Professor Tom Burns CBE is well-known in UK psychiatry. He is Emeritus Professor of Social Psychiatry at the University of Oxford and has made significant contributions to the field of community and social psychiatry through his numerous publications, research and teaching. The first time I met him I was a trainee on the St George's training scheme in London. His intellectual rigour and compassion were remarkable. I met him next when he was conducting the controversial OCTET study on community treatments orders (CTOs). As a new consultant, I was quite amazed by the scope of the study. It is well-known that the study results showed no difference in readmission between the CTO group and the control group. It was subject to many objections and CTOs are still in use. Professor Burns has also been writing for the public, including his excellent book Our Necessary Shadow: The Nature and Meaning of Psychiatry. And he is a harsh critic of the DSM!

Thank you very much for this opportunity, Professor Burns. I wanted to start our interview with a question on CTOs. How did you come up with the idea of the OCTET study?

In 1992 I was on the Royal College of Psychiatrists’ working party in which we proposed community treatment orders. At that time, I thought it was a good idea and remained interested in it. When the Mental Health Act was reviewed (what turned out to be dubbed the Mental Health Act 2007), I was the psychiatrist advisor on the scrutiny committee, formed of members of the House of Commons and House of Lords reviewing the Act. Community treatment orders were discussed and finally included in the Act. I was aware of two RCTs [randomised controlled trials] in the USA which both failed to show that community treatment orders worked. However, I believed the UK care system was much better and so had a better chance of delivering results. I wanted to conduct a trial to establish if they worked in the UK. There have been dozens of observational studies but I believed the best way to generate evidence for the effectiveness of CTOs was an RCT. And that took me to the OCTET trial.

I remember clearly the graphs you presented showing that CTOs did not provide any advantage over the control condition. There were, however, colleagues who disagreed with the results. Did you expect the reaction you got from other professionals?

I didn't expect the strength of reaction but I was not surprised. Interestingly, before community treatment orders were introduced, the majority of professionals were against them. However, after their introduction they continue to use them and seem unwilling to change. What surprised me was the low level of research understanding in many clinicians I encountered. My belief was that if others disagreed with the results, they could, and should, conduct another trial. The history of medicine has repeatedly shown trials that have negated earlier ones. The resistance from professionals disappointed me, but I was not surprised by it.

In this very journal, there was an article defending CTOs. What were your thoughts when you read it?

I remember a couple of articles defending CTOs. I have to say we did receive some valid criticisms. The Lancet published three letters which were excellent. However, many of the subsequent articles in favour of community treatment orders were merely opinion pieces. That includes the articles in this journal. I think the way forward is not by trading opinions but by more research.

One thing I have encountered is the worry that professionals who do not use a CTO in case a serious incident occurs could be criticised.

That I found very disappointing. Maybe it is a feature of my age, but I continue to believe that, as a doctor, you are primarily responsible to your patients and not to hospital management. Clinical decisions should be based on best evidence and not management policies. With regard to criticism in case of serious incidents, how can someone legitimately criticise you for not doing something that has no evidence for its effectiveness?

In your talk to the Royal Society of Medicine you queried whether the DSM was a friend of psychiatry. Is it?

It is not. I think we have unwittingly accepted a model of diagnosis which is not what we were trained for. There is a book called Making Doctors by Simon Sinclair where the training of doctors is studied in detail. He reminds us that training in medicine is an apprenticeship. We go through this apprenticeship working alongside experienced doctors. We learn, in this way, to recognise disorders – especially when they present in atypical forms. Up until DSM-III, diagnoses were made on this basis of pattern recognition. DSM-III took a model where diagnoses were tightly, and even legally, definable. In reality, probably 20% of diagnoses are a matter of judgement and that is where the apprenticeship is most important and effective. DSM-III aimed to get rid of such judgements. A strong initial motive was that defining diagnoses this way would reduce overdiagnosis. In the end it has contributed to increased overdiagnosis rather than restricting it.

I was glad to see you acknowledging the role of pattern recognition in making diagnoses. I recently read in the Oxford Handbook of Clinical Medicine about different ways to make a diagnosis, and diagnosis by recognition made by an experienced clinician was at the top of the list. Perhaps we need to ensure psychiatrists spend more time with patients and focus more on clinical patterns than simply applying criteria?

This is true. Strength in diagnosis is our unique contribution to the team. As we get more experienced we recognise the patterns more effectively and help our patients better.

I also wanted to ask you about your opinion of the state of psychiatry at present. There have been several changes leading to fragmentation of services.

I think the past 20 years have not been good for the quality of care. I think we have to take much of the blame because of weakness of leadership. It is always easy to make arguments for new services. But someone with the overview ought to have said, hang on a second, there are limited numbers of staff and limited resources. We cannot continually shift resources from generic community mental health teams to new services without balancing the costs with the gains. Core services have suffered a lot from the constant remodelling and the fragmentation of the services. There was a failure in recognising the impact of the new fragmented service on the existing core teams. Patients need continuity and generally hate fragmented services. The last thing they want is to have to explain everything over and over again to new people, which usually includes difficult memories. We have to fulfil our core obligation of treating severely mentally ill patients before we mandate desirable, but perhaps not essential, new teams.

Another disappointment for me was the abandonment of the term ‘patient’ in favour of ‘service user’. We did it in order not to disagree with anyone. It is not a trivial thing. Different surveys have shown that our patients prefer the term patient to service user. We all become patients at some point in our lives. In addition, when you are negotiating for resources and parity of esteem, we can argue that our ‘patients’ have the same rights as medical ‘patients’. So, I am delighted that the College is trying to reverse this and hope it can get the message out loud and clear.

Professionals have been caught in a very complex bureaucracy. I can't help but ask your opinion on outcome measures, especially clustering.

I think clustering is a disaster. A complete mistake which I think also arose by failing to argue back. We know our diagnostic systems are not perfect but that is no reason to replace them with untried alternatives such as clusters. Thankfully I have heard that some organisations are moving away from them.

About time! Finally, what do you think of the preoccupation with risk in psychiatry?

It is not just in psychiatry. It is permeating all other areas of our lives. However, preoccupation with risk is particularly damaging in psychiatry. What works in psychiatry is collaborating positively with our patients for a good outcome (a good, trusting therapeutic relationship) rather than an exaggerated focus on negative outcomes. There will always be some risk. We need, at some stage, to trust professionals. All these risk assessment forms protect nobody.

Thank you very much for your time.