(Vahram Muradyan for The Washington Post)

Sometimes, doctors and patients disagree about end-of-life care. I know, because it happened with me.

Not long ago, a frail-looking elderly patient appeared at my cardiac health clinic with a file full of hospitalizations stemming from a heart attack years before. He’d had three coronary stents put in, had had heart bypass surgery and was unable to walk for more than a block due to chest pain. I saw that a previous doctor had written “DNR” — do not resuscitate — in his chart, so I asked him to confirm his wishes.

No, he said, to my surprise. He actually wanted to be a “full code” — meaning that chest compressions, shocks and intubation were to be used if necessary to keep him alive.

I was taken aback. Someone had presumably told him that he would probably not survive a cardiac arrest even with cardiopulmonary resuscitation. (Only 1 out of 7 patients older than 80 do; this man was close to 90.)

“I don’t want anyone giving up on me,” he said. No matter how much I explained that CPR would probably only cause him to suffer longer without adding to his life, he remained adamant. He left my office visit with a yellow “full code” indicator in his chart instead of the red “DNR” one.

When I walked home that night, I couldn’t help thinking how badly our conversation had gone. I had discussed end-of-life planning with many patients, and usually we’d been able to come to decisions that made sense both medically and to the patient. But not this time.

Maybe I hadn’t explained the facts well enough; maybe there was a language barrier; or maybe he just didn’t trust me.

Medicare recently put in place new rules that will allow doctors to be reimbursed for discussing end-of-life preferences with patients.

Originally derailed by the “death panel” accusations of 2010, the measure has been hailed as a major victory toward a better, more transparent system of end-of-life care. Patrick Conway, the chief medical officer for the Centers for Medicare and Medicaid Services, remarked that such advance-care planning discussions were important to promote “patient- and family-centered care.”

But sometimes these discussions are anything but patient-centered. That’s because what the doctor wants doesn’t always align with what the patient needs.

As a resident, I approach most end-of-life conversations with a narrow focus: Confirm a code status and get the name of an emergency contact person. “Do you want to be resuscitated if you lost your pulse?” “Do you want to have a breathing tube if you were unable to breathe on your own?” “Whom should we contact in case of an emergency?” I needed to document answers to those questions to finish my patient-chart note. With my pager ringing and admissions arriving, I usually didn’t have time for anything else.

The patients I speak to, however, want so much more out of their end-of-life conversations.

One of the most common questions I heard was “How long do I have?” Indeed, about half of patients desire some knowledge about prognosis when discussing the end of life. They also wanted to know whether their symptoms — pain and nausea, for example — would ever go away, and what options would allow them to be at home during their final days. One woman even asked whether someone could set up her deathbed along the same beach where she had grown up. I frequently can’t answer these questions.

Yet even these potentially frustrating conversations aren’t that common: While 8 in 10 people of all ages think it is important to talk to their doctor about their end-of-life wishes, fewer than 1 in 10 report actually having had these discussions. But there are two participants in those end-of-life conversations: patient and doctor. Unless each realizes what the other wants, both risk coming away from those conversations feeling uncomfortable or sensing that things had not gone well.

Before I saw my elderly heart patient for his next visit, I came across the website of the Conversation Project, a group dedicated to helping people discuss their wishes for end-of-life care. Led by health professionals, media, clergy and others, the Conversation Project has published a “Starter Kit” to help patients and health professionals prepare for end-of-life conversations. The guide first asks patients to fill in the blank: “What matters to me at the end of life is ______.” Subsequent steps are based on that answer.

More resources for advance-care planning exist. A nonprofit called ACP Decisions has created videos to help physicians explain emotionally charged end-of-life concepts such as CPR, hospice and feeding tubes. At my hospital, we have used simulated conversations with patient actors to help residents approach the end-of-life conversation.

So at my next meeting with my patient, rather than beginning the discussion around his code status, I asked what mattered most to him. Over the next five minutes, he explained how his great-granddaughter in Ethiopia had just been born. The family was flying to America next month, presumably to see my patient for the last time. Even if it meant receiving compressions or shocks, he wanted to make sure he was alive to see that baby. He remained a full code until our next visit, which was two months later, after his family had left. At this point, he decided to revert to DNR. We are discussing hospice care now.

I no longer think that my first discussion with that patient was a failure. But it opened my eyes to how little I’d been listening. End-of-life discussions are opportunities to learn more, to start a conversation that we as doctors have been missing out on for quite a while. Once we listen enough to learn, maybe those “goals of care” discussions will start focusing on the goals of the patient, not the doctor.

Parikh is a resident in internal medicine and primary care at Brigham and Women’s Hospital in Boston. Follow him on Twitter @ravi_b_parikh.


Where in the world is the best place to die?

A nurse with fatal breast cancer says end-of-living discussions saved her life

Honest talk about dying