CBT for severe mental disorder: a good product that is in danger of being over-extended

Abstract

There often comes a time in the history of every successful treatment in psychiatry when the notion of diagnosis seems to evaporate and the treatment seems to be a universal panacea. This happened in the past when, in the heady days after the introduction of antidepressants, these drugs were not only alleged to be effective in the treatment of depression, anxiety and somatic disorders, but also in “depressio sine depressione”. Cognitive behavior therapy (CBT) is getting closer to this dangerous milestone, dangerous because it can lead to disillusion, so it is worth examining this in the new treatment areas described in the paper by Thase et al 1. Improvement in schizophrenia may be only a consequence of improvement in depression, anxiety and related symptoms. When training medical students I advise them that, if they are stuck when asked for a named treatment for any psychiatric disorder, they should reply “CBT”. When asked for the rationale for this, I suggest they reply: “because mood, symptoms, behavior and thinking are all closely inter-related and in (X) disorder mood and all other symptoms are made worse by cognitive distortion”. This comment is relevant to the use of psychological treatments in schizophrenia. Although some of the trials of CBT in schizophrenia have been carried out without antipsychotic drugs being included, most, as Thase et al indicate, have used the psychological therapy as an adjunctive one. This is not ideal, as schizophrenia is a heterogeneous condition associated with considerable mood disturbance. There is also evidence that those who have relatively “pure” schizophrenia have a much better outcome than those who have comorbid mood or substance use disorders 2. Anxiety in particular is a very prominent symptom in schizophrenia 3. It is therefore quite possible that the benefits of CBT in schizophrenia are dependent entirely on the mood component, and although target symptoms of schizophrenia such as command hallucinations may be improved, this is not necessarily an anti-schizophrenic effect, as it could be secondary to the effects of treatment on mood. It is perhaps worth reminding ourselves that a similar adjunctive therapy for schizophrenia proposed 35 years ago, beta-blockade in the form of propranolol, was similarly found to be effective as an adjunctive treatment to antipsychotic drugs 4, but not when compared directly with chlorpromazine 5, and we know now that beta-blocking drugs have very little role to play in the treatment of schizophrenia apart from their possible value in treating abnormal movements. A collaborative relationship may be therapeutic and independent of CBT. In chronic schizophrenia, where negative symptoms are prominent, a great deal can be achieved by establishing a good collaborative relationship with patients. This is one of the important aspects of the recovery model 6, but is not specifically linked to CBT. It is also an important component of systematic environmental adjustment, or nidotherapy, in the treatment of schizophrenia 7, 8. In further studies in this population, CBT should be compared with these other collaborative treatments. The evidence base is subject to publication bias. There has been a great deal of concern in recent years about unpublished trials of drug treatment creating bias and distorting subsequent systematic reviews and meta-analyses. We all need to be reminded that psychological treatments in general are subject to publication bias, very clearly shown in respect of studies in depression 9. This is partly because single-blind methodology is less rigorous than double-blind methodology, and we need to be cautious when enthusiastic investigators are comparing a psychological treatment with a pharmacological one. Although Thase et al have claimed publication bias is not present in studies of CBT in schizophrenia, that bias is highly likely to be present. One of the major problems with complex psychological treatments is that adequate numbers to test hypotheses in trials are relatively rare, and encouraging results with small trials are often not replicated in large ones with much more rigorous methodology, and better independent assessment 10, 11. Early results promising but must work harder. It is possible to summarize the position of CBT in the management of severe mental disorder in the same language as a school report on a precocious, talented, but somewhat self-satisfied adolescent. “C. Beatty has impressed everybody with his ability and flair, but in recent months has become a bit of a dilettante. He needs to knuckle down and concentrate on his core work a little more assiduously. We are sure he can do this, but he must not be distracted and needs to avoid spending too much time with his less critical friends”.