Reduction in the use of seclusion in a high secure hospital in the UK: a retrospective analysis
© Qurashi et al; licensee BioMed Central Ltd. 2007
Published: 19 December 2007
The main aim of this retrospective analysis is to ascertain whether a comprehensive seclusion reduction program was successful in reducing the use of seclusion within a high secure hospital in the UK.
The evaluation was conducted in Ashworth Hospital, UK. For the purposes of this paper the following strategies were identified post-hoc and analyzed as the independent variables namely, progressively lowering the threshold for a multi-disciplinary group review, providing targeted staff training dependent on patient needs and increasing the clinical staff-patient ratio. A multiple regression analysis was applied to monthly seclusion data covering the period from January 2002 – January 2006. As the total hospital census declined during the four-year period the monthly data for total hours of seclusion over the period divided by the inpatient census for that month served as the dependent variable.
A graphical analysis reveals there has been a 62% reduction in the number of seclusion episodes over the 4 year period. Results from the multiple regression analysis showed that the reduction in the threshold for an internal clinical review had a statistically significant effect (p < 0.05). There were no statistically significant results from the other variables included in the regression analyses although this should not be equated with an absence of clinical significance.
Our retrospective review confirms that reduction in the use of seclusion can be achieved when it is identified as both a managerial and clinical priority, supplemented by robust performance monitoring and effective peer reviews and case management. The formation of a hospital committee populated by senior clinical managers with the progressive lowering of thresholds to precipitate a formal peer review served as a clear reminder of this priority to all clinical staff. Further analysis of the data, including other variables, are intended to be published.
This article is published under license to BioMed Central Ltd.