Embed Size (px)
Citation preview
U.S. Department of Justice
Office of Justice Programs
BUREAU OF JUSTICE STATISTICS
Aging and mortality in the state prison population
Presented by
E. Ann Carson and Margaret Noonan Statisticians
Corrections Unit
U.S. Department of Justice
Office of Justice Programs
BUREAU OF JUSTICE STATISTICS
Aging of the state prison population, 1993-2013
Presented by
E. Ann Carson Statistician
Corrections Unit
3 www.bjs.gov
Disclosure
I have no actual or potential conflict of interest in relation to this presentation.
4 www.bjs.gov
Presentation outline
• How has the age of the state prison population changed over time?
• What is contributing to the change in the age distribution of the state prison population?
• Admissions
• Stock population
• What are the implications of an older prison population?
5 www.bjs.gov
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
50000
18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 72 74
State prisoners sentenced to more than one year by age, December 31, 1993, 2003, 2013
1993
2003
2013
Source: National Corrections Reporting Program data tool
6 www.bjs.gov
7 www.bjs.gov
Reasons for an increase in older offenders in the prison population • More older persons are being admitted
• Aging U.S. population
• Arrest rate of older persons
• Higher admission rate given arrest
• Older persons serving longer periods of time • Offense distribution
• Criminal history
• Enhanced sentences
8 www.bjs.gov
9 www.bjs.gov
Age distribution, U.S. resident and prison populations, December 31, 1993, 2003, 2013
Sources: BJS, National Corrections Reporting Program data tool, U.S. Census Bureau postcensal estimates
10 www.bjs.gov
11 www.bjs.gov
0
5
10
15
20
25
30
35
40
45
18-29 30-39 40-54 55+
Perc
ent
of
age
gro
up
Age at admission
1993 2003 2013
0
5
10
15
20
25
30
35
40
45
18-29 30-39 40-54 55+
Perc
ent
of
age
gro
up
Age at admission
1993 2003 2013
Sentenced state prisoners admitted for violent and drug offenses, by age, 1993, 2003, 2013
Violent offenses Drug offenses
Source: BJS, National Corrections Reporting Program data tool
12 www.bjs.gov
Sentenced state prisoners convicted of violent and drug offenses, by age, 1993, 2003, 2013
0
10
20
30
40
50
60
70
18-29 30-39 40-54 55+
Perc
ent
of
age
gro
up
Age on December 31
1993 2003 2013
Violent offenses Drug offenses
0
10
20
30
40
50
60
70
18-29 30-39 40-54 55+
Perc
ent
of
age
gro
up
Age on December 31
1993 2003 2013
Source: BJS, National Corrections Reporting Program data tool
13 www.bjs.gov
Age at prison release Years since prison release All offenses 1 year 5 year
24 or younger 51.3 % 84.1 % 25–29 45.4 80.3 30–34 43.4 77.0 35–39 44.4 78.1 40 or older 37.3 69.2
Violent offenses 1 year 5 year 24 or younger 45.6 % 80.4 % 25–29 39.6 76.7 30–34 38.6 72.0 35–39 42.1 74.0 40 or older 31.5 60.7
Drug offenses 1 year 5 year 24 or younger 50.0 % 85.4 % 25–29 44.3 80.4 30–34 41.2 76.1 35–39 40.7 77.0
40 or older 38.7 71.2
Cumulative rearrest rates for released prisoners from 30 states, 2005
Source: BJS, Recidivism of State Prisoners Released in 2005, NCJ 244205; http://www.bjs.gov/content/pub/pdf/rprts05p0510.pdf
14 www.bjs.gov
Reimprisonment of prisoners released from state prison in 1994, by age at release
Percent reimprisoned after --
Age at release 1 year 3 years
20 or younger 16.0 % 35.3 %
21 to 25 12.7 29.8
26 to 30 11.6 28.0
31 to 35 10.3 25.8
36 to 40 10.5 24.5
41 or older 6.7 17.9
Source: BJS Prisoner Recidivism Analysis Tool – 1994, http://www.bjs.gov/index.cfm?ty=datool&surl=/recidivism/index.cfm#
15 www.bjs.gov Source: A Primer:Three Strikes – The impact after more than a decade, Legislative Analyst’s Office, October 2005; http://www.lao.ca.gov/2005/3_strikes/3_strikes_102005.htm, accessed 2/29/16
16 www.bjs.gov
Percent of age group serving sentences enhanced under 2 and 3 strikes laws Age at census January, 2010 September 2015 18-19 5.6 % 7.1 % 20-24 11.4 19.7 25-29 17.3 27.2 30-34 22.1 29.5 35-39 27.0 32.6 40-44 30.8 31.3 45-49 34.0 35.9 50-54 33.7 38.8 55-59 31.5 42.5 60 plus 21.3 31.1
Number of offenders 41,009 39,436
Source: California Department of Corrections and Rehabilitation Second and Third Striker Felons and monthly population reports, http://www.cdcr.ca.gov/Reports_Research/index.html, accessed 3/2/2016
17 www.bjs.gov
Implications of an aging prison population • Costs
• From OIG report on federal prisoners:
“Aging inmates on average cost 8 percent more per inmate to incarcerate than inmates age 49 and younger….. In FY 2013, the average aging inmate cost $24,538 to incarcerate, whereas the average younger inmate cost $22,676. We found that this cost differential is driven by increased medical needs, including the cost of medication, for aging inmates….BOP institutions with the highest percentages of aging inmates also spent 14 times more per inmate on medication ($684) than institutions with the lowest percentage ($49).”
Source: “The Impact of an Aging Inmate Population on the Federal Bureau of Prisons”,
USDOJ Office of the Inspector General, May 2015,
https://oig.justice.gov/reports/2015/e1505.pdf
18 www.bjs.gov
Contact information
E. Ann Carson
Corrections Unit
Bureau of Justice Statistics
U.S. Department of Justice
202/616.2496
U.S. Department of Justice
Office of Justice Programs
BUREAU OF JUSTICE STATISTICS
Understand Mortality in State
Prison
Presented by
Margaret Noonan Statistician
Corrections Unit March 17, 2016
20 www.bjs.gov
2.2 Million Incarcerated
• State prisons and local jails hold the majority (90%) of the incarcerated population
• The majority of inmates are male • 93% of state prisoners
• 88% of jail inmates
• The majority are either White, non-Hispanic or Black, non-Hispanic • White (36% of prisoners; 36% of jail inmates)
• Black (40% of prisoners; 37% of jail inmates)
• Most are under the age of 35 • 49% of prisoners
• 60% of jail inmates
21 www.bjs.gov
Prison Populations: Admissions,
time served and offense
• Prisons admitted 630,000 sentenced prisoners in 2013
• About half of prisoners are violent felons • However, prisoners serve about 2 years before being released
• Most prisoners (95%) are eventually released
22 www.bjs.gov
Deaths in Custody Reporting
Program
• Began in 2000 with the passage of the Death in Custody Act • Created the DCRP
• Reauthorized at the end of 2014
• Collects individual level records on inmates who died in the custody of local jails and state prisons
• Nearly all jails participate (98% annually)
• All state departments of correction participate
23 www.bjs.gov
DCRP: Death overview
• Despite the majority of prison and jail inmates serving their time without incident, a small proportion of inmates die annually
• Approximately 3,000 state prisoners
• Prisoners and jail inmates have different mortality profiles • The majority (90%) of prisoner deaths are natural
• Cancer and heart disease are the top two causes of death
24 www.bjs.gov
DCRP: Inmate and resident mortality
• Despite having an estimated biological age that is 10 years beyond the chronological age of the inmate, inmates have a lower mortality rate than the resident population.
• Prisoners: 274 per 100,000
• Jail Inmates: 135 per 100,000
• U.S. residents: 822 per 100,000
• Risk factors limited for prisoners • Firearms, automobiles
• Inmate movement is controlled
• Demographic differences • Age is the largest contributor
25 www.bjs.gov
Race-Hispanic Origin Distributions in State Prison and U.S. Resident Population, 2007–2012
Sources: National Corrections Reporting Program, National Prisoner Statistics Program, NCHS WONDER
26 www.bjs.gov
What effect would a standardization
have?
• BJS had previously done small-scale standardizations for prisons and jails
• State prisoner mortality rate was 19% lower than the resident population
• White and Hispanic death rates were slightly higher than their resident population counterparts
• Black, non-Hispanic prisoners were 57% less likely to die than their resident population counterparts
• The jail standardization was limited to suicides and found that if the resident population resembled the jail population, the suicide rate in the resident population would increase
27 www.bjs.gov
Most recent standardization
project
• BJS established that adjusting for known demographic differences between the correctional population and the U.S. resident population reduced the mortality gap between residents and inmates.
• However, the earlier work was limited in scope • Prisons – overall mortality, e.g. not studied by cause
• Jails – limited to suicides
• BJS decided to revisit the topic with the ‘Understanding Mortality in State Prison: Do Male Prisoners Have an Elevated Risk of Death’ paper, which was published in ‘Justice Research and Policy’
28 www.bjs.gov
‘Understand Mortality in State
Prison’ paper
• The paper was limited to state prisoners and to males • State prisoners account for more than half of the incarcerated
population in the U.S.
• On average, state prisoner deaths account for 71% of deaths in incarcerated populations
• As a result, there is a larger number of cases to work with when looking at deaths by cause
• Between 2001-2010, there were 45,539 prisoner deaths v. 10,010 jail inmate deaths
29 www.bjs.gov
Decedent characteristics
30 www.bjs.gov
‘Understanding Mortality’:
Methods
• To address the mortality differential between inmates and the U.S. resident population, BJS compared the two populations after adjusting for known differences
• Adjusted for • Sex
• Race/Ethnicity
• Age
• Excluded MVA deaths in the majority of the analysis
• Data sources • DCRP
• CDC Wonder model
31 www.bjs.gov
‘Understand Mortality’: Methods,
continued
• Limited to male inmates • Mortality varies by sex
• Men can’t die of ovarian cancer
• Women can’t die of prostate cancer
• Male account for the majority of state prisoners (93%) and prisoner deaths (96%)
• Ages 18 to 75
• Less than 5% of prisoner deaths were of inmates age 76 or older
32 www.bjs.gov
Source: National Corrections Reporting Program, National Prisoner Statistics Program
Percentage Distribution of State Prisoners and Deaths: State Prisons, 2007–2012
33 www.bjs.gov
‘Understanding Mortality’:
Findings
• State prisoners who died demographically differed from other prisoners
• Of those who died – • Half were white, non-Hispanic
• More than one-third were 55 or older
• More than two-thirds were violent offenders
34 www.bjs.gov
‘Understand Mortality’: Findings
• The major finding is that even after standardizing the U.S. resident population to resemble the prison population, the adjusted U.S. resident mortality rate was higher than prisoners
• Mortality rates post adjustment • Resident population: 325 per 100,000 male U.S. residents
• Prison population: 259 per 100,000 male state prisoners
• Prior to standardization, male U.S. residents died from cancer (200) and heart disease (220) at three times the rate as their prisoner counterpart. Post-adjustment the rates were equal. Conversely, AIDS-related and liver disease deaths remained higher in the prison population post-adjustment
35 www.bjs.gov
36 www.bjs.gov
‘Understand Mortality’: Findings
• In addition to the analysis by cause, we compared prisoner mortality to U.S. resident mortality by age and race-Hispanic origin.
• Overall, prisoners regardless of age group or race-Hispanic origin died at a lower rate than their adjusted U.S. resident counterpart with one exception
• White males 45 and older
37 www.bjs.gov
38 www.bjs.gov
‘Understanding Mortality’:
Findings
• After assessing overall mortality by standardizing for age and race-Hispanic origin differences, we revisited mortality by cause and split the findings by race
• Once split by race-Hispanic origin, the mortality rates between male prisoners and their age-adjusted U.S. resident counterparts varied.
• For white, non-Hispanic prisoners: cancer, heart disease
• For black, non-Hispanic prisoners: disease of the kidney, AIDS-related causes and homicides
• For all inmates: diseases of the liver
39 www.bjs.gov
40 www.bjs.gov
‘Understanding Mortality’:
Discussion
• In most instances, the age-race adjusted resident population males had worse mortality outcomes than prisoners
• Cancer, heart disease and respiratory illnesses were about equal among prisoners and the U.S. resident pop, post-adjustment
• Suicide, drug/alcohol and homicides intoxication mortality rates increased in the resident population post adjustment
• Suggesting these causes maybe more due to being a male between the ages of 18 and 75 than being a prisoner
• Accidents were the only unnatural deaths whose mortality differed very little post-adjustment
41 www.bjs.gov
‘Understanding Mortality’:
Discussion
• Disease of the liver, septicemia and AIDS-related deaths were higher in the prison population even after adjusting the U.S. resident population by sex, race and age.
• Suggesting that these causes could be due to lifestyle lead prior to incarceration.
• Prisoners who died were different from the stock prisoner population
• Specifically, they were:
• Older than the stock population (73% were 45 or older)
• Older at admission (mid-to-late 40s)
• Disproportionate white and violent
42 www.bjs.gov
‘Understanding Mortality’:
Discussion
• Why do white male prisoners experience a greater die-off than other race-ethnic groups?
• Disproportionately violent offenders
• Disproportionately rapists and murderers
• On average, five years older at death than other prisoners
• Median age at death: 54 years
• Older at admission
• Median age at admission: 43 years
• White, non-Hispanic inmates: 46 years of age
• Despite being older at admission, they served about 5 years prior to death.
43 www.bjs.gov
Further areas of research
• We would like to do similar analysis with female prisoners
• We are currently working to reproduce the report with jail inmates
• We have begun work on assessing the effect of acute intoxication and history of substance abuse
44 www.bjs.gov
Contact information
Margaret Noonan
Statistician
Bureau of Justice Statistics
