Letters

All you need is cognitive behaviour therapy?

Critical appraisal of evidence base must be understood and respected

Integrated biopsychosocial approach will be treatment of choice

Political range of psychotherapies has emerged

Summary of responses

Critical appraisal of evidence base must be understood and respected

EDITORThe article by Holmes and the accompanying commentaries essentially aim to explore the evidence for the efficacy and effectiveness of the psychotherapies and how this should be applied in everyday clinical practice.¹ It is ironic that the arguments against cognitive therapy include numerous instances of idiosyncratic use of research evidence.

Firstly, for example, in his commentary Bolsover selects three studies to support his view that the evidence base for cognitive therapy is weak. We would challenge him to apply his arguments to the seven systematic reviews of cognitive therapy in the Cochrane Database and the Database of Abstracts of Reviews of Effectiveness.²

Holmes and also Bolsover cite a single paper to suggest that cognitive therapy is less effective in the real world of clinical practice than in university based clinical trials. This caricature may have applied to some trials conducted 30 years ago but is irrelevant now. To give just two examples, recent trials of cognitive therapy for chronic depression specifically recruited individuals who were depressed despite adequate trials of pharmacotherapy and psychotherapy.³ Also, the Cochrane review of cognitive therapy for schizophrenia includes examples of "real world" interventions.⁴

Holmes argues that leading cognitive therapists are starting to question aspects of their discipline. However, these critiques are quoted out of context; the criticisms actually refer to the need to adapt the basic cognitive model to enhance its effectiveness for other disorders. Far from being a weakness, the critical appraisal of cognitive therapy by its practitioners is an important reason why it has been systematically evaluated in such a wide range of conditions.

We agree with Holmes that it is unhelpful to evaluate psychotherapies by using only the research methods applied to drug trials. If clinicians and researchers aspire to an evidence based health service, however, they must accept two challenges. First is the challenge of evaluating what they think they do. We look forward to the evidence base of the psychodynamic psychotherapies developing, to allow more valid comparisons between the different approaches. Until then there is no escaping the robust evidence that exists for the use of cognitive therapy across clinical conditions and settings and the fact that much of this research is clearly applicable to the NHS. The second challenge is for clinicians and commentators to understand and respect the critical appraisal of the evidence base. Regrettably, some of the contributors to this series of articles have failed in this.

Tom Sensky, reader in psychological medicine. 
Imperial College of Science, Technology and Medicine, West Middlesex University Hospital, Isleworth, Middlesex TW7 6AF t.sensky{at}ic.ac.uk

Jan Scott, professor of psychological treatments research. 
Department of Psychological Medicine, Institute of Psychiatry, London SE5 8AF

Integrated biopsychosocial approach will be treatment of choice

EDITORDespite being a psychiatric trainee interested in ultimately doing further psychoanalytic training I was dismayed by the lack of balance in the debate presented by Holmes and the commentaries to his article.¹

Cognitive behaviour therapy is effective for a wide range of conditions and is patient friendly. Whether it is useful in severe mental illness depends largely on what is being attempted. Furthermore, it can be a useful intermediate step towards later psychoanalytic treatment.

Even if the popularity of cognitive behaviour therapy was the result of a superior marketing policy there is a substantial amount of research by which to judge it. Unfortunately, the same cannot be said about psychoanalytic psychotherapy. Although psychoanalysis has a vast literature base, it almost entirely consists of subjective opinion backed up through the use of selected case reports or material. Although this is clinically helpful, psychiatrists cannot generalise from it with any confidence. This is not only because we cannot say whether these patients are representative of our practice population but also because, even if they were, it is doubtful that we could replicate the intervention.

The time has come for the psychoanalytic camp to prove that a psychodynamic approach can work in NHS psychiatry. Bateman and Fonagy have shown that it can be done. ^{2 3} Moreover, although I agree that there is a limit to the extent that one can objectify the subjective, I do not think that this should stop us from attempting to do so. The time has come for us to clarify further what it is about the relationship that is therapeutic and to find a way of measuring relational change. Once found, let us apply this, not only to research in the psychotherapies but also to psychopharmacology, as there is little doubt that a drug prescription is more than simply the prescription of a drug.⁴

There will never be a time when there is a single panacea for mental illness; an integrated biopsychosocial approach will always be the treatment of choice.

Brian Darnley, senior house officer. 
Waterlow Mental Health Unit, London N19 5NX bjmdarnley{at}blueyonder.co.uk

Political range of psychotherapies has emerged

EDITORIt seems from the article by Holmes and the accompanying commentary by Tarrier that a political range of psychotherapies is now emerging (figure).¹ On the left wing we have traditional Freudian psychodynamic therapy, which addresses the causes of human distress through an understanding of unconscious conflicts rooted in the past. It contains an implicit assumption that gaining insight will lead to clinical improvement. Such "black couch" therapy, as championed by Holmes, usually takes many months or years. The message exhorted is that we should spend a lot of taxpayers' money on it despite a lack of evidence of benefit.

View larger version (50K): [in this window] [in a new window]   Political spectrum of psychotherapies

On the right wing lies cognitive behaviour therapy, as originally promoted by Beck.² It asks what can practically be done by addressing specific problems in the present. It is time limited (usually 8-20 sessions in total) and emphasises personal responsibility for change ("homework").

Where does this leave other psychotherapies? Cognitive analytic therapy occupies the centre ground as the New Labour of talking treatments ("Tough on the problem, tough on the causes of the problem").³ The Independent party (interpersonal therapy) and the Green party (family and group psychotherapies) emphasise that individuals depend on each other for effecting lasting change. The favourite, however, is surely non-directive counselling. Like the Liberal Democratic party, it has popular local support (counselling within primary care), although this "feel good" factor wanes in general elections (randomised trials of efficacy).⁴ Finally, stress debriefing after trauma has been widely embraced as the Monster Raving Loony party option, with good evidence of no benefit, and even possible harm.⁵

Psychotherapists and politicians have much in common. They presume a relationship of trust and believe that real change can be achieved by talking. They seldom answer questions directly, and use silence as a therapeutic tool. Grand inquisitors such as Archie Cochrane and Robin Day, if they were alive today, would be impressed with the range of talking treatments now available. For voters and patients alike, "You have never had it so good."

Paul Blenkiron, consultant in adult psychiatry. 
Bootham Park Hospital, York YO30 7BY paul.blenkiron{at}excha.yhs-tr.northy.nhs.uk

1.	Holmes J. All you need is cognitive behaviour therapy? [With commentaries by R Neighbour, N Tarrier, ED Hinshelwood, and N Bolsover.] BMJ 2002; 324: 288-294. (2 February.)
2.	Beck AT. Cognitive therapy and the emotional disorders. New York: International Universities Press, 1976.
3.	Rees H. Cognitive-analytical therapya most suitable training for psychiatrists? Psych Bull 2000; 24: 124-126[CrossRef].
4.	Harvey I, Nelson SL, Lyons RA, Unwin C, Monaghan S, Peters TJ. A randomised controlled trial and economic evaluation of counselling in primary care. Br J Gen Pract 1998; 48: 1043-1048[ISI][Medline].
5.	Wessely S, Rose S, Bisson J. Brief psychological interventions ("debriefing") for immediate trauma related symptoms and the prevention of post-traumatic stress disorder. In: Cochrane Library. Oxford: Update Software, 1998.

Summary of responses

EDITORThe debate on cognitive behaviour therapy and its accompanying commentaries prompted an outpouring of self examination by psychoanalysts and others who believe in the psychodynamic approach to psychotherapy. In a cluster of 18 letters they asked: What do we do? How can we measure it? How can we show that our treatments work? What is psychotherapy anyway?¹

Nick Totton, a psychotherapist, took it one step further: "psychological distress" is not a medical condition, he writes. "The unpalatable truth seems to be that if society really wants to address the causes of psychological distress it needs to look at its ways of dealing with emotion, with relationships, with work, and with sexuality."

He, and others, including Michael van Beinum, a child psychiatrist, also argued that generating evidence for psychodynamic psychotherapy is difficult when traditional comparative trials don't work. They offered a variety of reasons: clients have multiple problems, clients seek out the therapist that suits them, treatment is open ended and defined by the relationship between client and therapist, and randomised trials take no account of meaning or of language, both of which are important to psychoanalysis. Van Beinum suggested that qualitative research would be a reasonable alternative.

Should therapy be evidence based anyway? Defenders of cognitive behaviour therapy, sitting smugly on a mountain of their own evidence, were clear that it should. John Taylor, a clinical psychologist, echoed Tom Sensky and Jan Scott and Brian Darnley (above) when he accused traditional psychotherapists of feeble excuses and delaying tactics. Is it unreasonable, he wonders, for people to ask whether a client with a personality disorder is still cutting herself after two years' treatment and "good progress" in her relationship with her therapist?

The turf war doesn't end with the evidence, or lack of it. Both sides accused each other of squawking to attract the attention of policy makers to secure funding for their ideas and ultimately their jobs. Two letters, though, were more conciliatory. Rhona Sargeant, a senior registar in psychotherapy, writes: "Evidence based medicine appears to have given us something to argue about. Of course evidence is important, but I believe we are abusing it in order to fight an interprofessional battle, fuelled by pressure on resources." Patients who need psychotherapy are needy and challenging to treat, she continues. Managers see them as expensive. We need a united front. There are plenty of different needs to go around.

One correspondent from outside the United Kingdom (there were only two) was also dismayed by the squabbling. Christopher Booth, a psychiatrist from British Columbia, says patients all get on far better over there.

Several letters argued that the BMJ's "debate" was biased against cognitive behaviour therapy. The responses were, if anything, the opposite. Six of the best letters (good points tightly argued and supported with references) came down firmly in defence of cognitive behaviour therapy. For example, Colin Espie, a professor of clinical psychology, says that cognitive behaviour therapy is an effective treatment for insomnia according to 50 randomised trials, two meta-analyses, and a systematic review. John Taylor mentions using cognitive behaviour therapy to treat violence, sexual agression, and antisocial behaviour, and Steve Williams, a general practitioner, writes that cognitive behaviour therapy works, and works long term, because it teaches patients to be their own therapist.

Alison Tonks, freelance medical editor. 
Bristol