U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics Special Report Suicide and Homicide in State Prisons and Local Jails August 2005, NCJ 210036 --------------------------------------------------------------- This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.wk1) and the full report including tables and graphics in .pdf format are available from: http://www.ojp.usdoj.gov/bjs/abstract/shsplj.htm --------------------------------------------------------------- By Christopher J. Mumola BJS Policy Analyst -------------------------------------------- Highlights State prison, local jail suicide rates have fallen sharply since the 1980's * In 2002 the suicide rate in local jails (47 per 100,000 inmates) was over 3 times the rate in State prisons (14 per 100,000 inmates). * The suicide rate in the Nation's 50 largest jail systems (29 per 100,000 inmates) was half that of other jails (57 per 100,000). * Violent offenders in both local jails (92 per 100,000)and State prisons (19 per 100,000) had suicide rates over twice as high as those of nonviolent offenders (31 and 9 per 100,000 respectively). Homicide rates in State prisons dropped 93% from 1980 to 2002 * Homicide rates were similar in local jails (3 per 100,000) and State prisons (4 per 100,000). * 67% of homicide victims in State prisons had served at least 2 years; 37% had served 5 years. * Violent offenders were the victims of most State prison homicides (61%), and their jail homicide rate (5 per 100,000) was over twice that of nonviolent offenders (2 per 100,000). ------------------------------------------------ Data from new Bureau of Justice Statistics (BJS) data collections offer the first opportunity to analyze the personal characteristics, current offenses, and environmental factors surrounding inmate deaths in local jails and State prisons nationwide. To implement the Death in Custody Reporting Act of 2000 (PL 106-297), BJS began collecting inmate death records from all local jails in 2000 and expanded reporting to include State prisons in 2001. In this first report from the Deaths in Custody Reporting Program, data from 2000 to 2002 highlight inmate and facility characteristics related to high risks of suicide and homicide. Jail suicide rates declined steadily from 129 per 100,000 inmates in 1983 to 47 per 100,000 in 2002. In 1983 suicide accounted for the majority of jail deaths (56%), but by 2002, the most common cause of jail deaths was natural causes (including AIDS)(52%), well ahead of suicides (32%). Suicide rates in State prison fell from 34 per 100,000 in 1980 to 16 per 100,000 in 1990, and have since stabilized. State prison homicide rates dipped sharply from 1980 (54 per 100,000) to 1990 (8 per 100,000). By 2002 prison homicide rates had declined further, down to 4 per 100,000. Homicide rates in local jails were more stable, declining slightly from 5 per 100,000 in 1983 to 3 per 100,000 in 2002. ------------------------------------------ The Death in Custody Reporting Act of 2000 The passage of the Death in Custody Reporting Act of 2000 (DICRA, PL 106-297) dramatically altered programs collecting data on inmate deaths. Prior to the act, BJS conducted annual counts of State prisoner deaths. Counts of jail inmate deaths were collected in the Census of Jails, which is conducted every 5 or 6 years. For both populations, death counts were obtained by gender and by general cause categories, such as illness/natural causes, AIDS, suicide, and homicide. These aggregate counts of deaths did not allow for analysis of individual death cases. DICRA was attached as a grant requirement of the Violent Offender Incarceration and Truth-in-Sentencing (VOI/TIS) incentive grant program. Beginning in 1996, these grants provided over $2.5 billion to all 50 States and U.S. Territories for expanding prison capacity to house violent offenders for longer periods. Each State receiving VOI/TIS funds was required under DICRA to report: "on a quarterly basis, information regarding the death of any person who is in the process of arrest, is en route to be incarcerated, or is incarcerated at a municipal or county jail, State prison, or other local or State correctional facility (including any juvenile facility) that, at a minimum, includes -- (A) the name, gender, race, ethnicity, and age of the deceased; (B) the date, time, and location of death; and (C) a brief description of the circumstances surrounding the death." BJS developed a new series of collections to meet the mandates of the act. Aggregate counts of deaths were replaced by detailed, individual inmate death records, collected every 3 months from over 3,000 jail jurisdictions, 50 State prison systems, juvenile correctional authorities in all 50 States, and roughly 18,000 State and local law enforcement agencies nationwide. These new data collections were phased in over 4 years, with local jails reporting in 2000, followed by State prisons in 2001 and State juvenile authorities in 2002. A network of statewide law enforcement reporters began submitting arrest-related death records to BJS in 2003. With these new collections, BJS has enhanced both the frequency and scope of its data on inmate mortality. Among other improvements, BJS now collects information on specific medical causes of death, as determined by a coroner or medical examiner. BJS replaced a general category of "illness/natural causes," with specific categories of medical conditions related to mortality, such as cancer, heart disease, and hepatitis C. A detailed analysis of these fatal medical conditions will be the focus of the next report from this data collection series. -------------------------------------------------- -------------------------------------------------- Long term trends show sleep declines in rates of State prisoner homicide and local jail inmates suicide Over the past two decades, State prison and local jail inmate mortality rates have displayed some dynamic changes. Suicide was the leading cause of death among jail inmates in 1983 (129 per 100,000 inmates); by 1993 that rate had been cut by more than half (54 per 100,000 inmates), and illness/natural cause (67 per 100,000) had become the most common cause of jail deaths. By 2002 the jail suicide rate (47 per 100,000) had fallen to nearly a third of the 1983 rate. Rates of death from AIDS-related causes in jails also declined; the 2002 rate (8 per 100,000) was less than half of the 1988 rate (20 per 100,000). As a result of these reductions, the overall mortality rate in local jails dropped 37% between 1983 and 2002. State prison suicide rates have historically been much lower than those of jails, but these also dropped sharply from 34 per 100,000 in 1980 to 14 per 100,000 inmates in 2002. Even more dramatic was the decline in homicide deaths, from 54 per 100,000 inmates in 1980 to 8 per 100,000 inmates in 1990, and to 4 per 100,000 inmates in 2002. With the introduction of new therapies during the 1990's, AIDS-related mortality rates in State prison fell rapidly from 100 per 100,000 inmates in 1995 to 15 per 100,000 inmates 5 years later. Overall State prisoner mortality rates have grown slightly (6%) since 1980, mostly due to illness/natural causes (up 40% since 1980). ------------------------------------------ Nationwide, 337 State prisoners committed suicide during 2001-02 Suicide and homicide accounted for a combined 7% of all State prisoner deaths during 2001-02. The average annual suicide rate of State prisoners (14 suicides per 100,000 prisoners) was a third of that of local jail inmates (48). Prison suicide rates showed wide variation at the State level. New Hampshire, Nebraska, and North Dakota all reported no suicide deaths during the 2-year period. Another six States had suicide rates of 5 per 100,000 prisoners or lower. Thirteen States had suicide rates of at least 25 per 100,000 prisoners, led by South Dakota (71), Utah (49), Vermont, Alaska, and Arkansas (each with 36). In most State prison systems, suicides were rare events. Only 9 States reported as many as 10 prisoner suicides during this period, with 42% of all suicides taking place in four States. California (52), Texas (49), New York (21), and Illinois (20) reported 142 of the Nation's 337 State prisoner suicides. About half of all States (24)recorded 3 or fewer suicides. 87 State prisoners became victims of homicide over 2 years Most States did not have any prisoner homicides in the course of a year. During 2001, 31 States reported no prison homicides; 29 States did not report a homicide during 2002. Three States reported 43% of all homicides -- California (21), Texas (10), and Maryland (6). No other State reported more than 5 homicides during 2001-02. Homicide rates were low in most States, and 5 had a rate of at least 10 homicides per 100,000 prisoners, led by South Dakota (34) and New Mexico (17). However, even in these 5 States, a combined total of 13 homicides were reported over 2 years. Suicide rate in the Nation's 50 largest jail jurisdictions half that of all other jails There are over 3,300 local jails operated by county and municipal jurisdictions nationwide. Jails typically hold unsentenced offenders, those sentenced to less than a year, and offenders sentenced to longer terms who are awaiting transfer to State prison. As a result, almost every State prisoner has been through a period of jail confinement. Over a 3-year period (2000-02), the Nation's 50 largest jail jurisdictions reported a total of 1,037 deaths from all causes. This death count represented a higher overall mortality rate (167 per 100,000 inmates in the average daily population) than other jails (140 per 100,000). Mortality rates varied widely among the top 50 jurisdictions. Twelve of these 50 jurisdictions had overall mortality rates of fewer than 100 deaths per 100,000 inmates, led by Suffolk County, Massachusetts (29), Orange County, California (42), and Mecklenburg County, North Carolina (52). Another 16 of the top 50 jurisdictions had rates of 200 or more deaths per 100,000 inmates, led by Baltimore City, Maryland (381), Jacksonville City, Florida (341), and Davidson County, Tennessee (291). The 50 largest jail jurisdictions collectively had a comparatively low prevalence of suicide. Inmate suicides accounted for 17% of all deaths in these 50 largest jurisdictions but were the cause of 41% of the deaths in all other jails. The suicide rate of the 50 largest jurisdictions (29 per 100,000) was half that of all other jails (57). Eight of the top 50 jurisdictions reported no suicides during 2000-02, and another 4 jurisdictions had a suicide rate of 10 per 100,000 or less. Ten of these jurisdictions also had suicide rates of at least 50 per 100,000 inmates, led by Clark County, Nevada (107), Wayne County, Michigan (97), and Baltimore City, Maryland (88). During 2002 the Nation's smallest jails had a suicide rate 5 times that of the largest jails On an average day in 2002, over 40% of the nation's jails housed fewer than 50 inmates, while 2% of all jails held at least 1,500 inmates. Rates of inmate suicide were closely related to jail size, with the smallest facilities recording the highest suicide rates. The Nation's largest jail facilities recorded the lowest suicide rates (32 per 100,000 inmates). The suicide rate rose steadily as jail size decreased and was over 5 times higher (177 per 100,000) in jails holding fewer than 50 inmates. However, given their small populations, these jails accounted for 14% of all jail suicides. Jail suicide rates drop by over 90% when based on "at-risk" population BJS has usually based jail mortality rates on the average daily population of inmates (an ADP of under 700,000). A more sensitive measure of jail mortality would reflect the far larger number of persons admitted into these facilities over the entire year (nearly 13 million). All of these persons admitted are at risk of dying while held in jail. Past attempts to collect admission data for a whole year were unsuccessful, because many jail information systems do not keep cumulative counts of admissions. As part of the new Deaths in Custody records, BJS collected annual admission data, which can be used to calculate an at-risk measure of mortality for the Nation's largest jails. While the 50 jurisdictions had an average daily population of 207,471 over the 3-year period, these same jails had an average of 2,827,133 admissions each year. As a result, the at-risk mortality rates of these jurisdictions are far lower. The ADP rate of overall mortality in the top 50 jurisdictions (167 per 100,000)was 15 times the at-risk rate(11). The ADP-based suicide rate for these 50 jurisdictions (29 per 100,000)was 14 times the at-risk suicide rate for these facilities (2 per 100,000). Males and white inmates had the highest rates of suicide in jails Among local jail inmates, mortality rates varied across demographic subgroups. In terms of deaths from all causes, male inmates had a higher death rate (150 per 100,000 inmates) than females (130). Gender was a stronger factor in suicide rates: males (50 per 100,000)were 56% more likely to commit suicide than female jail inmates (32). The homicide rate of male jail inmates was low (3 per 100,000) and female inmates did not experience a single homicide during 2000-02. The most common cause of death among jail inmates was illness (48% of all jail deaths during 2000-02). As a result, the overall mortality rates of jail inmates steadily rose with age. Among jail inmates age 18-24, the mortality rate was 60 per 100,000; this rate was 3 times higher for inmates age 35-44 (179), and over 11 times higher for inmates age 55 or older (694). The only exception to this pattern was the death rate of jail inmates under age 18 (138 per 100,000), who made up less than 2% of all jail deaths. Jail suicide rates also increased with inmate age. Inmates age 18-24 were the least likely to commit suicide (38 suicides per 100,000 inmates); this rate increased 24% for inmates age 25-34 (47), and 39% for inmates age 35-44 (53). The oldest inmates, age 55 or older, had the highest rate of suicide (58 per 100,000) among adult inmates. The youngest jail inmates were the exception to this pattern; jail inmates under 18 had the highest suicide rate in local jails (101 per 100,000). Given their relatively small numbers, inmates under the age of 18 committed 35 of the 918 jail suicides recorded nationwide over 3 years. Inmate age did not have any clear relationship to jail homicide rates, which were no higher than 7 per 100,000 inmates for all age groups. Both the youngest (under 18) and oldest (55 or older) inmates had no homicide deaths during 2000-02. White jail inmates 6 times more likely to commit suicide than black inmates and 3 times more likely than Hispanic inmates Mortality rates displayed substantial differences by race and ethnicity. Death rates from all causes for both black (118 per 100,000 inmates) and Hispanic (98) jail inmates were at least 20% below the overall jail inmate mortality rate (148). But the death rate of white jail inmates (219 per 100,000) was 86% higher than that of black inmates and over twice as high as the rate for Hispanic inmates. Differences across racial/ethnic categories were more pronounced in jail inmate suicide rates. The suicide rate of white jail inmates (96 per 100,000 inmates) was more than triple that of Hispanic inmates(30) and was 6 times the suicide rate for black inmates(16). White inmates accounted for nearly three-quarters of all jail inmate suicides during 2000-02. Unlike the overall mortality and suicide rates, homicide rates were not related to race/ethnicity. White, black and Hispanic jail inmates were all equally likely to be victims of a homicide (3 deaths per 100,000 inmates). State prisoners age 45 or older made up 17% of inmates but 66% of deaths Just as in local jails, male State prisoners had higher overall mortality rates than female prisoners. While this difference was modest in local jails(the male death rate was 15% higher), males (251 deaths per 100,000 prisoners) were 79% more likely than females (140)to die in State prison during 2001-02. In contrast, male and female suicide rates in State prisons were similar (14 suicides per 100,000 males, compared to 10 per 100,000 females). In local jails men were over 50% more likely than women to commit suicide. The increase in mortality rates seen in older jail inmates was also evident among older State prisoners. The overall death rate was lowest for State prisoners age 18-24 (34 per 100,000). The death rate was over 5 times higher for State prisoners age 35-44 (182) and nearly 17 times higher for prisoners age 45-54 (571). The mortality rate of the oldest prisoners, age 55 or older, was highest (2,019 -- or 59 times higher than the rate for prisoners age 18-24). Deaths attributed to "illness/natural cause" made up 80% of all State prison deaths reported during 2001 -02. Two-thirds of all State prison deaths involved inmates age 45 or older, though such inmates represented 17% of all State prisoners held during 2001-02. Despite the close relationship between age and the overall mortality rates in State prison, inmate age was not related to suicide rates. State prisoner suicide rates ranged from 13 to 14 suicides per 100,000 prisoners for every age group over 18. The suicide rate of State prisoners under 18 was 4 times higher (52 per 100,000), but this age group accounted for less than 0.3% of State prisoners and had 3 suicides nationwide over 2 years. By comparison, 116 prisoners age 25-34 committed suicide during 2001-02. Age also showed no relationship to State prison homicide rates, with all age groups over age 18 recording a homicide rate of either 3 or 4 per 100,000 inmates. No reported homicides involved State prisoners under age 18 during 2001-02. Black inmates had the lowest suicide and homicide rates in State prisons As in local jails, white inmates had the highest overall mortality rate (327 deaths per 100,000 prisoners). While the mortality rate of white jail inmates was 86% higher than that of blacks and 123% higher than that of Hispanics, the differences in State prison were smaller. White State prisoners were 35% more likely than Hispanic inmates (243 per 100,000) and 58% more likely than black prisoners (207)to die during 2001-02. White inmates had the highest suicide rate of all State prisoners (22 suicides per 100,000 inmates). This rate was 22% higher than the Hispanic suicide rate (18 per 100,000). By comparison, white inmates in local jails were 3 times more likely than Hispanics to commit suicide. Black inmates had the lowest suicide rate of all State prisoners (8 per 100,000). Blacks were about a third as likely as whites to commit suicide in State prison and less than half as likely as Hispanics. Homicide rates were less than 10 per 100,000 State prisoners for all racial/ethnic groups during 2001- 02. Hispanic inmates were the most likely to be killed in State prisons (7 homicides per 100,000 inmates), which was over 3 times the homicide rate of black inmates (2 per 100,000). The homicide rate for white inmates (5)almost matched the rate for all State prisoners (4). Violent offenders committed suicide at nearly triple the rate of nonviolent offenders in jails The death rate of violent offenders in local jails (212 per 100,000) was 75% higher than that of nonviolent offenders (121), but this difference was larger in cases of suicide. The suicide rate of violent jail inmates (92 per 100,000)was nearly triple that of nonviolent offenders (31). Kidnaping offenders had the highest suicide rate (275), followed by those inmates held for rape (252) or homicide (182). Among all nonviolent offenders, only probation/ parole violators had a suicide rate of at least 100 per 100,000 (118). Drug offenders were found to have the lowest rates of mortality, particularly suicide. Drug offenders were the only group that had fewer than 100 deaths from all causes per 100,000 jail inmates (92). The suicide rate of drug offenders (18 per 100,000) was the lowest among offender groups. Violent offenders (92) were 5 times more likely to commit suicide than drug offenders, and public-order offenders were more than twice as likely to commit suicide (42). Local jails had an average of fewer than 20 inmate homicides each year Over 3 years (2000-02), there were 59 jail inmate homicides reported nationwide, resulting in a rate of 3 jail inmate homicide deaths per 100,000 inmates. Violent offenders were the most likely to be killed in local jail (5 homicides per 100,000 inmates), followed by property and public-order offenders (3 for both). Drug offenders (1 per 100,000) had the lowest homicide victimization rate of all offenders. Kidnaping offenders had the highest rate of jail inmate homicide (15 per 100,000 inmates -- 5 times the rate for all inmates), followed by inmates held for rape (9) and violation of parole/ probation (7). But even among these offenders with the highest homicide rates, a combined total of eight homicides took place nationwide over this 3-year period. Drug offenders had the lowest suicide and homicide rates of all State prisoners State prison mortality rates showed similar patterns by offense type. Violent offenders not only had the highest overall mortality rate (312 deaths per 100,000 prisoners), they were the only State prisoners with a death rate of at least 200 per 100,000 prisoners held. Property and public-order offenders each had a rate of 184 deaths per 100,000, followed by drug offenders (166). Compared to violent offenders in local jails (92 suicides per 100,000 inmates), the suicide rate of violent offenders in State prison (19 per 100,000) was much lower. But among State prisoners, violent offenders were more than twice as likely to commit suicide as nonviolent offenders (9 per 100,000). Kidnapers had the highest suicide rate (36 per 100,000 prisoners), followed by offenders held for homicide (29), sexual assault (23), and assault (20). Among nonviolent offenders, probation/parole violators had the highest suicide rates (18 per 100,000), followed by offenders held for arson (16), burglary (14), and obstruction of justice (14). Drug offenders recorded the lowest suicide rates of all State prisoners (6 per 100,000 inmates). The rate of homicide in State prison was 4 per 100,000 prisoners, and varied little across offense types. Three types of offenders had as many as 10 homicides per 100,000 prisoners -- arsonists (16), kidnapers (15), and probation/parole violators (12). Among these three categories with the highest homicide rates, the number of homicides was small, with a total of nine prisoners killed over 2 years. State prisoners convicted of fraud and driving while intoxicated had the lowest rate of homicide, with zero homicides reported for 2001-02. Nearly half of jail suicides occurred in the first week of custody Jail inmate suicides were heavily concentrated in the first week spent in custody. Forty-eight percent of all jail suicides during 2000-02 took place during the inmate's first week following admission. In particular, almost a quarter of all jail suicides took place either on the date of admission to jail (14%) or the following day (9%). The frequency of jail suicides slowed after the initial week, with the second week of custody accounting for 10% of jail suicides. The next 2 weeks in custody (days 15 to 30) accounted for even fewer suicides(8%). Despite this early concentration of suicides, more suicides took place after the 60th day in jail(24%) than during the first 2 days (23%). The median time served in jail prior to committing suicide was just over 1 week (9 days), but this period of time varied across demographic and criminal offense categories. Females spent less than half as much time as males in jail prior to committing suicide (median time served: 4 days for females and 10 days for males). The median length of time served by Hispanic inmates prior to suicide (23 days) was over twice as long as the time for white inmates (9 days) and nearly 4 times longer than that for black inmates (6 days). Of all offender groups, public-order offenders spent the shortest time in custody prior to committing suicide; half of their suicides took place in the first 3 days of custody. Property and drug offenders each had a median time served of about a week (7 and 8 days, respectively)prior to suicide. Violent offenders spent the longest time in custody prior to suicide; half of their suicides took place after spending 3 weeks in jail (20 days). 7% of State prison suicides took place during the first month In State prison, suicides were less concentrated around admission. Sixty-five percent of jail suicides occurred in the first 30 days, but 7% of prison suicides took place during the first month. Most State prison suicides (65%)took place after the inmate's first year of confinement, and 33% took place after the inmate had served at least 5 years in prison. The median time served in State prison prior to a suicide (30 months) was over 100 times longer than in local jails (9 days). Male (30 months) and female (29 months)State prisoners spent almost identical amounts of time in prison before committing suicide. However, race was related to the length of time served prior to suicide. Half of all suicides by white inmates occurred in the first 21 months of custody, while the corresponding figures for black (40 months) and Hispanic inmates (49 months)were twice as long. Violent State prisoners spent more time in custody prior to suicide than other offenders(median time served was 45 months). Drug offenders were the only other offender group who served a median of at least a year in State prison prior to their suicides (18 months), followed by property (10 months) and public-order offenders(9 months). At least 80% of suicides in prison and jail occurred in the inmate's cell; time of day not a factor The vast majority of both local jail (80%) and State prison (87%) inmate suicides took place within the inmate's cell or room. Temporary holding areas (lockups) were the next most common location of suicide events (10% in jails, 4% in prisons). Common areas such as cafeterias, libraries, and recreational areas were the scene of very few suicides (6% in jails, 3% in State prisons), as were areas outside of the correctional facility (2% of jail suicides, 3% percent of prison suicides). Suicide events in both local jails and State prisons showed little relationship to the time of day. Aside from morning hours (20% of jail suicides), the frequency of suicides in other parts of the day varied from 24% (during afternoon hours) to 28% (evening and overnight hours). Similar data were reported for suicides in State prisons. Twenty percent of State prisoner suicides took place in the morning, with all other times of day varying from 25% (evening and overnight hours) to 30% (afternoon hours). In both local jails (94%) and State prisons (89%), the majority of suicide events were followed up by the performance of an autopsy or post-mortem examination by a medical examiner or county coroner. Most jail homicides occurred at least 2 weeks after admission During 2000-02 the number of homicides in the more than 3,000 jail jurisdictions nationwide had an average of fewer than 20 each year. In State prisons, which held nearly 1.2 million inmates nationwide, there were fewer than 50 homicides each year during 2001-02. These homicide counts resulted in a rate of less than 5 homicides per 100,000 inmates in both State prison (4 per 100,000) and local jail (3 per 100,000 inmates, based on ADP). In the 50 largest jails nationwide, the at-risk rate of homicide averaged 0.4 per 100,000 inmates held during the year. Unlike suicides, homicides in local jails were not concentrated in the first few days following admission. Twelve percent occurred in the first 2 days in custody, but 54% took place after the inmate had served at least 2 weeks in jail. The median length of time served prior to a homicide death (29 days) was triple that of suicide deaths in local jails (9 days). Two-thirds of homicide victims in State prison had served at least 2 years in prison; nearly 40% had served 5 years or more The initial months following admission to prison accounted for a small percentage of State prisoner homicides. One percent of prison homicides took place during the victim's first month in prison, and less than a tenth of homicide victims had served fewer than 6 months (8%). A fifth of homicides involved State prisoners who had served less than a year. Among the 5,824 total prisoner deaths reported nationwide during 2001-02, fewer than 20 were homicides of a first-year inmate. Sixty-seven percent of State prison homicide victims had spent at least 2 years in prison, while 37% had served at least 5 years. The median time served in State prison by homicide victims was 44 months. Hispanic homicide victims (with a median time served of 22 months) were killed after serving less than half as much time as white (46 months) or black (55 months) prisoners. Public-order offenders were the most likely to be killed early in their prison terms, with a median time served of just under 2 years (23 months). The median term served by both drug (40 months) and property (45 months)offenders was about twice as high. Violent offenders had the longest amount of time served in prison prior to being killed, with a median term of almost 5 years (55 months). Prison suicides took place almost exclusively inside the deceased's cell or room (87%); no other location accounted for even 5% of suicide events. However, over a quarter of all prison homicides (29%) took place in common areas within prisons, such as cafeterias, libraries, workshops, and recreational yards. A small percentage of homicide events took place in either a temporary holding area or a location outside of the prison facility (3% for each). Prisoners' cells or rooms (61%) were the most likely scene of a homicide in State prison. State prison homicides were over 3 times more likely to occur during the morning (39% of homicides) than between midnight and 6 a.m. (11%). Nearly all State prison homicides (92%) resulted in an autopsy or post-mortem exam of the deceased. All but 8 of 87 prisoner homicides during the 2-year period were committed by other inmates (91%). Of those "other homicide" events, most involved escape attempts or cases in which assailant identity was not established. Homicide rate of U.S. residents, when standardized, 10 times the rate of jail inmates in 2002 According to rates compiled by the Centers for Disease Control and Prevention (CDC), the U.S. Resident population experienced 6 homicides and 11 suicides per 100,000 residents. The homicide rates for both State prisoners (4) and jail inmates (3) were lower than that for the U.S. population. Suicide rates for both State prisoners (14) and jail inmates (47) were higher than the rate for the resident population. However, reliable comparisons of such rates require closer analysis. The demographic compositions of inmate populations do not reflect those of the U.S. resident population. In 2002 the U.S. population was 51% female, 81% white, and 22% age 55 or older; by comparison, the State prison population was 6% female, 50% white, and 4% age 55 or older. The suicide and homicide rates of these demographic subgroups vary substantially. For example, the homicide rate of black males age 18-24 in the resident population (108 per 100,000) was over 8 times that for white males of the same age (13). As a result, the differing rates of death seen in the general population and correctional facilities reflect differences in demographic makeup as much as differences in the relative safety of these environments. To improve the comparison of mortality risks, the resident population rates can be standardized by age, race, and gender to match the proportions seen in prisons and jails. The resulting rates estimate what the resident population mortality rates would be if the U.S. resident population had the same demographic composition as prisons and jails. Standardized to match the State prison population, the resident population had a homicide rate (35 per 100,000) nearly 9 times the rate of homicide in State prisons (4). Standardizing to match local jail demographics yields a greater difference the resident rate (32 per 100,000) being nearly 11 times higher than the rate in jails (3). State prisoners had a higher rate of suicide (14 per 100,000) than the overall resident population (11). Once standardized to match the State prisoner population, the U.S. resident rate of suicide (18) exceeded that of State prisoners in 2002. The standardized resident suicide rate (17 per 100,000) was less than half of the jail suicide rate based on ADP (47). However, an at-risk rate of jail suicide would be a more appropriate comparison, but not all jails reported the needed admission data. Based on the at-risk measure of suicide for the top 50 jail jurisdictions, an at-risk jail suicide rate for all jails would likely be less than a tenth of the ADP measure. (See page 5.) Methodology BJS phased in data collection activity under the Death in Custody Reporting Act of 2000 (PL 106-297), with the first collection of death records covering only local jail facilities. The 2000 jail collection covered the entire calendar year (the act became law in October of 2000), while subsequent collections were done on the quarterly basis required by the act. BJS requires a quarterly report from all jails which had an inmate death during the period. All jails were instructed to complete an annual summary of death reports and population counts (to allow for calculation of death rates). Jail response rates for all 3 years were over 99%. In 2000, 3,063 jurisdictions responded, and 9 refused, for a response rate of 99.7%. Data for 2001 was submitted by 3,049 jurisdictions, with 2 refusals, for a response rate of 99.9%. In 2002, data were submitted by 3,030 jurisdictions and refused by 6, for a response rate of 99.8%. Quarterly collection of State prison inmate death records began in 2001. These records were collected from State departments of corrections, rather than from each prison facility. For all years, BJS has had 100% participation from all 50 State prison systems. Data were also collected from the District of Columbia for 2001 in which it still operated a prison system, prior to transferring sentenced felons to the custody of the Federal Bureau of Prisons. Copies of all questionnaires collected under the Deaths in Custody series can be found on the BJS website at . Standardized U.S. resident death rates Homicide and suicide rates for specific age, race, and gender groups within the U.S. population can be queried from the Centers for Disease Control and Prevention's online injury mortality reports http://webappa.cdc.gov/sasweb/ncipc/mortrate.html>. For example, the 2002 suicide rate for white females, age 35-44, was 7.8 per 100,000. These rates were then standardized to match the characteristics of State prison and local jail populations by weighting the rates by the proportion of all inmates represented by that subgroup. The sum of all of the weighted subgroup rates provides the standardized rate for the resident population. Population proportions for these gender, race and age subgroups of inmates were derived from the National Corrections Reporting Program (for State prisoners) and the 2002 Survey of Inmates in Local Jails(for jail inmates). Population bases for mortality rates Annual mortality rates were based on different population counts: 1. For prisons the custody population on June 30 of each year 2. For local jails the average daily population in each calendar year. Estimates of the population at risk for the largest jail jurisdictions combined the population on January 1 and admissions during the year. ------------------------------------------ The Bureau of Justice Statistics is the statistical agency of the U.S. Department of Justice. Lawrence A. Greenfeld is director. Christopher J. Mumola wrote this report, under the supervision of Allen J. Beck. Doris J. James, Lauren E. Glaze, and Rebecca L. Medway verified the report, and Tom Hester edited it. Christopher J. Mumola, under the supervision of Allen J. Beck, designed the survey, developed the questionnaires, and monitored data collection and data processing. Data collection and processing of State prison death records were carried out by Lara Reynolds. Data collection and processing of local jail death records were carried out by Pamela Butler, Margaret Ferguson, Patricia Torreyson, and Pearl Chase, under the supervision of Charlene Sebold, Governments Division, Census Bureau, U.S. Department of Commerce. August 2005 NCJ 210036 Office of Justice Programs Partnerships for Safer Communities http://www.ojp.usdoj.gov ------------------------------------------ End of file 08/12/05 ih