The problem with predicting the future.

By Dr Joseph Hawkins, Consultant in Palliative Medicine, Clinical lead for End of Life Care, Ashford and St Peter’s NHS Foundation Trust. Twitter: @JoeHawk75825077

we are entering a world in which we can’t expect easy prognostications of the future… we will see a rising number of sick survivors-frail individuals of all ages whose survival is predicated by a frequency of return to hospital and for whom hospital may be the resting place but when… is hard to know.”

Hospitals are the meeting points of the sick and dying and the invisible line between each is a cause of more distress and misunderstanding than almost any other quandary in medicine. As Palliative Medicine physicians we often find ourselves teaching on how to spot when people have crossed this line. Yet despite all of our experience it can seem sometimes that the more the topic of dying is discussed the harder it becomes to define. This isn’t truly surprising, after all the end may be the same but the path travelled is different for each of us. 

In recent times the ‘Would you be surprised’ question has been challenged and it has been suggested that it is less reliable than initially thought when grossly screening our patient population for a better understanding of their future. During a discussion about the would you be surprised question, a colleague suggested to me that the phrase ‘sick enough to die’, posed as a question might be a good alternative. Although I should add that they acknowledged that phrasing sick enough to die as a question is potentially no different than the intent of the would you be surprised question.

I think the phrase sick enough to die, whether used as a question or statement is an excellent one. It neatly encapsulates the sense of uncertainty and possibility in a few short and clear words. However, it’s not really a prognosticating tool. Whilst helpful for prompting one to think about uncertainty or acknowledge dying, there are times when sick enough to die may not fit the situation in the mind of the clinician or observer.

I’ve often considered how difficult it is to answer the question of when dying starts even when it might reasonably be considered part of one’s own job. There is an inherent fallacy in medical plans- the fallacy that x will lead to y. In reality clinicians are aware that bias leaks in to plans, unconsciously subverting the outcome for individuals. I suspect that in an attempt to compensate for these biases many clinicians will cling to those things that can’t be readily disputed. Like world building in a computer game each action completed in a rigid order to set things up for an expanded future. Only in real life the rules aren’t set and the future even less so.  

So if we can’t (shame on us), reliably look at a person and say they have less than a year to live, then how do we go about predicting dying? For me it’s a recognition of diminishing returns. The steady shape of entropy’s course on the map of life. With each disease there is a shape -often marked by sign posts written in pathology results, scans and courses of therapy no longer effective against their pre-designated destination. Whether it is the accumulation of diagnoses or the increase in dependency the tell-tale signs are there to be read. 

I need to point out my awareness that whilst I make the above statements with a sense of surety, it is a surety tempered by the knowledge that I can be wrong each time I discuss this topic with a patient. For accuracy I therefore start almost every explanation about prognosis with an acknowledgment of fallibility. Thankfully it’s generally a good thing when I’m wrong. 

Perhaps all of this means that when it comes to neat phrases that describe prognostic uncertainty we are simply beyond what can be wrapped up by the same handful of words. When it comes to communication the more succinct and clear the message the better the understanding and this is a complex message by virtue of having to encompass every form of dying. 

In 2019 I enjoyed reading Dr David Oliver’s piece on the subject of discussing dying 1 where he acknowledged the needed to both reflect a possibility of dying and surviving when discussing death with unwell inpatients as a consultant geriatrician. His own thoughts on this were informed by Dr Kathryn Mannix’s book the end in mind 2. Taking both of these respected communicators words forward with some thoughts of my own I suggest that we are entering a world in which we can’t expect easy prognostications of the future. Where instead we will see a rising number of sick survivors-frail individuals of all ages whose survival is predicated by a frequency of return to hospital and for whom hospital may be the resting place but when that time may come is hard to know. For these individuals the reflection of uncertainty when discussing the future is imperative in engendering choice and so it is important to ask all those who may be responsible for their care to recognise and consider having such a discussion that may lead on to advance care planning. 

The question remains-how do we phrase a term that accurately asks the question of our peers about whether the patient they are seeing is more or less likely to predictably die in the next period of months rather than years? I suspect that we don’t have the luxury of punchy phrases any longer for all of the reasons covered up to here. So maybe we simply have to ask: do you think the available therapies are showing diminishing returns, or if something sudden might predictably occur and not be likely to be reversible? If so-perhaps a conversation about the future and about dying would help make that future a little easier for the person in question. 



  1. Oliver, David. What to say when patients are “sick enough to die” BMJ 2019;367:l5917
  2. Mannix K. With the end in mind. William Collins, 2018.

Also by this author:

Welcome to the department of rebrands.

Creation of a new palliative care ward-a one year retrospective

How to perform a palliative exorcism

(Visited 1,176 times, 1 visits today)