Research helps understand how doctors recognise when a patient is dying
Press release published
Research published in the journal BMJ Open [SK1] has identified the key signs and symptoms that expert doctors use to recognise when terminally ill patients are imminently dying.
The findings come from researchers, led by Dr Nicola White, at the Marie Curie Palliative Care Research Department at UCL, whose previous study found that some doctors are frequently inaccurate when predicting how long those living with terminal illnesses will survive. The previous study revealed a wide variation in errors, ranging from an underestimate of 86 days to an overestimate of 93 days.
While the NICE guideline on care of dying adults in the last days of life describes the recognition of imminent death as an essential first step towards improving care for dying patients, it does not clearly explain how clinicians are expected to identify such patients – nor does it offer advice about how novice doctors can learn or improve their clinical skills in this area.
The team’s new study set out to, first, identify a group of palliative care doctors who had particular expertise at recognising when a patient is dying. Volunteer doctors completed an online test, in which they were presented with 20 real case studies of patients who had been referred to a hospital or hospice for palliative care. The doctors were asked to decide whether the patient was close to death or not, and the results enabled researchers to identify the top 20% of participants who were most accurate at predicting when the patients would die.
The ‘top-performers’ were then presented with an additional 50 cases of hypothetical patients and were again asked to estimate the probability of death within the next 72 hours. Based on the decision-making of this ‘expert group’, the researchers were able to determine the relative importance of different factors that unconsciously lay behind the doctors’ survival predictions.
The most influential factor was the Palliative Performance Score, which measures how much support someone needs based on their functional ability: ranging from fully independent (100%) to bed bound needing full support and unconsciousness (10% and below). The next most important factor was the presence of an abnormal pattern of breathing called Cheyne Stoke breathing.
Paddy Stone, Professor of Palliative and End of life care at the Marie Curie Research Department at UCL, said:
“Even at the very end of their lives, patients, their families and the staff looking after them want to know how long they may have left to live. Understanding that death is imminent can allow patients and their families adequate time to prepare. This research has shone a light on how expert doctors are able to recognise when terminally ill patients are nearing the ends of their lives and it is to be hoped that this greater understanding will help to educate medical students and less experienced doctors about how to improve this clinical skill.”
Dr Adrian Tookman, Medical Director at the Marie Curie Hospice, Hampstead, said:
“When someone is unwell and imminently dying, the question that is almost always asked is ‘how long has my loved one got?’ This is such an important issue since it allows people to make the very best decisions in, what for most people, is a very distressing time. This paper reinforces that expertise and knowledge in this area is needed, we need to understand how we can help people to manage the uncertainties that can occur at this time. Medicine is part art and part science, we must ask and explore the most challenging questions to improve our understanding of the dying process. It is only by doing this, that we can teach others to manage dying patients in the most compassionate and appropriate way.”
Julie Pearce, Director of Nursing and Quality at Marie Curie, UK, said:
“The role played by the multi-disciplinary team caring for patients who are dying is central to the experience of both the person and their family. The effectiveness of that care can be determined by a blend of the knowledge and experience of the team. This study demonstrates how difficult it is to predict that the person is imminently dying on clinical signs alone. It would be interesting to explore the value and contribution made by other members of the team to determine whether a blend of clinical signs and other subtler changes in the person’s condition could add further to the ability to predict how close to the end of life the person is. Improving knowledge in this area is key to providing better information and support to the family “
The researchers hope the study will help inform the development of educational training materials to teach novice doctors, medical students and other healthcare professionals how to model their own judgement policies on those of the ‘experts’.
The study titled ‘How do palliative care doctors recognise imminently dying patients? A judgement analysis’ was supported by the terminal illness charity Marie Curie and a UCL PhD studentship.
Notes to editor
About Marie Curie
Please note – we are now called ‘Marie Curie’ (not Marie Curie Cancer Care)
Marie Curie – care and support through terminal illness
Marie Curie is the UK’s leading charity for people with any terminal illness. The charity helps people living with a terminal illness and their families make the most of the time they have together by delivering expert hands-on care, emotional support, research and guidance. Marie Curie employs more than 2,700 nurses, doctors and other healthcare professionals, and with its nine hospices around the UK, is the largest provider of hospice beds outside the NHS.
For more information visit www.mariecurie.org.uk
Like us at www.facebook.com/mariecurieuk
Follow us on www.twitter.com/mariecurieuk
The Marie Curie Palliative Care Research Department at UCL
An internationally recognised centre of research with a team of over 20 full-time researchers. The department receives core funding from Marie Curie and also undertakes research funded by the National Institute for Health Research (NIHR), the Alzheimer's Society and other research funders. The department has particular research strengths in the areas of palliative care for people with dementia, prognostication in advanced cancer and the management of cancer-related fatigue.
About UCL (University College London)
UCL was founded in 1826. We were the first English university established after Oxford and Cambridge, the first to open up university education to those previously excluded from it, and the first to provide systematic teaching of law, architecture and medicine. We are among the world's top universities, as reflected by performance in a range of international rankings and tables. UCL currently has over 41,500 students from 150 countries and over 12,500 staff. Our annual income is more than £1 billion.
www.ucl.ac.uk | Follow us on Twitter @uclnews | Watch our YouTube channel YouTube.com/UCLTV
About the research
The study titled ‘How do palliative care doctors recognise imminently dying patients? A judgement analysis’ was published in the journal BMJ Open. It was supported by the terminal illness charity Marie Curie and a UCL PhD studentship.
The objective of the research was to identify a group of palliative care doctors who perform well on a prognostic test and to understand how they make their survival predictions.
All members of the Association of Palliative Medicine (APM) were eligible to take part. 99 doctors completed the prognostic test and were included in phase 1 analysis. The top 20% were invited to participate in phase 2 – 14/19 doctors completed the judgement task and were included in the phase 2 analysis.
Phase 1: an online prognostic test, developed from a prospective observational study of patients referred to palliative care. Participants were presented with 20 real case studies – these patient cases were collected by the researchers, who had observed the patients and recorded whether or not they died within 72 hours. The clinical characteristics of each patient were summarised on one page and presented to the doctors in a style that would be familiar to clinicians participating in a multi-disciplinary team meeting. Doctors were then asked to read through each case and make a numerical predication about whether or not each patient would die within 72 hours – 0% indicating no chance and 100% indicating certainty.
The survival estimates provided by the participants in the test were compared with the known outcome of each case. The accuracy of each doctor’s prognostic estimate was determined by calculating the Brier Score, which can range between 0 and 1; 0 representing perfect accuracy and a score of 1 representing perfect inaccuracy. Analysis of this data found that the average Brier Score of all participants was 0.237.
Phase 2: an online judgement task consisting of 50 hypothetical case studies. The top 20% doctors who most accurately predicted when the patients would die were asked to complete an additional test – the mean Brier Score of the ‘experts’ was 0.184. This time, 14 doctors were presented 50 cases of hypothetical palliative care patients and asked to estimate the probability of death within the next 72 hours. The doctors were presented with seven factors, which had been previously identified by a specialist palliative care team as the top indicators of imminent death.
Based on the results of the decision making of the ‘expert group’ the researchers were able to determine the judgement policy of these doctors, which identified six factors that were significantly associated with clinician predictions of death.
These results may be used to teach novice doctors how to improve their prognostic skills.