[Associations of short-term evaluation assessments and patient characteristics with the use of coercive measures]

Dutch mental health care institutes are currently making a tremendous effort to reduce the number of coercive measures they impose, such as seclusion. Despite this effort, a group of patients are still being subjected to enforced medication or seclusion.

To obtain more knowledge about which patient characteristics are associated with coercive measures and thereby to identify signs that could be acted upon at an earlier stage to prevent deterioration of the patient's condition and reduce the need for coercive measures.

We performed a prospective, naturalistic three-month study involving patients in two acute psychiatric admission wards. To collect the data we required, we used two short-term risk-assessment instruments: the Brøset Violence Checklist (BVC) and the Kennedy Axis V (KA-V). By means of statistical analyses we investigated which patient characteristics were associated with the use of coercive measures.

Of the 179 patients, 52 patients (29%) were subjected to a coercive measure during the admission procedure. The following patient characteristics were found to be associated with coercive measures: a bipolar disorder, involuntarily admission, display of physical violence just before admission, scores on the BVC and the KA-V items assessing 'social skills' and 'violence'. The two BVC items 'attacks on objects' and 'display of physical violence just before admission' were the ones most strongly associated with seclusion.

Although the sensitivity of the regression model was modest, the associated patient characteristics, combined with the two short-term risk assessment scores, may be helpful for identifying at an early stage those patients who run a high risk of being subjected to coercive measures.