Posted on December 4, 2020 at 9:55 AM
by Dominic Wilkinson @Neonatalethics
This week, the Care Quality Commission (CQC) published an interim report into resuscitation decisions during the COVID-19 pandemic. According to a number of media outlets, the report found that in the first wave of the crisis inappropriate and possibly unlawful ‘do not resuscitate’ orders were used “without the consent of patients and families” (see eg Telegraph, Sky).
There are real concerns and important questions to answer about policies and care for patients in care homes and in the community during the pandemic. However, the media stories, and the CQC report itself appear to illustrate two ethical misconceptions.
The consent misconception:
A number of the media reports perpetuate a common misunderstanding about ‘do not resuscitate’ orders. They imply, or directly state that lack of consent to DNACPR (Do Not Attempt Cardio-Pulmonary Resuscitation) would make such a decision ethically problematic.
But “Do not resuscitate” orders do not ever require patient or family consent.
Cardiopulmonary resuscitation (CPR) is a form of medical treatment provided to patients if their heart or breathing stops. It is routine to commence CPR if a patient suddenly collapses. However, contrary to the impression sometimes conveyed by television programs, the success rate of cardiopulmonary resuscitation is usually low. Only one in ten patients who have a cardiac arrest outside hospital and who receive CPR survive. The chance of survival for patients in residential care or a nursing home is one in 100 or less (eg see here, slightly higher at 2-3 out of 100 in Japan).
Any decision to provide a treatment to a patient requires the informed consent of the patient. If the patient can’t consent, then treatment may be provided on the basis of their prior wishes (where known) and ‘best interests’.
Decisions not to provide a medical treatment do not require the consent of a patient. There is no ethical requirement for doctors or other health professionals to seek the agreement of a patient not to offer or provide a treatment that they (the professional) think would not be appropriate.
Of course, that doesn’t mean that doctors shouldn’t find out the patient’s wishes about CPR. The patient may not wish for CPR (in which case it certainly shouldn’t be provided). Alternatively, the patient may have priorities or values that are relevant to the decision. For example, the patient’s views about what quality of life they would judge to be acceptable, or what they would be willing to undergo for the sake of a small chance of benefit, are relevant. A doctor should take these into account in thinking about whether CPR would be in the patient’s best interests. But the fact that the patient (or their family) would want CPR to be provided does not mean that it must be given.*
To be fair, the CQC report itself does not appear to be guilty of the consent misconception.** Its main criticism was that such decisions were made without involving patients or their representatives.
The best interests misconception
But the CQC report does appear to fall prey to a different misunderstanding. In several places it states that DNACPR is a “best-interests” decision. In other words, in the CQC view, the decision about providing CPR, or not, should be based on an ethical assessment of what would be best for the patient.
In many situations, decisions about medical treatment at the end of life are entirely focussed on what would be best for the patient. That is why it is important to know the patient’s wishes and to take them into account.
But there is an important separate reason for not providing medical treatment. In some cases, providing the treatment would be harmful to others. Then the treatment cannot be provided even if it would be best for the patient.
It is this second type of reason that was potentially at the front of health professionals’ minds back in March and April at the start of the pandemic.
We may have difficulty, now, remembering back to a time when the first wave of COVID-19 was starting to surge in the UK and we were all fresh into the national lockdown.
But at that point in time there were two very real concerns about providing cardio-pulmonary resuscitation to some patients in the community.
Firstly, there was a concern that with rapidly rising cases of severe COVID-19, there was intense pressure on intensive care units. If the situation in the UK ended up being similar to Northern Italy, there would not be enough ventilators and intensive care beds for patients with COVID-19. In that setting, it would be ethically inappropriate to admit to intensive care a patient who had had an out of hospital cardiac arrest and had a very low chance of survival. That decision would be based on resource allocation, not on ‘best interests’. However, if a patient is not able to be admitted to intensive care, it may then be pointless to embark on CPR if the patient’s heart or breathing were to stop.
Second, there were considerable (and justifiable) anxieties among health professionals about their personal risks of contracting and becoming seriously ill from coronavirus. Providing CPR to a patient whose heart has stopped (blowing air into their lungs, compressing their chests) is high risk in terms of spreading virus particles to anyone in the immediate environment. In order to safely provide resuscitation, health professionals would have put on a full set of personal protective equipment. This would likely mean a delay before any resuscitation could be started, and potentially make the chance of any benefit for the patient even smaller.
These two factors are ethically relevant, and might have provided a justification for a DNACPR decision at the peak of the first wave of the pandemic. They appear to have affected official policy relating to resuscitation in other parts of the world. For example, in Montreal, between April and August, paramedics were told not to provide CPR for any patient with asystole (a “flatline”). In normal circumstances, the paramedics would have provided or continued CPR until the patient arrived in hospital (although the chance of recovery for a patient with asystolic arrest is extremely low). But during the pandemic, the ambulance service protocol was modified “to protect the paramedic, to protect the health system”. They returned to a normal approach in August once the numbers of cases had fallen and there no longer seemed to be a high risk for paramedics.
There are three important things to recognise about these non-best interests factors relevant to DNACPR decisions.
First, if these factors were the basis for any decisions about resuscitation at the start of the pandemic, they should have been made clear to patients and to the wider community. They would not have been an excuse for decisions made in secret, behind patients’ backs. They should have been a prompt for and part of a conversation with patients and families about what was possible, and what was not possible, what might be an option, and what would not be, in the event of deterioration. (There should have been, though there was not, an honest discussion by politicians with the community about the limitations of the public healthcare system)
Secondly, related to the above: if a decision not to provide CPR was because of the risk to health providers and/or the non-availability of intensive care, that would have no implications for other forms of medical treatment. The CQC report raises legitimate concerns that health professionals sometimes misinterpreted a DNACPR order as meaning that a patient should not receive other forms of treatment (for example being admitted to hospital). But that is not the case. There are many medical treatments that may be helpful, appropriate and available, even if CPR is not. For patients who are approaching the end of life, or who might be facing a life threatening illness, it is important to talk about CPR. But that should be only a small part of a wider conversation about the person’s wishes and priorities.
Thirdly, if those factors were relevant to DNACPR decisions back in March or April, they may no longer apply. We are in a different phase of the pandemic. There remain large numbers of cases, but the concerns about intensive care capacity are less acute than they were 8 months ago. Furthermore, the greater understanding of ways to protect health and care professionals, greater availability of personal protective equipment, and familiarity with its use may mean that the risks to professionals are less of a concern. The issues have not gone away, but they might mean that a different approach is appropriate. It would be important to revisit decisions made back at the start of the year.
It is vital to learn from our mistakes. The CQC investigation and report is important. There appear to have been serious shortcomings in decision-making and in communication in some cases. There are important lessons to learn for the future, not least about the importance of clear honest communication about medical treatment towards the end of life and what may or may not help, what may do more harm than good.
But when we are thinking about the rights and wrongs of resuscitation decisions it is important to avoid misunderstandings and oversimplification. There is more to DNACPR than ‘consent’ and ‘best interests’.
* In this very helpful piece for Medium, Tor Butler Cole clarifies the legal position in the UK on DNACPR here.
**On page 8 of the report, it does potentially imply this: “Sometimes the family were told that the person had agreed to the DNACPR, but there were questions raised about whether this was informed consent”