Do specialist palliative care teams really have better crystal balls?

Dr Natasha Wiggins, Consultant in palliative medicine and clinical lead for end of life at the Great Western Hospital NHS Trust @Dr_N_Wiggins

Dr Ellen Haire, Wales Clinical Academic Track Fellow in Palliative Medicine, University of Cardiff @drellenhaire

Ellen Haire and Natasha Wiggins in a virtual meeting

Picture: Ellen Haire (left) and Natasha Wiggins in a virtual meeting


Google ‘crystal ball palliative care’ and over 2 million results will appear. Other than what appeared to be a successful charity fundraiser in 2019 – The Crystal Ball – it seems there is a lot of conversation around the dark art of prognostication.

Recognition of dying can be a rate limiting step for good end of life care and it is important. Important due to the time frames required to unlock levels of care e.g. in England someone must have ’weeks to short months’ left in order to qualify for Continuing Health Care funding, or less than 2 weeks for most hospice in-patient end of life admissions. Important to have the time to develop relevant and individual end of life care plans enabling focus on comfort, dignity and what matters to the person. These plans might include wishes like, “I want to get home to see my cat” or “I want to go on the longest zip line in the world, we can put the syringe pump in my backpack”. Unless someone is recognised as ‘being in the last days of life’, such plans may not be facilitated and a dying person may go through unnecessary procedures, perhaps without symptom control and dignity at the helm of their care. They may miss out opportunities to communicate key things to their loved ones – sentimental and practical. The value of early recognition of this time has been further thrust into relevance with the relaxing of COVID-19 induced visiting restrictions for people who are dying.


The National Audit for Care at the End of Life (NACEL)

This occurs annually to assess quality and outcomes of care of the dying in acute hospital trusts in England, Wales and Northern Ireland as compared to NICE guidelines1. One measured outcome is ‘Recognising the possibility of imminent death’.

The 2019 NACEL reported a national average of 88% of decedents were recognised as dying with a median time of recognition to death as 41 hours2. Using this as the standard, an audit was completed in an acute hospital trust in England reviewing time between recognition of dying (ROD) and actual death. To satisfy our curiosity around the question of crystal balls, we also looked at the role the specialist palliative care team (SPCT) played in this.

Data for the first cycle was collected via retrospective review of 49 notes randomly selected by an Excel algorithm from decedents in the acute trust April-May 2020.  88% (43/49) of people were recognised to be dying. The median time of recognition that someone was dying before occurrence of actual death was 48 hours.  60% (30/49) of these decedents were seen by SPCT, with 97% (29/30) recognised to be dying. The median time before death was 72 hours.

40% (19/49) were not seen by SPCT with 74% (14/19) recognised to be dying. The median time of ROD before death was 24 hours.


Did things improve?

No. After a host of education interventions, a second data collection demonstrated no change in the above figures. Some readers may recall that April/May/June 2020 was a very hectic time and results may not be generalisable to other months. Whilst ‘peak pandemic’ meant that good prognostication and palliative care was more important than ever, staff were in survival mode, which rarely lends itself to change in personal practice. However, it does follow a trend of unpublished audits with limited improvement following education over time.


What does this suggest?

Although we are of course biased(as SPCT members) it suggests that involving the SPCT in unplanned hospital patient care on a significant scale is a good idea.

The lack of overall improvement in ROD following predominately education-based interventions highlights the need for a change in process. What could this look like? Perhaps widening the referral criteria to the SPCT? Or do we teach all staff (palliative care is everyone’s business, after all, and should not just be specialist-delivered) to recognise dying sooner, however challenging this may be. Could these people have been identified earlier? Would a prognostic review during a previous hospital admission have highlighted a downwards trajectory? 67% of the decedents in our audit were > 75 years old. We know that in the last year of life for those older than 75, 80% of people have one or more hospital admissions3 . It would be beneficial to review the last year of decedents’ lives to identify earlier points in their illness trajectory for possible intervention and key conversations.

Various prognostic tools could be applied; the Palliative Prognostic score; Palliative Prognostic Index; Feliu Prognostic Nomogram and Palliative Performance Scale have all been proposed as prognostic tools for palliative cancer care, however none demonstrated superiority compared to clinical opinion4.

Reflecting on the recent death of Queen Elizabeth II, Dr Kathryn Mannix presented an excellent twitter thread highlighting the process of “Ordinary Dying”.5 Early in the process she felt “all of us in palliative care recognised what was unfolding” and anticipated her death. Ordinary dying happens every day, up and down the country but for many, their lives’ end in acute hospital, when they may not have wanted this, perhaps progression of illness towards the end of life having gone unrecognised.

With the acknowledgement of variation within acute specialties, results indicate that perhaps SPCTs really do have better crystal balls and their involvement increases the time between recognition of dying and death, giving the dying person and their loved ones perhaps more time to discuss wishes, organise finances and begin the bereavement process. Proposed reasons for this are increased experience in caring for dying people and having time for more in-depth discussions which may reveal information to guide prognosis. The less scientific “haven’t you grown?” effect should also be considered; a team who have been caring for a patient day in, day out may struggle to notice incremental deterioration, something that may be more obvious to the fresh and objective eyes of the SPCT – akin to a child growing not noticed by their daily care giver.

With the recognition that there are many potential prognostic touch points within the UK healthcare system, this audit has generated discussion around the need to look more closely at the impact of using prognostic tools for a specific cohort of people  during an acute hospital admission irrespective of diagnosis. Could this create increased opportunities for the possibility of poor prognosis to be discussed, and SPCT involvement can be used at key moments during a person’s care to guide these conversations?


Take home messages:
  • Specialist palliative care teams may be better than their acute hospital colleagues at recognising dying earlier on
  • Education around recognition of dying may not work and a change in process should be considered
  • Should a prognostic indicator tool be used for patients with 3 or more long-term conditions on admission to acute hospital?



  1. Healthcare Quality Improvement Partnership. National Audit of Care at the End of Life. Second round of the audit (2019/20) report. England and Wales. [Online] 2020 [Cited 10/04/21] Page 10.
  2. Healthcare Quality Improvement Partnership. National Audit of Care at the End of Life. Second round of the audit (2019/20) report. England and Wales. [Online] 2020 [Cited 10/04/21] Page 33.

(Visited 455 times, 1 visits today)