U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics ******************************************************** This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.csv) and the full report including tables and graphics in .pdf format are available on BJS website at: http://www.bjs.gov/index.cfm?ty=pbdetail&iid=5433 ********************************************************* Special Report Use of Restrictive Housing in U.S. Prisons and Jails, 2011–12 Allen J. Beck, Ph.D., BJS Statistician On an average day in 2011–12, up to 4.4% of state and federal inmates and 2.7% of jail inmates were held in administrative segregation or solitary confinement. Nearly 20% of prison inmates and 18% of jail inmates had spent time in restrictive housing, including disciplinary or administrative segregation or solitary confinement, in the past 12 months or since coming to their current facility, if shorter. Approximately 10% of all prison inmates and 5% of jail inmates had spent 30 days or longer in restrictive housing. This report is based on data from the National Inmate Survey (NIS), 2011–12, conducted in 233 state and federal prisons and 357 local jails, with a sample of 91,177 adult inmates nationwide. The NIS is part of the National Prison Rape Statistics Program, which collects reports of sexual victimization from administrative records and from allegations of sexual victimization directly from victims through surveys of inmates in prisons and jails. The inmate surveys contain a wide range of data beyond measures of sexual victimization, including items useful for describing inmates held in state and federal prisons or local jails and their confinement experiences. The Bureau of Justice Statistics (BJS) completed the third NIS between February 2011 and May 2012. ********************************************************** *************** Highlights *************** * Younger inmates, inmates without a high school diploma, and lesbian, gay, and bisexual inmates were more likely to have spent time in restrictive housing than older inmates, inmates with a high school diploma or more, and heterosexual inmates (figure 1). * Inmates held for a violent offense other than a sex offense and inmates with extensive arrest histories or prior incarcerations were more likely to have spent time in restrictive housing than inmates held for other offenses and inmates with no prior arrests or incarcerations. * Use of restrictive housing was linked to inmate mental health problems: 29% of prison inmates and 22% of jail inmates with current symptoms of serious psychological distress had spent time in restrictive housing units in the past 12 months. * More than three-quarters of inmates in prisons and jails who had been written up for assaulting other inmates or staff spent time in restrictive housing in the past 12 months. * Among inmates who had spent 30 or more days in restrictive housing in the past 12 months or since coming to the facility, 54% of those in prison and 68% of those in jail had been in a fight or had been written up for assaulting other inmates or staff. * Prison and jail facilities varied widely in their rates of use of restrictive housing. In 17% of prisons and 9% of jails, fewer than 5% of inmates spent time in restrictive housing. In comparison, in 38% of prisons and 24% of jails at least 25% of the inmates had spent such time. * Prisons with higher rates of restrictive housing had higher levels of facility disorder; lower levels of inmate trust and confidence in staff; higher concentrations of violent inmates (other than sex offenders) and inmates with longer criminal histories; higher percentages of inmates with mental health problems; and higher percentages of lesbian, gay, and bisexual inmates. ********************************************************** ***************************************** Measuring the use of restrictive housing ***************************************** The National Inmate Survey (NIS) is part of the National Prison Rape Statistics Program, which collects reports of sexual victimization from administrative records and from allegations of sexual victimization directly from victims through surveys of inmates in prisons and jails. The Bureau of Justice Statistics (BJS) has implemented this program to meet the requirements of the Prison Rape Elimination Act of 2003 (P.L. 108-79). However, the inmate surveys contain a wide range of data beyond measures of sexual victimization, including items useful for describing inmates held in state and federal prisons and local jails and their confinement experiences. This report examines data reported by inmates held in adult facilities on their current housing and any time spent in disciplinary or administrative segregation or solitary confinement in the past 12 months or since coming to the facility, if shorter. The use of restrictive housing is difficult to measure. Absent uniform definitions and information systems that classify inmates in comparable categories, estimates based on data reported by correctional officials are subject to variation and uncertainty, depending on the data collection. Nevertheless, almost every correctional system, at the federal, state, or local level, places inmates in some form of restrictive housing to separate some inmates from the general institutional population. Inmates may be held in restrictive housing for their protection or for the safety of other inmates. They may be held while awaiting classification or reclassification, while awaiting transfer to another facility or unit within a facility, or while awaiting a hearing or as a sanction for violating a facility rule. Inmates may also be separated from the general population to provide for their special needs (e.g., medical or mental health) or to ensure the safety, security, and orderly operation of the facility. Whether it is disciplinary segregation, administrative segregation (largely nonpunitive in nature), or solitary confinement (involving isolation and relatively little out-of-cell time), restrictive housing typically involves limited interaction with other inmates, limited programming opportunities, and reduced privileges. However, the use of restrictive housing varies widely in terms of duration and conditions of confinement. The NIS surveys, which involve separate samples of prisons and jails, collect information on the use of restrictive housing from the perspective of the inmates. Data from the most recent National Inmate Survey (NIS-3), conducted between February 2011 and May 2012, provide measures of prevalence beyond the housing status of inmates on a single day, including whether the inmates had spent any time in restrictive housing in the past 12 months or since coming to the facility, if shorter, and the total amount of time they had spent. The surveys of prison and jail inmates provide estimates of time in restrictive housing by inmate demographic characteristics, criminal justice status and history, current and past mental health status, and facility misconduct in the past 12 months. When aggregated at the facility level, the NIS-3 data also provide information on a representative sample of prison and jail facilities, including detail on variation among facilities in the use of restrictive housing by selected facility characteristics. The NIS-3 survey, conducted by RTI International (Research Triangle Park, NC), was administered to 91,177 inmates age 18 or older, including 38,251 inmates in 233 state and federal prisons and 52,926 inmates in 357 jails. The results are nationally representative of prison and jail inmates at the time of the survey and representative at the facility level for each sampled facility. (See Methodology for detailed description of the sampling and estimation.) ********************************************************** ************************ Individual-level rates ************************ -------------------------------------------------- On an average day in 2011–12, up to 4.4% of state and federal prisoners and 2.7% of jail inmates were held in administrative segregation or solitary confinement -------------------------------------------------- Based on inmate self-reports, an estimated 1.9% of state and federal prisoners and 2.2% of local jail inmates said they were housed in administrative segregation or solitary confinement at the time of the survey (table 1). However, the actual percentages held in restrictive housing may be higher. An estimated 2.5% of prison inmates and 0.5% of jail inmates completed a paper survey that did not inquire about their housing status. Many, but not all, of these inmates may have been in administrative or disciplinary segregation. Some were inmates whom staff were reluctant to bring to the interview room because they were considered too violent, and some were inmates whom staff determined should not have access to a computer. In addition, some inmates were unavailable for any contact by survey staff. Among the inmates selected for the survey, approximately 0.19% of prison inmates and 0.23% of jail inmates were unavailable because they were in segregation; 0.29% of prison inmates and 0.77% of jail inmates were considered too violent even for survey staff to contact for a paper interview; and 0.49% of prison inmates and 0.73% of jail inmates were considered to be mentally incompetent by facility or survey staff (not shown). Combined, these excluded inmates totaled approximately 1.0% of all selected prisoners and 1.7% of all jail inmates. While the survey results were adjusted for nonresponse through a series of weighting adjustments within each selected facility and nationwide, the adjustments were not linked specifically to the reasons for nonresponse. However, the impact on the national estimates was likely to be small because the excluded inmates represented about the same percentages among all selected inmates as found in the final survey estimates. The weighting adjustments were sufficient to provide an overall estimate without including these inmates who were not contacted by survey staff. The national estimates of percentage held in restrictive housing could be as high as 4.4% of prison inmates and 2.7% of jail inmates if most of the inmates completing a paper form were assumed to have been in some form of restrictive housing and a further adjustment for nonresponse was made to account for inmates who were held in segregation and unavailable to the survey staff. -------------------------------------------------- Nearly 20% of prison inmates and 18% of jail inmates had spent time in restrictive housing in the past 12 months or since coming to the facility, if shorter -------------------------------------------------- An estimated 18% of prison inmates and 17% of jail inmates said they had spent time in disciplinary or administrative segregation or solitary confinement in the past 12 months or since coming to the facility, if shorter. If combined with respondents for whom time in restrictive housing was unknown, the actual percentages may be as high as 20% among prison inmates and 18% of jail inmates. This experience was limited to the inmates’ current facility but increased steadily with time served. Overall, 49% of prison inmates and 92% of jail inmates had been in the facility for less than a year (not shown). During the 12 months prior to the survey, state and federal prisoners had been in the facility for an average of 8.6 months and jail inmates for an average of 3.5 months (table 2). Among inmates who had been admitted to the facility in the past month, 8% of both prison and jail inmates had spent some time in restrictive housing. Among inmates who had been in the facility for 2 to 3 months, 12% of prisoners and 14% of jail inmates had spent time in restrictive housing. Among those in the facility for 6 to 8 months, 20% of prisoners and 27% of jail inmates had spent time in restrictive housing. Among inmates who had served 12 months or more, 20% of prisoners and 35% of jail inmates had been in restrictive housing at some point. The total time inmates had spent in restrictive housing varied among prison and jail inmates. Approximately 10% of all prison inmates and 5% of jail inmates said they had spent 30 days or longer in restrictive housing. In comparison, about 3% of prisoners and 6% of jail inmates had spent less than a week. -------------------------------------------------- Time in restrictive housing varied among inmate demographic groups -------------------------------------------------- Younger inmates were significantly more likely than older inmates to report having spent time in restrictive housing. Among inmates ages 18 to 19, 31% of those in prison and 25% of those in jail had spent some time in restrictive housing (table 3). Among inmates ages 20 to 24, 28% of those in prison and 23% of those in jail had been in restrictive housing at some time during the past year. The percentages who reported time in restrictive housing were lower among persons age 30 or older in prison (20% or below) and among persons age 25 or older in jails (19% or below). Inmates without a high school diploma were more likely than high school graduates to have spent time in restrictive housing. Among prison inmates, 20% of those with less than a high school education had spent time in restrictive housing, compared to 15% of those who completed high school. Similarly, among jail inmates, 19% of those with less than high school education had spent time in restrictive housing, compared to 15% of those who had a high school diploma or more. In prisons, black inmates (21%) were somewhat more likely than white inmates (16%) to have spent time in restrictive housing; however, in jail they were equally likely to have spent such time (17% each). Inmates of other races (including American Indians and Alaska Natives; Asians, Native Hawaiians, and other Pacific Islanders; and those reporting two or more races) were more likely than white inmates to have spent time in restrictive housing (20% in prison and 22% in jail). Hispanic inmates (16% in prison and jail) were as likely as white inmates in prison and white and black inmates in jail to report having spent time in restrictive housing. Lesbian, gay, and bisexual inmates (28% in prison and 22% in jail) were more likely than heterosexual inmates (18% in prison and 17% in jail) to have spent some time in restrictive housing. -------------------------------------------------- Time in restrictive housing units linked to current offense and past criminal justice contacts -------------------------------------------------- * Inmates held for a violent offense other than a sex offense (25% in prison and 28% in jail) were significantly more likely than inmates held for other offenses to have spent time in restrictive housing (table 4). * Inmates with extensive criminal histories were also more likely than inmates with shorter criminal histories to have spent time in restrictive housing. Among inmates with 11 or more prior arrests, 24% of those in prison and 22% of those in jail had been in restrictive housing. In comparison, about 13% of inmates in prisons and jails who had been arrested once had been in restrictive housing. * Inmates who had a prior incarceration (20% of prison inmates and 19% of jail inmates) were more likely than other inmates (about 13% for both) to have been in restrictive housing in the past 12 months. * Among inmates who had been in prison or jail before, the percentage reporting time in restrictive housing increased with amount of time they had served in the past. Among those who had served 5 or more years on a prior incarceration, 21% of prisoners and 22% of jail inmates had been in restrictive housing at some time in the past 12 months. -------------------------------------------------- Use of restrictive housing linked to inmate mental health problems -------------------------------------------------- The inmate surveys collected data on the past mental health problems of inmates. Inmates were asked whether they had ever been told by a mental health professional that they had a mental health disorder, or if because of a mental health problem they had stayed overnight in a hospital or other facility, used prescription medicine, or received counseling or treatment from a trained professional. (See Methodology for more detail.) On every measure of past mental health problems, inmates who reported a problem were also more likely than other inmates to report that they had spent time in restrictive housing in the past 12 months or since coming to the facility, if shorter (table 5). Time in restrictive housing was reported by-- * 26% of prison inmates and 23% of jail inmates who had been told they had a mental health disorder * 31% of prison inmates and 25% of jail inmates who had stayed overnight in a hospital or other facility during the 12 months prior to their admission for mental health problems * 26% of prison inmates and 23% of jail inmates who at the time of the current offense were taking prescription medicine for mental health problems * 26% of prison inmates and 23% of jail inmates who had ever received counseling or therapy from a trained professional—such as a psychiatrist, psychologist, social worker, or nurse—for mental health problems. Lower percentages of inmates without a mental health problem had spent time in restrictive housing. Overall, about 14% of prison inmates and 12% of jail inmates who reported no past mental health problems had spent time in restrictive housing. -------------------------------------------------- A high percentage of inmates with current symptoms of serious psychological distress had spent time in restrictive housing units -------------------------------------------------- The surveys included the K6 screening scale to determine whether inmates had a current mental health problem. The K6 was previously developed by Kessler and others for estimating the prevalence of mental illness in non- institutional settings as a tool to identify cases of psychiatric disorder***Footnote 3 Kessler, R.C., Green, J.G., Gruber, M.J., Sampson, N.A., Bromet, E., Cuitan, M., Furukawa, ... Zaslavsky, A.M. (2010). Screening for serious mental illness in the general population with the K6 screening scale: results from WHO World Mental Health (WHM) survey initiative. International Journal of Methods in Psychiatric Research, 19(1), 4–22***. It has been used widely in epidemiological surveys in the United States and internationally, including with prison populations***Footnote 2 Schneider, K., Richters, J., Butler, T., Yap, L., Richards, A., Grant, L., et al. (2011). Psychological distress and experience of sexual and physical assault among Australian prisoners. Criminal Behaviour and Mental Health, 21(5), 333- 349***. Since 2008, the K6 scale has been used in federal epidemiological studies to measure symptoms of serious psychological distress (SPD). Although the K6 has been demonstrated to be a good predictor of serious mental illness in prior studies, a technical advisory group, convened by the Center for Mental Health Services at the Substance Abuse and Mental Health Services Administration (SAMHSA), recommended that it should be supplemented with questions on functional impairment to improve statistical prediction and validity. (See Methodology for detail on specific items and scoring.) Consistent with other measures of mental health or emotional problems, the K6 revealed that prison and jail inmates identified with SPD were more likely than inmates with no mental health symptoms to have spent time in restrictive housing. Among inmates with symptoms of psychological distress, 29% of those in prison and 22% of those in jail had spent time in restrictive housing in the past 12 months or since coming to the facility. In comparison, among inmates with no symptoms of mental health problems, approximately 15% of those in both prisons and jails had been in restrictive housing. While these differences may reflect a variety of factors related to the use of segregation by correctional authorities, including sanctions imposed for violations of facility rules, they may also reflect the need to provide protective custody (for nonpunitive reasons) and placement in administrative segregation (while assessing treatment needs and appropriate classification). Moreover, time in restrictive housing—especially longer periods of time—may trigger symptoms of SPD. Data also showed that inmates who had not spent any time in restrictive housing had lower levels of SPD than other inmates. Among prison inmates who had not spent any time in restrictive housing, 13% were identified with SPD. Among jail inmates without any time in restrictive housing, 25% were identified with SPD (table 6). Among both prison and jail inmates, rates of SPD were significantly higher among those who had spent time in restrictive housing; however, the rates did not increase with the length of time they had been in such housing. An estimated 24% of prison inmates and 35% of jail inmates who had spent 30 days or longer in restrictive housing had SPD. Nearly identical rates of SPD were reported among inmates who had been in restrictive housing for only a day (22% of prison inmates and 35% of jail inmates). Overall, the data revealed no relationship between the length of time in restrictive housing and rates of SPD. -------------------------------------------------- More than three-quarters of inmates written up for assaulting other inmates or staff had spent time in restrictive housing -------------------------------------------------- The prison and jail surveys included five items that measure the prevalence of inmate misconduct during the past 12 months or since coming to the facility, if shorter. Inmates were asked whether (1) they had been in a fight with another inmate, (2) they had been written up for assaulting the inmate, (3) they had been in a fight with staff, (4) they had been written up for physically assaulting a staff member, and (5) they had been written up for verbally assaulting a staff member. (See Methodology for detail.) Time in restrictive housing was reported by-- * 49% of prison inmates and 43% of jail inmates who had been in a fight with another inmate * 56% of prison inmates and 52% of jail inmates who had been in a fight with staff * at least three-quarters of those who had been written up for fighting with other inmates (77% in prison and 79% in jail) or verbally assaulting staff (74% in prison and 80% in jail) * more than 80% of both prison and jail inmates who had been written up for physically assaulting a correctional officer or other facility staff (table 7). Approximately 10% to 11% of inmates who had no mention of fighting or being written up for assaulting other inmates or staff had spent some time in restrictive housing. Among inmates serving the longest amount of time in restrictive housing (i.e., 30 or more days in the past 12 months or since coming to the facility, if shorter), 54% of those in prison and 68% of those in jail had been in a fight or had been written up for assaulting other inmates or staff (not shown). --------------------------- Facility-level rates --------------------------- When aggregated at the facility level, the inmate self-reports provide reliable facility-level estimates of the use of restrictive housing. Unlike most BJS surveys, the NIS-3 was designed to provide facility-level estimates with sufficient precision to accurately describe facilities. Within each facility, the number of inmates sampled was based on criteria related to the expected prevalence rate of sexual victimization (4% in prisons and 3% in jails), with a desired level of precision (a standard error of 1.75% in prisons and 1.40% in jails) and an expected response rate (70% in prisons and 65% in jails). Due to the size of the samples within each facility, the NIS can provide estimates for other facility-level characteristics. See Sexual Victimization in Prisons and Jails Reported by Inmates, 2011–12 (NCJ 241399, BJS web, May 2013.) The data reveal significant variation in the use of restrictive housing across prison and jail facilities. On average, prisons and jails used restrictive housing at similar levels: 17% of inmates in the average prison and 17% in the average jail had spent time in restrictive housing in the past 12 months or since coming to the facility (table 8). However, the distributions of facility-level rates were quite different: * Fewer than 1% of their inmates had spent time in restrictive housing in 7% of the jails and less than 1% of prisons. * 25% or more of the inmates had been in restrictive housing in 38% of the nation’s prisons, compared 24% of jails. Prisons had higher rates than jails of inmates held in restrictive housing for 30 days or more. Approximately 27% of jails had held less than 1% of their inmates for 30 days or more, compared to 13% of prisons. At least 30% of prisons had held 10% or more of their inmates in restrictive housing for 30 days or more, compared to 17% of jails. -------------------------------------------------- Use of restrictive housing associated with indicators of facility disorder -------------------------------------------------- In the absence of administrative data on assaults, gang activity, seizures of weapons, or other security incidents, the inmate self-report data may also be used to provide independent measures for each facility. The prison and jail surveys asked inmates to report on-- * fighting within their facility and whether they had been in a fight with other inmates or staff * whether they worried about being assaulted by other inmates * whether they had seen other inmates with weapons * if there was lots of gang activity in the facility * if some of their possessions had been taken by other inmates. When aggregated at the facility level, the inmate responses provided a series of indicators of facility disorder. (See Methodology for more detail.) The data revealed a clear relationship between the use of restrictive housing in facilities and these indicators of facility disorder. On every measure, prison facilities with higher percentages of inmates reporting disorder had higher rates of inmates held in restrictive housing in the past 12 months (table 9). In prisons, the correlation with the use of restrictive housing was the highest for the percentage of inmates reporting having been in fights with other inmates (r=0.65) and in fights with staff (r=0.59). In prisons, the same pattern was found for the percentage of inmates held in restrictive housing for 30 days or more. Facilities in which a high percentage of inmates reported having been in fights with staff (r=0.61) or other inmates (r=0.53) also had high rates of using long-term segregation. Among jails, five of the seven measures of facility disorder were associated with greater use of restrictive housing. Although the correlations in jails were generally lower than those observed in prisons, the percentage of inmates who had been in restrictive housing at some time in the past 12 months or since coming to the facility remained strongly related to the percentage who had been in fights with other inmates (r=0.37) and staff (r=0.37). The same relationships were found for the percentage of inmates who had spent 30 days or more in restrictive housing. -------------------------------------------------- Facility-level rates of restrictive housing associated with characteristics of inmates housed -------------------------------------------------- Selected facility-level measures were calculated based on inmate-level characteristics associated with reports of time in restrictive housing. Facilities’ use of restrictive housing varied in the types of inmates that they held. These measures of facility composition included the percentage of inmates-- * with serious psychological distress * with a past mental health problem * held for violent crimes other than sex offenses * with 11 or more prior arrests * with a prior incarceration * with less than a high school diploma * who were lesbian, gay, or bisexual * who were ages 18 to 24. (See Methodology for more detail.) Among prisons, five of the eight measures of facility composition were associated with greater use of restrictive housing (table 10). The correlation with use of restrictive housing was the highest for the percentage of inmates with current symptoms of SPD (r=0.65) or past mental health problems (r=0.48) and the percentage held for violent offenses other than sex offenses (r=0.50). While facilities with an increasing percentage of inmates with 11 or more prior arrests also had higher rates of restrictive housing, the association was not as strong (r=0.20). The data also revealed a clear link between the percentage of inmates who were lesbian, gay, or bisexual and the greater use of restrictive housing (r=0.33). The same patterns were found for the percentage of inmates held in restrictive housing for 30 days or more, although the correlations were somewhat weaker. The percentage of inmates held with symptoms of SPD (r=0.51) and the percentage held for violent crimes (excluding sex offenders) (r=0.46) were the most strongly correlated with the percentage of inmates who spent 30 days or more in restrictive housing. The composition of inmates held in local jail facilities was largely unrelated to the use of restrictive housing. Among jails, only two of the eight measures of facility composition were positively associated with the greater use of restrictive housing. The percentage of inmates with past mental health problems had the strongest correlation (r=0.42) with the percentage of inmates who had spent any time in restrictive housing. One of the composition measures for jail facilities was associated with the percentage of inmates who had spent 30 days or more in restrictive housing. In constrast, the percentage of inmates with 11 or more prior arrests was negatively correlated with the rate of long-term restrictive housing (r=-0.19). -------------------------------------------------- Lack of inmate trust and confidence in staff linked to greater use of restrictive housing -------------------------------------------------- The prison and jail surveys also included items that asked inmates to characterize staff, including whether the staff at the facility-- * are generally fair * do their best to make this facility safe and secure * try to meet the needs of the inmates * break up fights quickly * use physical force only when necessary * let inmates know what is expected of them * generally treat inmates with respect * follow facility rules when handling inmate complaints and grievances * often write up inmates who don’t deserve it. As with the measure of facility disorder and inmate composition, the inmate self-report data may be used to provide an independent measure of the overall level of trust and confidence that inmates have with the staff in each facility. (See Methodology for more detail.) The responses to these nine items were combined at the inmate level and then weighted to provide a summary score for each facility. To account for item nonresponse, inmate-level scores were calculated only if an inmate had responded to two or more items. Their score reflected the percentage of items for which the inmate had provided a negative response (“disagree” for all items except the last, and “agree” that staff “often write up inmates who don’t deserve it”). In each facility, the inmate responses were then averaged (weighted by the inmate weight adjusted for nonresponse) to provide an overall facility-level score. The facility-level scores in prisons ranged from 5% to 72% (with a weighted mean of 46%), and in jails from 8% to 73% (with a weighted mean of 37%). The higher the percentage, the more negative the inmates’ assessments of the facility staff. There was a clear relationship between the use of restrictive housing and this summary indicator of facility climate. Prison facilities with higher percentages of inmates reporting negative assessments of staff had higher percentages of inmates held in restrictive housing (r=0.38) and higher percentages of inmates being held for 30 days or more (r=0.33) (table 11). Among jail facilities, the association with negative perceptions of staff fairness and trust was limited to the percentage of inmates who had spent 30 days or more in restrictive housing (r=0.25). Other indicators of a facility’s climate included measures related to crowding and sufficient staff at the facility. Inmates were asked about crowding in their housing unit and in areas outside of their housing unit (such as in the dining hall, classrooms, gym, or work areas). Inmates who reported “very crowded” were weighted and summed to provide a measure of crowding. In addition, inmates were asked if the facility had enough staff to provide for the safety and security of inmates. Again, when their responses were aggregated at the facility level, they provided an indicator of facility climate. (See Methodology for more detail.) The use of restrictive housing was unrelated to prison crowding. Neither measure was associated with the percentage of inmates held in restrictive housing. In jail facilities, crowding of areas outside of the housing units was correlated with greater use of restrictive housing (r=0.50) but not with the length of time in such housing (r=0.17, which was not statistically significant). As with measures of facility disorder, lack of sufficient staff was associated with a facility’s use of restrictive housing. Among prison facilities, the higher the percentage of inmates reporting that the facility did not have enough staff to provide for the safety and security of inmates, the higher the percentage of inmates who reported time in restrictive housing (r=0.31) and the percentage who spent 30 days or more in such housing (r=0.20). In jail facilities, only the percentage of inmates reporting a lack of staff to provide for safety and security was associated with the percentage who spent 30 days or more in restrictive housing (r=.25). A consistent pattern was observed at the facility-level. Facilities with higher rates of restrictive housing had higher levels of facility disorder; lower levels of inmate trust and confidence; higher concentrations of violent inmates and inmates with longer criminal histories; higher percentages of inmates with mental health problems; and higher percentages of vulnerable populations (i.e., lesbian, gay, and bisexual inmates, and younger inmates) *************** Methodology *************** ----------------------------------- National Inmate Survey-3 (NIS-3) ----------------------------------- The National Inmate Survey is part of the National Prison Rape Statistics Program, which collects reports of sexual victimization from administrative records and from allegations of sexual victimization directly from victims through surveys of inmates in prisons and jails. BJS has implemented this program to meet the requirements of the Prison Rape Elimination Act of 2003 (P.L. 108-79). The inmate surveys contain a wide range of data beyond measures of sexual victimization, including items useful for describing inmates held in the state and federal prisons and local jails and their confinement experiences. This report examines data reported by inmates held in adult facilities on their current and past time spent in disciplinary or administrative segregation or solitary confinement. BJS completed the third National Inmate Survey (NIS-3) between February 2011 and May 2012. The survey, conducted by RTI International (Research Triangle Park, NC), was administered to 91,177 inmates age 18 or older, including 38,251 inmates in 233 state and federal prisons and 52,926 inmates in 357 jails. The results are nationally representative of prison and jail inmates at the time of the survey. The NIS-3 consisted of an audio computer-assisted self-interview (ACASI) in which inmates used a touch-screen to interact with a computer-assisted questionnaire and followed audio instructions delivered via headphones. Some inmates (733 prison inmates and 255 jail inmates) completed a short paper form instead of using the ACASI. Many of these inmates were housed in administrative or disciplinary segregation or were considered too violent to be interviewed, some were inmates who refused to come to the interview room, and some were inmates who staff were reluctant to bring to the interview room for other reasons. For approximately the first two minutes, survey interviewers conducted a brief personal interview to obtain background information and the date of admission to the facility. For the remainder of the interview, respondents interacted with a computer-administered questionnaire using a touch-screen and synchronized audio instructions delivered via headphones. Respondents completed the ACASI portion of the interview in private, with the interviewer either leaving the room or moving away from the computer. -------------------------------------------------- Selection of prisons and inmates within prisons -------------------------------------------------- A sample of 241 state and federal prisons was drawn to produce a sample representing the 1,158 state and 194 federal adult confinement facilities identified in the 2005 Census of State and Federal Adult Correctional Facilities, which was supplemented with updated information from websites maintained by each state department of corrections (DOC) and the federal Bureau of Prisons (BOP). Of the 241 selected prison facilities, seven had closed prior to the start of data collection, and one had transitioned from holding males to females during the data collection period and was considered a closed facility, based on the original sampling criteria. All of the other selected prison facilities participated fully in the NIS-3. The NIS-3 was restricted to confinement facilities—institutions in which fewer than 50% of inmates were regularly permitted to leave, unaccompanied by staff, for work, study, or treatment. Such facilities included prisons, penitentiaries, prison hospitals, prison farms, boot camps, and centers for reception, classification, or alcohol and drug treatment. The NIS-3 excluded community-based facilities, such as halfway houses, group homes, and work release centers. A roster of inmates was obtained just prior to the start of data collection at each facility. Inmates who were age 15 or younger and inmates who were released prior to data collection were deleted from the roster. Inmates who were ages 16 to 17 were sampled separately and have been excluded from this report. Each eligible adult inmate was assigned a random number and sorted in ascending order. Inmates were selected from the list up to the expected number of inmates determined by the sampling criteria that reflected the projected response rate (70%), a desired level of precision, and size of the facility. A total of 74,655 adult prison inmates were selected. After selection, 2,233 ineligible inmates were excluded. Overall, 60% of the selected eligible prison inmates participated in the survey. Approximately 90% of the participating prison inmates (38,251 adults) completed the sexual assault survey from which the data on restrictive housing were drawn. -------------------------------------------------- Selection of jail facilities and jail inmates -------------------------------------------------- A sample of 393 jails was drawn to represent the 2,957 jail facilities identified in the 2005 Census of Jail Inmates, which was supplemented with information obtained from inmate surveys (NIS-1 and NIS-2) conducted in 2007 and 2008–09. The 2005 census was a complete enumeration of all jail jurisdictions, including all publicly operated and privately operated facilities under contract to jail authorities. The NIS-3 was limited to jails that held six or more inmates on June 30, 2005. These jails held an estimated 720,171 inmates age 18 or older on June 30, 2011. Jail facilities were sequentially sampled with probabilities of selection proportionate to size (as measured by the number of inmates held on June 30, 2005). Of the 393 selected jails in the NIS-3, 20 facilities refused to participate, 2 were excused due to construction or lack of space at the facility, and 14 were determined to be ineligible. All of the other selected jail facilities participated fully in the NIS-3. A roster of inmates was obtained just prior to the start of data collection at each facility. Inmates who were age 15 or younger and inmates who had not been arraigned were removed from the roster. Inmates who were ages 16 to 17 (juveniles) were sampled separately and have been excluded from this report. Each eligible adult inmate was assigned a random number and sorted in ascending order. Inmates were selected from the list up to the expected number of inmates determined by the sampling criteria. Due to the dynamic nature of jail populations, a second roster of inmates was obtained on the first day of data collection. Eligible adult inmates who appeared on the second roster but who had not appeared on the initial roster were identified. These inmates had been arraigned since the initial roster was created or were newly admitted to the facility and arraigned. A random sample of these new inmates was chosen using the same probability of selection used to sample from the first roster. A total of 112,594 jail inmates were selected. After selection, 11,342 ineligible inmates were excluded: 9,479 were released or transferred to another facility before interviewing began, 1,036 were mentally or physically unable to be interviewed, 25 were age 15 or younger or their age could not be obtained during the interview process, 296 were selected in error (i.e., an inmate was incorrectly listed on the facility roster), and 484 were on unsupervised work release or only served time on weekends. Of all selected inmates, 22% refused to participate in the survey, 1.1% were not available to be interviewed (e.g., in court, in medical segregation, determined by the facility to be too violent to be interviewed, or restricted from participation by another legal jurisdiction), and 8% were not interviewed due to survey logistics (e.g., language barriers, releases, and transfers to another facility after interviewing began). Overall, 61% of the selected eligible jail inmates participated in the survey. Approximately 90% of the participating jail inmates (52,926 adults) completed the sexual assault survey from which the data on restrictive housing were drawn. For a more detailed description of the NIS-3 sample, see Sexual Victimization in Prisons and Jails Reported by Inmates, 2011–12 (NCJ 241399, BJS web, May 2013). ---------------------------------------- Weighting and nonresponse adjustments ---------------------------------------- Responses from interviewed inmates were weighted to provide facility- and national-level estimates. Each interviewed inmate was assigned an initial weight corresponding to the inverse of the probability of selection within each sampled facility. A series of adjustment factors was applied to the initial weight to minimize potential bias due to nonresponse and to provide national estimates. A final ratio adjustment to each inmate weight was made to provide national- level estimates for the total number of inmates age 18 or older who were held in prisons at yearend 2011 or in jails at midyear 2011. These estimates for state prisons were 1,154,600 adult males and 83,400 adult females; for federal prisons, 190,600 adult males and 13,200 adult females; and for jails (with an average daily populations of six or more inmates), 628,620 adult males and 91,551 adult females. -------------------------------------------------- Standard errors and tests of significance -------------------------------------------------- As with any survey, the NIS estimates are subject to error because they are based on a sample rather than a complete enumeration. A common way to express this sampling variability is to construct a 95%-confidence interval around each survey estimate. Typically, multiplying the standard error by 1.96 and then adding or subtracting the result from the estimate produces the confidence interval. This interval expresses the range of values that could result among 95% of the different samples that could be drawn. The standard errors in appendix tables 1 through 8 have been used to compare estimates of the prevalence of restrictive housing among selected groups of inmates that have been defined by demographic subgroup, criminal justice status and history, mental health status, and indicators of inmate misconduct. To facilitate the analysis, differences in the estimates of percentage of inmates reporting any time in restrictive housing have been tested and notated for significance at the 95%-confidence level. For example, the difference in the percentage who reported time in restrictive housing among white prison inmates (16.0%), compared to black prison inmates (20.8%), was statistically significant at the 95% level of confidence (table 3 and appendix table 3). In all tables providing detailed comparisons, statistically significant differences at the 95% level of confidence or greater have been designated with two asterisks (**). The standard errors in appendix tables 9 through 11 have been provided to test the significance of the linear association (based on the Pearson product-moment coefficient) between selected facility characteristics and the percentage of inmates who experienced time in restrictive housing. By weighting by the inverse of the probability of selection of each facility, the facility-level estimates and standard errors take into account the sampling variability at the first stage in the NIS-3 sampling design. To construct a 95%-confidence interval around each correlation coefficient, the standard error may be multiplied by 1.96 and then added or subtracted from the estimated coefficient. The coefficient is considered statistically significant when the confidence interval excludes zero. For example, the estimated correlation between the percentage of inmates in prisons who were frequently in fights and the percentage of inmates who reported having spent some time in restrictive housing was 0.52. Based on a standard error of 0.08, the 95%-confidence interval was 0.36 to 0.68, which was statistically significant (table 9 and appendix table 9). In all tables providing estimated correlations, statistically significant coefficients at the 95% level of confidence or greater have been designated with two asterisks (**). -------------------------------------------------- Measures of time in restrictive housing -------------------------------------------------- Prior to the start of the ACASI portion of the survey, interviewers conducted a brief personal interview using CAPI to obtain background information and the date of admission to the facility. The CAPI interview included an item on current housing: A9. Which of the following best describes the housing unit where you spent last night? 1. An open dorm 2. A dorm with cubicles 3. A unit with cells 4. A unit with rooms 5. An area not originally intended as housing, such as a gym, classroom, or day room 6. Administrative segregation or solitary confinement 7. None of these This item was not administered to inmates who received the paper form. Because many, but not all, of these inmates may have been held in restrictive housing, they were designated as “don’t know” and included in the estimates. Based on the final national weights, 2.5% of prison inmates and 0.5% of jail inmates were in this category. The ACASI portion of the interview included two additional items on restrictive housing: X9a. [FILL ITEM], have you spent any time in disciplinary or administrative segregation or solitary confinement? 1. Yes 2. No X9b. [IF X9a=1] [FILL ITEM], how much total time have you spent in disciplinary or administrative segregation or solitary confinement? 1. 1 day or less 2. More than 1 day but less than 7 days 3. At least 7 days but less than 14 days 4. At least 14 days but less than 30 days 5. 30 days or more As with item A9, these items were also not administered to inmates who received the paper form. Because many, but not all, of these inmates may have been held in restrictive housing while in the facility, they were designated as “don’t know” and included in the estimates. The FILL ITEM specified a reference period provided automatically by the computer based on how long the inmate had been held at the facility. For inmates who had been held in the prison or jail for less than 12 months, the reference period was “Since you arrived at this facility.” For inmates who had been in the facility for 12 months or more, the reference period was “During the past 12 months.” -------------------------------------------------- Screening for serious psychological distress -------------------------------------------------- The K6 consists of six questions that ask inmates to report how often during the past 30 days they had felt-- * nervous * hopeless * restless or fidgety * so depressed that nothing could cheer them up * everything was an effort * worthless. The response options were (1) all of the time, (2) most of the time, (3) some of the time, (4) a little of the time, and (5) none of the time. Following Kessler, the responses were coded from 4 to 0, with 4 assigned to “all of the time” and 0 assigned to “none of the time.”***Footnote 1 Kessler, R.C., Green, J.G., Gruber, M.J., Sampson, N.A., Bromet, E., Cuitan, M., Furukawa, ... Zaslavsky, A.M. (2010). Screening for serious mental illness in the general population with the K6 screening scale: results from WHO World Mental Health (WHM) survey initiative. International Journal of Methods in Psychiatric Research, 19(1), 4–22.*** A summary scale combining the responses from all six items was then produced with a range of 0 to 24. The summary score was then reduced to three categories: 0 to 7 indicates no mental illness, 8 to 12 indicates an anxiety or mood disorder, and 13 or higher indicates serious psychological distress (SPD). See Sexual Victimization in Prisons and Jails Reported by Inmates, 2011–12 (NCJ 241399, BJS web, May 2013) for a discussion of K6 scaling rules and past applications. -------------------------------------------------- Measures of past mental health problems -------------------------------------------------- The NIS-3 included four items to measure the prevalence of any problems with emotions, nerves, or mental health an inmate may have had in the past: R24. Have you ever been told by a mental health professional, such as a psychiatrist or psychologist, that you had… a. Manic depression, a bipolar disorder, or mania? b. A depressive disorder? c. Schizophrenia or another psychotic disorder? d. Post-traumatic stress disorder or PTSD? e. Another anxiety disorder, such as panic disorder or OCD? f. A personality disorder, such as antisocial or borderline personality? g. A mental or emotional condition other than those listed above? R27. During the 12 months before you were admitted to [this facility/any facility to serve time on your current sentence], did you stay overnight or longer in any type of hospital or other facility to receive treatment or counseling for problems you were having with your emotions, nerves, or mental health? R30. At the time of the offense for which you are currently [being held/serving time], were you taking prescription medicine for any problem you were having with your emotions, nerves, or mental health? R33. Have you ever received counseling or therapy from a trained professional—such as a psychiatrist, psychologist, social worker, or nurse-- for any problem you were having with your emotions, nerves, or mental health? ---------------------------------- Measures of inmate misconduct ---------------------------------- The NIS-3 included five items that measure the prevalence of inmate misconduct during the past 12 months or since coming to the facility, if shorter: S17. In the past 12 months, have you been in a fight, assault, or incident in which another inmate tried to harm you? S21. In the past 12 months, have you been in a fight, assault, or incident in which a correctional officer or other facility staff person tried to harm you? X6a. In the past 12 months, have you been written up or charged with assaulting another inmate? X7a. In the past 12 months, have you been written up or charged with physically assaulting a correctional officer or other facility staff? X8a. In the past 12 months, have you been written up or charged with verbally assaulting a correctional officer or other facility staff? ---------------------------- Facility-level measures ---------------------------- In the absence of administrative data on characteristics of each facility, the inmate self-report data have been used to provide independent measures of facility disorder, composition, and climate of each facility. Because the NIS sample was designed to produce prevalence estimates of sexual victimization for each sampled facility, it also has the capacity to provide reliable estimates for other facility characteristics. Each facility sampled in the NIS is self-representing, and consequently the inmate responses, when weighted and summed at the facility-level, provide an overall indicator for each sampled facility. Facility disorder--The NIS included seven items that provide a facility-level estimate of the percentage of inmates reporting some indication of facility disorder: S13. In the past 12 months, how often are inmates at this facility hit, punched, or assaulted by other inmates? (Percentage reporting “frequently.”) S14. In the past 12 months, how often do you worry about being hit, punched, or assaulted by other inmates in this facility? (Percentage reporting “frequently.”) S15. In the past 12 months, how often have you seen other inmates with some type of weapon? (Percentage reporting “frequently” or “sometimes.”) S16. In the past 12 months, how much gang activity has there been at this facility? (Percentage reporting “a lot.”) S17. In the past 12 months, have you been in a fight, assault, or incident in which another inmate tried to harm you? (Percentage reporting “yes.”) S21. In the past 12 months, have you been in a fight, assault, or incident in which a correctional officer or other facility staff person tried to harm you? (Percentage reporting “yes.”) S25. In the past 12 months, have any of your personal possessions or belongings been taken by another inmate without your permission? (Percentage reporting “yes.”) Facility composition--Facility composition measures are based on inmate subgroups that have a high percentage who reported time in restrictive housing. When weighted, summed, and converted to percentages at the facility- level, the characteristics provide an overall indicator of the percentage of inmates in the facility with the characteristics. These measures include the percentage of inmates-- * with serious psychological distress * with a past mental health problem * held for a violent offense (excluding sex offenders) * with 11 or more prior arrests * with less than a high school diploma or equivalent * who were lesbian, gay, bisexual, or other sexual orientation (other than heterosexual) * who were ages 18 to 24. Facility climate--Facility climate measures are based on three items related to crowding and sufficient staff and nine items related inmate perceptions of staff: * S6. How crowded is it in your housing unit? (Percentage reporting “very crowded.”) * S7. How crowded is it outside of the housing unit—for example, in the dining hall, classrooms, gym, or work areas? (Percentage reporting “very crowded.”) * S29. In the past 12 months, do you think there has been enough staff at this facility to provide for the safety and security of inmates? (Percentage reporting “no.”) * S9. Please indicate whether you agree or disagree with each of the following statements. Staff at this facility… * S9a. Are generally fair * S9b. Do their best to make this facility safe and secure * S9c. Try to meet the needs of the inmates * S9d. Break up fights quickly * S9e. Use physical force only when necessary * S9f. Let inmates know what is expected of them * S9g. Generally treat inmates with respect * S9h. Follow facility rules when handling inmate complaints and grievances * S9i. Often write up inmates who don’t deserve it. The entire ACASI questionnaire (listed as the National Inmate Survey-3) and the shorter paper and pencil survey form (PAPI) are available on the BJS website. **************************************************************** The Bureau of Justice Statistics of the U.S. Department of Justice is the principal federal agency responsible for measuring crime, criminal victimization, criminal offenders, victims of crime, correlates of crime, and the operation of criminal and civil justice systems at the federal, state, tribal, and local levels. BJS collects, analyzes, and disseminates reliable and valid statistics on crime and justice systems in the United States, supports improvements to state and local criminal justice information systems, and participates with national and international organizations to develop and recommend national standards for justice statistics. William J. Sabol is director. This report was written by Allen J. Beck. Jessica Stroop and Tracey Snell verified the report. Jill Thomas and Morgan Young edited the report. Barbara Quinn and Morgan Young produced the report. October 2015, NCJ 249209 **************************************************************** ************************************************* Office of Justice Programs Innovation * Partnerships * Safer Neighborhoods www.ojp.usdoj.gov ************************************************* ********************************************** 10/15/2015 1:050pm JER **********************************************