End of life care: improving DNACPR conversations for everyone

Conversations often happen too late and in emergency settings 

“We need to be having these discussions earlier as a nation rather than leaving them for an urgent situation. This is where misunderstanding occurs due to high emotions on both sides.”  

Survey response of doctor working in geriatric medicine  

Most of the poor experiences, from focus group feedback, conversations with doctors and our complaints, related to acute hospital settings. Though this evidence shows these are not the ideal settings for conversations about end-of-life, we acknowledge this will sometimes be necessary. But all audiences who took part in our research felt there could be: 

• a better balance with earlier planning in other settings 
• support for better conversations to take place in emergency settings.  

There is excellent experience within palliative, geriatric and learning disability healthcare staff, which could be shared with staff working in emergency settings. 

“People expressed that, at these times, emotions such as shock, fear and anxiety were already running high, the conversations were rushed, and they were not given time to reflect on the information. This contributed to a feeling of having something ‘done’ to you rather than it being an individualised, considered and sensitive discussion.”  

Compassion in Dying, page 10 

The cases below are examples of poor communication of DNACPR decisions in acute settings, including emergency departments and high dependency units. 

Case study 2: Trust did not communicate with patient’s family 

Mr A was admitted to hospital in June 2018 for planned surgery to remove a tumour from his kidney. After surgery, Mr A went to the Trust’s high dependency unit and then moved to the urology ward. Shortly after, he became unwell and was diagnosed with sepsis. His condition got worse and he sadly died. On the date of Mr A’s death, doctors had spoken to his wife, Mrs A, and advised that her husband had rapidly deteriorated and was likely to die. But there was no reference to DNACPR. Shortly after this conversation, doctors made a DNACPR decision and completed the relevant documents.  

While we found no failing in the clinical decision of the DNACPR, we found a failing in the Trust’s communication with Mrs A about the DNACPR. This meant Mrs A was unprepared and in shock when she found out the Trust had not attempted to resuscitate her husband. This caused Mrs A considerable distress. She felt this made the grieving process worse. 

Case study 3: Inaccurately recording family member’s wishes 

In April 2020, Mrs A’s husband, Mr D, was admitted to hospital from the care home where he had been getting respite care. At the time, Mr D was in his seventies and had Lewy body dementia (a condition that affects thinking, memory and movement) and Parkinson’s disease (a condition in which parts of the brain become damaged). Mr D was admitted to hospital with suspected COVID-19 symptoms. A doctor rang Mrs A to advise that CPR would not be in Mr A’s best interests. Mrs A told the doctor she firmly believed the Trust should attempt resuscitation for her husband. Over the next few days, Mr D’s condition got worse and he sadly died. Mrs A received Mr D’s medical records and found they said she had agreed with the proposed decision not to attempt CPR.  

While we found no failing in the clinical decision made about CPR, we found a failing in the Trust inaccurately recording Mrs A’s wishes. This caused Mrs A further upset and frustration. 

Time pressure and lack of experience leads to poor communication 

It is clear the main problem with having and reviewing these conversations in acute settings is time, especially where rapid deterioration in health can happen. 

“Unfortunately, as much as people may be aware of the guidelines, we struggle to find time to have these discussions ahead of time, before people become unwell. We also don’t continue the conversation as things change with time.”  

Survey response of doctor working in emergency medicine 

“Colleagues are still reluctant or too time-pressured to have proactive conversations.”  

Survey response of doctor with a non-clinical background 

Where time was available, people told us they had more positive and empathetic conversations. 

“The difference is humanity. It’s being willing to align yourself with a fellow human who’s in more trouble than you are.”  

Compassion in Dying research, page 16 

“The doctor wasn’t talking as a doctor he was talking as a fellow human being. We’d been talking about music.”  

Compassion in Dying research, page 16 

As highlighted in the graph below, doctors point to time and training as issues that need to be addressed in the DNACPR process to enable better quality, rights-respecting conversations to take place. Half of doctors who responded to our survey, across different specialisms, asked for more training on how to have sensitive and clear conversations. This suggests experienced doctors accept this is an area for development and welcome support. 

Figure 1: What might help address the issues and barriers to managing the DNACPR decision-making process effectively? 

Source: PHSO’s survey of doctors, 2023 

Advance care planning 

Our research points to advance care planning as a partial solution to improving communication around DNACPR decisions. Our evidence suggests that opportunities to have these conversations early are often missed and that the tools available such as the ReSPECT forms and advance care principles are not always used. Trusts are not encouraging thinking about end-of-life conversations in enough settings. 

As part of the DNACPR survey we asked doctors if they were aware of the advance care principles, which had been produced in response to the CQC report and overseen by the MOG on DNACPRs. 

53% of doctors had heard of the Universal Principles for Advance Care Planning and 47% had not heard of them at all. Awareness levels were highest among respondents specialising in end-of-life care and those working in palliative medicine. Those who were newer in their roles (less than two years) had the lowest awareness levels (22%), suggesting that more work is needed to embed these principles. 

Figure 2: What impact, if any, do you think the Principles have had on how DNACPR notices have been applied since March 2021? 

Source: PHSO’s survey of doctors, 2023 

“There are issues around awareness, understanding and education that need addressing, and more and better guidelines help to some degree, but primarily there are systems issues that mean that these guidelines cannot be actioned in practice; therefore it is a systems approach rather than educational/guidelines approach that requires implementation.”  

Survey response of doctor working in general practice 

Where conversations happen ahead of time, the ReSPECT form does seem to enable these discussions more holistically than a DNACPR form.  

“ReSPECT does allow a more nuanced approach to what is right for the patient.”  

Survey response of doctor working in cardiology 

But there needs to be adequate time and training to complete these. 

“ReSPECT gives the opportunity to provide much more detailed recommendations about the patient’s wishes and priorities, and what specific medical treatments they would and wouldn’t want. Unfortunately, this is not always completed as well as it could be … this is down to the clinician not the form.”  

Survey response of doctor working in geriatric medicine 

Nurses already play an important role as interpreters and challengers around DNACPR decisions. They are vital to conversations taking place in more appropriate formats and settings. Nursing staff are a crucial part of accountability in all care settings, where needed, clarifying and challenging conversations and decisions as people who usually know patients and their individual needs on a different level from doctors. 

While all the nurses we spoke with felt they would be able to challenge a poor DNACPR decision, they acknowledged that this is unlikely to be the case for more junior healthcare staff, some of whom are likely to have closer relationships to the patient involved in the decision.  

Nursing staff in one trust developed their own solution to this. Junior nurses felt opportunities for advance care planning conversations were being missed for their patients. Senior nurses produced a document containing criteria which, if matched with the patient, empowered junior nurses to recommend an advance care planning conversation to the doctor. This encouraged a ‘speak up’ culture. It also promoted good multidisciplinary team working, which was considered as important in making sure well-rounded decisions can be made about advance care planning. 

Including people with different communication needs in advance care conversations 

People with learning disabilities may need more time to process what end-of-life conversations will mean for them. Advance care conversations are vital to making sure people with learning disabilities are included in decisions about their healthcare and their rights are upheld. 

“People with learning disabilities need sufficient time to process information and make informed decisions about DNACPR. Rushing the decision-making process is viewed as problematic and may prevent individuals from being able to make an informed decision about crucially important healthcare.”  

BIHR, page 19 

“People with learning disabilities should still have the same rights as people without a learning disability, it might just be that it takes a little bit longer for them to be able to understand what you’re actually talking about.”  

BIHR, page 6 

Learning disability nurses also said that conversations often happen too late or at an inappropriate time for people with learning disabilities. Moving to an advance care planning approach would make sure conversations can happen at the right time, in the right context (such as community settings) and with someone the patient knows. This perspective was also reflected by nurses working with older people and in palliative settings.  

In 2017, annual health checks became a way to offer a separate end-of-life conversation for people with learning disabilities. While we encourage annual health checks as a way of having end-of-life conversations in a more appropriate setting, our evidence suggests that this is not always happening. If effective, this could cover 320,000 people (NHS England, ‘A summary and overview of the Learning Disability Annual Health Check electronic clinical template’). 

For older people, nurses suggested considering the benefits of having advance care conversations in residential settings for those living with frailty. Where possible, this would avoid these conversations happening for the first time in acute settings. 

More pathways for conversations 

Overall, evidence suggests that there is no quick fix. There need to be whole-system changes to make sure these important conversations happen in settings where patients feel comfortable and doctors have time. 

There is no one clear pathway for these conversations to take place. Our research showed there needs to be more options for end-of-life conversations to happen at the right place and time for individuals, with more healthcare staff equipped to have them. Confining conversations about end-of-life to one healthcare setting (such as care homes) would not account for people’s individual needs and could reinforce discrimination. 

If more pathways open up for patients to have discussions about their end-of-life wishes, cultural issues would still need to be addressed to encourage proactive conversations. These issues prevent conversations about death and dying from happening more naturally between the health service and patients earlier in people’s lives. This leaves people and their families unprepared when the time comes for conversations about end-of-life care (K. Mannix, ‘With the End in Mind’).