Longitudinal trends in using physical interventions to manage aggression and self-harm in mental health services.

Objectives: Repetitive aggression by a patient receiving mental health inpatient care is likely to elicit various patterns of response from care staff over time. This study sought to examine patterns of coercive physical intervention in relation to repeated episodes of aggression by particular patients.

Methods: A data set of 9,945 aggression and self-harm incidents over a five-year period in one mental health service in England was constructed. Incidents by a specific individual were categorized according to their position in a sequence (first, second, and so on) and by the use of physical intervention by staff to manage the incident.

Results: Trends in the use of physical intervention varied across settings. There was a significant tendency in general (nonforensic) services for use of physical intervention to increase in response to physical aggression (physical intervention in first versus subsequent incidents: odds ratio [OR]=.69, 95% confidence interval [CI]=.54-.90) and to decrease in response to threats (physical intervention in first threat versus subsequent threats: OR=1.62, CI=1.09-2.39).

Conclusion: There were significant trends over time in the use of physical intervention to manage violence and self-harm. However, the dynamics behind this finding will remain unclear without further research.