Developing new treatments: on the interplay between theories, experimental science and clinical innovation
Introduction
From the early days of behaviour therapy, it has often been argued that one of the great strengths of behaviour therapy, and subsequently cognitive-behaviour therapy, is that it has a sound scientific basis. Many early behavioural treatments were based on animal learning theory and the emphasis on changing cognitions that is central to cognitive therapy is often justified by reference to the now substantial literature that demonstrates that many psychiatric disorders are associated with characteristic cognitive biases. However, the link between theories, experimental science and treatment developments in behaviour therapy and cognitive-behaviour therapy is complex. In addition, the nature of the relationship seems to have changed over time.
In the early days of behaviour therapy, learning theory was seen as the theoretical basis of most treatments and the therapy procedures were often seen as direct analogues of procedures that had been demonstrated to be effective in animal experiments. For example, Wolpe (1958) derived the procedures of systematic desensitization from procedures that he had successfully used to overcome conditioned fear reactions in cats. Similarly, later advocates of more intensive fear reduction procedures (e.g. flooding, in vivo exposure, exposure and response prevention) saw their procedures as direct analogues of procedures that had been used in other animal experiments that explored ways of extinguishing conditioned fear.
With the advent of cognitive therapy, the relationship between theories, experimental science and treatment development seemed to change. Cognitive theories focused on specifying therapy targets (negative thoughts and images, faulty patterns of attention, problematic memory processes, etc.), rather than therapy procedures. There was an experimental background for some of the treatment procedures. For example, the suggestion that it is generally better to get patients to generate the evidence against their negative beliefs than to directly confront them with such evidence was antedated by work in social psychology (for example, Brehm, 1966, Ross, 1977). However, it is unclear whether the early pioneers of cognitive therapy were aware of these experiments and most of the procedures in cognitive therapy’s first major success (the treatment of depression) seem to have been based on Beck’s insightful clinical observations and the highly creative pilot work of a clinical team with a mixture of psychodynamic and behavioural backgrounds (Beck, Rush, Shaw, & Emery, 1979).
The fact that, unlike learning theories, cognitive theories do not directly suggest treat ment procedures has meant that a considerable amount of work has to be done to convert a promising cognitive theory into a successful cognitive treatment. Treatment procedures that have proved useful in changing problematic cognitive targets in one disorder are not necessarily powerful interventions in another disorder. The way in which clinical researchers set about creating new cognitive-behavioural treatments is rarely discussed in the literature. The present article takes the occasion of a Festschrift in honour of John Teasdale to describe the way in which our group have approached the task. For the past 20 years, we have used an integrated programme of phenomenological, experimental and treatment development studies to try to create new and, hopefully, effective cognitive therapy programmes for anxiety disorders. We are, of course, not unique in combining various types of basic research and treatment development work. In anxiety disorders, the groups led by David Barlow, Edna Foa, Richard Heimberg, Lars Goran Ost, Jack Rachman, Ron Rapee, and Marcel van den Hout, among others, have been major exponents of such a strategy. However, like ours, their basic research and treatment studies are published separately and the way in which the basic research and treatment studies mutually interact to develop more effective interventions is rarely described in print. For this reason, we thought the topic might be of some interest to the reader.
Section snippets
John D. Teasdale
It is fitting that discussion of the interrelationship between theories, experimental science and treatment development appears in a special issue honouring John Teasdale’s outstanding achievements. Without question, John is one of the field’s leading figures in the simultaneous advancement of theory, experimental science and treatment. He has made major, and interlinked, advances in each area. His classic studies on the effects of thinking on mood (Teasdale and Bancroft, 1977, Teasdale and
A research strategy
Our general strategy for developing new treatments has been to: (i) use clinical interviews and cognitive psychology paradigms to identify the core cognitive abnormality in an anxiety disorder; (ii) to construct a theoretical account which explains why the cognitive abnormality does not self-correct; (iii) to test the hypothesised maintaining factors in rigorous experimental studies; (iv) to develop specialised cognitive treatments which aim to reverse the empirically validated maintaining
Attention to phenomenology helps identify fruitful therapy targets
Seligman (1994: p. 64) pointed out that the key intellectual move in the development of cognitive therapy was Beck’s proposal that something that had previously been seen as a symptom of depression (negative thinking) was in fact a phenomenon that had a key role in driving the disorder. Close attention to phenomenology and its possible significance has proved similarly helpful in our work with anxiety disorders.
Experimental tests of cognitive models have often generated powerful treatment procedures
In order to test our cognitive models of anxiety disorders, we have had to develop experiments in which the cognitive abnormalities and maintaining processes specified in the models are manipulated in order to see whether they have their predicted effects on symptoms. The experiments, when successful, have often become incorporated into the full treatment programme. Some prominent examples are given below.
The cognitive model of panic disorder proposes “that individuals who experience recurrent
Therapy experiments are a good way of testing whether treatment procedures work
Randomised controlled trials are the best way to determine whether a treatment works. However, they are not well suited to determining how the treatment works. The “how” question can be broken into two components; (i) Which of the many procedures involved in a treatment are effective? and (ii) What targets (beliefs, behaviours, attentional deployment, memory processes) need to be changed for symptomatic recovery to occur? With respect to the identification of effective procedures, early
Limitations in therapeutic outcome lead to refinements in both theory and treatment
Trying to work out why some patients failed to respond to early versions of our cognitive therapy programmes has proved valuable in generating new treatment procedures and has also helped elaborate the cognitive models on which the procedures are based. In our early work on panic disorder, we noticed that patients who failed to respond to verbal discussion techniques and behavioural experiments tended to experience stereotyped visual images during their panic attacks. In the images, their worst
If at first you do not succeed, try, try again (and only then do the randomised controlled trial)
The fact that cognitive models only specify treatment targets, not effective treatment procedures, means that extensive treatment development work is usually required before one considers conducting a randomised controlled trial. Some investigators have tended to assume that the cognitive therapy procedures that have been used with success in one disorder (for example, depression) can be transferred with little modification to other disorders. While this may be true in some instances, it has
Limited availability of evidence-based psychological treatment
As Barlow and Hofmann (1996) point out, evidence-based cognitive-behavioural treatments are not available to the majority of patients with anxiety disorders. There are several reasons for this regrettable state of affairs. First, for psychological treatments there is no equivalent of the vast promotion budgets that drug companies understandably set aside to ensure that their expensively developed medications are used by health care providers once the medications have been shown to be effective.
Concluding remarks
The way in which psychological treatments are developed is rarely discussed in the literature. For this reason, it is hoped that the present account will have been of some interest to readers. The close interplay between theory, experiments and treatment development that has characterised our group’s work is something that I was privileged to observe while one of John Teasdale’s doctoral students. Any errors that our group have made in subsequently following a master’s strategy are entirely our
Acknowledgements
The Wellcome Trust funds the group’s research on social phobia and posttraumatic stress disorder. The studies on panic disorder and hypochondriasis were funded by both the Wellcome Trust and the Medical Research Council of the United Kingdom.
References (71)
- et al.
Treatment of childhood memories: theory and practice
Behaviour Research and Therapy
(1999) A cognitive approach to panic
Behaviour Research and Therapy
(1986)- et al.
A prospective investigation of the role of cognitive factors in persistent posttraumatic stress disorder (PTSD) after physical or sexual assault
Behaviour Research and Therapy
(2001) - et al.
A cognitive model of post-traumatic stress disorder
Behaviour Research and Therapy
(2000) - et al.
Community based cognitive therapy in the treatment of posttraumatic stress disorder following the Omagh bomb
Behaviour Research and Therapy
(2002) - et al.
Recurrent images and early memories in social phobia
Behaviour Research and Therapy
(2000) - et al.
Cognitive preparation enhances the beneficial effects of video feedback following a stressful social performance
Behaviour Research and Therapy
(2000) - et al.
Self-images play a causal role in social phobia
Behaviour Research and Therapy
(2003) - et al.
The enhancement of video feedback by cognitive preparation in the treatment of social anxiety. A single session experiment
Journal of Behavior Therapy and Experimental Psychiatry
(2002) - et al.
Posttraumatic stress disorder after motor vehicle accidents: 3-year follow-up of a prospective longitudinal study
Behaviour Research and Therapy
(2002)