Intensivists and trainee intensivists of the world, brace yourselves.  You’re about to re-live a  telephone conversation that you have nightly and that makes your blood boil every time….

Ring, Ring…
You: “Hello intensive care, registrar speaking.”
Caller: “Hello, this is the most junior doctor in the emergency department, I have a patient who needs to come to ICU.”
You: “OK, what’s the story?”
Caller: “Mrs. Smith is a 103 year old lady from a nursing home where she’s bed-bound with an MMSE of 1 and is doubly incontinent.  She has metastatic cancer, end stage COPD and all of her joints are prosthetic.  She’s come in with pneumonia.  She has a respiratory rate of 40, sats of 80% on a non-rebreather, pulse of 140, BP of 70/40 and a lactate of 10. Oh, she’s also got a urea of 30 and a potassium of 7.  She needs to come to ICU for ventilation and dialysis. Thanks.”
You: “Hmmm.  This poor lady sounds like she’s dying. I think you should speak to the family and tell them that.”
Caller: “Oh I’ve already spoken to the family.  I asked them if they want us to do everything and they said yes, they want everything done, so you have to take her.”

It’s painful because it’s so close to reality.  The goal of intensive care medicine is to save the lives that can be saved, and simultaneously to recognize when these people can’t be saved, and provide them and their families with a good death.   Intensive care is all about reversibility.  And unfortunately the gradual trajectory towards death that our hypothetical centenarian is on is most certainly not reversible.  Intensive care can prevent people from dying of sepsis, trauma, postoperative complications and myriad other reversible problems.  But we can’t prevent people from dying of cancer, dementia, end stage chronic illness and frailty.

As I see it, there are 2 problems with the “the family wants everything done” scenario.  A communication problem and a recognition problem.  The communication problem is that the poor junior doctor who’s ringing the grumpy intensive care doctor in the middle of night is trying to deal as best they can with a situation that they have probably had no training for whatsoever.  They’ve gone into a little room full of emotional family members who probably know that their loved one is dying but, like any normal person, wish it weren’t so.  When the untrained junior doctor has asked the question “Do you want us to do everything?” they’ve said yes. Who wouldn’t?  However to us everything is ventilation, dialysis, inotropes, CPR.  To them everything is fluids, antibiotics, food, water, pain relief, good nursing care.  I see this time and again.  Families are labelled as being angry and unreasonable when requesting that their loved one have “everything done” and then when asked again, by a more experienced doctor, it becomes clear that the family certainly don’t think that CPR or ICU are appropriate.  They just want their grandmother to have pain relief, dignity and a nurse on hand to answer their questions.  The issue is that the opposite of everything is nothing.  And logically if a family answers no to “do you want everything?” then they must want nothing.  And no family in their right mind would want that.

So how should the question be asked?  Well before we get there, we first need to look at the second part of the problem.  Recognition.  We need to be able to recognize the patient who is coming to the end of their natural life and differentiate that patient from someone who is critically ill with a reversible problem and who may benefit from intensive care.  This is difficult.  It takes experience and involves gestalt.  There’s no scoring system or decision rule.  It’s obviously always better to err on the side of caution when you’re unsure, but often it’s fairly obvious.

The point of recognizing the dying patient is that it changes the meeting with the upset family in the little room significantly.  Here’s why.  In this age of increased patient autonomy and decreased medical paternalism, more and more doctors (at all levels) are unfairly pushing the decision making about end of life issues onto a distraught group of surrogate decision makers.  A one size fits all set of questions have arisen.  I think this somewhat defensive approach of deputizing a difficult medical decision onto a surrogate decision maker has arisen from a lack of knowledge and a fear or of potential legal consequences.  So I think it’s appropriate to highlight some of the debate that’s occurring in this field from  ethical and legal scholars.

Barbara Hayes is an ethicist from the University of Melbourne.  She published an amazing paper in the Internal Medicine Journal in January in which she put forward a model to assist clinicians with decision making around providing CPR(1).  To devise the model she interviewed 33 doctors and nurses of varying seniority.  Their views seem to encompass many of the issues that are commonly raised when this thorny topic is discussed;

  • Who’s decision is it to perform CPR? Is it the doctor’s or the patient’s? (or family’s?)
  • Why can’t we just treat CPR like any other treatment – with indications and contra-indications.  We don’t take someone’s appendix out if it isn’t indicated, why do we have to do CPR when it isn’t indicated?
  • What do you do when families have strong religious (or other) beliefs?

Dr. Hayes then uses the themes that arise in these interviews to construct a framework for ethically sound CPR decision making. To do this she splits patients into 4 categories;

  • The obviously dying patient
  • The medically unwell patient who is not imminently dying but who would not survive CPR
  • The patient expected to have a poor outcome from CPR
  • The patient for whom the outcome from CPR is uncertain

She then constructs a simple pathway to guide decision making asking 2 simple questions;

  • Is the patient likely to survive CPR?
  • If so, are they likely to have a good outcome?

It’s outlined in a simple model in the article which is a must reasd. I really like this model.  It’s simple and gives guidance to all practitioners.  Most importantly, it emphasizes that for the first two patient groups, its OK just to tell the family that CPR will not be offered, and then tell them why.  No more silly questions asked incorrectly.

Blinderman et al put forward a similar framework in JAMA last year(2), in which they state that offering CPR to all-comers is an “ethically unjustifiable practice.”  They use a similar categorization of patients, but with only 3 groups;

  • Those in whom CPR is a plausible option
  • Those in whom we should recommend against CPR
  • Those in whom CPR will not be offered.

The authors re-iterate that in the case of the patient who is obviously dying of an irreversible process, there is no option.  Not offering CPR is the ethically correct thing to do.

So if you use these frameworks, the discussion is now much easier to have.  If CPR isn’t going to work, you tell the family that and then tell them about all the amazing palliative care you’re going do.  Tell them how their loved one is going to get pain relief, hydration, possibly antibiotics, expert nursing care and time to spend with family and friends. If you’re pretty sure that CPR won’t work or, even worse, will work but will leave the patient in worse position than where they started from. Tell the family that.  Tell that that if it was you or your loved one you wouldn’t have CPR.  But again, tell them that all other treatment will continue.  If CPR is a plausible option but the situation isn’t entirely clear you need to take a history.  You need to sit down with the family and take a detailed social history about the patient. Not all 85 year olds are created equal You may even find out what the patient’s wishes are, thereby trumping anything the family says.

What’s happening now is that instead of asking, you’re telling.  You’re using your knowledge and experience to guide a group of people through one of the most stressful experiences of their life.  You’re giving them the opportunity to have their voice heard, while reinforcing that CPR is a medical treatment and the decision to do it or not is made by the doctor at the end of the bed.  Most importantly, you’re not offering them the so called “Chinese restaurant menu” of options.

Some may criticize this approach for not adhering to one of the key ethical principles of modern medicine – autonomy.  They will say that the days of the paternalistic doctor are long gone, and that’s a good thing.  But I wonder if there’s still a role for paternalism in medicine.  Doctors and nurses have knowledge and experience that patients and families don’t. I think it’s ethically dubious not to tell families what we believe should be done (or not done.)  A recent article in the NEJM refers to this as “Freedom from the tyranny of choice.”(3) which I think is very apt.  Sometimes we think we’re helping people by giving them options, but we’re not.

OK, so we now know that it’s ethically OK to not offer CPR to obviously dying patients.  What about the law? An interesting pair of  (open access) articles in the MJA last year looked at the legality of surrogate decision making and of doctors unilaterally making a patient not for resuscitation(4)(5).  The articles look at the legal aspects of a coroner’s court decision in the Australian state of Queensland where the coroner criticized emergency department staff for initiating a not for resuscitation order on an 82 year old woman with respiratory failure secondary to pulmonary fibrosis.  While the coroner didn’t disagree with emergency physician’s management of the patient, or indeed making her NFR, his criticism was that he interpreted Queensland guardianship legislation as requiring consent for surrogate decision makers before any treatment (including CPR) is withheld.

The 2 articles take both sides of the coroner’s argument.  One argues that this interpretation of the legislation puts medical staff in a very awkward ethical position.  Getting consent to withhold CPR implies that there is choice, when in this case clearly there was not.   The counter argument (from a professor of Law) makes some interesting points.  His argument is perhaps summed up by one of his opening statements, in which he describes futility as “a subjective notion masquerading as a form of professional, objective and scientific assessment.”  He argues that the views of the surrogate decision makers be taken into account before any treatment is withheld or withdrawn.  I’m not sure I agree with all of his arguments, but he does make a good point that most healthcare professionals are blissfully ignorant of health law in their states and jurisdictions.

I think the central issue here is good communication with patients and their families.  I think that the decision to offer CPR or not is a medical one but that patients and their families must be informed that this decision has been made.  They can be given opportunities to contribute to the discussion, but they must be aware of the fact that the decision is a medical one.  If they strongly disagree with the decision they must have a route of appeal open to them.

Talking about dying and end-of-life decision making should be thought of as a procedure like any other (except that it’s probably the hardest one we do.)  We get fixated on intubation and sticking in central lines.  We all know (roughly) how many tracheostomies we’ve done.  But we don’t have prolonged arguments on Twitter about how best to have an end of life conversation.  We don’t keep a logbook of how many difficult conversations we’ve had.  Looking after a dying patient is meant to be a core skill for every doctor, nurse and paramedic, but most of us get taught the 5 stages of grieving by someone who’s never told a family that their beloved matriarch is going to die and it’s left at that.

So what’s the best way forward?  I’m keen to know what formal training, if any, people have had in this area?  I’m very keen to argue with people who disagree with the ethical or legal arguments I’ve put forward. But most of all I want end of life communication to be done better so I can stop getting that phone call in the middle of the night.

1. Hayes B. Clinical model for ethical cardiopulmonary resuscitation decision-making. Intern Med J. 2013 Jan. 17;43(1):77–83.

2. Blinderman CD, Krakauer EL, Solomon MZ. Time to revise the approach to determining cardiopulmonary resuscitation status. JAMA: The Journal of the American Medical Association. 2012 Mar. 7;307(9):917–918.

3. Lamas D, Rosenbaum L. Freedom from the tyranny of choice–teaching the end-of-life conversation. N Engl J Med. 2012 May 3;366(18):1655–1657.

4. Lawrence S, Willmott L, Milligan E, Winch S, White B, Parker M. Autonomy versus futility? Barriers to good clinical practice in end-of-life care: a Queensland case. Med. J. Aust. 2012 Apr. 2;196(6):404–405.

5. Stewart CL. A defence of the requirement to seek consent to withhold and withdraw futile treatments. Med. J. Aust. 2012 Apr. 2;196(6):406–408.


This post represents my perspective on an area of medicine that is still open to some ethical and legal debate.  These words are meant to encourage thought and debate but are not intended to be used to make decisions in the management of patients.  Local laws and policies should be adhered to at all times.

8 thoughts on “It’s ethically, morally and legally OK to not do CPR (sometimes.)

  1. Thank you David for a very well articulated commentary.
    Broadly I couldn’t agree more and felt my own frustrations mirrored in your words. It is an area I feel could be managed so much better in a large percentage of cases.
    I think one of the biggest “communication problems” lies in the lack of discussion about end of life care by primary health carers and their specialist teams prior to the patient reaching a terminal phase of the illness. Even if this is discussed, it is difficult to access this documented information after business hours.
    So yes, we emergency Doc’s could certainly approach this situation differently as described above in the emergency situation. However there are certain situations where a decision about whether to offer or not offer CPR (and other invasive therapies) as a “treatment” could be made and communicated well before the time that action is imminently required.
    I think as a community there is room to move in many areas about how we manage end of life care and the discussions we have with patients and their families including the time and place those discussions take place.

  2. I like to think I have not made one of those phone calls requesting an admission…

    But I don’t think phone calls to ICU are always useless, especially when the question of futility is grey, or when the discussion with families becomes difficult. e.g. If the family has not yet been primed for this discussion as per Bec’s comment above (and how often do we seen this in the patient with advanced malignancy or COPD!). If a family/partner is having a hard time coming to terms with the limitations of modern medicine, it may be helpful to have representatives from other specialties supporting your decision not to offer a particular treatment option.

    I have had a NFR/not for cardioversion order I made in ED be revoked by an inpatient team. The patient made it to CCU before then being made palliative. It made me realise that good communication amongst specialties as well as families is also important. I can’t imaging how confusing it must have been for the family.

    The relatively new acute resuscitation plan (ARP) forms in Queensland are good in that they prompt discussion with families and documentation of this. However they are confusing if there is no NOK etc.

    Great post highlighting also some of the grey areas in Australian law regarding end of life, thanks.

    1. Thanks Kath. Please don’t get me wrong. I think phone calls from colleagues are great. Medicine is a team sport and especially in the ethically, legally and emotionally fraught world of end of life decision making, a unified approach is ideal. I used to work in a hospital where it was the norm for ICU and ED (and other specialties) to sit down together with families and give a medical consensus view. Unfortunately medicine is becoming increasingly silo-ised and this doesn’t seem to happen much any more. What isn’t so good is to get phone call from a colleague who has already made their mind up and may or may not have made the right decision.

      Shame about your NFR order being revoked. I think it’s important to be aware of the fact, however, that such orders can change and that patients and families are allowed to change their mind (of course the final decision still lies with the medical staff.) I suspect your case highlights the fact that medical and surgical teams often have a long term relationship with a patient and need a bit of time to gain enough evidence for themselves that the patient is not responding to curative measures.

      There’s a great paper on this by Hinton et al in this month’s Critical Care which I’ll talk more about soon.


  3. Eloquently written, David.
    “Our purpose in Intensive Care is not to prolong life, but to restore health” – MO’L

  4. Very well-needed discussion that’s only beginning. I’m a resident and wish we had more education about death and dying. In most medical schools, end of life care is an hour to 2 hour lecture, or there is a 2 week hospice and palliative care elective rotation some choose to take in med school or in residency. There is not a standardized approach to teaching this topic. However, with the AAFP, we’ve started that process of standardizing end of life care education – check out – an award winning point of care tool for death and dying discussions. Cheers.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s