Re-admissions to the State Hospital at Carstairs, 1992-1997.

OBJECTIVE: To examine the incidence and characteristics of re-admissions to the State Hospital over the six year inclusive period of January 1992 to December 1997. RESONS FOR THE STUDY: RISK ASSESSMENT: A re-admission to a high-security setting may be the result of a failed risk assessment at the time of discharge. An analysis of re-admissions might reveal facts that will improve risk prediction. In this context every re-admission to maximum security should qualify for critical incident appraisal. A re-admission could also indicate prudent risk management where the patient is readmitted prior to the occurrence of any new incident. Our analysis may give an indication how risk prediction and risk management is functioning in the Forensic Psychiatry setting.

HEALTH ECONOMICS: Patient numbers at the State Hospital have exceeded planning predictions. In the early 1990s the prediction was that numbers would level off at around 200 patients. Instead admission rates continued to rise and the State Hospital in-patient population has on occasion reached its maximum. This trend has continued throughout the decade with the daily patient numbers averaging 236, 248, 252 and 249 in the last four years (1997-2000). In this context the role played by re-admissions is an important health economics issue.

DESIGN: Data gathered from the medical records department and examination of medical case files, entered into a database, and subsequently analysed using a variety of tools. Descriptive statistics for the total State Hospital patient population between 1992-1994 were taken from a published survey. Other comparative analyses were undertaken using data from previously published studies.

SETTING: The data originate from the State Hospital, Carstairs, Lanarkshire, which is the only psychiatric hospital in Scotland to offer maximum-security care. Patients are admitted from ordinary psychiatric hospitals as well as from a penal setting. The maximum number of staffed beds is currently 257. These comprise an admission unit, two rehabilitation units including one with flatlet accommodation, a specialist learning difficulties unit, an all female unit and six male wards. The Hospital is staffed by psychiatrists, GPs, psychologists, social workers, psychiatric nurses including nurse specialists, occupational therapists, and other specialist staff employed in the education, occupations and recreation departments.

RESULTS AND CONCLUSIONS: Annual re-admission rate, averaged over 1992-1997 is 22.3% of all patient discharges in the hospital. This is broadly in line with other studies quoted in the literature. The re-admission rate co-varied with first admission rate at an almost constant ratio of 0.38 over the six-year period of study. There is no apparent systematic reason for this phenomenon. The most apparent risk factor was a previous admission to the State Hospital of less than a years duration. Compared with the current average length of stay of five years, this must be considered a short admission. Patients with an above average length of stay were rarely readmitted. Most re-admissions returned during the first year after discharge and nearly all had returned within a three-year period. State patients on a Restriction Order who will be subject to conditional discharge with ongoing supervision fared best and were least likely to be re-admitted. However such patients are also likely to have an above average length of stay. Convicted prisoners and remanded offenders were most likely to be readmitted. There was almost no crossover between the civil and criminal populations when their status was reviewed on re-admission. This raises the question, whether these two patient groups should be perceived and managed differently. The recent move to build medium secure units in Scotland, such as the Orchard Clinic in Edinburgh, now makes this a practical option. This finding also suggests that, in Scotland, bed closures in Mental Hospitals are not leading to a decant of mentally disordered patients into the penal system, at least at the level of more serious offending. Violence was overall the most common reason for return to the State Hospital. Violence was the predominant reason for re-admission from a local hospital whereas relapse of symptoms, where such symptoms were associated with dangerous behaviour in the past, was rarely given as a reason. In contrast, relapse was the predominant reason for return from prison. This may be related to inadequate prison hospital facilities rather than good risk prediction.