The problem of prognostication in assisted dying.
30:48 Share

Welcome to the Medicine and Science Podcast from the BMJ. I'm Cameron Abassi, Editor-in-Chief. This week, we hear from a roundtable of experts on the Assisted Dying Bill moving through the UK Parliament. I mean, I think it's a real problem that this is the Terminal Ill Adults Bill. I think the public weren't looking for that. They were looking for an Assisted Dying Bill and really a bill that was about unbearable suffering.

And we take a dive into true crime, talking about the doctors who've committed murder. Shipman maintained his innocence to an astonishing extent in the trial, but he didn't know that morphine stays in a body in the ground for hundreds, if not longer, years. The UK government is debating legislation to allow assisted dying in England and Wales.

Other jurisdictions in the UK are also considering similar legislation. Scotland and the Isle of Man have similar bills moving through their parliaments. The bill being considered in Westminster puts doctors at the forefront of deciding if their patient will be eligible for a medically assisted death. The key criteria being a six-month prognosis. But is making a six-month prognosis clinically possible? Is it even realistic?

I'm joined by a panel of experts to discuss this question. Could you all please introduce yourselves? Hi, I'm Nancy Preston. I'm a Professor of Supportive and Palliative Care and Co-Director of the International Observatory on End-of-Life Care at Lancaster University. And I study how we deliver palliative care and also the end-of-life decisions people make, including around assisted dying.

Hello, I'm Scott Murray, an Emeritus Professor of Primary Palliative Care at Palliative Care in the Community at Edinburgh University and also a retired GP with a special interest in looking at the typical trajectories of decline at the end of life. Hello, my name is Simon Etkind. I'm a palliative care consultant and assistant professor of palliative care at the University of Cambridge with an interest in clinical uncertainty.

Hello, I'm Professor Suzanne Ost from the Law School at Lancaster University. Assisted dying is one of my main areas of research and I was the expert advisor for Jersey Citizens Assembly on assisted dying and the Nuffield Council Citizens Jury as well. Scott, perhaps I could come to you first and ask you about this business, about six months prognosis.

Thank you. Well, the bill does suggest that implicitly, implicitly suggests actually that it's possible. But as a clinician, as a family doctor, I would find it very difficult indeed. And really, I'm interested in illness trajectories or how people do spend the last month's years. And in fact, most people nowadays take many months, often years to die in that process. And it's not just getting iller and die, but it

any ups and downs on that process. And when you start there, they may die in the next six months or six months after that. Esther Ransom is still alive after two years. And so someone will likely die in some six-month period from the beginning, but which we don't know. So therefore, it's really impossible to say the person will probably die. We can't expect them to die in the six months. We expect them to be spread out over that period of time.

I mean, I guess that's something most doctors will understand and people who have good insight into this will understand how difficult it is to make that particular prognosis that someone's likely to die in under six months. But could you explain for the people who aren't as close to the issue as all of you are, why it's so difficult? It's just the nature of the illness that people get recovery and get better again and the various treatments as well.

But in the old days, people died quickly and predictably. But now that's not really happening. So as soon as you're at a certain stage, no one knows. The person will possibly pass away suddenly during a longish period of time.

And so from first principle, I'm not surprised that it's totally impossible to find any instrument to actually prognosticate at six months. We could do it at six weeks, possibly. Yeah. When someone is declining at the end there. And most people identified who will likely die in six weeks will die, 90% will die by six months. Yeah. Or even at four years, it's easier. But six months is particularly problematic. Yes. Yes. Okay. So Simon, if I'd come to you, if...

Six months is very hard to predict. Six weeks is much more, you know, it can be more solid in your prediction. How do we resolve this? How might we deal with this problem? Well, I think it's really challenging. And prognostication is not just about formulating how long we think someone has to live. It's also about...

communicating that to patients and families and having a and and reassuring them and giving them the information that they can then use to plan and the evidence is that whilst doctors make lots of errors in both directions with prognosis we on average as it stands tend to be optimistic and we we like to give people hope and we on average say that people might have longer and

what might change for example if if this sorry can i just be clear what you said we like to give people hope so we tend are you trying are you saying that clinicians tend to be more optimistic when they're talking to relatives and and patients is that what you said

So the evidence suggests that clinicians make errors in both directions when we're prognosticating. But we tend to err on the side of optimism. So whereas when we give a prognosis, some people...

may live much longer than that prognosis. On average, people will live shorter than we prognosticate. What this legislation may do is it may change that balance because if you have somebody who you think their prognosis is longer than six months but wants to access assisted dying, then there may be a pressure then to underestimate prognosis. So it changes the equation of how we communicate. In some Australian data of experiences,

of Australian doctors estimating prognosis over six months for access to assisted dying, they did find the cutoff arbitrary and found it difficult to operationalize. So either way, whatever target you have for this, the point is it will always be difficult to estimate. And as Scott was saying, we don't really, we don't have any reliable instruments to help you prognosticate in this way. So what do we do then, Scott and Simon? Do we, if,

assisted dying is to be implemented does that need to be taken out or does do some other approach need to be taken well the evidence that is there is that the

people who we would expect to die within six weeks, 90% of that group of people will have passed by six months. So that's using the Altitude Progress Indicator. And that one has evidence. So that could be used. Although I think a number of people advocating for this bill, you know, would not be content with that. No. Looking for longer prognoses. Although that's the bill that the public, general public, may be looking for, you know, people in tractable suffering at the end of life. Yeah. Yeah.

Okay, let's just switch tack a little bit. I mean, the other issue is talking about the disease paradigm. I mean, a lot of thinking around assisted dying tends to focus on one disease. You've got one illness, how long have you got to live with that? And it's all complicated by multimorbidity and frailty. Nancy, what's your thinking around this?

I mean, I think that's exactly right. And it's not just the complication of doing the prognostication, but what is the driver of the person seeking the assisted death? So what we've started to see in places like Canada, someone may have a terminal condition or conditions that meet the requirements.

But actually what's driving them might be social inequity. They can't afford the house or something like that. So the driver isn't the terminology. It's actually much wider even than healthcare concerns. So people are almost using that as a way to get the assisted death. So what do we do about that?

I mean, I think it's a real problem that this is the terminally ill adults bill. I think Scott's right. The public weren't looking for that. They were looking for an assisted dying bill and really a bill that was about unbearable suffering. But it's probably easier to get a bill through that is about people who are dying because somehow it's more palatable and it's a different type of suffering. Now, if you work in palliative care, I think many of us would say a lot of that suffering can be managed, can be helped, can be supported. And

You know, if you work in mental health and you have someone with untractable suffering there, they might feel similarly to us. But who are we to say what is unbearable suffering? And lots of countries where they have legalized this, it is not about terminology. It is about unbearable suffering. So what you're saying is fundamentally a different approach. The approach is being taken.

I mean, I think most of the world is seeing terminology as problematic. And I think everything that Scott and Simon are saying, it's difficult. And they've had problems in some jurisdictions where trying to find the right specialist to make that decision, who's a specialist in the condition or conditions, who is able to make a palliative care assessment, who's of sufficient level, really delays people getting an answer and getting through the system. So everything about this is incredibly complicated.

whereas if you're talking about someone's decision to say, well, actually, my life is unbearable in the same way it was accepted about withdrawal of treatment, withdrawal of ventilation. I'm not saying this is the answer, but I think this is what the public were looking for in the bill, and it's a very different bill to the bill we're looking at. Yeah, the other point is around how people...

will seek an assisted death. I mean, how does that happen? What about informed consent? What's the mode of death? These things don't seem to be specified at the moment.

I think the bill is purposely vague on this. And I think partly because we don't know what drugs are going to be used. They say this is based on the Oregon system. But if it's that, then this would be given orally and the drugs would take quite a while to take, so an hour, hours. If you go for something like the Swiss system, they set up an IV and the person's

self-administered by opening the port on the IV, and then it's a matter of minutes. So that's not specified. And again, I think purposely so. So not the drugs nor the route of administration. So that would need discussing with the person who is seeking the assisted death. And Suzanne might want to comment on that from a legal perspective. Yeah, I was going to come to Suzanne. I mean, clearly with the issue now, the way Nancy's described it,

that, you know, the whole consent process and the other uncertainties is that this opens up a whole kind of raft of sort of medical legal implications. Well, I mean, on that issue of informed consent, obviously valid consent at law has to be given by someone with capacity. Somebody has to give that voluntarily. And importantly, it has to be informed. And one of the things that we know that the bill would require of doctors under one of the clauses is

is that they should explain and discuss with the patient a number of different things. And one of those is about the nature of the substance that might be provided to assist the person to end their life, including how it would bring about death as well. So the patient would need to be given information about that. And the law would require doctors to give the patient information about that in a way that the patient can understand as well. So that could well be challenging. Okay.

The way you're both describing this is that, again, there are problems with the bill. So how do we get from where we are now to something that's actually workable?

I mean, I think that's what they hoped would happen in the committee stage. But I think Stephen Kinder, who's the Minister for Palliative Care, has said that in recognition that the bill is vague, they've given this extension from the two years to the four years implementation period. So my guess is that's when that is going to happen. And we're already getting these hints from the NHS and from others that this is going to go outside of healthcare, which is very much, I guess, the BMA's position and

And almost what happens in a lot of countries to an extent, there are certain doctors or nurses who are willing to be involved. It's the minority, but probably a lot of the challenge is identifying who those people are. Yeah. The other bit, sorry, I'd hope they looked at was explaining what reasonably expected does it mean? Should it be that means on the balance of probability this will happen?

or not because it suddenly seems to open the gate more further than a probabilistic one yeah and Suzanne I wanted to ask you um one of the issues is I mean Nancy just mentioned it as well around somehow she described it as happening outside healthcare another way to describe it is it's

is demedicalizing it. And I'm just wondering whether Suzanne, you know, how do we go about the medical? What's the non-medical model of assisted dying? Yeah,

So, well, there's a number of different models. There's inside health, there's a more hybrid model, and then there's completely outside of health as well. And one example of a jurisdiction where it is completely outside would be provided by private providers would be Switzerland, where you have the help from the right to die organizations who carry out assisted dying there. You could have a fully private model that was funded by charities, right

Right to die organisations sometimes will contribute in Switzerland, Dignitas does that. So that might be one way to go. Questions there would be raised of course about equality of access and how easy it would be to pay for that, although presumably having an assisted death here is going to be cheaper than travelling over to Dignitas and I think the figures I saw is that that's in excess of £12,000 to do that.

The more hybrid model might be one where you have NHS funding, but it's a separate service outside of the NHS. I'm thinking perhaps about, you could equate this to abortion and BPAS and the services that BPAS, for example, provides for early abortions outside the NHS, but funded by the NHS. And that also raises issues as well and concerns

If you were to go fully inside, then obviously you have concerns there about resources. And either way, whether it's hybrid and it's partially funded by the NHS or inside the NHS, you've got that real difficulty of how we then afford assisted dying services alongside all the other end of life services we have. And of course, palliative care as well.

Yeah, I mean, I just want to ask all of you this before we finish, which is the point about palliative care. And another issue, central issue here is, is palliative care optimised? There's an argument that we should be spending more on palliative care, improving palliative care services, in which case there wouldn't be the need for assisted dying. And I just wonder...

between all of you, I mean, where do you stand on this? And BMJ, you know, we're on record as saying we're supportive of assisted dying legislation and that we're also supportive of optimising palliative care services. But I wonder what your thoughts are. Simon?

Yeah, so I think there is a way to go with palliative care. I've just come off the wards to do this podcast and I was talking to somebody who wanted to go home and I had to explain to them the limited services that were available to support them.

when they go home to die. And we had a really difficult conversation about, well, what if I need urgent support out of hours and that isn't there? And I think there are very limited palliative care services and there are disparities in access. So I think there's certainly plenty of

room for improvement there lots of people who don't have access to palliative care who could but alongside that we do know that there are a a smaller number of people who who even if they have access to optimal symptom control may still feel they want to pursue assisted dying so it's how do you bring those together but i don't think we can we can't pursue one and not the other one and there's a a lot that needs to happen with palliative care yeah nancy

I think there's an element where palliative care becomes a bit of a red herring in this because I think a lot of this isn't always about a failure of a system. It's a fear of what my death might be like and it's wanting to maintain control as long as I can, including controlling my death. And we definitely need to improve palliative care services. That's a given.

But for a lot of these people, it's just a choice they might have made at the moment they were diagnosed. It's something they really think about. It's something they share with one or two people and then they go to access it. So it's not always about a failure of palliative care. It's about a fear of what dying might be like and a desire to be cured.

in control and what is crucial there is people need to be have access to someone with palliative care knowledge who can really explain what their death might be like because it might be very different to their mothers or their fathers or a bad death they've seen and they need to be shown how we can support them at the end of life so that they are making the most informed decision they can

Thank you, Suzanne. I think for me, it's particularly a funding issue. And my concern is that given that we have about two thirds of palliative care that's funded through charities, if we do go down a route where assisted dying is fully state funded, that really, to me, sends the wrong symbolic message. So we need to get the funding for both right. That would be my view on that. Okay. And finally, Scott.

I think it's a great opportunity, maybe in the next two, four years to take palliative care forward, especially in primary care, not just in hospices. But also, as Nancy was saying, letting people know, understand what it might be like for them the course of their illness, so they're not so scared about it. And also the assistance that might be available through palliative care and the broader community sector.

Okay, well, I think we'll all happily agree that we really haven't solved anything here today. It's just very, very complicated. But it's important that we get clarity on the issues that we've been discussing. And I'm sure that conversation is going to, well, we need to try and make sure that conversation does continue because it seems that the debate isn't actually pinpointing these issues in the way that it should be. Thanks for your time today. Thank you very much. Thanks. Thanks.

And Scott and Simon's editorial, Assisted Dying and the Difficulties of Predicting End of Life, as well as Nancy and Suzanne's opinion piece, Breaching the Stalemate on Assisted Dying, It's Time to Move Beyond a Medicalised Approach, are both available on our website at bmj.com. Dim the lights and raise the curtain.

There is a public fascination with Doctors Who Kill. It's deeply rooted in the public psyche and to give us a taste of their true crime stage show, An Appointment With Murder, I'm joined by Harry Brunges and Andrew Johns. MUSIC PLAYS

It's the year 2000 and Dr Harold Shipman stands in court accused of murdering 15 of his patients with the suspicion he may have murdered many more. It takes the jury six days to reach their verdict. We find the accused Dr Harold Shipman guilty of 15 counts of murder. The judge passes down 15 life sentences with the recommendation that Dr Harold Shipman is never released from prison.

The year is 1957 and Dr John Boccan-Adams stands in the dock accused of the murder of one patient with the suspicion that he may have murdered many more.

It takes the jury 44 minutes to return a verdict. We find the accused, Dr John Buckingham Adams, not guilty. Dr Buckingham Adams walks from the dock a free man. The judge later remarks, the rigorous standards of the law sometimes mean that the guilty walk free. Two doctors. Both loved and trusted by their patients. Two of the greatest serial killers in British history. Buckingham Adams is alleged to have murdered over 150 of his patients. Dr Harold Shipp.

is estimated to have killed over 250 of his patients. So what happened? Why did they do it? How did they get away with it for so long? Why was one imprisoned whilst the other walked free? And could it happen again?

That was Harry Brunges and Andrew Johns with an extract from their stage show and podcast, An Appointment With Murder, which is a name that speaks for itself. And they're both here in the studio to tell us more about how it all began. Harry. Well, this all started now a decade ago, which we can't actually...

actually believe it's that long ago. I was a junior surgeon, SHO, at St Mary's Hospital Eastbourne in 1983 when this notorious GP called John Boggan Adams was admitted as a patient and died. And in the last few years, I've lived in a house which he was a regular visitor to. He was the doctor for the family who lived there. So I read about him and his alleged multiple serial murder

murders of patients, which he was actually acquitted. And Andrew, who's had a career as a forensic psychiatrist, initially abroad more than the Maudsley, gave evidence to the Shipman Inquiry. So Andrew and I wrote an essay for the Royal Society of Medicine on the comparative psychopathology of these two. Did we not, Andrew? We did, because we both realised that we knew rather a lot from our respective professional careers.

placements about two medical serial killers. So we thought this could be slightly interesting. Shipman was all over the national press but not fully understood. Most of the British public these days would have forgotten about Bock and Adams who was 1958. So we thought there's mileage in explaining this to the great British public. So that's how we started. You wrote an essay and it became a stage show. Well, I suppose I will probably take the blame from that having theatrical antecedents. We wrote it and I

I remember saying to you, Andrew, if we actually had to pace this and screenplay it, it could actually work as an hour's piece and we could do it as a two-hander. We never thought we would be approaching 60 years old and we'd have our inaugural performance at the Edinburgh Festival. Given that we're not trained as actors, that was an interesting experience. But Harry would claim to be an actor. Come on. Well, actually... You might not be trained. Well, I've...

been in and out of the perilous world of show business alongside medicine my whole life but Andrew is naturally an entertainer and a wit and a raconteur so we were both quite comfortable with it Were you filling a 150 seat theatre? That wasn't the worry the worry for me was sticking onto the script so like Marilyn Monroe I had to have bits of cues pasted around the stage for me to walk around and look at

But you're used to that from medical school. Well, I know, and learning all these side effects of, you know, drugs. There were some funny moments. In one show, all the lights, all the electricity in the building went down. It was a five-minute gap. And eventually, everything came back on again. And Harry and myself had lost our places in the script. Harry turned to me and said, where are we? And I said, Edinburgh. LAUGHTER

And there was another, on a slightly more sombre note. We're very careful in the show not to be respectful of the relatives of the deceased and the deceased people themselves. We take the piss out of the doctors, if I'm allowed to say that. You can say that, Mbiondje. And then after our first show in Edinburgh, a rather lugubrious woman came up to us and said...

Dr. Shipman killed my mother. Oh dear. And we can cope with that. Finding humour in some, you know, in problematic, emotionally troubling situations is how you get through medical school and your life as a doctor. And also these two doctors were almost comic characters. Bocan Adams was famously bumbling, not good enough as perceived for hospital medicine, sent to be a GP in Brighton.

He was a jolly... Eastbourne, I heard. Eastbourne. He was a jolly chap who enjoyed shooting and ingratiating himself with widows. But he was a hopeless doctor. During the war, he applied to be the part-time anaesthetist for the wounded folk. But he wasn't allowed to because he was so incompetent. Oh, God. Oh, God. Yeah, just as well. About Bodkin Adams, people know more about Shipman than Bodkin Adams.

There was a strong suspicion by the sounds of it that he had... How many people did he murder? At the time, he was alleged to have murdered a patient called Edith Murrell. And the prosecution at the Old Bailey reserved the right to proceed with a second murder of a woman called Bobbie Hullett. The first case, the prosecution went significantly wrong and they lost and they didn't proceed. But as history has gone on,

the general view is that he could have been responsible for slowly killing about 150 patients. And he benefited from an extraordinary number of wills of the old ladies who he's looking after. Do you recall the number? Well, I can do that for you because when the Scotland Yard looked at the case, they investigated 150 deaths between 1946 and 1956. And Bodkin Adams had been a beneficiary 131 times. Gosh.

The story of Bodkin-Adams then does have echoes with Chipman, very clearly. Yes, it does. Because Bodkin-Adams, we have two high-profile doctors going for high-profile trials. In the case of Bodkin-Adams, as Harry has hinted, the prosecution was flawed and there is a suggestion that the government stepped in to

Kibosh the prosecution it was very confident. With Shipman entirely different. The police exhumed 15 bodies They found lethal doses of morphine so they had a solid case. Yeah, the CPS wanted to prosecute more than that but this They took the view the 15 was enough. So there are 15 on the indictment and Shipman maintained his innocence to an astonishing extent in the trial the experts saying he was

poisoning people with morphine, he stuck to his view that he was a good doctor, he criticized the trial experts,

but he didn't know that, for example, morphine stays in a body in the ground for hundreds, if not longer, years. Oh, years? Yes, because morphine has been found in the body of Egyptian sarcophagi. And we're disappointed, Dr. Barsi, that you don't know that. I should have watched your show. I mean, one theme that you're touching upon, I mean, you didn't actually mention it in Race and Shipment, but he was also very well thought of

By his patients. And similarly with Bodkin Adams, if you're saying that all these people left their money to him in their will, then that must have been a similar situation. I mean, that shows the extent of trust, sort of blind trust, patients have in their doctors. I guess those were more paternalistic times. And Shipman...

Practiced as a single-handed GP. His patients were very loyal to him. I think he was single-handed so he wouldn't be observed by a colleague. But when the police started investigating Shipman, the police officer concerned was spat out on the streets in Hyde by people who said, you're being nasty to our good doctor.

And a fellow doctor offered to pay his bail money. So he was astonishingly well thought of. Yeah. So if people tune in to your podcast, what will they get? The podcast is, they will be getting aspects of our actual show, but we've now got the show runs, including a Q&A for an hour and a half. We've actually put 12 hours in the can now for the podcast. But is it sort of themes or how does it? Yes, we go. So we do a podcast.

podcast on the early life of Balkan Adams. Then followed, moves on to how were their crimes detected, how they come to notice, how they were investigated, the trials themselves. I do a whole programme on the psychopathology of the doctors. What do we think was going on that allowed them to commit such crimes? We do another programme on could it happen again? And a big one on the diagnosis. And we do another episode on previous doctors who were murdered.

We actually do the trials in great detail as well. Okay. It sounds really fascinating. If people want to hear the full show, yeah, where is it? It's called Appointment with Murder. If you want to listen to it, all you have to do is go to the crime desk at Daily Mail and it will tell you everything you need to know. Thanks to Harry and Andrew for joining us today and you can find more information about their show in the programme notes.

That's it for this episode of Medicine and Science. We'll be back in a fortnight, so subscribe wherever you get your podcasts. I'm Cameron Abassi. Thanks for listening.