A register of violence incidents showed a progressive increase in in-patient violence from 1976 to 1984, followed by a slight decline to 1987. The case notes of all 137 in-patients committing assaults during 1982 were analysed and compared with those of matched controls. Violence was often repetitive. The most common victims were nurses and then patients. The violent patients were more likely to be schizophrenic, deluded and hallucinated, and to have been repeatedly admitted. The violent and control groups had similar backgrounds, and both showed high levels of social isolation and unemployment. Violence tended to be repetitive, and the violent patients were usually identifiable from previous aggressive and disturbed behaviour. There was a high proportion of Afro-Caribbean patients who tended to be younger, more psychotic, and more seriously violent. Afro-Caribbeans were particularly likely to be detained compulsorily and treated in a locked ward.

The increasing complexity of wildland fire management highlights the importance of sound decision making. Numerous fire management decision support systems (FMDSS) are designed to enhance science and technology delivery or assist fire managers with decision-making tasks. However, few scientific efforts have explored the adoption and use of FMDSS by fire managers. This research couples existing decision support system research and in-depth interviews with US Forest Service fire managers to explore perspectives surrounding the Wildland Fire Decision Support System (WFDSS). Results indicate that fire managers appreciate many WFDSS components but view it primarily as a means to document fire management decisions. They describe on-the-ground actions that can be disconnected with decisions developed in WFDSS, which they attribute to the timeliness of WFDSS outputs, the complexity of the WFDSS design, and how it was introduced to managers. We conclude by discussing how FMDSS development could address concerns raised by managers.

The literature on violence by psychiatric patients is reviewed. Emphasis is given to recent papers relevant to clinical management. M ost studies are cross-sectional.. The M audsley/Bethlem Violent Incidents Register is used to provide longitudinal data over a period in excess of 20 years. In acute units most violent actions are carried out by patients suffering from schizophrenia, personality disorder and alcohol or substance abuse. In longer stay units violence tends to be associated with either intractable schizophrenia or with organic syndromesÐ particularly mental impairment. Most in-patient violence is minor and repetitive. Patients at risk are usually easily identi® able by a past history of violence or aggression. Most psychiatric patients are non-violent, but a small proportion of patients account for a large number of incidents. Violence by psychotic patients is particularly likely to occur in the period leading up to and immediately following admission, when the patient is acutely disturbed. The prevention of violence is largely a matter of good psychiatric and nursing care for patients at risk. Much of the violence committed by recently admitted psychotic patients could be prevented if after care facilities were more effective in preventing non-compliance and relapse.

No abstract