Readmission after discharge from a medium secure service between 1999 and 2017: A retrospective cohort study

Previous research into outcomes after treatment in medium secure psychiatric hospitals has mostly relied on pre-millennium data.

To examine patient and inpatient service-related factors associated with readmission within 2 years following discharge to the community or open conditions from conditions of medium security.

A retrospective cohort study of 137 patients discharged either to the community or to open conditions from one 64-bed South Wales Regional Medium Secure Hospital Unit between July 1999 and November 2017 was completed using data from healthcare records to document demographics, diagnosis and nature of index offences together with researcher-completed ratings of inpatient progress using the Dangerousness Understanding Recovery and Urgency Manual (DUNDRUM) Programme Completion and DUNDRUM Recovery scales. Binary logistic regression analyses were used to identify independent associations between inpatient progress according to these measures and readmission.

Forty-three patients (31%) were readmitted within 2 years of discharge and 23 (17%) in breach of legal conditions on discharge. Most readmitted patients (n = 29, 67%) returned directly to medium secure care. There were significant binary level associations between readmission and severity of the index offences (lower), number of adverse childhood experiences (higher), history of drug misuse (more likely) and number of previous psychiatric admissions (higher). Binary logistic regression confirmed that these relationships were not independent. No inpatient service-related variables, according to DUNDRUM scale scores, showed an independent association with readmission within 2 years post-discharge.

The proportion of medium security hospital patients who were readmitted within 2 years of discharge aligns with estimates found in earlier national research using the same follow-up period. Since levels of inpatient progress bore little if any relationship to longer term outcomes, our findings highlight the need for investigating factors in the discharge environment that are linked to readmission. It is possible that readmission may indicate effective monitoring and responsive care to the changing needs of patients, but a better understanding of this is essential.