Case and Commentary
Sep 2006

Discussing Code Status with Patients and Their Families

Lorraine M. Stone, MD, MSPH and James A. Tulsky, MD
Virtual Mentor. 2006;8(9):559-563. doi: 10.1001/virtualmentor.2006.8.9.ccas1-0609.


George Johnson was admitted to the hospital with nausea and abdominal pain. His blood levels of amylase and lipase were markedly elevated, and he had a history of alcohol abuse. Dr. Jones, the intern on call, was confident that the 45-year-old Mr. Johnson was suffering from a routine case of acute pancreatitis. As morning rounds began it became clear to the inpatient team that Dr. Jones was utterly exhausted. The senior resident, Dr. Smith, felt a bit uneasy about the situation, having seen “routine pancreatitis” degenerate into systemic inflammatory response syndrome and a trip to the intensive care unit. Although she was sympathetic to Dr. Jones’s fatigue, she considered this important enough to merit a brief discussion and a learning point for the students.

After a short talk about the potentially serious complications of pancreatitis, Dr. Smith asked the intern about Mr. Johnson’s code status, pointing out that it was not mentioned in the chart. Dr. Jones replied that he believed it was unnecessary to discuss resuscitation since Mr. Johnson was stable and would surely be discharged within a couple of days given the nature of his illness. Dr. Smith emphasized the importance of discussing code status with all patients, even those who appear stable, since it is not possible to predict who will do well and who will not. She encouraged Dr. Jones to return to Mr. Johnson’s room after rounds to discuss this issue more formally.

Entering Mr. Johnson’s room, exhausted and feeling the time pressures of his other duties, Dr. Jones raised the subject abruptly. “There’s something important we need to discuss, Mr. Johnson. If something bad were to happen to you, would you want us do CPR?”

Shocked at the mention of this topic seemingly out of the blue, Mr. Johnson worriedly exclaimed, “What? Am I dying? Am I going to die? What’s wrong with me doc? Tell me what’s going on!”

Startled by his response, the intern replied, “No, no, I’m sorry, I didn’t mean that. It’s just that…well…we really have to get this information in everybody’s chart, just in case.”

Visibly upset at both Dr. Jones and the subject matter, Mr. Johnson yelled, “What are you, crazy? I’m not going to die, and this is hardly the time to ask me to think about it. Now get out.”

Dr. Jones left the room quickly, angry at Dr. Smith and her advice. Was it really so important to discuss code status with this man or with other hospitalized patients who had simple, curable conditions? What could he do in future encounters to better broach this difficult subject?


Dr. Jones is learning by experience how difficult it is to start conversations about CPR preferences out of context. The intention of asking all hospitalized patients about their CPR preferences is a good one, but in most cases it requires more than a quick, casual question. In the case before us, George Johnson is, at 45, a young member of today’s hospitalized population. He has a history of alcohol abuse, but there is no other report of chronic medical illness or repeated hospitalizations. This may be the first time Mr. Johnson has been approached with a conversation regarding his mortality. The situation is exacerbated by Dr. Jones’s exhaustion and his inability to introduce the subject properly. Mr. Johnson responds with obvious shock and confusion, thus turning a potentially meaningful discussion into an angry outburst.

One can easily understand why Dr. Jones was hesitant to ask Mr. Johnson whether he would want CPR; few 45-year-old men without multiple comorbidities will refuse CPR or other life-sustaining treatments. The question is really not about what to do in the event of an acute decompensation. Rather, the physician’s goal is to gain greater understanding of the patient’s underlying values and goals for care and how these might influence difficult treatment decisions should his condition deteriorate significantly over time. This conversation can be had in a less abrupt manner by providing a more explicit context [1].

Approaching thePpatient to Discuss Care Preferences

Initiating a conversation about CPR preferences warrants thoughtfulness and preparation on the part of the physician. Next time, Dr. Jones should pause before entering his patient’s room, reflect on his own physical and emotional state and assess whether these will impact negatively on the conversation. While fatigue and haste cannot always be avoided, recognizing their presence can help prevent their deleterious effect. Dr. Jones can also make sure he appears professional and shows the patient respect, with his shirt tucked in and white coat on. Once in the room, Dr. Jones should sit at the patient’s level and, tired as he may be, focus entirely on the patient [2-3].

Connecting with the Patient

Most likely this is the first time Dr. Jones has interacted one-on-one with Mr. Johnson, so they do not really know each other. Taking a moment to understand who his patient is by asking questions about his life outside of the hospital shows interest and helps develop trust. Understanding more about the patient’s personal background can also provide insight into care preferences. Dr. Jones should pay attention to his patient’s affect and make an effort to empathize with his fear, sadness or distress.

Introducing the Subject of CPR

It is often most effective to enter into discussions of this nature by using the technique of “ask-tell-ask.” Dr. Jones can first ask the patient to describe his understanding of his medical situation. If the patient does not fully comprehend the situation, Dr. Jones can tell him his understanding of the illness and attempt to correct any misperceptions. This may be followed by another ask to clarify if the patient now understands correctly. If Mr. Johnson does have a correct understanding of his illness, then Dr. Jones can repeat the information back to him, assuring him that he has grasped the situation. In this case, Dr. Jones has the luxury of providing appropriate reassurance to Mr. Johnson by reinforcing that Mr. Johnson’s recovery is going well and that he expects him to continue on that course.

After discussing Mr. Johnson’s current medical situation, Dr. Jones may wish to ask him whether he has ever thought about what would happen if he didn’t get better or if some future illness were to take a turn for the worse. Again, the patient can be reassured that this is something that the doctor discusses with all of his patients. Patients who have had loved ones in similar situations, or who have themselves been in an ICU, may have very clear ideas on the subject. If this question does not elicit the information needed, Dr. Jones must get more specific. He might say, for example, “Have you ever thought of what kind of care you would want if you got so sick that you had to be in the intensive care unit on a ventilator or life support to stay alive?” The patient’s answer to this question is less important than his answer to the follow-up question: “Can you tell me why you feel that way?” These answers provide insight to the values and reasoning underlying the stated preference and offer a foundation for insight into what Mr. Johnson may want in other situations. Finally, the most useful information Dr. Jones can obtain in this setting is Mr. Johnson’s choice of a health care surrogate. Dr. Jones should encourage Mr. Johnson to discuss his preferences with his chosen surrogate; without discussion, the surrogate’s assessment of Mr. Johnson’s preferences is unlikely to be accurate [4-6].

General Strategies

Conversations about CPR preferences vary greatly depending on the age, health and health literacy of the patient. With all patients it is important to avoid vague or overly technical terminology and to use vocabulary they understand. The question “Would you want us to do everything?” is not helpful for eliciting preferences; it confuses patients and implies that less than optimal care may be offered if the patient answers “no” to the question. CPR should not always be the standard of care, so it is crucial that patients understand the difference between withholding CPR and withholding treatment of their underlying illness and its associated symptoms.

Dr. Jones should also learn to give his professional opinion regarding the appropriateness of CPR just as he would for other procedures. Physicians are trained to give opinions and recommendations so that patients can make informed decisions. After Dr. Jones takes the time to understand his patient’s general care preferences, he can make a recommendation about the value of CPR in achieving the patient’s goals. The recommendation ought to be based on the expected efficacy of CPR in achieving the patient’s stated values—not the physician’s.

A More Complicated Conversation

Medical residents learn the importance of knowing a patient’s code status early in their training. The term is found on history and physical forms and daily rounding sheets. Unfortunately this simple term ignores the complicated nature of the conversation. A 45-year-old’s opinion on whether he wants CPR is going to depend on his chance for recovery and baseline functional status as well as on his underlying values. Mr. Johnson’s case illustrates the challenge of discussing CPR preferences with members of a young, relatively healthy population. In situations such as this, the patient’s values and goals need to be discussed in the context of present and possible future illnesses to shed light on what his or her preferences may be as the medical situation changes. With practice, Dr. Jones will make conversations about care preferences, rather than code status, a routine part of every new patient admission.


  1. Tulsky JA. Beyond advance directives: importance of communication skills at the end of life. JAMA. 2005;294(3):359-365.
  2. von Gunten CF, Weissman DE. Fast Fact and Concept #023: DNR Orders in the Hospital—Part 1. Available at: Accessed June 14, 2006.

  3. von Gunten CF, Weissman DE. Fast Fact and Concept #024: DNR Orders in the Hospital Setting—Part 2. Available at: Accessed June 14, 2006.

  4. Balaban RB. A physician’s guide to talking about end-of-life care. J Gen Intern Med. 2000;15(3):195-200.
  5. Fischer GS, Arnold RM, Tulsky JA. Talking to the older adult about advance directives. Clin Geriatr Med. 2000;16(2):239-254.
  6. von Gunten FC. Discussing do-not-resuscitate status. J Clin Oncol. 2003;21(9 Suppl):20s-25s.


Virtual Mentor. 2006;8(9):559-563.



The people and events in this case are fictional. Resemblance to real events or to names of people, living or dead, is entirely coincidental. The viewpoints expressed on this site are those of the authors and do not necessarily reflect the views and policies of the AMA.