This site is intended for health professionals only

When a patient asks ‘How long have I got?’, it’s time we gave an honest answer

GPs are too shy about addressing patients’ questions on dying and end-of-life care, says Dr Paul Barker.

It’s fascinating and rare to understand a moment in time captured through the eyes of a great writer and historian. Whilst in hospital in France in 1929 George Orwell diarised his experiences in an essay he entitled ‘How the poor die’. He summed up his experiences like this:

‘However great the kindness and the efficiency, in every hospital death there will be some cruel, squalid detail, something perhaps too small to be told but leaving terribly painful memories behind, arising out of the haste, the crowding, the impersonality of a place where every day people are dying among strangers.’

I started to think just how much things had changed in the intervening seventy-odd years. Orwell described ‘natural death’ as ‘almost by definition something slow, smelly and painful’. I wonder how the harmonics of this concept resonate throughout our practice today?

As GPs we pride ourselves on end-of-life care. Most of us strive to make dying a natural part of life. As ordinary an occurrence as being born or paying taxes. We are uniquely placed in the trust of the patient and their families to allow patients who die at home, dignity and genuine care in the last days of life. Death at home and in a hospice is usually the choice of a patient and their families. Yet, so many of our patients are admitted to a general hospital to die in the last days of life.

Preferred place of care includes nursing homes and rest homes and despite the best advanced decisions and living wills the tidal flow of patients at the end of lives does not seem to be halting. Let’s look at why this might be.

There seems to be a rift between primary and secondary care in the way we talk about prognostication and end-of-life care.

Imagine the situation in which a patient asks a question with regard to their prognosis. Of course many of us will defer to a higher authority, in this case an oncologist, specialist nurse, radiotherapist, radiologist, interventional radiologist, MDT, or anyone in the increasingly long chain involved in our patients care.

Sometimes, despite explicitly asking ‘How long have I got Doc?’, we might simply say something along the lines of: ‘It’s difficult to know’ or ‘we’ll know more after the MDT’.

The perception seems to be that this approach fragments care and the patient may take months to swing through the ‘specialist cycle’, seeing a different expert on each occasion. This gives rise to the concept of ‘prognostic paralysis’ whereby patients – despite asking relevant and pertinent questions – are deferred to the next expert without ever receiving an answer. Rightly or wrongly, secondary care tends to talk about prognosis in aliquots of time months or weeks. Patient’s often say I was ‘given’ six months or 12 months.

Diagnosing dying

We in primary and palliative care tend to use language differently. Of course we have the undoubted benefit of opinions from our expert colleagues. Linguistically and culturally the ‘gifting’ of a certain time period is a norm, which is written into our programming and is surprisingly immutable.

However, a far more personal approach is to ask how patients were a month or a week or a day ago. If they are becoming increasingly unwell month by month patients tend to have months to live. If things are changing week by week, they tend to have weeks to live.

Likewise if things change day by day, patients tend to have days to live. This leads us to ask questions along the lines of ‘how were things compared to this time last month, have they changed much?’

If so, then there is an opening to talk about end-of-life care. Indeed putting an understanding of prognosis back to the patient and framing it in terms of feedback from their own body enables patients to understand their prognosis and personalise their end of life care.

However, I’m sure that by now you’ve spotted one of several flaws in the argument. Firstly, what constitutes a palliative care patient or a terminal illness?

We know for example that many chronic diseases will have a very similar prognosis to many incurable cancers. Yet we rarely talk about our patients with heart, renal or respiratory failure as palliative. Similarly, many frail elderly patients will come towards the end of their lives without ever having had the opportunity to talk about their end of life wishes. Death seems to come ‘like a bolt from the blue’ or as ‘such a shock’.

So is the phenomenon of ignoring dying simply the willing suspension of disbelief? Or are we simply terrible at diagnosing the last years, months or days of life?

Well actually, the answer is multi-factorial. We as health professionals are absolutely terrible at diagnosing dying. We tend to overestimate prognosis by three to five times and get it wrong most of the time (63%) and indeed so do our patients [2].

Armed with this misleading information, patients may well decide to undergo inevitably futile sometimes painful and unnecessary interventions. This has an obvious health impact for the patient and their family not to mention the health economics in wasted drugs and hospital hours.

So why are some deaths so damnably difficult to predict?

Patients with a cancer, often have a relatively straightforward disease trajectory. That is a graphical representation of ‘wellness’ plotted against time. Cancer is a well understood disease and is extensively resourced. Again, culturally and linguistically, many people still equate cancer with dying. The myth and folklore surrounding cancer are confusing, to say the least. Ill-informed comment from the media serves to baffle us and our patients still further.

However patients deaths from cancer do not tend to come as such a shock when compared to patients with chronic illness. Most cancer patients still die in hospital but many will die in the place of their choosing – often at home or in a hospice – many will understand the nature of their disease and plan for their end of life choices.

Patients who have chronic disease like COPD or heart failure. Have a very different trajectory. In this case the movement of the graph from top left (pretty good thanks) to bottom right (dead) is interspersed with often frequent exacerbations of either infection or decompensation. Any of these exacerbations may lead to death but predicting the final one is very difficult. Patients with these diseases tend to have relatively less resource to community care and often death comes as a surprise. Mostly these deaths occur in hospitals and nursing homes. Dying is rarely discussed and end of life choices often change with the condition of the patient.

Likewise, patients on a frail, elderly trajectory. A gradual dwindling of function until death supervenes. Death may be sudden or come after a long final illness. Mostly these deaths are in nursing homes and hospitals. Again, few of these patients will have made end of life decisions or choices. We all have examples of where attempted resuscitations have gone wrong. I recall a lady who was taken into a porch, her chest exposed to the public and a scene from ER re-enacted upon her body. Imagine just how traumatic for all those left behind to have that memory etched upon their consciences. Attempted resuscitation was degrading and futile and the final indignity to an otherwise peaceful or ‘natural death’.

It’s obvious to say that diagnosis dying is difficult and health professionals and patients are equally bad at making the diagnosis. However, understanding the disease trajectory at least allows us to know the natural history of dying from a particular disease. Now enter the surprise question, namely ‘would you be surprised if your patient was to die in the next year or three months’ (or whatever time period chosen). If the answer to that question is that you wouldn’t be surprised, this opens up the ideal opportunity for end of life care discussions.

Most of end-of-life care is formalised in the Holy trinity of Gold Standards Framework , Liverpool Care Pathway and Preferred priorities of care concepts. However, best palliative care will always be an amalgam of what we do best i.e. communication, team working honesty and a genuine caring and affection toward our patients. You know as well as I do that the primary care team provides an excellent forum for identifying and caring for patients toward the end of their lives and avoiding Orwell’s view of ‘natural death’ as something rather wretched.

 I think that it’s easy to look at our colleagues in hospices as somehow distant from primary care. Perhaps we forget that nearly all patients’ journeys toward the end of their lives have primary care involvement to a greater or lesser extent. The concept of the hospice and their gold standard end of life care is just as alive and well today within primary care. It’s good to get away from the concept that hospice care refers to a specific building or the bricks and mortar that surround some patients toward the end of lives. In fact the concept of hospice care is something that we as primary care physicians continue to do well. Whilst some deaths in hospital ‘surrounded by strangers’ are inevitable, we can only attempt make a change by offering best palliative and end of life care to everybody towards the end of their lives.

Dr Paul Barker is a GPSI in palliative medicine in Bournemouth. He wrote this article to promote Hospice Care week, earlier this month.


1) Lamont, E. (2001) Prognostic Disclosure to Patients with Cancer near the End of Life

2) Allen, L et al 2008. Discordance Between Patient-Predicted and Model-Predicted Life Expectancy Among Ambulatory patients with heart failure.