When a patient is admitted to the hospital in the U.S., there’s a standard question physicians like me are supposed to ask: “If your heart stops beating, do you want us to do CPR?”
On the surface, this may seem like a mechanic asking a customer, “If your car stalls, do you want us to jumpstart the engine?” Who would say no to this, especially in a hospital?
The problem is that this exchange, which we call asking about “code status” in medicine, centers around a closed-ended question. Talking to a patient about their preferences for cardiac resuscitation, intubation, and/or other life-sustaining treatments needs to be a complete, often lengthy discussion, not just a box to check.
A single question doesn’t give context, ask what’s most important to you to guide this decision, or share why we may recommend do-not-resuscitate (DNR) when CPR likely won’t work and cause further suffering. Popular shows like “Grey’s Anatomy” and “House” show 70% of patients surviving in-hospital cardiac arrests, but in reality, that number is just 17% . A popular recent New York Times article highlighted what can happen when there is confusion about resuscitation preferences.
However, what many readers may not have understood is that physicians receive very little training on how to actually have these important conversations. Communication skills training, especially around serious illness topics like code status, isn’t routinely taught in medical education. That must change.
U.S. health care started embracing shared decision-making between patients and clinicians in the early 2000s. Yet most physicians are not trained in discussing end-of-life care preferences. Studies show that patients who discuss their medical goals and treatment preferences are more likely to have a higher quality of life. However, the status quo of physicians learning to have code status discussions on the job may be harming our patients, their families, and our health care system.
Less than half of doctors report knowing what to say during these conversations, and less than a third feel they received enough skills training to navigate them. We bury our discomfort beneath euphemisms like “heroic measures” or “doing everything,” and medical jargon like “resuscitation.” Resident physicians spend only about 10 minutes discussing code status with each patient, while attending physicians devote roughly one minute of discussion with patients. One study found more than half of participants with advanced cancer weren’t asked about code status on admission to the hospital, so they were entered in the computer as “presumed full code,” assuming they wanted CPR. Yet, on further clarification, 48% of these patients said that they would have been DNR if someone had asked.
As a fellow physician in palliative care, doctors often ask me to help them have these conversations with patients and caregivers. My role isn’t to talk people out of wanting CPR. It’s to assess their understanding of it and their illness, provide necessary context, and learn what’s important to them before making recommendations. I’ve seen the unintended pain inflicted when these conversations are delayed or deferred. A man at his wife’s bedside in the ICU grappling with her wish to avoid life support, not knowing she’d need to be on a ventilator as part of CPR. A daughter wishing she knew CPR was unlikely to help her mom with stage 4 cancer, and worrying her mom’s final moments were painful after seeing the physical force of well-performed chest compressions.
Electing to forgo CPR is a tough decision but one that a physician may recommend if the risks outweigh the possible benefits. Unfortunately, we reach points when we’ve addressed all potentially reversible causes of illness, and no further medical treatment options remain. Comfort measures only (CMO) is another code status option that prioritizes interventions to keep patients comfortable at the end of their lives and would not include receiving CPR.
Some patients tell me even if CPR worked, they wouldn’t want the things that commonly come after it like artificial nutrition through gastric tubes, breathing through a tracheostomy, or requiring care in a nursing home. If CPR restores their life but they can’t keep playing with their grandchildren, cooking for their spouse, or fishing, they might feel like they aren’t living. You can’t escort patients into these emotionally challenging hypotheticals in a one-minute or one-question conversation.
As a new palliative care fellow, I have room to grow in having these conversations, but I’m afforded two things I didn’t have as a resident. The first is training to have these code status discussions. Through lectures, simulation training, and feedback, I’ve learned that code status determination shouldn’t be the focus of the conversation even if that question needs clarification. Who is the patient outside of the hospital? What do they understand about their illness? What are their care goals and what’s most important to them? These questions need to be answered first. Does CPR align with what’s most important to them, and do we think realistically CPR will get them to their goals?
The second is ample time to do so. My team usually spends an hour with patients or families discussing the nuances of CPR and care at the end of life. We revisit these conversations as we get to know patients better, partnering in hoping for the best while also planning for the worst.
However, we can’t expect every physician to spend another year in training to learn these skills. Medical students need opportunities to learn how to have code status discussions. Practicing clinical skills with standardized patient actors is a commonly used teaching modality in medical education and could be easily adapted to include code status discussions. Early skills training in having code status discussions should be considered like learning to do procedures. Both are high stakes with serious consequences that require training and opportunities for feedback. Our patients and their caregivers count on us to elicit and respect their wishes in the medical care we provide. We must include code status discussions in the medical education curriculum.
Lindsey Ulin is a palliative care physician fellow at Massachusetts General Hospital and The Dana-Farber Cancer Institute.
