Transcript Pathways to Desistance - National Juvenile Justice Network

Pathways to Desistance:
A Longitudinal Study of Serious
Adolescent Offenders
Edward P. Mulvey, Ph.D.
Law and Psychiatry Program
Department of Psychiatry
University of Pittsburgh School of Medicine
National Juvenile Justice Network
Teleconference
January 28, 2010
Pathways to Desistance Study
Supported by









Office of Juvenile Justice & Delinquency Prevention
National Institute of Justice
National Institute on Drug Abuse
John D. & Catherine T. MacArthur Foundation
Pennsylvania Commission on Crime & Delinquency
Arizona Governor’s Justice Commission
Robert Wood Johnson Foundation
William Penn Foundation
William T. Grant Foundation
Pathways to Desistance Study
Working Group Members

Edward Mulvey

Laurence Steinberg

Elizabeth Cauffman

Laurie Chassin

George Knight

Carol Schubert

Sandra Losoya

Robert Brame

Jeffrey Fagan

Alex Piquero
Clarion soldier to get Medal of
Honor
19-year-old threw himself on
grenade to save comrades
Saturday, May 24, 2008
By Milan Simonich, Pittsburgh Post-Gazette
"My intent was to portray Ross as an average boy who made mistakes early in his
life and then surprised everybody by doing an extraordinary thing after
undergoing a transformation that started with his probation and continued with
his Army discipline. It's important to me to tell people that they can't count a child
out because of his mistakes," Tom McGinnis said.
Pathways to Desistance Study
Reasons for the study

Richer information about serious adolescent
offenders

Picture of the desistance process
 Individual maturation
 Life changes
 Systems involvement

Improved practice and policy in juvenile
justice
 Risk assessment
 Targeted interventions and sanctions
Study Design
 Two sites: Philadelphia and Phoenix
 Enroll serious adolescent offenders
 1,354 felony offenders, aged 14 -18
 Females and adult transfer cases
 Regular interviews over eight years
 Initial interviews
 Time point interviews
 Release interviews
 Other sources of information
 Collateral interviews
 Official records
Who are these adolescents?





16 years old on average
86% male
Average of two prior court appearances
About half appearing for a felony against a person
Ethnically diverse
2%
25%
29%
44%
Caucasian
African American
Hispanic
Other
What we look at
Background Characteristics
•
•
•
•
•
•
•
•
•
•
Personal characteristics (e.g. family, marital
relationships)
Academic achievement and commitment
Routine activities
Offense history
Alcohol and drug use/abuse
Exposure to violence
Psychopathy
Emotional reactivity
Acculturation
Personality
Psychological Mediators
•
•
•
•
•
•
•
•
•
•
Psychological development
Mental health symptoms and threat-control override
Head injury
Use of social services
Perceptions of opportunity
Perceptions of procedural justice
Perceived thrill of doing crime
Moral disengagement
Religious orientation
Costs and rewards of offending
Family Context
• Parental Monitoring
• Parental Relationships
• Parent orientation
Personal Relationships
• Relationships with romantic partner & friends
• Peer delinquency and gang involvement
• Contact with caring adult
Community Context
• Neighborhood conditions
• Community involvement
• Personal capital and social ties
Life Changes
Monthly data available regarding:
Living arrangements
School involvement
Legal involvement
Work
Romantic relationships
Social service involvement/sanctions
Living Situation Calendar
Month 1
Month 2
Month 3
Month 4
Month 5
Month 6
Subject 1 926 West
St Gabe’s
Hall
926 West
Huntington
St
St Gabe’s
Hall
Vision Quest
Youth
Forestry
Camp
Subject 2 2829 W.
Madison
Street
Jail
1008 S.
Wilmot
1008 S.
Wilmot
1008 S.
Wilmot
PO Box
3400
5003 Master
Subject 3 St
2nd and
Norris
2nd and
Norris
2nd and
Norris
House of
Corrections
PO Box
1059
Huntington
St
Augusta
Progress so far

Entire sample past the 72 month follow-up point

About 90% of interviews completed at each time
point

Over 24,000 interviews completed
Examples of topics being
investigated









Procedural justice
Perceptions of risk/benefit of crime
Psychosocial maturity and criminal offending
Effects of substance use treatment
Acculturation/enculturation
Family functioning
Perceptions of opportunities
Neighborhood effects
Service Provision/Institutional Care
Self-Reported
Offending over Time
Self Report Variety
Score
Self Reported Offending over Three Years
Males only
Group 5
(8.5%)
8
6
Group 4
(15.1%)
4
Group 3
(18.3%)
2
0
Group 2
(33.8%)
0
Group 1
(24.2%)
6
12
18
24
30
Months after Initial Interview
36
Mean Number of Re-arrests
Through 36 Months for Each Group
3
2.5
Mean
2
number of
re-arrests 1.5
1
0.5
0
1
2
3
4
Self-reported offending group
5
Percent of time in institution
over three years for each group
0.60
0.50
Mean prop_lock
Percent
time in
institution
0.40
0.30
0.20
0.10
0.00
1
2
3
GROUP
Error bars: 95.00% CI
4
5
Average Percent
in Each Setting at Each Time Point
100
80
Percent
60
Group 4
Group 5
40
20
0
Jail/Prison
Detention
YDC/ADJC
Setting Type
Contracted
Res
Conclusions

In serious offenders, small group (8-9%) with
high and continued offending, and larger group
with high and declining offending (15%)

Largest group (about 58%) reports low levels of
offending, but still spends about 30% of follow
up period in institutional care

Can’t predict the high end persisters from
desisters very well from baseline characteristics

Placement history is very similar for these
groups
Placement in a
Juvenile Institution
Data



Juvenile Court cases in both sites
N = 921
 probation = 502
 institutional placement = 419
Outcomes are:



rate of re-arrest (by year)
level of reported antisocial activity
66 variables measured at baseline, including
demographic, familial, peer, legal, psychological,
mental health related, substance abuse, psychosocial maturity and prior adjustment
Research Question #1
Is there a treatment effect of placement
vs. probation on subsequent rate of rearrest or self-reported antisocial
activity?
Placement vs. Probation
Naïve Comparison
Mean Yearly Rate of Re-Arrest,
by Placement Status
rate
1.4
1.2
1.0
0.8
0.6
0.4
1.20
0.63
0.2
0.0
probation
placement
Propensity Score Matching

Two step process:

A propensity score is calculated for each case. It is
the predicted probability that you get placed given all
of the background characteristics considered

Take each placed case and match it to one or more
probation case with similar propensity score

We then can look to see if the placed group looks
similar to the matched probation group on a
variety of characteristics that might affect the
outcome

If the groups look alike, we can attribute any
difference in the outcomes to the fact that they
were placed
Getting Balanced Groups using
Propensity Scores

Overall, 42 of 66 baseline variables were
significantly different between the placed and
probation groups

After matching, 64 out of 66 variables were NOT
significantly different between the placed and
probation groups

In other words, we have ruled out these 64
variables as potential causes of group differences
in the outcomes
Treatment Effect of Placement
Matched Groups
Mean Yearly Rate of Re-Arrest,
by Placement Status After Matching
1.4
1.2
rate
1.0
0.8
1.06
1.20
0.6
0.4
0.2
0.0
probation
placement
No significant differences between groups in rate of re-arrest
Research Question #2
Is there a community safety benefit for
a longer length of stay in a juvenile
institution?
Methodology for Length of Stay

Length of stay is broken up into discrete
“doses”

Methods to get similar cases across different
levels of the “dose”
 65
of 66 variables show no difference
among the groups, meaning we can rule
them out as causes of differences in
outcomes

Response Curve is estimated
.08
Dosage Categories
.04
0
.02
Density
.06
Histogram of LOS in months
0
10
20
30
length of stay, in months
40
50
Doses roughly correspond to quartiles: 1) 0-6 mo., 2)
6-10 mo., 3) 10-13 mo., 4) > 13 mo.
Dose-Response Curve
Quartiles as Doses
Expected Rate of Re-Arrest,
by Quartile Dose Category
1.4
1.18
1.12
1.26
1.2
0.92
rate
1.0
0.8
0.6
0.4
0.2
0.0
0-6
6-10
10-13
> 13
Dose-Response Curve
3 Month Intervals as Doses
Expected Rate of Re-Arrest,
by 3 mo. Dose Category
3.0
2.55
2.5
rate
2.0
1.11
1.5
1.35
1.08
1.04
1.0
0.5
0.0
0-3
3-6
6-9
9-12
> 12
Conclusions



For intermediate lengths of stay (i.e., 3-13
months), there appears to be little or no marginal
benefit for longer lengths of stay in a juvenile
institution
Inferences about the impact of shorter and longer
stays are less certain
 < 3 mo. – too little power
 > 13 mo. – too much variability
Caution about need to account for treatment in
these settings
Substance Use
Treatment
High Rates of Substance Use Disorders
(Past Year Diagnoses)
Some substance use
diagnosis:
Males
Females
37%
35%
50
40
30
20
10
0
A-A
C
H
Alcohol
-
A-A
C
H
Drug
A-A = African-American C= Caucasian, non-Hispanic H = Hispanic
Sum of Substance Use
Frequency Scores
There is Variability in Substance Use
Over Time (Males)
12
High
Stable
(11.0%)
10
8
6
High Declining
(7.6%)
4
Moderate
(33.4%)
2
0
Low
(34.6%)
base 6mo 12mo 18mo 24mo 30mo 36mo
Abstainers
(13.4%)
Months after Initial Interview
Treatment Effects



With family involvement, significant short term (6-month)
effects of treatment on
 Alcohol use
 Marijuana use
 Offending
Strengths of the analyses
 Test of treatment as actually provided
 Controlling for “street time”
 Above and beyond drug testing
Not a one-shot “Inoculation”
 Chronic, relapsing, remitting disorder
 68% of males with an initial disorder got treatment in
the first year
Summary and Implications

Substance use is a prevalent, strong
predictor of offending

There is variability in substance use over
time (not very predictable from initial
factors)

Treatment had short term (but not longterm) effects on substance use and
offending

Justice system involvement can be an
opportunity for treatment
Themes so far

These adolescents are not uniformly “bad” kids on
the road to adult criminal careers. Instead, a large
proportion report low levels of offending after
court involvement.

Longer institutional stays do not appear to reduce
offending.

Substance use is a major factor related to
continued criminal activity in serious adolescent
offenders. Fortunately, treatment for substance use
seems to work to reduce offending.
“Be of use”
Contact Information
Principal Investigator
Edward P. Mulvey, Ph.D.
[email protected]
Study Coordinator
Carol A. Schubert, M.P.H.
[email protected]