Why do this?
This paper, published by Professor Patrick Pullicino in the Catholic Medical Quaterly (Volume 62(4) November 2012, online journal), represents the intellectual foundation on which the current, frequently destructive debate about the LCP is based. The research into prediction, prognostication, treatment withdrawal and misapplication formed the basis of a lecture at the Royal Society of Medicine in July 2012. This meeting was convened by the Medical Ethics Alliance.
As a general physician, gastroenterologist and hepatologist I have been and always will be involved in the care of dying patients. Although I am not a palliative care specialist, and although I do not have detailed knowledge of the evidence for or against the LCP, I do feel that I am qualified to make this response. For I am one of thousands of doctors who have been accused, implicitly, of practising euthanasia.
I have not attempted to submit this critique to the Catholic Medical Quarterly as correspondence.
In this article I summarise each of the paper’s section in turn and then present my own criticisms in italics. I have deliberately avoided trying to mount a full scale defence or overall justification for the LCP, choosing to limit myself to the points raised in Pullcino’s paper. Excellent articles extolling the LCP have been published elsewhere. It may be helpful to open the original paper in a separate window while reading this.
Introduction No comments.
Prediction and Prognostication The difference between these two terms is discussed. Prediction represents an individual clinician’s ‘educated guess’ of a patient’s expected survival, whereas prognostication is based on objective data and statistical modelling. It is emphasised that data does not exist to allow accurate prognostication in the ‘short term’, ie. the ‘final hours and days’ for which the LCP is designed.
Seeking to highlight prognostic scores that address patients with very short term survival prospects, he describes the Palliative Prognostic Score. This study split patients into three groups according to their estimated survival, median duration being 76, 32 and 14 days. 30 day survival probability for the latter group was 17%. He then mentions a nomogram (by Felieu J et al, 2011) which gives a 15 day survival probability, but points out that it was inaccurate a third of the time. ‘In this study, in a quartile (99 patients) of mean survival 10 days, over 10% survived much longer, with survival up to 200 days.’
These important observation force those of us who use the LCP to examine our thought processes. Are our prognostic/predictive skills really as bad as the literature would suggest? I know of no-one who uses prognostic scores and objective criteria on the ward. The diagnosis of ‘dying’ is indeed a subjective exercise, its accuracy increasing as patients display more and more typical clinical features. On further reflection however I concluded that the studies described are not particularly relevant to clinicians treating patients at the very end of their lives. What we are required to is recognise dying and manage it, not predict that dying will occur a week, two weeks or two months in the future. The question we must ask ourselves is ‘do we diagnose dying accurately?’, not ‘are we good at determining how long this currently stable patient will survive with this illness?’
Literature search The author states that no sources could be found to describe the use of prognostic scores within a ‘very early time frame’
Clinical factors associated with withdrawal of care This brief review concentrates on a study of organ support withdrawal in 15 intensive care units. It was found that subjective factors related to physicians’ perception of survival probability, potential cognitive deficit and substituted judgment of the patients’ view on resuscitation, but not age, prior functional status, illness severity or organ dysfunction were independently associated with the decision to withdraw. An accompanying editorial explored this phenomenon, raising concerns that physicians were ‘creating a self fulfilling prophecy’; ie. deciding that death was inevitable and facilitating it’s evolution.
The author then describes a study of patients with neurosurgical emergencies, reporting that survival was improved with aggressive surgical management and intensive care support. He concludes the section with ‘Practitioners tend to be overly pessimistic in prognosticating outcome based upon data available at the time of presentation.’
Although a discussion about the power of physicians’ opinions in end of life scenarios is valuable, and the danger of the ‘self fulfilling prophecy’ is of particular interest, I did not feel that studies of ICU patients were relevant to the LCP. The vast majority of patients put on the LCP do not have organ support withdrawn.
Care of a patient put on LCP A patient was admitted under the author’s care and put on the LCP by a trainee in the context of apparently intractable seizures. Next day the author determined that the relatives were not in agreement with it. LCP was withdrawn. The patient was discharged and survived for three months at home with maximal support for activities of daily living.
The error here appears to have been a lack of communication. The LCP was not applied correctly, and the author’s subsequent cancellation of it is not therefore surprising. The fact that the patient survived also demonstrates that the prediction, or prognostication, was inaccurate. My take on this case was that ‘intractable seizures’ is an unusual circumstance for the use of LCP, and the ‘diagnosis of dying’ all the more difficult. It is not surprising that the author was motivated to investigate the LCP following this experience. I wondered if the author’s experience of LCP in a non-medical specialty had led to a skewed view of its benefits and risks.
Conclusions The author states that there is no scientific evidence ‘to support a diagnosis that the patient is in the last hours or days of life.’ He then moves on to state that without an evidence base use of the LCP equates to an Assisted Death pathway. He highlights the very subjective decision making process, and recalls the problem of the self fulfilling prophecy.
The author then makes this hugely controversial statement: ‘If we accept to use the LCP we accept that euthanasia is part of the standard way of dying in the NHS. The LCP is now associated with nearly a third of NHS deaths. Very likely many elderly patients who could live substantially longer are being killed by the LCP including patients with “terminal” cancer, as the above research shows. Factors like pressure of beds and difficulty with nursing confused or difficult-to-manage elderly patients cannot be excluded as biases towards initiating the LCP.’
The full import of this statement is explored below, but even if the accusation of euthanasia is overlooked, it must be emphasised that no evidence has been provided to support an overall rise in mortality since the LCP was introduced.
Other statements include:
‘Starting a patient on the LCP, is an abandonment of evidence-based medicine in a critically-ill section of the hospital population’
Patients reaching the natural end of their lives are not critically ill.
‘Nursing of elderly patients who are on the LCP in proximity to those in whom evidence-based medicine is determining care, is confusing to junior medical staff and nurses alike’
Although deserving of attention, there is no evidence for this.
‘Use of the LCP is likely to have negative effects on elderly patients in particular, who are not on the LCP and to undermine the doctor-patient relationship’
This has certainly come to pass: trust between patients and doctors has been eroded during the LCP debate.
Defining prognosis and prediction is useful, although the difference between the two may seem rather semantic to many. As I have mentioned already, I am not sure that an analysis of our skill at prognostication is relevent to how we use LCP, the use of which is triggered by signs of possible dying. Nevertheless, if the LCP is perceived to guarantee death, it is very important that we identify dying patients accurately. Is this an achievable aim? Probably not. Should this admission result in abandonment of the LCP? Probably not. No methods of medical assessment, and no therapies, are 100% accurate or successful. As long as patients are reviewed regularly, to ensure comfort and to confirm the impression that the they are in fact dying, we should be able to minimise the risk of erroneous diagnosis while ensuring that the vast majority of patients benefit in terms of comfort.
It is the lack of evidence supporting the exercise of prognostication that drives this paper. The evidence that does exist in support of the benefit that patients derive from the LCP is not discussed. I think recognition that some evidence exists supporting the LCP would have added balance to this paper (for instance this ‘cluster trial’ – courtesy of Katherine Sleeman, Clinical Lecturer in palliative care, KCL, Cicely Saunders Institute).
Another area deserving discussion is that of communication. Prof Pullcino’s paper touches on this only briefly, during his description of the man with siezures, and I wonder if a greater focus on family conversations would have increased its relevance in this regard.
My overriding objection to this paper centres on the use of the word euthanasia. It is suggested that widespread use of the LCP equates to institutionalised euthanasia, and implicit in this is an accusation that individual practitioners have killed their patients. To read this, as a doctor who has used the LCP, is very difficult. The accusation is made in the conclusion without any supporting evidence. The ‘evidence’ that is reviewed in the paper does not touch upon intentional killing. If we are regularly making inaccurate predictions (or prognoses), that is of course unacceptable and must be addressed, but the term euthanasia suggests that we are intentionally killing our patients. There is absolutely no evidence for this.
This paper, and the thoughts behind it, sparked a huge controversy over end of life care in this country. I think it is methodologically weak and structurally flawed. I think it contains baseless conclusions, and is excessively liberal with emotive, hurtful accusations of intentional killing.
Acknowledgment: Dr Rita Pal alerted me to the paper’s online publication and has been helpful in researching details on the original RSM presentation.