International comparative studies show that Dutch seclusion rates are relatively high. Therefore, several programs to change this practice were developed and implemented. The purpose of this study was to examine the impact of a seclusion reduction program over a long time frame, from 2004 until 2013. Three phases could be identified; the phase of development and implementation of the program (2004-2007), the project phase (2008-2010) and the consolidation phase (2011-2013). Five inpatient wards of a mental health institute were monitored. Each ward had one or more seclusion rooms. Primary outcome were the number and the duration of seclusion incidents. Involuntary medication was monitored as well to rule out substitution of one coercive measure by another. Case mix correction for patient characteristics was done by a multi-level logistic regression analysis with patient characteristics as predictors and hours seclusion per admission hours as outcome. Seclusion use reduced significantly during the project phase, both in number (-73%) and duration (-80%) and was not substituted by the use of enforced medication. Patient compilation as analyzed by the multi- level regression seemed not to confound the findings. Findings show a slight increase in number and seclusion days over the last year of monitoring. Whether this should be interpreted as a continuous or temporary trend remains unclear and is subject for further investigation.

Keywords: Inpatient psychiatry; Organizational change; Program evaluation; Seclusion.