5. 4. 2013
The Liverpool Care Pathway Review
Dear Baroness Neuberger
I am a Consultant working with older people with dementia. I run both inpatient wards and a Liaison service providing advice to other physicians about the care of acutely medically ill people with dementia and other mental disorders of older life. I first wrote to the BMJ about the LCP in 2008 . As well as seeing the LCP used first hand, in the wake of my 2008 letter to the BMJ I was contacted by a number of relatives of patients, from across the country and around the world, whose family members were being managed on the LCP. I have therefore a wide perspective of the LCP and its use in a variety of settings. Very sadly, the development of version 12 of the LCP does not appear to have allayed the serious problems that were arising with its use back in 2008.
What has surprised me about this debate is the extent to which it has not been driven by doctors who think there is a problem with it, but rather it has been driven by the distress and outrage of bereaved relatives. I understand that thousands of the public have complained. For my part, I have seen some families whose loved ones were on the LCP, and who, were given advice which enabled them to have the LCP stopped. Generally, moribund people who were sedated and neither drinking nor eating improved and survived as a result. I have heard many more compelling stories of recovery and survival after relatives acted to get people off the LCP. There are now, within the public domain stories of people who were clearly not dying who / and recovered after having been taken off the LCP; this, if nothing else, questions the reliability of a diagnosis of dying in an individual.
So it does appear that the LCP is unsafe and causing grave difficulties in clinical care. I recently wrote again to the BMJ expressing my view that the LCP is not safe . Although it is popular with physicians who appear to be convinced that they can reliably diagnose imminent dying, the LCP appears to be encouraging bad practice as well as making doctors believe that they can diagnose dying when in-fact that diagnosis is most unreliable.
Key problems with the LCP
- Diagnosing dying is very hard and unreliable, more so in non-cancer cases.
- The LCP mandates routine prescription (prn) of sedatives morphine etc even if they are not needed
- Thus, the most junior member of staff with least experience etc can start them at 3am etc.
- Recommended doses also appear high especially for frail elderly and opiate naive people and
- Non cancer patients may be more sensitive to medicines recommended by the LCP
- Therefore its use can clearly lead to death;. Which is what the many complaints about it suggest.
- When a death occurs on the LCP, it is hard to see how that would show up as variance in the national audit.
- The issue is then raised that “it’s not the pathway, it’s bad clinicians”, or “The pathway cannot be bad as it’s only a guideline”. My view is that that is nonsense. By predicating care primarily upon prognosis and not need and encouraging clinicians to do what it states, the LCP inevitably sets up situations where those who are wrongly diagnosed as dying, die as a result of the use and guidance of the LCP
Each key point is discussed briefly below
1. Diagnosing dying is very hard and unreliable, more so in non cancer cases.
This is clearly true and is accepted wisdom within medical circles. It does appear that it is easier to diagnose rapid pre-mortem deterioration in people with terminal cancers etc, but in the acutely physically ill and people with dementia where the pathway is also often used, this is very difficult and the evidence base is against the LCP. There is an excellent article by Pullicino on this matter .
2. The LCP mandates routine prescription (prn) of sedatives and morphine etc even if they are not needed.
The National LCP audit finds that compliance with prescribing opiate like and sedative medicines in all those who are placed on the LCP is very high drugs (99% for opiates and 96% compliance for other medications that sedate etc). It requires anticipatory prescribing. That means that even if medicines are not needed now they should be prescribed on an as required basis in case they are needed, It is entirely clear that these drugs can cause respiratory depression and inability to drink etc and therefore it is clear that if they are used they can make death more likely. The doctrine of double effect has rightly been accepted for years by the medical and legal professions. But if sedatives or strong painkillers are used without good reason, double effect cannot apply.
3. So that the most junior member of staff with least experience etc can start them at 3am etc.
Once the “as required” medicines a written up, even very junior members of staff can simply give these medicines at night. And this is an even greater problem with modern hospital care, where continuity of care is so challenged.
4. Recommended doses also appear high especially for frail elderly and opiate naive people and
Support documents for the LCP  have suggested standard initial (as
required) doses for people starting on the LCP. These does do seem high
and if given are likely to have substantial side effects for frail and
elderly opiate naïve patients. And high doses make it more likely that the
LCP will cause death. Supporting documents for the LCP recommend 5-10mg of
sub-cutaneous morphine  and morphine injected by that route is at least
twice as potent as oral morphine. Such doses effectively sedate and
compromise swallowing etc in young people. Although there is a tiny prompt
to “Review dose/drug/frequency for the
elderly, frail and in renal failure” the suggested doses are too high and will harm some people.
5. Non cancer patients may be more sensitive to medicines recommended by the LCP
It has long been said that morphine in appropriate doses does not shorten life but there are two very important caveats to that statement. Firstly it does not apply to the other sedative medicines that the LCP expects to see written up. And secondly, the evidence base for that was derived for the treatment of cancer p[ain. Frail medically ill elderly people are not the same and their survival may be compromised [5,6].
Therefore the LCP is more likely to shorten life or even cause death in frail older people who do not have a clear diagnosis of terminal cancer.
And high doses make it more likely that the LCP will cause death,
especially in the frail ,
elderly and those with dementia.
6. Therefore its use clearly can lead to death. Which is what the many complaints about it strongly protest.
The claim is made in a consensus statement that the LCP does not hasten death. It is obvious that in a weak person whose fluids have been withdrawn and who is sedated, then death will be hastened. Thus I struggle with the consensus statement. It does seem hard to argue the validity of the consensus statement on that point.
7. When a death occurs on the LCP, it is hard to see how that would show up as variance in the national audit.
The National Audit records a decision not to prescribe opiates or sedatives as variance. But in fact if they are given unnecessarily, or in too high a dose I cannot see how that would be recorded as variance. Indeed, in the national Audit such a thing would pass as good practice. Even if the patient dies. And I note that median survival is only 29 hours, meaning that doctors are either far more proficient at prognostication than the evidence base suggests is possible, or the LCP does indeed have the ability to hasten and even cause death. And it does not appear that that wold be recorded as variance under the National Audit.
8. The issue is then raised that “it’s not the pathway, it’s bad clinicians”, or “The pathway cannot be bad as it’s only a guideline. “
My view is that that is nonsense. By predicating care primarily upon prognosis and not need and encouraging clinicians to do what it states, the LCP inevitably sets up situations where those who are wrongly diagnosed as dying, die as a result of the use and guidance of the LCP
Specialist palliative care physicians do indeed tend to see less
difficulties with the LCP than others. But the ones with whom I have
spoken tell me that they although they see difficulties with it, they
often work around or adapt the LCP with their expertise to enable good
care. Most, but not all, of the difficulties appear to arise in acute
care. Very many doctors, nurses and leaders within the NHS (at local and
national meetings) have either raised their concerns with me , or shared
concerns when I raised them . And junior doctors and nurses have told me
that it makes them confident that they are doing the right thing: which is
problem especially if that confidence is misplaced and encourages inappropriate care.
Much more could be said including about poor information to families, the absence of patient consent procedures and also the imperative to review the administration of fluids. As far as we can tell from the National Audit, the majority of people on the LCP do receive opiates and sedatives etc via a syringe driver. And fluids are hardly ever started for symptomatic relief of hydration.
Has the LCP improved care of the dying?
Claims are made that the LCP has improved care of the dying. The LCP is said to improve practice by enabling and encouraging holistic care at the end of life. But the evidence on communication, consent and spiritual care is that those components are poorly done, while medication cessation and opiate and sedative prescribing show high rates of compliance.
Painful reports of death with severe dehydration have only really been heard since the LCP was introduced. The LCP does not encourage the symptomatic relief of dehydration with iv or sc fluids and version 11 of it did not encourage oral fluids either.
Has the LCP made care of the dying worse?
My concern is that the real cultural change being brought about in medicine by the LCP, while unintended, is far more sinister. Firstly, the LCP may be giving clinicians a false confidence in their ability to diagnose dying accurately. Then if that leads to prescription of sedatives etc (as is required by the LCP) we see that median survival is only 29 hours one can at least understand the view that the LCP can cause death.
But we have also seen descriptions of painful deaths, over days and weeks of people who are placed on the LCP and who have no fluids etc. While fluids are not banned by the LCP, very few are started on them and dehydration is a real problem. Dehydration looks painful and is distressing at least for relatives. And in those who struggle with it over days, it is clear that the initial premise of death within 72 hours was false.
Many relatives would suggest that the LCP has made death more painful for some, and failed to improve communication with families. Dehydration and oversedation have been particular worries.
Some young medics and nurses have told me that they see intensive care being given to the point when the LCP is started and rapid decline after that. That is worrying as perhaps what we are seeing is a reduced ability of the medical profession to just treat dying people gently and to see how they do. We used to do that when I first qualified and it worked well. It allowed a more balanced and expectant approach to the very ill and those who might be dying. But if in fact the LCP encouraged clinicians to see treatment as “full on” or “end of life”, with no in-between position of gentle care while we wait to see what will happen.
So the LCP, with its clear construction around prognosis is deeply problematic. A needs based palliative approach is needed instead.
What needs doing?
We do need excellent palliative care of the frail, sick and those who are approaching the end of their life. None of us wish to see the dying suffer. But nor should we see such distress as is seen on the LCP or people dying as a result of poor diagnoses. We need a new approach that is based primarily and explicitly upon need, and not primarily upon prognosis.
- Treloar A, Continuous Deep Sedation. Dutch research reflects problems with the Liverpool care pathway. BMJ. 2008 April 26; 336(7650): 905.
- Treloar A. The LCP is not safe. http://www.bmj.com/content/346/bmj.f1184/rr/634583
- Pullicino P. The dangers of abandonment of evidence-based medicine in the use of the Liverpool Care Pathway. Catholic Medical Quarterly Volume 62(4) November 2012. http://www.cmq.org.uk/CMQ/2012/Nov/LCP-Pullicino.html
- Gomes T et al. Opioid dose and drug-related mortality in patients with non-malignant pain. Arch Intern Med. 2011;171(7):686-691; 691-693
- Trescot A et al. Opioids in the Management of Chronic Non-Cancer Pain: An Update of American Society of the Interventional Pain Physicians’ (ASIPP) Guidelines. Pain Physician 2008: Opioids Special Issue: 11:S5-S62
Dr Adrian Treloar MRCPsych, FRCP, MRCGP
Consultant and Senior
Lecturer in Old Age Psychiatry