferring information to patients cited in the literature is to have patients’ recall back to the physician what they were just told. Yet it feels cruel to me to ask patients, “Can you tell me what I just told you about your prognosis?”

There are gentler ways of eliciting that information. Often, when I have these conversations with patients, and I am worried they may not have fully grasped what I said, I will say: “Sometimes, people want to talk to their loved ones about what was discussed at our visit. If you talk to your spouse today, what would you tell your spouse?” That is a gentler way of finding out what the patient heard and comprehended about the conversation.

Some clinicians just assume patients are misunderstanding what they have been told, and they keep repeating the prognosis over and over. How do you check for comprehension when the information you want the patient to understand is a terribly sad fact? It’s a real dilemma.

Helping Patients Maintain Quality of Life

How have advancements in cancer treatment that extend life expectancy but do not cure the disease complicate end-of-life decision-making and conversations between physicians and their patients?

End-of-life discussions are more complicated now than they used to be. On the one hand, advances in cancer therapies that offer some benefit to some patients are a very good thing. On the other hand, these conversations now are more nuanced. Clinicians have to make it clear that although there may be some life extension, the treatment is not curative and most likely will come with a range of adverse side effects.

More work needs to be done to help patients and their family members think through all the considerations of risk/benefit of continuing on third-, fourth-, and fifth-line agents; these measures may offer only a few weeks or months of life extension and may result in functional impairments and reduced quality of life. These discussions are some of the most complicated, skill-requiring conversations in medicine.

Guest Editor

Jamie H. Von Roenn, MD, FASCO

Jamie H. Von Roenn, MD, FASCO

Dr. Von Roenn is ASCO’s Vice President of Education, Science, and Professional Development. Palliative Care in Oncology addresses the evolving needs of cancer survivors at various stages of their illness and care.

Making Difficult Choices

A recent study comparing aggressive end-of-life care among 146,000 older patients with advanced cancer living in nursing homes and community settings found that both groups—58% of community dwellers and 64% of nursing home residents—received aggressive treatment in the last 30 days of their lives.8 Why are so many patients with metastatic cancer still receiving aggressive care at the end of life?

Patients often simultaneously hold two beliefs. They do not want to be put on life support or undergo intensive treatment near the end of life. However, many also say they do not want to die, so there is a real conflict in that situation for both patients and physicians.

Some patients have a very strong desire to stay alive and are willing to endure a lot of aggressive care, even if it has very little chance of providing a life-prolonging benefit. Clinicians are required to be honest with their patients and offer their best judgment on the risk/benefit of therapies that have uncertain likelihood of providing some life extension.

DISCLOSURE: Dr. White reported no conflicts of interest.


1. Silveira MJ, Kim SYH, Langa KM: Advance directives and outcomes of surrogate decision making before death. N Engl J Med 362:1211-1218, 2010.

2. Zier LS, Burack JH, Micco G, et al: Surrogate decision makers’ responses to physicians’ predictions of medical futility. Chest 136:110-117, 2009.

3. White DB, Carson S, Anderson W, et al: A multicenter study of the causes and consequences of optimistic expectations about prognosis by surrogate decision-makers in ICUs. Crit Care Med 47:1184-1193, 2019.

4. Epstein AS, Prigerson HG, O’Reilly EM, et al: Discussion of life expectancy and changes in illness understanding in patients with advanced cancer. J Clin Oncol 34:2398-2403, 2016.

5. Gilligan T, Coyle N, Frankel RM, et al: Patient-clinician communication: American Society of Clinical Oncology consensus guideline. J Clin Oncol 35:3618-3632, 2017.

6. Gesme DH, Wiseman M: Advance care planning with your patients. J Oncol Pract 7:e42-e44, 2011.

7. Weeks JC, Catalano PJ, Cronin A, et al: Patients’ expectations about effects of chemotherapy for advanced cancer. N Engl J Med 367:1616-1625, 2012.

8. Koroukian SM, Douglas SL, Vu L, et al: Incidence of aggressive end-of-life care among older adults with metastatic cancer living in nursing homes and community settings. JAMA Netw Open 6:e230394, 2023.