Seclusion, Restraint, Therapeutic Alliance, and Legal Admission Status: What Really Matters?

Abstract

To determine the relationships, if any, between use of seclusion and restraint and factors such as demographic parameters, diagnosis, legal admission status (voluntary or involuntary), symptoms, cognitive function, global functioning, therapeutic alliance, attitudes toward medication, and insight, among psychiatry inpatients in Ireland. We used validated tools to perform detailed assessments of 107 adult psychiatry inpatients admitted to acute psychiatry units at 2 general hospitals in Dublin, Ireland over a 30-month period, between September 2017 and February 2020. The most common diagnoses in our sample were affective disorders (46.7%), schizophrenia and related disorders (27.1%), and personality and behavioral disorders (11.2%). Over a quarter (n=29, 27.1%) of the participating patients had involuntary legal status. Of the 107 patients, 11 patients (10.3%) experienced sedation and/or physical restraint, with 9 patients (8.4%) experiencing at least 1 episode of seclusion and 10 patients (9.3%) experiencing at least 1 episode of physical restraint. On the basis of multivariable analyses, seclusion was associated with younger age and involuntary status, while physical restraint was associated with involuntary status. Each multivariable model explained just over one third of the variance in the distribution of these seclusion and restraint practices. Use of seclusion and restraint was most strongly associated with involuntary admission status and, in the case of seclusion, younger age, rather than sex, diagnosis, symptoms, cognitive function, global functioning, therapeutic alliance, attitudes toward medication, or insight. The network of interactions between involuntary status and use of seclusion and restraint merits much closer attention, especially as use of seclusion and physical restraint appears to be associated with involuntary legal status independent of level of symptoms, therapeutic alliance, or insight.