The new long stay in an inner city service: a tale of two cohorts.

We assessed the long-stay inpatients (length of stay > 6 months) from a deprived inner-city catchment area with a population of 210,000 in 1993 and 1995 on a variety of measures, following up both cohorts after 24 months. Total numbers of long-stay inpatients were reduced from 56 (26.7 per 100,000 total population) to 35 (16.7 per 100,000) between 1993 and 1995, in line with the closure of dedicated long-stay beds. The 1995 cohort were more symptomatic according to the BPRS (t = 2.8, p = 0.007, 95% confidence interval 18.1, 3.0), more commonly detained under the Mental Health Act (chi 2 = 6.07 p = 0.05) and more commonly from an ethnic minority (chi 2 = 3.7 p = 0.05). At 2 year follow-up 57% of the 1993 cohort were living out of hospital, compared with 60% of the 1995 cohort. Patients were discharged to a variety of settings, some highly supported. For the combined sample the presence of certain challenging behaviours (absconding, disturbance at night, noncompliance with treatment and violence) predicted continuing inpatient status as did three items on the Social Behaviour Schedule (bizarre behaviour, laughing to oneself and violence). Only five (9%) of the original sample remained inpatients for the entire four year follow-up.