Secondary Logo

Journal Logo

Special Report

The Big Picture

Communicating Prognosis in the ED

Highet, Bridget MD

doi: 10.1097/
Special Report



They surround us every day. The haunting frames of cancer, the tired breaths of COPD. We walk among them. We try to erase the writing on the wall with IV fluids and antibiotics, but we know that many of our patients are dying. Perhaps not today, perhaps not tomorrow, but soon.

Seventy-five percent of Medicare decedents are seen in the emergency department in the last six months of their lives. Fifty-one percent are seen in their last month. (Health Aff [Millwood]. 2012;31[6]:1277; Some would argue that these visits are critical, slivers of time when these patients belong to us. We have an opportunity in these small moments to make a profound difference on their disease trajectories.

Perhaps the first step is sharing what we know. When we walk behind the glass of our workstation to whisper to the nurse, “He isn't going to make it to Christmas,” could we muster the courage to turn back around, pull up a chair, and tell the patient what we think?

We emergency physicians receive no formal training in prognosis. We pride ourselves on our ability to thin-slice: sick or not sick. We steer away from conversations about the end of life, feeling unprepared and worrying we will take away hope.

I faced these same challenges during my last year as a palliative medicine fellow. When paged with a consult, I would scour a chart that often spans weeks or months, and then stumble—a stranger—into the room of a family in crisis. The tools I have acquired in my training have changed my practice of emergency medicine. It has made me listen more closely, look more carefully, and speak less often. In many moments, our specialties run parallel. Armed with palliative tools, I believe that we emergency physicians can improve our conversations with seriously ill patients.

Back to Top | Article Outline

Finding the Words

Assess Understanding: We all have shared the astonishing experience of a patient being unable to tell you the purpose of a surgery he had—one week earlier. Often our chronically ill patients suffer the same guileless ignorance. Sixty-nine percent of patients with stage IV lung cancer and 81 percent of those with stage IV colon cancer believed their chemotherapy was possibly curative. (N Engl J Med. 2012;367[17]:1616;

Before jumping into the chasm of a serious conversation, I recommend first asking the patient what she knows about her illness. The answers you receive will be illuminating. You will discover that patients don't know what their medicines are for. They don't realize that their COPD, CHF, or cancer is incurable. By starting here, you can lay the groundwork for what needs to be discussed and gain tremendous insight into a patient or family's perspective.

Ask Permission: A study of older adults demonstrated that 75 percent would want to know if they had a year or less to live. That being said, 25 percent don't want to know. (J Gen Intern Med. 2012;27[5]:568; So ask, “How much do you want to know?” Starting this dialogue empowers patients and families to take control of the conversation. Sometimes the patient will turn you down when you broach the topic of prognosis, but providing the option allows us to maintain an open and honest channel of communication with our patients. (J Clin Oncol. 2006;24[25]:4209.)

Use Ranges: Discussing prognosis feels unnatural to emergency physicians. It is not a part of our education, and our lack of longitudinal relationships with our patients makes the conversation difficult. The good news is all physicians are inaccurate at disclosing prognosis, and physicians with longer relationships with patients tend to be overly optimistic. (West J Med. 2000;172[5]:310;

I challenge you to change your perception of prognosis. We are not fortune-tellers, and patients will prove us wrong even when we are armed with the best tools. The exact day a patient will die is between them and the divine, and it is paramount to acknowledge this to ourselves and to them.

But what if I ask you if a patient is likely to die in hours to days, days to weeks, weeks to months, months to a year, months to years? Does your comfort start to settle? A patient who is actively dying has hours to days while a patient with advanced dementia on his third hospitalization for severe sepsis probably has months to a year. Delivering information in ranges allows us to communicate a usable paradigm that can assist with the difficult decisions that our patients often face. (Hallenbeck, J., Palliative Care Perspectives. 2003: Oxford University Press.)

Pairing Hopes and Worries: We are always hopeful for our patients; otherwise, we wouldn't have chosen this field. We often steer away from serious conversations because we don't want to upset our patients or their families. We don't want to do harm. Lakin described how we can communicate prognostic information to our patients without taking away hope. (JAMA Intern Med. 2019 1;179[1]:5.) “I am hopeful that your mother will recover from the bleeding in her brain, but I am worried that she may deteriorate.” Using “I” language and pairing our hopes and worries can ally us with our patients and their families while still honestly disclosing our concerns.

Best Case/Worst Case: Take the case of a 90-year-old bedbound woman with advanced dementia who comes to the ED from a nursing home. She is pale, wan, with labored respirations, a cloudy Foley catheter, and a blood pressure of 70 systolic. We are faced with uncertainty when predicting what the future will bring for this patient. Sometimes in palliative care, we use an approach derived from a tool designed by surgeons at the University of Wisconsin entitled best case/worst case. (J Am Geriatr Soc. 2015;63[9]:1805;

What is the best-case scenario for her if we pursue critical care interventions for this patient? She may have a short ICU stay, and ultimately be discharged back to her nursing home where she will have the same level of function as she did before. What is the worst case? She could have a prolonged ICU stay punctuated by interventions like central lines and mechanical ventilation. She could contract C. diff or pneumonia, get sicker, and ultimately live out her final days in the hospital. What is the most likely scenario on this continuum?

What if her family pursues comfort care? Perhaps her best-case scenario is being enrolled in hospice, having her symptoms managed aggressively, and then dying, likely within days to weeks surrounded by her family.

The bottom line of this exercise is the realization that the best-case scenario for this patient in choosing ICU admission is to return to her nursing home with the same quality of life she had before her admission. The likelihood is that she will deteriorate at least some. Painting this picture for the family allows them to make an informed decision on the best course of action for the patient based on their understanding of her values and what she deems an acceptable quality of life.

Back to Top | Article Outline

Prognosis in Cancer Patients

Even if we don't talk about it, most of us know the feeling of cancer-riddled ribs cracking beneath our compressions. One of my first encounters with moral distress in medicine was as an intern tasked with resuscitating a woman with stage IV breast cancer. A new doctor, I was focused on my ACLS algorithm, commanding the room, trying to expertly time my epinephrine doses, and ensuring compressions were deep and rhythmic. I can still picture her bony chest heaving with every forced breath from the bag, and her vacant, lifeless eyes staring at the ceiling of the critical care bay. What were we doing?

The United States still struggles with high rates of hospitalization at the end of life despite guidelines and efforts to reduce aggressiveness in end-stage cancer care. Forty percent of Medicare decedents with cancer are admitted to an intensive care unit at least once in the last six months of life, 27 percent in their last month. (JAMA. 2016;315[3]:272.)

The typical trajectory of cancer is one of steady disease progression until a precipitous decline in function marks the end of life. (BMJ. 2005;330[7498]:1007; Many of our cancer patients arrive to us at the tipping point of their disease. If we can identify and communicate this decline to our oncology colleagues, perhaps we can help to swing the pendulum when our patients have the greatest need.

If I had one question to ask a patient with cancer, it would be this: “How much of your day is spent in bed or in a chair? If the answer is more than half, there is cause for concern. The American Society of Clinical Oncologists chose as its first guideline for the Choosing Wisely campaign in 2012 that cancer-directed therapies are unlikely to benefit patients with poor performance status. This is defined by a limited ability for self-care or confinement to a bed or chair for more than half of a patient's waking hours. (J Palliat Med. 2012;15[2]:175.)

Patients with cancer at the end of their disease undergo a precipitous decline punctuated by loss of function. They sleep more, eat less, and become unable to take care of themselves. Patients spending more than half a day in bed or a chair have a median survival of six months or less and cancer-directed therapy no longer improves survival. (J Palliat Med. 2012;15[2]:175.) It is common for patients to present to the ED during this decline, and if we identify it, we can bring it to the attention of the patients, their families, and their oncologists.

Certain cancer presentations are also associated with a poor prognosis. Malignant pericardial effusion or malignant hypercalcemia (except for newly diagnosed breast cancer or myeloma) portends a median survival of eight weeks or less. (Ann Intern Med. 1990;112[7]:499; Heart. 1996;75[1]:67; Multiple brain metastases or venous or pulmonary thromboembolism predicts a median survival of six months or less. Malignancy-related small bowel obstruction, pleural effusion, and ascites also suggest a poor prognosis, particularly when combined with other signs of advanced disease or functional decline. (J Palliat Med. 2012;15[2]:175.)

A discussion regarding goals of care may be appropriate if a patient has poor functional status or several of these high-risk conditions. A reasonable approach is to get the patient's oncologist on the phone, and ask, “What's the big picture here?”

You may get an answer explaining that the patient is in a promising clinical trial of immunotherapy, and the oncologist believes it's worth fighting. Alternatively, you may discover that the oncologist has been seeing the same patterns that you are. Perhaps it would be appropriate to get hospice involved and consider a transition to comfort care.

My hope is that emergency physicians can use these tools to feel empowered to engage in difficult conversations with seriously ill patients. We are the gatekeepers of the hospital, and our words and choices affect our patients' trajectories. We can confidently inform and facilitate the complex decisions our patients are faced with in serious illness with compassion and honesty.

Share this article on Twitter and Facebook.

Access the links in EMN by reading this on our website:

Comments? Write to us at

Dr. Highetis an emergency medicine and hospice and palliative physician at Halifax Health in Daytona Beach, FL. She recently completed a palliative medicine fellowship at the Mayo Clinic in Jacksonville, FL. Follow her on Twitter @bhighetEDPALL.

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.