Tuesday, 20 December 2011

Exploring the Underlying Processes of Mental Illness - An interview with Dr Warren Mansell on the Transdiagostic Approach



Hi Warren. First, congratulations on winning the May Davidson Award 2011.  Can you tell us a bit about yourself?

I completed a DPhil at University of Oxford on cognitive processes in social phobia with David M. Clark and Anke Ehlers in 1997. Since then, I have developed an interested in processes shared across psychiatric disorders, how to explain them using control theory and also how to understand the mood swings in bipolar disorder and develop new forms of cognitive behavioural therapy. I have been based at the University of Manchester since 2005, where I do a mixture of research, teaching, training and delivering therapies.

How did you become interested in mental health?

I was always interested in why some people have very different emotions to others. Why do some experiences, like a racing heart, make people shaken to their bones with fear, and other people relish the feeling? I wanted to understand this properly and help people change the way they think about their feelings – if they want to!


A lot of your research is on the Trans-diagnostic approach to mental disorders. Can you explain what this means?

It is a hypothesis that the processes that are shared between all the different psychiatric disorders are more important to identify and address in treatment, than the processes that are specific to certain disorders. You could apply this to biology (e.g. the COMT gene) or family processes (e.g. hostility) just as much as to cognitive processes (e.g. worry, avoidance, self-criticism).

Can you tell us a bit about your research into and the evidence base for the Trans-diagnostic approach?

Allison Harvey, Roz Shafran, Ed Watkins and myself conducted a systematic review of the cognitive and behavioural processes implicated in maintaining Adult Axis 1 psychiatric disorders. This is a book published by Oxford University Press in 2004. We actually found evidence for between 12 and 15 different processes that are transdiagnostic! More importantly, we couldn’t find any consistent evidence that there were processes specific to a specific disorder! It appeared from our analysis that psychiatric disorders are largely different because people have different contents of their concerns (e.g. evaluation by other people; contamination; weight and shape) rather than there being any differences in processes between disorders. Some processes seem more exaggerated in certain disorders – e.g. worry in generalised anxiety disorder – but this process is more prevalent in other disorders too.


As well as identifying these 12-15 trans-diagnostic processes, you have also identified a single mechanism that these seem have in common – Inflexible control. What is inflexible control?

I am not sure whether ‘identified’ is quite it. I have a hypothesis, guided by Perceptual Control Theory (PCT; Powers, 1973) that what these different processes all share is that they are ways that people are trying to control their experiences that are particularly ‘rigid’ or ‘inflexible’. Essentially this means that these processes are carried out with little regard or awareness of their effects on important personal goals that the person holds. For example, what makes the suppression of emotions a problem is not doing it per se – as a doctor or as a psychological therapist we sometimes need to put our own feelings to one side when imparting bad news, or listening to other people’s concerns. It is only when suppressing emotions starts to encroach on one’s life goals that it is a problem. For example, if a person chronically suppresses their anger this might not be a problem until they get bullied at work and need to assert themselves to stop it happening. So, all of this implies that the way to help people with mental health problems is to help them notice what are the ways of controlling that are working for them, and what are the ones that are truly impacting on their life. 

We help people to monitor them, and regulate them in a way that allows them to achieve their life goals, rather than being stuck in controlling things at a more concrete level. A lot is said about ‘living in the moment’ but this is no use if you are doing things in the moment that are jeopardising your future plans. Truly adaptive thinking involves focusing on the present moment to fulfil your broader values and life goals that extend into the future and extend to other people. As you might imagine our therapy sessions can be quite expansive dialogues! We call the therapy Method of Levels – because our goal is to help people to shift their attention to different ‘levels’ of goals they may have, rather than inflexibly being stuck in a rut.

As you note in your book, 'Cognitive Behavioural Processes Across Disorders', you must still account for the differences in mental disorders, how would you explain these?

Ah, I did this above. The irony is that the classification system could be very accurate and yet still not tell us a great deal about what is the correct psychological mechanism to target in therapy – because this mechanism is shared across everyone and does not account for their differences. It would account for differences in severity but not in type of disorder. So, to test this properly, disorders would need to be matched for severity I imagine.


What are the implications for treatment, based on this knowledge? Do you think we should be moving towards metacognitive therapy and method of levels?

I think I have mentioned this above. In fact, most therapies can be adapted to apply them on a transdiagnostic basis, even cognitive therapy and behaviour therapy from the 1960s! In terms of actually efficacy, if you have all the time in the world as a therapist and trainer, and you are tactful in how you use diagnostic information, there is no point shifting to transdiagnostic therapies – they will be no more (and possibly less) effective than disorder-specific therapies. This is because, even though the processes are shared across disorder, you still get to apply them more directly to the client in front of you if you know their diagnosis. However, think about it differently. There are over 300 different disorders, most of them co-occur, and there is rarely time (over 90 minutes) and training to diagnose patients in services, especially most of them who are in primary care. Transdiagnostic therapies come into their own with this heterogenous population because it promises a more focused way to train and deliver therapies. 



Whilst it may be the best we have now, do you see a future for categorical classification or should it be replaced by, say by a trans-diagnostic or dimensional approach?


I think there is a much likelihood of the psychiatric classification system disappearing as there is of scientists giving up on classifying the chemical elements, or different species of animal and bird. However, over time, mental health researchers might realise that the underlying mechanisms are shared across disorders, just as protons, neutrons and electrons are shared across elements and evolution through natural selection is shared across species. Both a classification system and an underlying theoretical model are entirely compatible with one another, if their roles are clearly specified and differentiated. So, I don’t have any interest in a different classification system. I have an interest in understanding the underlying shared mechanisms in exquisite detail. 


What do you see as the important research areas for yourself in the future?

The fundamentals of Perceptual Control Theory through a mixed methodology from qualitative interviews through experimental studies to computer modelling. We are using it to formulate and help understand people’s rich network of goals and ways of coping. We are also using it to model processes such as the interplay between approach and avoidance in exposure therapy for phobias. In tandem with this basic science, I need to do the basic bread and butter of validating, adapting and disseminating our CBT for bipolar disorder and Method of Levels as a transdiagnostic cognitive therapy.


If you could recommend one mental health book, what would it be?

Short of recommending any one of my own! - ‘Hold That Thought’ by Tim Carey


What advice would you give to students interested in going into the field of mental health?

I think you need a blend of the ‘emotional’ and ‘rational’ to be a good practitioner or researcher in the mental health sphere, and comfortable and confident in using both. If you feel that you want to use both in your work, and are keen on enriching your experience and knowledge of both, you will do well, helping other people along the way…


Many thanks from MancPsychSoc for finding the time to talk to us.

Warren Mansell is a Reader in Psychology, Clinical Psychologist and Cognitive Behavioural Therapist at the University of Manchester. 

A full list of his publications is available here

For more information watch Warren's lecture at the May Davidson Award 2011.


1 comment:

  1. Since I am currently revising it, very briefly, and basically and ungramatically...

    The 12 processes can be divided into 5 broader processes of attention, memory, reasoning, thought and behavour.


    1) Transdiagnostic 'attentional processes' consist of 3 domains...
    first is 'selective attention to external stimuli'. this attentional bias means that individuals with mental disorders will selectively attend more to stimuli that reflect their current concerns. for example, GAD patients will do worse in the stroop test when related to socially or physically threatening words.

    second is the 'selective attention to internal cues'. Patients will attend to internal cues more than non-clinical samples, focussing more on shaking or sweating for example.

    third is 'attentional avoidance', whereby patients will deliberately avoid attending to stimuli relating to their current concerns --> avoidance or safety behaviour. for example avoiding eye contact.


    2) Transdiagnostic memory 'processes' consist of 2 domains...
    The first is 'explicit selective memory'. this refers to the increased recall of specific information that relates to a patient's current concerns. For example, presented with a list of words, recall for depressed patients will be biased in favour of negative words.

    Second is 'recurrent memories'. whilst controllable mental imagery is used in everyday problem solving, patients imagery may manifest in recurrent intrusive memories of a nature that is specific to their worries. this memories occur in all modalities and are often misinterpreted or appraised (reasoning biases)


    3) transdiagnostic 'Reasoning biases' consist of three domains
    first, 'interpretation bias' for example patients may mistake ambiguous stimuli as threatening or neagtive' this is is known as personalization

    second is 'expectancy reasoning', or more specifically 'representation heuristic'. e.g. gamblers fallacy --> unrealistic beliefs about the future

    third is 'emotional reasoning' - emotions are interpreted as an indicator of reality 'I am nervous thefore there is danger'


    4) Transdiagnostic 'Thought processes'consist of 2 domains...
    first, 'recurrent thinking' this is actually an example of the second domain which is (regulatory) 'metacognitive' beliefs. in response to intrusive thoughts patients may ruminate worry and focus on them --> affecting overall higher level goals (see MOL)

    second, 'metacognitive beliefs' consist of regulatory as above, and metacognitive knowledge. this refers to our beliefs about our thinking. e.g.' ruminating and worrying will help solve my problems '

    5) Transdianostic behavioural process consist of 2 domains...
    first 'avoidance behaviour' e.g. social phobic not going out. one adverse outcome means the opportunity to disprove current negative schema or concerns is lost.

    second is 'saftey behaviour'. this is used within-situation e.g. avoid talking eye contact etc.

    NB I HAVE USED MAINLY EXAMPLES FROM DEPRESSION AND ANXIETY DISORDERS BUT ALL PROCESSES ARE ACROSS AT LEAST 4 IN MOST CASES MOST AXIS 1 DISORDERS (SEE WARREN MANSELL'S PAPER)

    the relavence of this research is that TD approach is to be used in cases where there is comorbidity, a lacking model for a specific disorder, no specialist, or where disorder based therapy has failed. it is NOT to replace disorder based therapy.

    ReplyDelete