How Do We Talk About the Worst-case Scenario?

This post was originally published in the American Academy of Family Physicians’ Fresh Perspectives blog — a venue that frequently highlights Dr. Edwards’ writing.  In this post, Dr. Edwards ponders the best way to broach the subject of end-of-life care when it can be both uncomfortable and not top-of-mind for many of us.

Back in December — before COVID-19 — my phone rang in the middle of the night. My hand shot out of a dead sleep, silencing the ringer as I fumbled to force the call to voicemail. Awake but groggy, I quietly sneaked out of the bedroom I share with my spouse.

[power of attorney and living will]

This call was from the wife of a patient. As a direct primary care doctor, I invite my patients to call with urgent things any time of day. Even with the open invitation, I have only had three middle-of-the-night calls since opening my practice in January 2017.

I had been speaking with both her and my patient frequently during the past few weeks. My patient, William, had had a tough go of it since I had started caring for him in early 2017. We had gotten through renal cell carcinoma resection despite his uninsured status. We had worked time and again to figure out why he had persistent leukocytosis — and finally got the diagnosis of chronic myelogenous leukemia once he was able to get a bone marrow biopsy. While I medically managed his coronary artery disease, diabetes and hypertension, we had had to coordinate several different hospital and clinic systems to get him the specialty care he needed: hematology, oncology and ophthalmology. Even with all that, he had been fairly stable until last fall.

He qualified for Medicaid due to disability but had missed jumping through some hoop or hadn’t crossed a T, and he had been dropped. He was having trouble figuring out exactly how to do the right thing — despite the fact that he had legal help and is one of those people who can recite his medication list from memory. He can tell you the date and time of all his upcoming appointments. And his wife’s, too. But he couldn’t quite figure out how to get coverage.

As we were sorting out his insurance mess, his CML seemed to morph and we watched as his white count climbed into the 50s, then the 60s. His oncologist — who was still seeing him in spite of his perpetual “Medicaid pending” status — had arranged for bosutinib through the manufacturer’s patient-assistance program, but the medication was sitting at William’s house, unused, because the oncologist’s health system had cut him off from further care unless he could pay. His oncologist had pleaded William’s case as high as the CFO, but there it was summarily dismissed. William was on his own.

Desperate, I had placed numerous calls to his oncologist’s office to figure out if I could monitor his treatment — with the oncologist’s help — until we got his financial situation sorted out. I couldn’t stomach the idea of my patient’s treatment sitting in his house, unused — simply because he needed a quarterback.

There was a back-and-forth between me, William, his wife and the oncologist, and we finally decided that we couldn’t wait. We would start treatment. I would watch for side effects, and our clinic would get labs.

But only a day after starting treatment in December — a few days before the middle-of-the-night phone call — he took a turn. His wife had called me a few days before, and I recommended he present to the ER.

The first time I met William, back in the spring of 2017, he was slightly unkempt, with straw-like, strawberry-blonde hair poking out from under a baseball cap. He shuffled into my office in a baggy T-shirt and sweatpants, his whole demeanor suggesting life hadn’t been easy. His chief complaint that day was that he just didn’t feel good — nothing specific — just that he was “off.” As it turned out, he was septic with pneumonia and ended up hospitalized for several days. So, when he feels off — nonspecifically — he gets full clearance from me to head to the ER.

As I brought the cold glass of my phone to my ear that night, I could hear the tinny recording of William’s wife sobbing, speaking in breathless fragments. She said that William’s heart had stopped. That they had to do CPR. That he was in the ICU with a tube helping him breathe. That she didn’t know what to do. She wanted me to call her.

The next day I came by William’s ICU room. Always a pragmatist, his wife tearfully shared that she just didn’t know how to answer the ICU team’s seemingly straightforward, simple question: What was his code status?

When William was discharged from that first hospitalization under my care — years ago — I had printed out a Physician Orders for Life-Sustaining Treatment form to discuss with him. But each visit following his discharge had brought a more urgent need, and the POLST form — with his name literally on it — collected dust in my clinic room and eventually was shredded.

In the intervening 2 1/2 years, I had never asked him his wishes. His wife never asked, either.

Since that moment in the ICU, I have wrestled with the uncomfortable realization that I am not good at figuring out my patients’ end-of-life desires. I will ask when I’m in the ER, and I will ask when I’m in the hospital. I will also ask — sometimes — when a patient has life-threatening, progressive organ failure and they’re in my care at my clinic.

When I first opened my clinic, I would discuss a patient’s code status at their initial visit, but after numerous times of getting turned down or dismissed with a simple “I’ll think about it,” I stopped asking. And I stopped following up. Further abetting my complacency, the EHR I use doesn’t even have a way for me to track if I know my patients’ end-of-life preferences.

As COVID-19 has slowly spread through our community, I’ve thought about this deficiency again and again. I just haven’t talked to my patients about their end-of-life preferences.

And I don’t quite know how to broach the subject at this point. It seems odd to reach out to them via random email and nonchalantly ask about their code status. I just can’t get away from the feeling that this would be weird:

Hi, Mr. X:

I just wanted to reach out to see if you have ever considered what you’d want if you were hospitalized in critical condition and couldn’t speak for yourself. Let me know! 🙂


Allison Edwards, M.D.

It would also be odd to just call out of the blue with a casual, “Hi, Mr. X, it’s Dr. Edwards, your primary care physician. I’m calling all of my patients to talk about their end-of-life plans, you know, given COVID and all that. Just give me a call whenever you have a minute to chat.”

I can’t push aside my very American (or maybe human?) feeling that abruptly bringing up this conversation is uncomfortable. Moreover, my patients are young. Twenty percent of our practice is under the age of 18 — and even if you take our pediatric population out of the data set, our mean age at the clinic is about 40.

But maybe that makes this conversation more important; maybe I need to have this conversation because of how unexpected — but potentially predictable — critical illness is in the context of COVID-19. And maybe I just need to state things clearly: With concern for spreading infection top of mind in the hospital setting, we may not be able to have this conversation with our loved ones at the exact moment when it matters most.

I’m keenly aware that there are great tools out there to facilitate the conversation. A POLST form is a worksheet that explicitly asks people to indicate their preference when it comes to life-sustaining treatment. It has a standard format and is easily and quickly understood by any health care professional. The biggest drawback is that it doesn’t allow for the nuance and situational changes that may change a person’s choice.

To capture that side of a person’s preferences, the Five Wishes living will operates a little more broadly and shifts the focus from the explicit actions a person desires (“Do you want chest compressions?”) to the more expansive medical, personal, emotional and spiritual dimensions that make a person whole.

The most nuanced resource I’ve found in the time of COVID-19, though, is The Conversation Project’s Being Prepared in the Time of COVID-19 worksheet. It guides people in legally creating a durable power of attorney for health care decisions — the most flexible action individuals can take when it comes to planning for uncertain and scary health care situations — and further prods the user to have a conversation with that person about their wishes. It links to resources on how to structure a conversation about end-of-life care and even has a few prompts to get people thinking about what they’d want to have happen during critical illness.

Finally, American Family Physician offers a collection of its best content and other AAFP resources regarding end-of-life care, including advance directives for physicians who are looking for more resources.  And the Academy also has a self-study course that specifically addresses having difficult conversations related to COVID-19.

Any — or all — of these resources are useful. As impersonal as it seems, I may just send out a mass email to all of my patients to gently invite them to opt in to the conversation. I have no idea how many people will engage with me on the subject, but I will happily have that discussion if they are interested. Because none of us knows when the worst will happen.

In January, when William was discharged from first the ICU and eventually from the hospital, I finally was able to ask him about his end-of-life preferences. He joked a bit about the agonizing rib pain from the CPR and the hoarseness from the ET tube, but he simply said, “Let my wife decide. I trust her.”

And for him, that was as much as he wanted to say. But at least we knew. Because we asked.