Generally, administrative segregation involved inmates who had a pattern of disruptive behavior in prison that was more serious and habitual than that of most inmates. Rarely were inmates placed in segregation only on the basis of the nature of the crime for which they had been sentenced. In the absence of a national standard or guideline, it is not clear what level or type of disruptive behavior triggers segregation. Demographically, inmates placed in administrative segregation tended to be young males. A disproportionately high rate of mental health needs were found in inmates placed in administrative segregation. For many of the inmates, these conditions apparently existed prior to their imprisonment. Two cohorts were used in this study, one to examine placement reasons and one for population comparisons. All administrative segregation hearings from January 2004 through December 2005 were used to analyze placement decisions; 1,614 hearings were held during this period, which involved 1,459 inmates. The Colorado incarcerated population at the end of fiscal year 2005 was used in the profile analyses. Administrative segregation inmates were identified among the inmate population through their classification levels; 1,210 inmates were classified as administrative segregation. Data on inmates in segregation were obtained with instruments that measure proficiency in reading, math, language, and spelling; general mental capacity; recidivism risk; personality disorders and clinical syndromes; and psychiatric disorders. 2 tables, 1 figure, and 43 references