Palliative care works best when it is considered early. Not only does it make patients more comfortable, it seems to extend life expectancy (famously shown in an RCT involving patients with lung cancer). I don’t know the reasons for this – maybe pain and distress left unaddressed draw energy out of people and hasten their demise, or maybe patients who are being monitored more holistically have their complications managed earlier. So, if you are a doctor who deals with seriously ill patients, you have to consider referral to palliative care days and weeks before the ‘writing is on the wall.’ Maybe, in the words of Kathryn Mannix, quoted in David Oliver’s recent BMJ column, when patients are ‘sick enough to die’.

Yet, as we learned in the era of the Liverpool Care Pathway (LCP), we can be wrong. Although patients may be sick enough to die, some, many, will not. While it is not wrong to refer to palliative care when a patient is not dying, such an action does represent, in the final analysis, a misjudgement. In a BMJ case report (£) we described how repeated conversations with a relative about the likelihood their loved one dying, and the likely need for a palliative approach, were in the end disbelieved, because the patient kept surviving against the odds. In trying to achieve good practice, we sowed confusion.

What happens when you, as a member of a specialist team, refer to palliative care? Well, before you start, you need to tell the patient that they are terminally ill. The palliative care team shouldn’t be expected to do that for you. The patient may then ask, ‘How long?’, and you will answer with words you will have grown comfortable with over the years. Some provide time-scales, many avoid such ‘hard’ estimates studiously. Often, patients and families ask if you are really sure – ‘Is there any chance?’ In many disease states, but especially cancer, there really is no ‘hope’ of cure. But outside cancer, some diseases fluctuate. Acute deteriorations in chronic illnesses may bring the patient near to death, and it is only right to prepare them for the worst. Then, inexplicably, a corner is turned. Nature plays a benign trick, one that makes you, as the clinician, look overly pessimistic.

The danger here is that referral to palliative care becomes a self-fulfilling prophecy. After referral, and having established that the patient is dying in the short term, you may agree not to perform any more blood tests, not to do physical observations, not to re-insert venous cannulas or prescribe antibiotics. In this case, signs of improvement may be missed. The ailing body’s first steps on the long, steep path to ultimate recovery may be overlooked. When those steps falter, the assistance that might have sustained this surprising change in trajectory will not be provided. The medical team, the family, the patient him- or herself, will fade to death unaware that for a short time there was a chance. Of course, such opportunities for recovery are rare, and to maintain ‘full, active treatment’ for every patient until they are clearly at the end of life would be to deprive an overwhelmingly large proportion of appropriate palliative care.

This balance is clearly described in NICE guidance on end of life care (written after the LCP was phased out).


There is a tension between 1.1.3 and 1.1.4. Improvements in sign and symptoms can be detected from the end of the bed, but other indicators of improvement may require blood tests, maybe even scans. But if the patient has been established on an end of life care plan, such tests might seem perverse. The decision to reverse must be prompted by a purely qualitative assessment. Is the patient more alert? Have they started taking food again? It’s medicine at its most traditional.

Doctors fear missing opportunities for recovery. Our main role in life is to extend the lives of others. Yes, it also to improve the lives of others – to make them feel better – but bottom line, it’s to stop people dying. That is what our years of training in acute medicine are focussed on. When we fail to do that, when we miss a sign or withhold a treatment, we have failed in that primary mission.

A generalised example, to illustrate the thought processes:

The team and I made a decision that a female patient in her mid-40s was at the terminal stage of her chronic condition. Everything that could have been done had been done, including a period of multiple organ support on the ICU. We scaled down care, ‘de-medicalised’ the situation, focussed only on interventions that would preserve or improve comfort, had several difficult but necessary conversations, and waited. The nursing team understood and agreed. It was sad to see her every day, knowing that the end was not far away. She expressed no great desire to be at home. All were prepared for a death in hospital, which, contrary to reports, is often as comfortable and satisfactory as any other location.

Then, five days later, we discussed her progress on the ‘board-round’, a meeting that often takes place before ward rounds. It allows doctors and nurses to update each other, away from the bedside. Nothing had changed. Is she comfortable? Yes. Fine, we carry on. Then, an hour later, we saw her. Her level of consciousness was fluctuating; at best, there were some groans of acknowledgment if her name was called out. Once or twice, a drink was taken through a straw. We reviewed the few remaining medications. We discussed whether there was an opportunity for her to die at home. We documented everything, and began to move on to the next patient.

Nothing had changed.


How long, I wondered, does dying take? What are the sure signs of dying? Is she dying?

We regrouped, and re-thought. Confidence in our diagnosis was collapsing. What if we are wrong? Let’s get a blood test, see how bad the kidneys are. Let’s correct any obvious abnormalities… yes, put in a cannula if that is required to reverse dehydration or give essential electrolytes. I was asked – So are changing our minds here? Are we changing the plan? What do we tell the family?

I thought about how to respond to this. The plan? There is no plan. There is only a life, the deep mechanics of which we do not truly understand. It’s not all on us, this process of deciding if a patient is dying or not. It’s not our decision. The patient decides; or their body does, at any rate. We observe, we make calculations, but we follow, we do not force. If the patient shows signs of improvement, or fails to follow the path that we predicted, and if there is a theoretical possibility that they can survive this episode, we must respond.

She continued not to die. Slowly, she strengthened. Eventually she was discharged. This was not her time.

Our preparedness to accept the likelihood of death and involve our palliative care colleagues must be accompanied by a willingness to re-assess and change our minds. When we do this, we are not wrong. Medicine is not always a right/wrong activity, despite the banks of MCQs that must be surmounted for us to achieve the right to practise. It is about maintaining attention over time, detecting subtle changes and preserving the mental flexibility (call it humility) to change your mind.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.