Patients deserve more than a DNR form

The use of DNR forms during the COVID-19 pandemic has revealed underlying problems with the way patients and the healthcare workers who care for them are treated by an underfunded system, argues junior doctor Stacey Williams.

Art by Rebecca Vine via Activists at Work.

I’m writing this article to respond to a number of news reports and social media posts discussing DNR (Do Not Resuscitate) orders, or as I refer to throughout, DNACPR (Do Not Attempt Cardiopulmonary Resuscitation) forms. There have been multiple news reports of DNACPR forms being sent in the post, or en masse. For some people, their doctors have discussed DNACPR as a form of resource allocation, or written the form solely because they have a disability. This has understandably caused outrage, and as a doctor, I’ve been upset to see friends treated with such little respect.

I’m angry about working in a system that focuses on the form, and not the conversation: I don’t want to be the junior doctor who in a single shift needs to again and again start DNACPR discussions with people I have not met before, who I have had no chance to build a rapport with, and then need to break to them that they may die during their hospital stay. At the same time, I feel a responsibility to explain why DNACPR forms exist, what they represent, but to also examine why they have become so toxic, and not to disregard this as just ‘hysteria’ and misunderstanding.

In this time of COVID-19, DNACPR forms have come to replace the meaningful conversation about what death looks like in hospital, and how serious a disease COVID-19 is. Amidst pre-existing systemic underfunding and a staffing crisis, but also a historically hierarchical healthcare system, what is the substance of respect we deserve, and how do we develop a collective struggle to demand it?

What are DNACPR forms?

DNACPR stands for Do Not Attempt Cardiopulmonary Resuscitation. It is a piece of paper which documents a person’s identifiable information, their medical conditions, and whether they are able to make and communicate decisions for themselves, or if this needs to be discussed with their next of kin. It lists this person’s main medical problems and reasons why one specific medical intervention, Cardiopulmonary Resuscitation (CPR) would be ‘inappropriate, unsuccessful or not in that patient’s best interests’. It also describes how this information has been discussed with the person, their next or kin, or in some cases legal representatives. In rare cases, even a discussion of CPR might be thought to cause harm, and in these exceptional circumstances it can be signed without discussion.

DNACPR forms document a patient is not for one specific medical intervention: CPR. This is performed when a person’s heart has already stopped. It requires chest compressions, electrical shocks and mechanical ventilation, where a patient has a breathing tube is inserted for a machine to take over the work of breathing. It means treating everything else, and most of the time early recognition, intervention and treatment stops short of requiring CPR.

DNACPR forms are only signed by a doctor, and not by a patient or their next of kin. However, the form requires a discussion, which should involve a description of what CPR is and why, from a medical perspective, it would be inappropriate, unsuccessful or not in that patient’s best interests. It should offer patient’s the choice of whether to decline CPR, but a patient cannot demand CPR. Once the form has been signed, the person is given a copy to keep at home called a ‘community DNACPR’ form. This makes it clear to any first responders to an emergency, like paramedics, not to start chest compressions if the person’s heart has stopped.

DNACPR does not mean ‘do not treat.’ The supportive treatment possibilities short of intubation (for instance oxygen, intravenous fluids, antibiotics), are all potentially available to patients with a DNACPR form in place. DNACPR does have a real meaning for healthcare workers though: if it is the medical opinion that a person will not benefit from CPR or ventilation, they must have a DNACPR in place, otherwise if their heart stops healthcare workers have a responsibility to start chest compressions. The healthcare workers would have to break the ribs, put needles into and give electrical shocks to people who have already died, and would be unlikely to be revived despite the indignity of these aggressive interventions. No doctor or nurse wants to do this needlessly.

There are some ambiguities in DNACPR forms that do not help in their interpretation. There are a number of templates and variations in the amount of detail included. In all places I have worked, the document is called a Do Not Attempt Cardio-Pulmonary Resuscitation, rather than Do Not Attempt Resuscitation. That tries to make it less ambiguous, that the discussion and its documentation does not mean ‘no treatment’.

Some hospital Trusts have adopted a more nuanced version of these forms, for instance the ReSPECT form, which not only details a discussion about CPR, but also the level of intervention in other medical treatments, such as dialysis, which do not always need an Intensive Care Unit (ICU). Patients may have already made legally standing Advanced Decisions about intervention, where they can deny, but not demand, medical treatments. For older, frailer patients, Advanced Care Plans are sometimes used to document more in depth discussions of reasons that person might come to hospital, for instance after a fall to check for a fracture or for pain control, or when instead to avoid a hospital admission and be treated in the community, like for a chest infection.

Discussions about dying and death, and what medicine can and cannot do to prevent it, should be respected as a right. This is now reflected across medical education and formalised as an expectation of doctors by the General Medical Council.

In the case Tracey v Cambridge University Hospitals NHS Foundation Trust 2014, Lord Ryder judged that: ‘[the] fact that the clinician considers that CPR will not work means that the patient cannot require him to provide [CPR]. It does not, however, mean that the patient is not entitled to know that the clinical decision has been taken.’ In not discussing CPR with Janet Tracey, and in writing a DNACPR form without adequate evidence of discussion with her or with her family, the hospital Trust was found to have denied Janet Tracey Article 8 of the European Convention on Human Rights, the ‘right to respect for [her] private and family life’.[1]

The conclusions from Tracey v Cambridge did change clinical practice. Doctors are required to ensure that patients ‘know that the clinical decision has been taken’: this is represented by a doctor filing a DNACPR form (often the same form as before the Tracey v Cambridge judgement) in a patient’s record, and a doctor telling them about it. In the National Health Service, medical interventions aren’t commodities provided to consumers, but a public good provided on assessment of likely benefit over possible harm. However, the form often becomes a token in place of a meaningful discussion around death, dying and the extent of what modern medicine can and cannot achieve.

Why are DNACPR forms being written now?

I am a doctor at a cancer hospital. I am not on the front line of the Emergency Department, or in the ICU, but I work shifts looking after COVID-19 patients both with and without cancer. I am seeing large numbers of cancer patients and patients who have been previously coping well with underlying medical conditions, who for the first time are being told about CPR and having DNACPR forms discussed with them. For many of these patients, it is only the COVID-19 crisis that has precipitated the forms. It feels to people like they are being given up on. It’s incredibly distressing for them, and also for the healthcare staff who do their best to convince them otherwise.

Why are these forms only being mentioned now? One reason relates to the biology of COVID-19. It is unlike most medical conditions we treat in hospital: doctors know, and openly acknowledge, that there is no cure for the infection currently. Hospital treatment can only support the organs of patients with the infection until their own body finds a cure: their immune system produces effective antibodies that destroy the virus and infected cells.

Patients with cancer, especially those who are receiving chemotherapy, older people or those with underlying health conditions such as diabetes have weaker immune systems, described as immunocompromised. There is growing evidence to suggest that these people develop more aggressive infections, causing difficulty in oxygenating the blood through the lungs and the body’s organs to shut down, and more likely to need hospital admission.

When a person comes to the Emergency Department with symptoms of COVID-19, if a discussion about CPR and intubation hasn’t already happened, doctors assess how bad the infection is for this patient, discuss what support options they feel will be beneficial, and when doing this right, listen to the patient about what their thoughts are about those options. These could include oxygen through a mask or nasal prongs, additional help with a machine that blows air into a person’s lungs while they are awake (called Non Invasive Ventilation, or NIV), intravenous fluids and additional antibiotics. It might also be to take part in a clinical trial. Up until here, these are treatments that can be provided even with a DNACPR form in place.

For some COVID-19 patients, for whom these all-but-ICU measures do not improve their condition, or who have features suggestive of aggressive infection, they may be taken to ICU to support them while their body mounts an immune response, generating the only treatment available now: antibodies. There is national guidance about how to assess who is likely to benefit from ICU care and who is not. There has been a lot of controversy about these assessment tools, and all guidance requires them to be interpreted, to use judgement and not just apply the numbers coldly. The numbers are stark, however: 68% mortality for patients in ICU over 70. Anecdotal reports from one ICU suggests that survival in their department is only 35%, for all ages. If the medical opinion is that ICU care is felt to not benefit a person, or likely to cause more harm, then a DNACPR form has to be written.

In intensive care, a person will be sedated and have a breathing tube inserted to allow a machine to control their breathing. They will have a plastic tube attached at their neck to administer medications. They will have a pressure monitor attached to the radial artery in their wrist to monitor blood pressure. They may be turned on their front to help increase oxygen levels. This can be for a number of days, averaging between 5 and 7 days, all spent in an ICU department away from loved ones, unable to communicate fully with the array of shift staff who, if lucky, have full PPE which obscures their faces and voices with masks and visors.

Each day in ICU is like running a marathon: the adrenaline, the sedation, the mechanical stress on a body is physically exhausting. It can also be very lonely. While it can be relatively easy to start this process in any patient, the more frail an individual, the harder it can be to take away the ventilator and the patient survive, either in the short term from being off the breathing machine, or survive to leave hospital. And for patients with COVID-19 who are so unwell their heart stops, early evidence suggests CPR is not very effective. In one hospital in Wuhan, of the 136 patients recorded who received CPR, four (2.9%) survived for at least 30 days, and one patient achieved a ‘favourable neurological outcome’, i.e. went back to some normality for them, after 30 days.[2] When asked about how and where someone would want to die, no one says an ICU. For patients whose frailty, or underlying medical conditions make it very unlikely that they will survive even with support in ICU, it may instead deny them the ability to spend their last days as they would want, at home or with friends and family.

When do we want to talk about death?

The second reason that these forms are being filled in now is that the conversation about what happens when someone dies, particularly in hospital, hasn’t happened already. That is obvious on one level, but for many who are now facing this conversation, their existing health conditions, prior to COVID-19, would have made it unlikely that CPR would be successful, that it was unlikely they would have been revived after their heart had stopped. It was just that they hadn’t been told that before now.

This is in part due to the fact that doctors find it hard to discuss death with their patients, particularly when it is not imminent. ‘I learnt about a lot of things at medical school’ writes surgeon Atul Gawande in Being Mortal, ‘but mortality wasn’t one of them.’[3] Gawande’s book, prompted by a reflection on his own father’s mortality, argues that that the dominant medical philosophy sees death as failure, making many doctors incapable of having meaningful discussions about death. This highlights a broader problem that public anger over the DNACPR forms has brought into the foreground. I am meeting patients with advanced conditions on the final, or beyond all available therapies, being told for the first time that they would not survive an admission to ICU. But why wasn’t this conversation started before? For these patients, their condition has already made them aware of their mortality; they wake up, they eat, they lie sleeplessly because of it. But why are many doctors incapable of talking directly about death, apart from when death is obvious and imminent? We must acknowledge that all people die, and that in some cases that will because of a cause that we can’t treat, either effectively or without causing more harm. What does a natural death mean? It is certainly not CPR.

There are structural reasons why these conversations are uneasy. These are nuanced, personalised decisions. They involve a medical opinion, which is formed from the privilege of a university education, a certain class position, access to medical knowledge of the condition or the system, often inaccessible or opaque to non-healthcare workers. The idea of your life being in someone else’s hands is uneasy, even more so when you feel unequipped to challenge that person. Doctors have a professional duty to prevent harm, and act with integrity in what they believe to be a person’s best wishes. They should not offer treatments that they do not think bring any benefit. But ultimately a ‘best interests’ decision is a decision being made is about another person’s life and death, it should champion that person’s wishes and respect their ability to make ‘unwise’ decisions if about themselves. These medical decisions are often laden with value judgements about what makes a life worth living and, without care, do little to challenge ableism, seeing ‘non-disability’ as the standard of a ‘normal’ life. In the last weeks, there has been a marked increase in DNACPR forms being applied to autistic people and those with learning disabilities, where there is no clinical basis for the decision.

There are also the material implications of funding cuts to the NHS and Primary Care. When do we have the time to talk about death? In a seven-minute consultation, where you can talk about one problem only? And do you want this conversation to happen with a doctor you know very little, the local Practice having merged with several other local Practices into a conglomerate where you never see the same GP twice? This is a process of alienation for both the person sat in front of the doctor (when does anyone have just one, clearly articulated physical or mental health need?), but also for the doctor. ‘We really do work with people,’ wrote the anti-colonial revolutionary Frantz Fanon of his work as a psychiatrist. ‘So if people are what constitutes the goal […] of our daily action, then it becomes clear […] that no dose of habituation or automatism can intervene’.[4] The lack of time for building and sustaining relationships, between doctors and the people who are only sometimes their patients, denies doctors that goal.

Secondary care, outpatient appointments with specialists, have faced the same issue. We are in the midst of a staffing crisis. There were over forty thousand nursing vacancies at the end of 2019. In cancer services, Cancer Research UK predicts that the numbers of oncologists needs to triple over the next seven years. All the while, there is no meaningful strategy to retain and recruit nurses and doctors to fill the gap; student nurses still pay to work. Overbooked clinics without adequate trained staff to have conversations about DNACPR mean the discussion is never started. You might see your doctor for a number of years, go through the pains of disease progression and set backs, but not talk about the possible event of dying, or DNACPR, because that would be a failure, and there would be too little time to address it as anything else. In not discussing DNACPR in those clinics or those GP practices over the past years, the forms are now discussed only at the door to the hospital by doctors who patients have never met before, or worse being told without time to discuss, or worst of all, just as a letter in the post. What respect is there in this?

Are DNACPR forms a way of triaging patients?

The third reason DNACPR forms are being filled in now is because the government has messed up on COVID-19, and has systematically underfunded the NHS for decades. There aren’t enough doctors and nurses, there aren’t enough beds and there aren’t enough ventilators. Even after 2016’s Exercise Cygnus – a nationwide simulation of the likely effects of a viral pandemic on the UK health system – the government did not increase ventilator capacity or PPE reserves. Based on this model of a less contagious and less deadly virus, the government continued with the strategy of developing ‘herd immunity’. This strategy only changed once the Imperial College COVID-19 model suggested the UK unmitigated response mortality could be as high as 510,000.[5] This prompted the lockdown we are all enduring, but did not change the strategy of no testing and no contact tracing, leaving the virus to propagate in the population – which looks suspiciously like herd infection.

This means that all patients, with COVID-19 or not, who need hospital admission and ICU will be triaged, as happens in conflict, in resource restriction, and now as deliberate strategy by this government. But this isn’t just happening because of COVID-19. We have 6.6 ICU beds per 100,000 of the population, whereas across Europe the average is 11.5: Italy, for example, has an average of 12.5. The NHS that has seen ‘efficiency savings’ that have put strain on the whole system, with Emergency Department waiting times before March the worst since the records began, and 1 in 5 patients waiting over a year for hospital treatment in the North of Ireland, all before the pandemic. It is not unusual for patients to have operations delayed, or treatment continued on a ward rather than in ICU because there are no beds, or not enough staff to look after patients safely. This is the reality of funding cuts to the NHS. In this time of COVID-19, doctors are left making the difficult decisions about who will benefit most from ICU and ventilation. It isn’t ultimately doctors’ fault that they are planning for this hell, but the government’s.

What is the solution?

People are being left to feel their lives are a price the government is willing to pay for its rogue public health policy, with some going so far as to describe it as a form of euthanasia. No one should be told a DNACPR decision is final without any respect shown them for it to be discussed, in all the ways I have tried to outline in this article.

The DNACPR form is an important documentation of a discussion about the limit of medical treatment, but it alone is not enough. We should demand a healthcare system that values relationships, equality and openness, with the funds and structure that respects time, reduces administrative burden and trains all healthcare workers to continue the important conversation about death and dying that a DNACPR form only starts. There is a limit to what medicine can achieve, and it can cause harm as well as do good. Death is not failure.

The most important conversations about living and dying are just not being had. We should recognise the structural and material inequalities in healthcare and move forward in a shared understanding about the limits of modern medicine and what we want beyond it.



[1] Tracey, R (On the Application Of) v Cambridge University Hospitals NHS Foundation Trust & Ors [2014] EWCA Civ 822 (17 June 2014). [2014] 3 WLR 1054, [2014] WLR(D) 277, [2014] EWCA Civ 822, [2015] QB 543, [2014] Med LR 273, [2015] 1 QB 543, (2014) 138 BMLR 1, 138 BMLR 1. URL:

[2] Shao F et al., In-hospital cardiac arrest outcomes among patients with COVID-19 pneumonia in Wuhan, China, Resuscitation (2020), doi:

[3] Atul Gawande, Being Mortal: Medicine and What Matters in the End (New York: Metropolitan Books, 2014.

[4] Frantz Fanon, Editorials of the weekly ward journal of the Blida-Joinville Psychiatric Hospital, March 1956, reprinted in: Frantz Fanon, Alienation and Freedom, edited by Jean Khalfa and Robert Young, translated by Steve Corcoran (London: Bloomsbury, 2018), pp. 338–9.

[5] Neil M. Ferguson et al., ‘Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand. WHO Collaborating Centre for Infectious Disease Modelling, MRC Centre for Global Infectious Disease Analysis, Imperial College London’, 16 March 2020. URL:


Please enter your comment!
Please enter your name here