Dr. Garrett is vice-chairman for clinical services and Dr. Lerman is clinical assistant professor, Department of Psychiatry, SUNY Downstate Medical Center, 450 Clarkson Ave., Box 1203, Brooklyn, NY 11203 (e-mail: michael.garrett@down state.edu). Dr. Lerman is also clinical director of Kingsboro Psychiatric Center in Brooklyn.

In New York State individuals who commit felony assault or homicide and who are judged not guilty by reason of insanity are typically admitted to a high-security inpatient facility and then transferred to a lower-security hospital for continued treatment and assessment. This patient group is diagnostically diverse, although most have a diagnosis of chronic paranoid schizophrenia. Some in this group have not committed acts of violence in many years, and some have not been violent at all since the felony that prompted their original admission.

To be discharged from the hospital facility, patients must pass through a series of stepwise increases in hospital "privileges" approved by clinical staff and the Forensic Committee. Despite good pharmacological treatment and good behavior, some patients fail to progress toward discharge because they lack insight into their disorder, remaining confined years longer than they would have had they been convicted of a felony and served jail time. Our facility sought an intervention that might increase insight in this group and revitalize progress toward discharge.

In the past two decades in Great Britain cognitive-behavioral therapy (CBT) has been adapted to the treatment of psychosis by Kingdon and Turkington and others and has been shown to increase insight among patients with psychosis. It builds on familiar CBT techniques but employs specialized interventions tailored to the psychotic patient. It departs from traditional conceptualizations of psychosis in assuming a continuum between psychosis and ordinary mind, which allows the therapist to foster a therapeutic alliance by "normalizing" aspects of the patient's experience. Using a patient-centered approach, the patient and clinician identify experiences that are sources of distress to the patient and identify patients' beliefs about these events. Then as "coinvestigators," they examine "evidence" for and against the patient's beliefs.

Eight patients, all of whom had been hospitalized for more than ten years, were the focus of a pilot intervention in 2006 that employed CBT for psychosis in individual sessions. In most cases, sessions occurred once a week for 45 minutes, continuing for an average of 20 sessions. Experienced therapists trained in CBT for psychosis treated most of the patients, but other experienced staff were recruited to the effort by providing a weekly seminar and peer supervision focused on CBT for psychosis.

A number of obstacles to implementation were immediately apparent. All psychotherapy requires an atmosphere of trust and openness between patient and therapist. This is difficult to achieve in a forensic setting. Patients expressed concern that what they said might be used as "testimony" against them in court. Furthermore, most psychiatrists in state facilities are biologically oriented, with limited training in individual psychotherapies. Even when interested in psychotherapy, psychiatrists occupied with large caseloads have little time for individual psychotherapy. Psychology and social work staff with prior training in CBT for depression and anxiety disorders may see CBT for psychosis as nothing new despite their lack of familiarity with advances in techniques specific to a psychotic population. In a system with limited resources, lonely patients are so hungry for individual attention that in some cases a time-limited CBT treatment was difficult to terminate. Patients who improved may have been responding both to nonspecific supportive aspects of the treatment and to the CBT intervention itself.

Results were encouraging. Despite the chronicity of illness, six of eight patients were judged to have benefited from the CBT intervention. In the case of Mr. A, the improvement was dramatic. Fifteen years ago he had committed a double homicide while in an acute psychotic state, but he had little prospect of discharge because he lacked insight into his disorder. He believed that a cult group had cast a spell on him, which led to the murders. He showed no interest in exercising off-ward privileges and thought frequently of suicide. As a result of the CBT intervention, Mr. A came to doubt his cult delusion and to acknowledge that mental illness may have played a role in the murders. He was able to express feelings of guilt, remorse, and worthlessness.

A clear improvement in mood ensued as these feelings were dealt with in therapy. Mr. A expressed an interest in exercising his pass privileges but feared he would be attacked by other patients if he left his ward. Patient and therapist conducted a behavioral experiment in which they walked together around the hospital grounds. No threat occurred on the walk and on subsequent walks. His "safety behavior" of socially isolating himself gave way to an active interest in passes and off-ward activity. He is currently making progress toward discharge.