top of page

Uncertainty and forensic psychiatry


Interview with Professor Andrew Forrester


Andrew Forrester is professor of forensic psychiatry at Cardiff University and a consultant forensic psychiatrist at Oxleas NHS Foundation Trust, covering prison services in the southeast London metropolitan area. He is the editor-in-chief of the British Psychiatric Bulletin and of Medicine, Science and the Law, Chair of the quality network prison mental health services to name a few of his responsibilities. Andrew has worked for over 20 years within prison mental health services. I got him on a Teams-call to discuss how the element of uncertainty is a key factor in forensic psychiatry.

Páll Matthíasson: Dear Andrew, thank you for taking the time to speak to TNP. We are writing about uncertainty and no discussion on that topic and mental health would be complete without thinking about it in relation to forensics and forensic psychiatry. So it would be great to get your thoughts on that?

Andrew Forrester: Let´s try to unpack that a bit. Thinking about uncertainty within prisons, I‘ll talk firstly about what I´ve seen among people in prison. I‘ve seen a lot of people coming to prison in a very chaotic, disordered state, often using drugs, sometimes with major mental illness, after a period of offending behaviour, sometimes including violence. In that state there is often chaos, really. The prison itself or the criminal justice system sets about trying to make order out of that chaos. And it does that in a number of different ways. Of course health services have an important part to play in that too, by screening people, and identifying health conditions as they come through. The system sets about it in a very straight forward way which is to put people into situations in which they are subsequently detained. And healthcare does it in a number of ways, most obviously by identifying acute health conditions, including mental illness – and where it exists, sometimes transferring these people off to other parts of the system, including health facilities, hospitals and so on. But also, through the treatment of alcohol and drug withdrawal.

 Professor Andrew Forrester: "I‘ve seen a lot of people coming to prison in a very chaotic, disordered state, often using drugs, sometimes with major mental illness, after a period of offending behaviour, sometimes including violence." Image by Unsplash.
Professor Andrew Forrester: "I‘ve seen a lot of people coming to prison in a very chaotic, disordered state, often using drugs, sometimes with major mental illness, after a period of offending behaviour, sometimes including violence." Image by Unsplash.

And I would say that I have seen a lot of people, particularly, who have been alcohol-dependent and come into prison, who then have a period of detox and withdrawal and then are subsequently transformed by it. So in that rather straight forward way, the prison and health system working together, can deal with chaos and uncertainty at arrival into prison. But beyond that it becomes more difficult, because of course detaining people in prison cells, holding them there, is both containing, but at the same time puts restriction upon peoples´ freedom that can have an adverse effect upon them and their mental health. What I would say about that is that in my experience people respond very differently to imprisonment. I´ve seen some people who decompensate in prison, become acutely unwell, psychotic and so on. Probably as a direct result of the stress of imprisonment and all that comes with it. But then there are other people who seem to respond quite well to what we might call detention or detainment and to the rules that are imposed in prison. You get up at a certain time, you have breakfast at a certain time, you are allowed out of your cell at certain times, you have to follow the rules when you are in there and so on. As regards uncertainty more generally, I mean, it is certainly true that when you think about risk of offending behaviour going forward it is important to look to peoples´ background and what they have done in the past. So previous offending behaviour does to some extend predict future offending behaviour. That is not to say people can´t change, of course they can, but when it comes to predicting certain serious events, such as homicide, we are really not very good at that at all. So therefore there is a degree of uncertainty that persists and is inherent in any kind of system that involves health and mental health. That is of course what we have to tolerate. What I am saying really, is that in the end, when it comes to individuals it is about an assessment and reflection on a case by case basis, because every case ends up being complex, once you get down to the details of it.

So, there is certainly uncertainty both in the presentation that people have but also uncertainty about what will happen next, which is what you are often asked to predict.

To an extent we all face uncertainty, none of us knows what will happen next in our lives. Although we can have a degree of certainty, in the sense that one day follows another, life can be unpredictable for all of us. I think that for people in prison, and people in the forensic systems, of course the thing that´s taken away from them is their ability to self-determine – or, to put it another way, their ability to self-determine is constricted within the parameters allowed within forensic services or prison. And that can be difficult for many.

What about the idea that some people after being contained within a very structured environment for a long time get anxious when faced with more freedom, where the constraints are reduced. Is that something that you see?

I´ve seen a lot of that in my clinical work. It´s a dominant theme. My most recent work was in a category C prison, a sentence prison and I´ve seen a lot of people coming up to being released and being scared about being released back into the community, not knowing how to function there. Things such as how to manage a bank account, how to get a GP, and also, in some circumstances, facing hostility from the wider community because of their offending behaviour – for example, perpetrators of extreme violence or sex offenders returning the community. So yes, that´s difficult, there is a phenomenon known as „gate fever“ which is exactly that. Coming out of the prison people struggle in that period, and we know from the literature that the period after release from prison up to 6 months is a very difficult period indeed with increased self-harm, suicide, return to drug use and so on. So for example, people who have been off opiods may return to using opiods, and that can be very dangerous when they go straight back to the level of drug use that they had before. People with psychotic illnesses that are controlled, we know that they may struggle to get back into services on release, so there is a real-life component to that difficulty during that period. I would say that if you´ve got used to the prison regime and especially if you´ve been in for a long time, there´s an element of institutionalisation. So moving beyond that to greater freedom is certainly difficult for many, for understandable reasons.

"My understanding is that suicide risk assessment is not an area where scales are useful, and instead clinical judgement is important."

That makes sense. Now a third aspect of uncertainty would be the question of the profession knowing what it´s doing. The uncertainty of the efficacy and effectiveness of treatments. Is there an element of that in the practice of forensic psychiatry?

I think there certainly is, I think we have a long way to go in terms of our evidence-based treatments. Why I say that is that until now in forensic services and certainly in prisons we have tended to co-opt and assume that treatments that are relevant within general adult psychiatry are relevant within the forensic and prison services. That´s great, in a way that makes sense and in the UK we have tended to follow NICE-guidelines, e.g. if someone has schizophrenia, we have tended to follow the same guidelines as we would in the community. However, there is an element of uncertainty around that. The first one is that I see many prisoners who end up with multiple diagnoses, and complexity is the norm rather than the exception. So if you have personality disorder, plus psychosis, plus PTSD, plus substance misuse, then we are pretty poor at knowing when and how to sequence treatment. Where do we start, what do we do next and what do we do after that? We´ve tended to assume that there is a treatment hierarchy, you tend to start with the psychosis and then move onto other bits, but I think that we are not very good at deciding as a profession how best to do that. There´s another thing, when we have made assumptions about treatment within forensic and prison services and occasionally, much less commonly than we would like, we have submitted it to research, that research has often given us answers that we haven´t expected. Showing for example that the treatments that we expected to be effective amongst prisoners haven´t been as effective as we expected them to be. So I think there are big questions to be answered there. Psychiatry, understanding of the mind, understanding of treatments, and applying them to people in prisons and forensic services, we have a long way to go. I suppose this is a plea for more research amongst prison and forensic services to improve the existing evidence base for treatment.

All sorts of scales and ways of quantifying life are a way to change uncertainty to risk, something you can measure and predict. Do you think we have a long way to go there, in forensic psychiatry?

I think we´ve come far. There was a long period of focus on risk assessment within forensic services, with tools like the HCR-20 being dominant instruments and it´s certainly been helpful and effective and supported by a large body of literature, which is valuable. However, there is a downside as well, which is that sometimes there can be a focus on getting the risk assessment documents right, at the risk of the relationship with the patient, and I think that actually, these instruments are useful as long as they are adjunct to the rest of the care that happens. But when they become the main thing you focus on, then it is possible to lose the relationship with the patient inside all of that. So I think that balance is required - yes, risk assessment instruments like the HCR-20 are helpful, but at the same time they need to be a part of the overall package of care, not the overall care itself. Do we have further to go with that? Almost certainly. There are limitations with all of these instruments, and if you are not careful, you can have the feeling that you´re going on and on and collecting more and more information, but not getting anywhere with it. So another criticism with these instruments is that they contain a lot of historic information, and while that is fine in predicting risk and so on, how does an individual that has a history of this or that escape that history moving forward? How does one move on from that? So individuals can feel trapped sometimes by the risk assessment instruments and have difficulty demonstrating that they´ve changed – and yet, surely we can all change. That´s where clinical judgement comes in. Ultimately, therefore, there is still a role for clinical judgement, and I suppose the way I think about these instruments is that they are important adjuncts to clinical care and decision-making. They are not deciding vehicles themselves, they just help make decisions in a more structured way.

"Individuals can feel trapped sometimes by the risk assessment instruments and have difficulty demonstrating that they´ve changed – and yet, surely we can all change."

It makes me think of suicide assessments, and especially a paper that Keith Hawton from Oxford wrote in the Lancet in 2022. His conclusion was that we should get away from the scales because they cannot predict, it is about clinical judgement at the end of the day.

Yes, I think that message has definitely got out about suicide risk assessment, I don´t think any one is using scales to think about suicide risk, unlike risk of violence, where people are using scales. So there is a distinct difference, there is a body of evidence that supports the use of scales, especially the HCR-20 for violence risk assessment. My understanding is that suicide risk assessment is not an area where scales are useful, and instead clinical judgement is important. An important difference is often that violence risk assessment may be planning further ahead, thinking about what might happen over the next 10 years for example, whereas suicide risk assessment is often considering the more imminent, this evening or perhaps tomorrow.

Thank you for your time.


Commenting has been turned off.
bottom of page