During Covid, I was frightened of another breathing admission.
Not frightened in the ordinary way. We had lived with breathing admissions for years. We knew the ward, the route, that particular fear. We walked a line between two risks: the nebulisers put her heart at risk, but stop them and her breathing crashed.
What frightened me then was different.
It was knowing how easily, if she crashed, my daughter could be seen first as a category.
My beautiful, vibrant, so very, very alive daughter, who sometimes struggles to breathe.
If her lungs failed. If her breathing stopped. That one moment could define whether she had the chance to live the rest of her life.
A DNACPR—do not attempt cardiopulmonary resuscitation—is a decision recorded in someone’s notes that, if their heart or breathing stops, cardiopulmonary resuscitation will not be attempted. It does not, by itself, mean that any other treatment should be withheld. It is about one particular intervention in one particular moment: the moment of arrest, and whether anyone will attempt CPR.
Properly made, it is a clinical judgment about whether CPR could work and whether it would benefit the individual person. It is not meant to be a judgment on the worth of their life.
But when that decision is made without the person or those closest to them being heard, the distinction can become frighteningly thin.
Because who gets to decide whether the attempt is worthwhile?
And if the patient is a child, how could the parents who know that child more deeply than anyone else not be at the centre of the conversation?
That does not mean a patient or parent has a veto. Clinicians do not have to provide CPR if they judge that it will not work. But “you do not have the final say” is not the same as “you do not have the right to be heard.”
The courts have been clear about that.
In Tracey in 2014, the Court of Appeal held that there is a presumption that a patient with capacity should be consulted before a DNACPR notice is placed in their records. A doctor may decide not to discuss it if they believe the conversation is likely to cause the patient physical or psychological harm—not simply distress, but harm.
In Winspear in 2015, the High Court applied the principle where the patient could not decide for himself. Where it is practicable and appropriate, the duty is to consult those close to the patient before the decision is recorded, not afterwards.
Current guidance is equally clear: DNACPR decisions must be made individually. They must be free from discrimination. They must not be based on a professional’s subjective view of another person’s quality of life.
And there it is.
Who gets to say which life is worth the attempt to save?
The reason the Winspear judgment exists is that a doctor placed a DNACPR notice on Carl Winspear’s record during the night without speaking to his mother, Elaine, who had cared for him throughout his life. The note said that the family should be spoken to in the morning.
When his mother was told, she disagreed. The notice was cancelled. Carl died that evening.
The court did not say that his mother had a veto. It said she was entitled to be consulted before the decision was placed on his record. Speaking to her after it had already been made did not fulfil that duty.
When I read that, my first thought was: how would I have known?
How would any of us know?
You are sitting at a bedside. It has gone from stable to not stable, so fast. Questions arrive in the middle of alarms, observations, medication and fear. How does a mother take in what a DNACPR means in those moments?
Would I have agreed because I was frightened of causing my daughter more pain?
Would I have had the strength to challenge it?
And would my challenge have been read as emotion—as a mother unable to accept what was happening—and therefore given less weight?
I said that what frightened me was my daughter being seen first as a category. I want to be clear that this was not only a feeling, there is evidence to back it up.
On 20 March 2020, NICE published its rapid guideline on critical care in adults. As originally written, it said that on admission to hospital every adult should be assessed against a nine-point Clinical Frailty Scale, and it drew a clear line between those who scored below five and those who scored five or more. It did not explain what frailty was. The British Geriatrics Society describes frailty as a concept linked to ageing, and says it should be distinguished from disability.
Most people with a learning disability would score five or more on that scale. Disabled people are far more likely to score seven than non-disabled people. The scale measures whether you need help with everyday tasks. It does not measure whether you are dying.
Five days later, on 25 March, after a threatened judicial review and public pressure from disability organisations, NICE amended the guideline. The Clinical Frailty Scale was not to be used for younger people, for people with stable long-term disabilities such as cerebral palsy, or for people with a learning disability or autism. An individualised assessment was to be used instead.
Five days. At the start of a national emergency, when everything was moving quickly and clinicians were reaching for whatever guidance existed. Desperate to find a way to choose between patients, when they didn’t have enough resources to save them all….
The mother who brought that legal challenge said afterwards that while she welcomed the amendment, she remained deeply concerned that the guidance had been issued at all without any consideration of what it would obviously mean for disabled people.
She had no authority over what was published. She had knowledge of what it would do. She had to threaten litigation to be heard—and it was corrected in five days, which tells you how visible the problem was to anyone who looked at it.
During the pandemic, the Department of Health and Social Care asked the Care Quality Commission to examine how these decisions were being made. Its review, published in March 2021, heard about the experiences of more than 750 people.
It found people who had not been properly involved. It heard from people who did not know that a decision about their care had been recorded at all. It also heard evidence of decisions being applied to groups rather than individuals. Blanket decisions of that kind may be discriminatory and unlawful under the Equality Act 2010. Older people and people with a learning disability were among those particularly affected.
But this did not begin with Covid.
In 2016, the Royal College of Physicians published a national audit of end-of-life care in English hospitals. It reviewed the case notes of 9,302 adults who died in participating acute hospitals.
A DNACPR was recorded in 94% of those notes at the time of death. Once sudden deaths were excluded, a senior doctor’s discussion about CPR with the patient was documented in 36% of cases. In 16% of the relevant cases, there was no recorded reason why a discussion had not taken place. A discussion with the person nominated as important to the patient was documented in 81% of cases.
That means that in roughly one in five cases, there was no documented discussion with the person closest to them.
The audit can only tell us what was written down. A conversation may have happened without being recorded, and this was an audit of adults who died in hospital, not children or every patient admitted in crisis. That matters. But so does what it found: documented involvement was far from universal, years before the pandemic began.
So the answer to my question—how would I have known?—is that some people did not.
Not knowing was not necessarily a failure of their vigilance. Sometimes, they were simply not told.
There is another problem beneath that one. We place enormous faith in the record, but a signature saying that something happened is not proof that it happened.
In April 2020, Eliot Harris died at Northgate Hospital in Great Yarmouth. The Trust accepted that there was a culture at the hospital of retrospective recording: observations carried out but not recorded at the time, observations carried out by one person and signed by another, and observations that nobody completed at all but which were signed as completed. It acknowledged that this was encouraged and expected by ward managers, and that recording was sometimes done days later. The inquest jury recorded this, while also finding that it did not cause or more than minimally contribute to Eliot’s death. That matters. The records were false either way.
More than two years after Eliot died, the senior coroner wrote a Prevention of Future Deaths report stating that quality audits carried out since his death showed observations were still not being carried out and recorded in line with the Trust’s own most recent policy.
In October that year, Rowan Thompson died at Prestwich Hospital. An inquest jury found that five of the eight staff on the ward had falsified observation records. On one morning, 24 observations were signed off; nine had been carried out. The senior coroner directed the jury that there was no evidence that this caused or contributed to Rowan’s death. That qualification matters. The signatures were there all the same.
Neither case was about resuscitation. I am not saying that CPR conversations were falsified. I am saying that a signature is not sacred simply because it is in a medical record. A record can say that a thing happened when it did not. It must not be allowed to end the conversation when the people who were there say something different.
The same 2016 audit found that only 4% of patients had a documented advance care plan made before they came into hospital. For most, these questions were arising inside the hospital, under pressure, when time and capacity to take anything in were already disappearing.
I can see how easily, in the moments we lived through, a decision could have been made in the background and, on the turn of a dime, my daughter might not be here.
I want to be precise about what that means. Not that all treatment would have stopped. Not that nobody would have cared for her. But if her heart or breathing had stopped while a DNACPR was in place, nobody would have attempted CPR.
I would have been told to sit with her and say goodbye.
And all that has been her wonderful life between then and now—none of it would have existed. All she has been, all she has done and the all the light she has brought to this world.
That is what lands in my body. As you are fighting for your child’s life, you discover that someone else has already decided what will happen if their heart or breathing stops. It would take those last moments from you and fill them with a helplessness almost impossible to describe: being unable to act while your child’s life ended, your brain trying to understand what is happening while you are still trying to be there for them.
I have come to call the nervous-system imprint this kind of responsibility leaves behind Post-Responsibility Trauma—PRT.
You are responsible for keeping someone alive. The authority to act sits elsewhere.
You are expected to notice every change. To remember every medication. To hear the difference in their breathing from another room. You are expected to stay calm enough to be taken seriously while the person you love is slipping away.
And if you fight too hard, would that fight be read as emotion? As a mother not coping? Would the very response created by the danger be used as the reason not to listen to you?
In my work, I have sat with people for years afterwards, still holding questions no one will answer.
What was decided? When? By whom? Were they told? Would the person they loved have died in the same way before Covid? If they had known, could they have changed what happened?
They may never know. That uncertainty is part of what they carry.
This did not end when the pandemic ended. It is still being carried.
After the CQC report, a Ministerial Oversight Group was established to oversee the promised changes. The Parliamentary and Health Service Ombudsman has since pointed to an accountability gap: the group has not met since May 2022, and the organisations responsible for implementing the recommendations are no longer reporting their progress through it.
The oversight group has not met since 2022.
That should trouble all of us.
The final LeDeR national report, published in July 2026, found that the proportion of reviewed deaths of adults with a learning disability where a DNACPR had been recorded rose significantly between 2021 and 2024, from 72.2% to 77.4%. The report found the orders had been completed correctly in around two-thirds of cases. It also found that people with severe or profound learning disabilities, and people with Down syndrome, were more likely to have one recorded than others: 83.6% of people with Down syndrome, against 71.7% of other people with a learning disability.
The report’s chief investigator said it was difficult to say whether the increase was appropriate. Discussing resuscitation is good practice where someone has a life-limiting condition and may die soon. But, as the pandemic showed, it can also be applied to younger people with significant disabilities who have no terminal condition at all.
A rise on its own tells us nothing. A rise that falls most heavily on the people most likely to be judged by their disability rather than their prognosis, alongside an accountability gap and a documented history of poor consultation, is a reason for scrutiny rather than reassurance.
It is awful regardless of whose child, parent, partner, sibling or friend it is. And it lands differently—more viscerally—when it is your own. I know that about myself, and I am not going to pretend otherwise. That the true horror of this lands fully when its your child, your loved on that this might happen to.
So here is where it lands for me.
The record is treated as the fact. The people who were actually in the room are treated as the interpretation.
Yet the people in the room are often the ones who know.
The mother who has listened to this child breathe for eighteen years and can hear the difference from another room. The father who remembers which medication made things worse last time. The sister who knows what that particular sound means. The person themselves, communicating pain or fear in a way that may not look familiar to someone who met them that morning.
Families hold years of knowledge about a person. Too often, that knowledge is renamed anxiety. It is called being too close to be objective. It is weighed against a set of notes written by someone who arrived at the start of the shift.
Why don’t we listen to parents?
Why?
Why?
Why?
Why do we ask parents to notice every change, every medication, every breath—and then call them emotional when they tell us something is wrong?
Why is the person who knows this child best so often the first person whose knowledge is explained away?
Why does authority count for more than knowledge, even when the person with the authority only arrived that morning?
A family may not have a veto. I understand that. But consultation after the decision is not consultation. Being informed is not being heard. A name in a box is not the same as a conversation.
The people who know are too often the people we listen to least.
Because sometimes there is one moment on which the whole rest of a life turns.
No parent should discover, after that moment has passed, that the decision had already been made in a conversation they were never invited into.
---
Post-Responsibility Trauma (PRT) is a concept I have developed to describe the nervous-system adaptation that follows sustained responsibility without the authority to act. It is offered as a proposed framework, not a diagnosis. © Lindsay Reynolds Dobson 2026.
Principal sources
R (Tracey) v Cambridge University Hospitals NHS Foundation Trust [2014] EWCA Civ 822 — Court of Appeal judgment
Winspear v City Hospitals Sunderland NHS Foundation Trust [2015] EWHC 3250 (QB) — case summary
NICE: COVID-19 rapid guideline—critical care in adults (NG159), March 2020, and the amendment of 25 March 2020
CQC: Protect, respect, connect—decisions about living and dying well during COVID-19
CQC guidance on DNACPR decisions
Royal College of Physicians: End of Life Care Audit—Dying in Hospital, 2016
Parliamentary and Health Service Ombudsman: End of life care—improving DNACPR conversations for everyone
Judiciary: Eliot Harris prevention-of-future-deaths report (2022-0260)
INQUEST: Rowan Thompson inquest findings
King’s College London: Learning from Lives and Deaths (LeDeR), 2024 report, published July 2026