Although there has been no truly satisfactory study of the association between antisocial behaviour and psychosis, a variety of sources suggest a greater than chance association between the two (Taylor, 1987), and in particular there is evidence of a substantial increase in violent behaviour in association with schizophrenia (Taylor & Gunn, 1984). There is also evidence of the possibility of over-representation of mentally ill people in arrest statistics, because of an increase in police concern (Teplin, 1985; Taylor, 1987). Studies in this area may concern rates of psychiatric disorder among persons in judicial circumstances (e.g. Taylor & Gunn, 1984), but the Harrow study provided an opportunity of examining the extent and nature of behaviour leading to police contact in a large, unselected sample of schizophrenic patients followed up for up to 13 years.