U.S. flag

An official website of the United States government, Department of Justice.

Administrative Segregation for Mentally Ill Inmates

NCJ Number
221699
Journal
Journal of Offender Rehabilitation Volume: 45 Issue: 1/2 Dated: 2007 Pages: 149-165
Author(s)
Maureen L. O'Keefe
Date Published
2007
Annotation
This study examined the prevalence of offenders in the Colorado Department of Corrections with mental illness (OMIs) in administrative segregation (AS) relative to the general population to determine the frequency at which they are being held in a highly restrictive environment.
Abstract
In general, segregated offenders were discernibly different from their non-segregated peers, especially on criminal history indicators and institutional behaviors. In particular, offenders involved with security threat groups or engaged in gang-related activities were targeted for administrative segregation (AS) placements at a rate 4.5 times of those who had no known involvement. Specifically, offenders with mental illness (OMIs) were found in AS at a disproportionately high rate, a disparity which has grown over recent years in conjunction with severely restricted funding for program services. Mental illness was found to be the third strongest predictor of AS classification. OMIs were responsible for a significant number of disciplinary violations within the prisons, but mental illness was a better predictor of AS classification than disciplinary infractions. OMIs were less likely to be identified with security threat groups, suggesting that their acting out might be directly related to symptomatic expression of their illness. This is supported by clinician ratings of OMIs as having more severe psychopathology paired with high activity levels and low impulse control. Largely the result of prison officials needing to safely and efficiently manage a volatile inmate population, administrative segregation, or supermax facilities are criticized as violating basic human needs, particularly for mentally ill inmates. This study compared Colorado OMIs to non-OMIs in segregated and non-segregated environments. Tables, figures, references