Family intervention in psychosis: who needs it?

The interest in families as part of the social environment of people suffering from psychosis dates from the war. The early ideas took the form of speculation without adequate research underpinning, and were often couched in terms that suggested that families might be to blame for the sufferers' illness. Nevertheless, the speculations were a first attempt to appreciate the impact of the social environment on psychotic disorders. Formalized attempts to modify the family environment for the benefit of patients date only from the past two decades. They received a considerable impetus from the development of a reliable way of evaluating family relationships. It also proved to be a strong predictor of outcome in schizophrenia, as measured by symptomatic relapse or readmission. This was the measure of expressed emotion (EE). It was developed in the 1960s by George Brown and his colleagues (Brown & Rutter, 1966; Brown et al, 1972) as an evaluation of key attributes of the way in which family members describe their interaction. The predictive capacity of EE in psychosis has now been established beyond doubt (Bebbington & Kuipers, 1994). It has been demonstrated in a wide range of locations and cultures. In the 1960s and 70s, psychological interventions with patients suffering from psychosis were regarded as at best ineffective, and at worst actually