Dr. Turkington Replies

To the Editor: Dr. Marlowe makes a number of valid points. We agree that the dissemination of cognitive behavior therapy for treating schizophrenia in the United States and elsewhere deserves careful consideration and further research. The accepted practice of cognitive behavior therapy for treating schizophrenia in the United Kingdom has been endorsed by the National Institute for Clinical Excellence (1) , which sets standards for the National Health Service and is monitored by the health commission, but this does not mean that it is equally relevant in other countries. Studies have been similar in their diagnostic inclusion criteria, which consist of patients from the schizophrenia group and therefore include patients with schizoaffective and delusional disorders. None of the major studies include patients outside this diagnostic cluster. I have one question pertaining to Dr. Marlowe’s comment that studies with the poorest methodology have reported the highest effect sizes: Is this not a recognized phenomenon throughout clinical research?

There are indeed some differences between the cognitive behavior therapy manuals currently in use (2 , 3) , but there is a consensus around the key components and order of application of techniques. The basic cognitive model for positive symptoms of schizophrenia has been developed and described by Garety and colleagues (4) and has been widely accepted. Long-term follow-up of studies, after therapy has been discontinued, has not yet demonstrated an enduring effect, but neither has this been the case with other interventions, whether pharmacological or psychosocial.

In terms of prognosis, we refer Dr. Marlowe to the 20-year follow-up study conducted by Harrison and colleagues (5) in which the outcome for patients with schizophrenia was nowhere near as negative as that implied by shorter follow-up periods and gives reasonable hope for recovery for many, especially to those who are offered treatments as promising as cognitive behavior therapy for schizophrenia.