Patient characteristics related to length of stay in a UK forensic inpatient sample: a systematic review; and, Examining the validity of the CORE-OM as a measure of distress in a forensic population with severe and enduring mental illness

Abstract

Introduction There is concern that individuals may experience prolonged stays in secure inpatient settings, and research highlights little convergence on factors associated with length of stay in the forensic inpatient population. Routine outcome measures are valuable in monitoring patient progress, including patient self-report measures such as the Clinical Outcomes in Routine Monitoring – Outcome Measure (CORE-OM). Research regarding use of the CORE-OM in UK forensic settings is limited. Previous research found CORE-OM scores for this population fell below clinical cut-offs; the reasons for which have not yet been examined. A systematic review was conducted to identify patient factors associated with length of stay in UK medium and high secure settings. An empirical study investigated the validity of the CORE-OM in a high secure setting, examining whether poor insight is associated with low self-report distress, as measured by the CORE-OM, and whether the CORE-OM can adequately capture distress arising from psychotic experience. Method A systematic search of studies investigating factors associated with length of stay in UK secure settings, between 2000 and 2020, was conducted. Eight studies from combined search results of 4738 articles, met inclusion criteria. Included studies were subject to full-text review. Quality assessment was subject to two independent ratings. In the empirical study, routinely collected CORE-OM and two measures of insight and symptomatology were retrospectively extracted from case files for 246 males in high secure inpatient care. Results The systematic review found a diagnosis of schizophrenia, previous psychiatric and forensic admissions, admission from other secure or prison settings, and Section 37/41 under the Mental Health Act (MHA) were significantly associated with longer stays in secure care. However, significant methodological weaknesses and variability defining ‘long-stay’ were present across studies. The empirical study found symptom subscales significantly predicted CORE-OM all item mean scores. Poor insight predicted higher self-report distress on CORE-OM all item mean scores and did not mediate the relationship between symptomatology and distress (CORE-OM). Cut-off scores fell below clinical reference samples and revised normative scores are provided for a male high secure population. Discussion The systematic review found mixed evidence regarding factors associated with length of stay in forensic secure settings, with clinical variables and MHA status providing the most consistent associations with length of stay. Research in high secure settings and with other samples out with England are urgently required to increase generalisability. Future research should focus on standardisation of predictor variables and prospectively designed studies that incorporate factors beyond static variables, including patient self-report measures and dynamic variables. Long-stay definitions should be considered in line with policy changes. The empirical study found the CORE-OM is sensitive to changes in clinician-rated symptomatology. Mean CORE-OM scores were below clinical reference samples in the community and revised normative scores for a high secure forensic sample are provided. Insight did not explain low self-report as poor insight predicted greater distress on CORE-OM scores. Future research considering factors associated with change in CORE-OM scores over time is recommended to investigate the symptoms-insight-distress interactions across the patient journey.