Decisions to detain: an interview with Dr Elizabeth Fistein
Dr Elizabeth Fistein is a psychiatrist whose research interests include clinical, legal and ethical perspectives on compulsory treatment. She completed her PhD at the Cambridge Intellectual and Developmental Research Group http://www.ciddrg.org.uk/people/ research-associates.html.
Interview by Lorna Rouse.
Could you start by briefly describing your research interests?
My research interests developed out of my clinical experiences. When I was training as a psychiatrist I got very interested in the way decisions were made about whether or not people would be committed against their consent. There was a big debate going on because the law around the safeguards for compulsory treatment was being reformed, so it was quite a live issue and I got very interested in how that played out in practice. I wanted to research the way compulsory treatment might be justified because I felt personally that the way the law was being reformed didn't make a great deal of sense. I wanted to see if there might be a better way of arriving at a framework with some reasonable justifications for intervening in people's lives with compulsory psychiatric treatments.
What were the reforms?
These reforms grew out of a couple of incidents where people were harmed or killed by people who had a history of mental illness. There was a lot of uproar in the media and a sense that something must be done. In fact these are very rare incidents, not necessarily predictable, and I was a bit worried that they were driving a whole raft of reforms that would have an effect on a lot of people.
What was the topic of your dissertation?
I'd started out looking at the way compulsory treatment was regulated in other countries, so that gave me a way of comparing frameworks here with those in other countries. I wanted to build on that and find out a bit more about the way that frameworks here were actually applied in practice and the way the concepts behind the frameworks might be conceptualised. I then wanted to explore how differences in that conceptualisation might lead to different interpretations. The central psychology aspect of my research was seeing how, in the real world, people made the decision whether or not to detain a patient.
Was that your main research question?
It was probably one of two main research questions: how, in theory, might compulsory detention be justified and how is it justified in practice? I wanted to identify the gap between theory and practice. I think there's probably some kind of practical wisdom in the way experienced practitioners go about what they do, which could inform theory. Additionally there are probably things that practitioners might do unthinkingly and theory could inform that too. My idea was to get an understanding of the real and the ideal in a way that would hopefully improve both our theoretical reasoning and what we do in practice.
Was there much existing literature on the topic?
Not a lot. I think that's partly because it's actually quite difficult research to undertake. It's a hard to reach phenomenon. There's been survey research polling clinicians and asking them how they make decisions and there's survey research using vignettes asking clinicians, 'would you detain a person under these circumstances and why?' That's interesting research and it throws up certainly the suspicion that people have their own extra legal reasons for making decisions that go beyond the criteria. Sometimes they will hold themselves to a higher standard and won't necessarily detain people who legally would be eligible for detention. At other times it looks like they might be tempted to lower the bar for themselves, for perhaps clinical concern or perhaps concern about comeback. With the survey research all we have is the clinicians' accounts and I was very interested in seeing what actually happened. I wondered if after the event people might be, quite understandably, reconstructing a justification and wanting to present themselves in a certain way. I decided to take a more ethnographic approach and go out and observe people actually deciding whether or not to detain somebody at a Mental Health Act Assessment. Before someone can be detained or compulsorily treated they need an assessment involving two doctors and someone called an Approved Mental Health Act Practitioner (AMHP). That used to always be a social worker but now people from other clinical backgrounds who aren't doctors can train to become AMHPs. The new legislation actually got enacted just as I was beginning the project, so I was able to observe people putting this new legislation into practice.
How did you carry out your research?
I arranged to go along to Mental Health Act Assessments, make notes while the clinicians were interviewing the person they were thinking about detaining and then afterwards record the discussions in which they made their final decision. I used a three stage approach to my analysis of those transcripts. I went through and did a kind of thematic analysis looking at the topics that were raised, so the kind of things that were discussed and therefore made relevant to the detention decision. Probably unsurprisingly most of those did relate to categories in the legal criteria, but interestingly an extra topic that was discussed was whether or not someone had capacity to be detained. This actually, and slightly controversially, isn't a criterion for detention under the Mental Capacity Act, but the clinicians were using it as an extra legal criterion.
I then drilled down deeper using discourse analysis and some elements of conversation analysis. I wouldn't say that I did a proper conversation analysis on my data, but I looked at the way in which themes were discussed in order to get a richer description of the process; the way in which they were raised and then discounted or made relevant. By looking at the response to the way the topic was raised you could gain an understanding of the function of raising that topic. As an example, someone raised the idea of whether the person had a personality disorder. From the way the next four turns of the discussion went it became evident that the idea of personality disorder was raised because all the clinicians had this idea that they wouldn't detain the person if they felt they had a personality disorder. This was because they felt it wasn't a treatable condition, so they weren't justified in detaining someone if it wouldn't make a difference for them. It was raised in order to be discounted and to then move on to justifying the decision to detain. It was really interesting, taking a broad approach to looking at the themes, then focusing in closer to see how those clinicians' criteria to detention were really shaped in the conversation and what was truly relevant to the decision to detain.
Then I conducted a series of narrative interviews to get a bit more context to allow me to analyse my transcripts of the decision making process. I used an interview technique called biographic narrative interview methodology. This was great for me as a novice qualitative researcher because whilst it's a very open qualitative research technique, it's a very clear process for conducting the interview. It's a full methodology, so in fact if you're doing it properly it's not just a process for conducting the interviews but also for interpreting the data. I made a decision only to use the data collection aspect and then use a kind of thematic analysis and discourse analysis approach to analysing my data.
What is different about that interview style?
The interview style is based on a two stage process. You start by very carefully constructing an initial question designed to generate a narrative which is relevant to the research question. The aim is to generate a biographical narrative. You can get someone to tell you their life story, that wasn't really relevant to my work. What I wanted was their professional life story. I could then focus in on their experiences of being involved in compulsory treatments. In the second stage you pick the first and the last thing the person talked to you about with the idea that this is relevant because that is how they've chosen to frame their story. You then hone in at points along the story which either seem relevant to your research question or relevant to the person. Those points in the narrative are picked out and a line of questioning is followed designed to generate a narrative for a particular incident, to almost get them re-experiencing that point in their life story and telling you what the situated subjective is for them at that point in time. This is on the understanding that you're not really getting that moment; you're getting a recreation of that moment. Putting all that together you can create a picture of the person, what's important to them, what seems to have shaped their behaviour and the values that shape their behaviour. It's a psychosocial research methodology. You think about the individual themselves as well as their context in society.
That's interesting. How did you find using that approach?
I found it a really a helpful method to use, it really did generate interesting interviews and it was a reassuring method to use.
Did it inform the other analysis?
It did, yes. Fairly consistent across my interviews was the idea that for some cases it was a straightforward yes you would detain a person for compulsory treatment, or no you absolutely wouldn't. Then there was this grey area in the middle. When people talked about the decisions that they found difficult, there were common themes. It seemed that within the group of people I was interviewing, there was an idea of a sort of framework of things that were relevant and if those things were clearly there or clearly absent the decision was easy. When it was a bit more equivocal, those were the difficult decisions. There was a fairly clear list of things that people had in mind, unsurprisingly a lot of them did relate to legal criteria. Perhaps a bit surprisingly there were these other ideas, like the idea of decision making capacity or whether the condition was treatable, which has been actually taken out of the legal criteria.
Why was it taken out of the legal criteria?
That was a controversy that had arisen from a particular public case involving a man with a personality disorder considered untreatable who committed a murder. It was felt that the action that ought to be taken was to amend the Mental Health Act. The unforeseen consequences of that would be possibly bringing a lot of people under the remit of the Mental Health Act, impacting on services, resource issues and justice issues. It seemed that the people I was interviewing were mindful of that point, particularly with decisions that they characterised as difficult. My interview data allowed me to generate a framework of things that people appeared to be taking into account when it was clearly met, not clearly met or in a grey area and how they handled that grey area. Then I could go back to my observational data and see how that appeared to be handled when people were making decisions in real time. The interesting thing was that in the interview data the clinicians presented themselves in a particular way, which I suppose would be consistent with fairly widely held values in the profession. In the actual observation data you can see that it's much more difficult. In the heat of the moment, with concerns about safety, people would be swayed by things that in the interview they would characterise as a clear-cut decision.
That's interesting. Do you think that the reactions of the media had an impact on clinicians' decisions?
I think it did and one or two people did say that in their interviews. One psychiatrist said she felt that they all got to a point where they felt they had to watch their backs with a particular decision - it was a gentleman with schizophrenia who had said something inappropriate to a teenage girl on a bus and she felt there was a lot of anxiety coming from the other doctor at the interview that eventually communicated itself to all of them. With that heightened anxiety she changed her position; initially thinking she didn't want to detain this man and then deciding that she ought to detain him.
Did you have any difficulty getting people to take part in your study?
I was surprised at the number of people who were quite willing to take part in lengthy interviews which were sometimes up to three hours. In terms of going out to observe the assessments, that was more difficult. I was impressed by the courage of people who were quite happy to lay their practice on the line and have it analysed. It was notable that the kind of assessments that I was able to get to were almost all female patients. A lot of them were in more contained settings like in a hospital setting or designated place of safety and one took place in a police custody suite. With those taking place in community settings or where there were concerns a person was more agitated or distressed there was a lot of reluctance to have an extra person there. That was another reason for incorporating the interview data, to at least get some understanding of how people characterise what it's like taking part in those more chaotic assessments.
Did people contact you when they were in the kind of situation that you were interested in?
Yes, I advertised by sending out emails, going along to regular meetings that doctors and social workers or AMHPS had within the Trust and said 'if you've got a Mental Health Act assessment coming up and you're happy for me to come along, give me a call,' and I was happy to go out any time of day or night. In practice they were almost all daytime, I think things just get more frantic with out-of-hours assessments. I spent a lot of time hanging around what they call the 'section 136 suite' at Peterborough. That's where people are brought if they are picked up by the police because they appear to have a mental disorder and they need to be brought to a place of safety for further assessment. I think that's the thing about ethnographic research, you have to be prepared to do a lot of hanging around and do some work building trust with the groups involved so they are prepared to have you there. Coming from a clinical background was an advantage. Potentially it had an effect as well because often meetings involved people I already knew, so I wasn't a totally objective observer. For that reason it was good that I had an objective recording of what people actually said. Just using my own notes would have introduced a significant component of bias into the data. I think it was very worthwhile, although it was a struggle getting permission to record these meetings.
Do you need permission from the patient?
That was really tricky and I still feel that was quite a controversial aspect of the research. Initially I wanted to record the patient interview as well as the clinician discussion and when I went for NHS research ethics approval that was rejected. They felt that it would be inappropriate to record in those settings because a person being assessed under the Mental Health Act wouldn't have the capacity to consent to be recorded. I agree that it is difficult; I wouldn't agree necessarily that you can make a blanket decision that a person in those circumstances wouldn't have the capacity to consent to the situation being recorded. I wouldn't agree necessarily that even if they didn't have the capacity there wouldn't be the potential to make the recording anyway because I think this kind of research is very important. What goes on in these assessments is a black box and making it more explicit would, in the long term, probably benefit people with mental disorders. It would improve the quality of these assessments, potentially making them fairer. It would be a way of making explicit potential problems that might arise, like the way that the people doing the assessment might be swayed by other considerations and improve training to reduce the risk of that happening. There is a good argument for saying that that is a worthwhile thing to do, but I can see the argument on the other side as well.
The NHS research ethics committee in some ways was really quite helpful to me because they said they thought what I was proposing was service evaluation rather than research. I've got mixed feelings about this. I have a worry about this process because once you hive off projects into service evaluation they're subject to much less regulation. There's a potential for removing safeguards for people who really are research participants. I think there are some really understandable reasons why those suggestions would be made and also, now that I am a member of a research ethics committee myself, I can see the burden that these committees are under; wading through all the paperwork that's required and trying to make a decision is a really difficult process. So I can totally understand why the situation developed but I still have some on-going concerns about it. The upshot of all of this was I conducted my project as a service evaluation project within an NHS Trust and I think it's quite good in a way because it meant I needed to develop a report of my findings to feed directly back to the Trust. People had the potential to see good practice and learn from that and also to see some potential pitfalls.
What were the pitfalls?
One I identified was that patients were being potentially admitted informally to hospital because people were bending over backwards not to detain them due to the stigma of being detained under the Mental Health Act, emphasising that they could then leave hospital if they wished to. In practice, I felt those people wouldn't have been able to leave hospital, they would have been detained if they'd tried to and were being potentially deprived of some of the safeguards that are in place for detained people having their detention reviewed by an independent panel. Doing the work as a service evaluation tied it in much more closely with the clinical practice that I was observing, but equally I wonder about the safeguards for the participants. As a clinician employed by the Trust it was felt legitimate for me to be at clinical assessments, it didn't involve a breach of confidentiality and I was observing with the purpose of collecting data to drive service improvement. In fact for me, my primary purpose was collecting data for my PhD.
So it sounds like your findings had implications for the participants, do they have implications for the legal processes?
I think they do. I think there's some practical wisdom in the fact that so many practitioners take into account factors like the person's capacity to make their own decision whether or not the treatment to be imposed on the person is likely to actually have some benefit for the person. It's got an intuitive appeal and I think the legislators could take notice of that and incorporate that into our legal framework. It also ties in very nicely with the theoretical justifications for intervening with someone's liberty. I think there are potential implications for legislators. The difficulty then is disseminating that information. I don't think there is going to be much will to go through a further process of reform of the Mental Health Act in the very near future because it is such a drawn-out painful process. It also potentially has implications for mental health law reform in other jurisdictions.
So your design sounds quite complex. Were there advantages and disadvantages of bringing all those different methods together?
Designing the project was a long drawnout process. I had a lot of help from my supervisors and we went through lots of iterations. It was all based on a model that's termed empirical ethics, which is a model for bringing together practical observational empirical data about the way ethical values are enacted, theoretical reasoning about the underpinnings of those values and iteratively running around a cycle of comparing the two. That was a really useful model and I based the design on work by Tony Hope and John McMillan. Then I had to develop some way of fleshing out that model practically. How was I going to collect the kind of data that would be useful to my research question? That is why this kind of monster of a complicated model for the research developed. I'm glad I did though because I think it enabled me to do something that had the subtlety required to approach my research question in a way that satisfied me. A simpler project design would have felt like too much of a blunt tool. It did create nightmares for me the whole way through! I had to go and get a lot of training in different methodologies to draw it together and when writing up my thesis I had to really think about how I described the work I pulled together.
You must have had an awful lot of material.
I think there were 144 pages of transcribed data to wade through! I then had to think about how I could really clearly justify this quite complex model that I had developed, which had mixed a lot of different approaches together. I think there was a unifying approach underneath it all and that was this idea that's developed from philosophies like Wittgenstein and Austin around how the way we use language shapes our concepts and our conceptions and then shapes how they have an effect in the social world. Both the philosophical aspect of my work when I was looking at the theoretical justifications and then the clinical work shared that common underpinning and I think that's the logical coherence to the project as a whole. I was looking at how the relevant concepts were developed in theory and at how those different conceptions of those concepts played out and were used by the practitioners in real world practice. There was a single underpinning idea that brought together aspects of looking at theory, looking at different laws and different jurisdictions and then looking at practice in one area of jurisdiction. On the face of it they look like three really quite different projects, but in my mind they were all about thinking about the way we use words to create a justification for action.
Did any questions come up from the project that you would really like to investigate further?
I've got really interested in the idea of choice and patient choice. It's clearly a really important concept in policy. Intuitively it's really important to incorporate choice into practice, but when you look at it choice seems to mean very different things to different people. Depending on the kind of meaning you give to choice it can have very different effects in the world. Saying, 'well it's someone's choice' can almost be a way of abandoning responsibility for that person. Even what looks like a choice, it's a very bare idea of choice - someone's done that so it's their choice. When you look at the kinds of choices that we respect and the kinds of choices that we intervene in, it becomes incredibly complex. So I'd really like to do some work exploring the idea of choice as used in health policy and then the way respect for choice is actually practically enacted in the mental health sector.
That's been really interesting, thank you.