Pathways to Desistance Pathways to Desistance

Supported bySupported by

� Office of Juvenile Justice & Delinquency Prevention

� National Institute of Justice

� John D. & Catherine T. MacArthur Foundation

� National Institute on Drug Abuse

� Pennsylvania Commission on Crime & Delinquency

� Arizona Governor’s Justice Commission

� Robert Wood Johnson Foundation

� William Penn Foundation

� William T. Grant Foundation

Reasons for the studyReasons 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 DesignStudy Design

� Two sites: Philadelphia and Phoenix

� Enroll serious adolescent offenders

� 1,355 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

Sanctions/

Interventions

Antisocial

Behavior

Psychosocial

Development

Mental health

Transition to

adult roles

Social

context

Individual

Characteristics

What we look atWhat we look at

� Individual functioning

� Background characteristics and history

� Mental health

� Psychosocial maturity

� Antisocial behavior (crime, substance use)

� Attitudes and beliefs

� Social contexts

� Peer

� Family

� Community

� Characteristics of sanctions and interventions

Life Event CalendarsLife Event Calendars

� Living arrangements

� School involvement

� Legal involvement

� Work

� Romantic relationships

� Social service involvement/sanctions

Progress so farProgress so far

� Entire sample past the 36 month follow-up point

� Over 90% of interviews completed at each time point

� Approximately 18,000 interviews

Who are these adolescents?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

25%

29%

2%

44%

Caucasian African American

Hispanic Other

Over the first three yearsOver the first three years……....

� At least one institutional stay 82%

� Had at least one job 89%

� Average length of each job 6 weeks

� Report being in a gang 18%

� Re-arrested (thru 24 months) 57%

� Died 2% (30 adolescents)

0

10

20

30

40

50

60

70

80

90

100

Percent in

each setting

type

Contracted

residential

Contracted

Residential-MH

YDC.ADYC Detention Jail/Prison Other

Type of setting

Percent in each setting over three yearsPercent in each setting over three years

Contacts with Supervision and Services at Each Follow-up point

0

5

10

15

20

25

30

35

40

45

Number of

contacts

6 mos 12 mos 18 mos 24 mos 30 mos 36 mos

Months after Initial Interview

Supervision Community-based services

SelfSelf--Reported Reported

OffendingOffending

Mean Self-Reported OffendingAcross Time by Site

0

1

2

3

4

5

12mos

24mos

36mos

Months after Initial Interview

Average Number

of acts

Philadelphia

Phoenix

Self Reported Offending over Three YearsSelf Reported Offending over Three Years

Males onlyMales only

0

2

4

6

8

0 6 12 18 24 30 36

Months after Initial Interview

Self Report Variety

Score

Group 4 Group 4

(15.1%)(15.1%)

Group 5 Group 5

(8.5%)(8.5%)

Group 2 Group 2

(33.8%)(33.8%)

Group 3 Group 3

(18.3%)(18.3%)

Group 1 Group 1

(24.2%)(24.2%)

Do Groups 4 and 5 differ Do Groups 4 and 5 differ

at baseline?at baseline?

� Demographics

� History of antisocial behavior

� Drug and alcohol problems

� Psychological maturity

� Attitudes toward the legal system

� Contextual factors

� Parental influences

� Peers

� Neighborhood

Do Groups 4 and 5 differ Do Groups 4 and 5 differ

over three years?over three years?

�� Justice system interventionJustice system intervention

�� Social service interventionSocial service intervention

�� EmploymentEmployment

�� Ordered lifestyleOrdered lifestyle

�� Number of movesNumber of moves

�� Routine activitiesRoutine activities

�� Alcohol useAlcohol use

54321

GROUP

0.60

0.50

0.40

0.30

0.20

0.10

0.00

Error bars: 95.00% CI

Percent of time in institution Percent of time in institution over three years for each groupover three years for each group

Percent

time in

institution

Average Percent Average Percent in Each Setting at Each Time Pointin Each Setting at Each Time Point

0

20

40

60

80

100

Percent

Jail/Prison Detention YDC/ADJC ContractedRes

Setting Type

Group 4

Group 5

Average Percent Receiving Supervision & Community-based Services at Each Time Point

0

20

40

60

80

100

Average

Percent

Supervision Community-based Services

Group 4

Group 5

ConclusionsConclusions

� In serious offenders,

� Small group (8-9%) with high and continued offending

� Larger group with high and declining offending (15%)

� Largest group reports low levels of offending, but stillspends about 30% of follow up period in institutional care

� Can’t predict the high end persisters from desistersvery well from baseline characteristics

� Over the follow-up period,

� systems involvement and work history don’t differfor these groups

� level of substance use and stability of dailyroutine do differ

Substance UseSubstance Use

Why Should Substance Use Predict Offending (or Desistance)?

Deviant

Peers

Need for

Income

Intoxication

Impaired Coping and Competence:

*Effects on Developing Brain

*Reward + Self-RegulationAdult Roles

0

10

20

30

40

50

60

70

80

90

M A C E H M A C E H

Pathways

MTF

Offenders are More Likely to Have Used Substances than their Same Age Peers

(compared to Monitoring the Future 10th graders)

Males Females

M = Marijuana A = Alcohol C = Cocaine E = Ecstasy H= Hallucinogens

Offenders Have High Rates of Substance Use Disorders (Past Year Diagnoses)

0

10

20

30

40

50

A-A C H - A-A C H

Alcohol Drug

Males Females Some substance use

diagnosis:

Community

Samples

37% 35% 5-17%

A-A = African-American C= Caucasian, non-Hispanic H = Hispanic

Males with Substance Use Disorders at the Beginning Report More Non-Drug Offenses (at each Follow-up)

Non-drug Offenses

N=1129

0

0.5

1

1.5

2

2.5

3

3.5

4

BL 6mo 12mo 18mo 24mo 30mo 36mo

Diagnosed Undiagnosed

There is Variability in Substance Use There is Variability in Substance Use

Over Time (Males)Over Time (Males)

0

2

4

6

8

10

12

base 6mo 12mo 18mo 24mo 30mo 36mo

Months after Initial Interview

Sum of Substance Use

Frequency Scores

Abstainers Abstainers

(13.4%)(13.4%)

Low Low

(34.6%)(34.6%)

Moderate Moderate

(33.4%)(33.4%)

High Declining High Declining

(7.6%)(7.6%)

High High

Stable Stable

(11.0%)(11.0%)

Patterns Over Time in Substance Use Are Related to Patterns Over Time in Offending

( Males)

0

10

20

30

40

50

60

70

80

Abstainers Low Use Moderate Use High

Declining

High Stable

Low SRO Low Moderate SRO Moderate SRO Desistors High SRO

Decliners and Stable/Frequent User Groups are

NOT Different (at Initial Interview) in:

� Age

� Race/ethnicity

� Family Structure

� Baseline Substance Use Disorder

� Parental Alcohol or Drug Problems

� Aspirations for Work and Family

� Substance Use Treatment

� Perceptions of Procedural Justice

� School Enrollment at 6-month Assessment

�� ButBut--at the first interview, compared to at the first interview, compared to

““Stable/Frequent UsersStable/Frequent Users””

““DeclinersDecliners”” were:were:

��Less likely to be in a gang Less likely to be in a gang

��Less impulsive Less impulsive

��More committed to staying out of future More committed to staying out of future

troubletrouble

A Word About TreatmentA Word About Treatment

� Significant Short Term (6-month) effects on

� Alcohol Use

� Marijuana Use

� Offending

� 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:

� Prevalent

� a Strong Predictor of Offending

� There is Variability in Substance Use Over Time(not very predictable from initial factors)

� Treatment had Short Term (but not long-term)effects on Substance Use and Offending

� Chronic, Relapsing, Remitting Disorder

� Justice System Involvement can be anOpportunity for Treatment

(NIDA: Principles of Drug Abuse Treatment forCriminal Justice Populations, July 2006)

Legal SocializationLegal Socialization

Attitudes Toward the Law

�� How adolescents experience and evaluate the law influences: How adolescents experience and evaluate the law influences:

�� whether they cooperate with legal authorities whether they cooperate with legal authorities

�� whether they obey the lawwhether they obey the law

�� how they react to punishmenthow they react to punishment

�� Four influences on legal socialization: Four influences on legal socialization:

�� Punishment costs Punishment costs –– loss of freedomloss of freedom

�� Punishment risks Punishment risks –– will you be caught and punished?will you be caught and punished?

�� Crime payoffs Crime payoffs –– social rewards, thrillssocial rewards, thrills

�� Procedural justice Procedural justice –– fair and respectful treatment byfair and respectful treatment by

police and judgespolice and judges

What Influences Legal Socialization?

Procedural Justice

Legitimacy Legal

Cynicism

Social CostsPunishment CostsPunishment Risks

Social RewardsPersonal Rewards

Attitudes about the Legal System

Legitimacy

Social Costs

Social RewardsPersonal Rewards

Procedural Justice

Cynicism about Law

Legal Cynicism

Punishment Risks

Social RewardsPersonal Rewards

How Does Legal SocializationInfluence Offending?

LegitimacyLegal Cynicism

Social CostsPunishment

CostsPunishment Risks

Social Rewards Personal Rewards

Self-Reported Offending

Procedural Justice

Legal Socialization and Crime

Social CostsPunishment Costs Punishment Risks

Personal Rewards

Self-Reported Offending

Procedural Justice

What Police Do Matters

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Low Medium High

SRO Variety

Punishment Risk Procedural Justice

The Importance of Legal SocializationThe Importance of Legal Socialization

� Harsh punishment may have unintended effects

� Crime rates increase when youthexperience harsh punishment

� Police matter

� Fair and respectful treatment by the policehelps to reduce crime

� Social relationships with family and other adults also matter

� Loss of status is a not a real cost to youthcaught doing crimes, and it may backfire

� Adolescent offenders are rational

� They do weigh costs risks and rewards,but not always in a way that leads to desistance

� Immature youth are less rational

Services for Serious Services for Serious

Adolescent OffendersAdolescent Offenders

0

1

2

3

4

5

6

Jail-Prison YDC-ADJC Contract Res Contract Res-

MH

Average Number of different services

=PA = AZ

Level of services provided to

youth differs across residential settings

Is there a link between

risk/need

and getting

appropriate services

in residential care?

ApproachApproach

� Identified two “treatable” types of problems

� Mood/anxiety problems

�Mental health-related treatment:

individual or treatment on a mental health

unit

� Substance use problems

�Drug and alcohol services

� Control for site, gender, ethnicity, days in

setting

� Focus on four most common settings

0

20

40

60

80

100

Percent Receiving M

H Tx

Detention Jail-Prison YDC-ADYC Contracted

Residential

Percent Receiving Mental Health Treatment

in Different Setting Types

No mood/anxiety

problemsHad mood/anxiety

problems

�In detention : 3X

�In YDC-ADJC: 4X

� In Jail-Prison: No Difference

�In Contract. Res: No Difference

0

20

40

60

80

100

Percent Receiving SU Tx

Detention Jail-Prison YDC-ADJC Contracted

Residential

Percent Receiving Substance Use

Treatment in Different Setting Types

No Substance Use

Problems

Had Substance use

Problems

�In Detention: 5X

�In Jail-Prison: 3X

�In YDC-ADJC: 4X

�In Contract. Res.:

No Difference

What is the relationship

between supervision/services

and adjustment in the

community after release

from an institution?

� Define “aftercare” period

� Six month following release from placement associated

with the study index petition

� Examined outcomes that indicate signs of negative and

positive adjustment

� Engagement

� Antisocial activity

� System involvement

� Considered duration and intensity of Supervision and

Community-based services

� Duration = Number of aftercare months with at least 1

contact

� Intensity = Number of contacts per month

Approach

ResultsResults

�� Duration of supervision reduces negative Duration of supervision reduces negative

outcomes by about outcomes by about ½½

�� Effects of intensity of services are less clearEffects of intensity of services are less clear

�� Effects hold when other factors consideredEffects hold when other factors considered

�� Parental monitoringParental monitoring

�� Peer associationsPeer associations

�� Caring adultsCaring adults

In our sample of serious offenders:

ConclusionsConclusions

�� Majority of youth in each setting report Majority of youth in each setting report

receiving multiple types of treatmentreceiving multiple types of treatment

�� Detention and stateDetention and state--run facilities do a better job run facilities do a better job

matching mental health and substance use matching mental health and substance use

treatment with those who need ittreatment with those who need it

�� Duration of supervision is important for Duration of supervision is important for

reducing the chances of negative adjustmentreducing the chances of negative adjustment

Family and Peer Family and Peer

Influences on Influences on

Offending and Offending and

ReRe--OffendingOffending

Effective Parenting CombinesEffective Parenting Combines

Warmth and ControlWarmth and Control

NeglectfulNeglectfulAuthoritarianAuthoritarian

IndulgentIndulgentAuthoritativeAuthoritative

High

High

Low

Low

WARMTH

CONTROL

Offenders’ Parents Are More Diverse

Than Most People Think

0

5

10

15

20

25

30

35

40

45

50

Authoritative Authoritarian Indulgent Neglectful

Community

RPD Sample

Community data from: Lamborn, Mounts, Steinberg, & Dornbusch, 1992

Problem Behavior Is Linkedto Quality of Parenting

0

0.2

0.4

0.6

0.8

1

1.2

1.4

Aggressive

Offending

Income

Oiffending

Substance

Use Problems

Substance

Dependence

Authoritative

Authoritarian

Indulgent

Neglectful

Ineffective

Parenting

Ineffective Parenting Is More Common in Distressed Neighborhoods

Low Social

Capital

Neighborhood

Disorder

Concentrated

Poverty

Residential

Instability

OffendingDeviant Peers

Ineffective Parenting Leads to Offending Mainly Because It Increases Affiliation with Deviant Peers

Ineffective

Parenting

Re-OffendingDeviant Peers

Role of Context Is Similar in PredictingRe-Offending After Re-Entry from Placement

Ineffective

Parenting

Low Social

Capital

Implications for Aftercare

� The context to which offenders return

matters

� Programs should attempt to

�Minimize affiliation with deviant peers

after re-entry

� Improve quality of parenting

� Increase family’s integration into the

community and access to community

resources

The Pathways Girls

Who are these girls?

9.2 years10.4 yearsAge of first self-reported

antisocial act

16%7%Sentenced as adults

37%11%Proportion of time spent

“locked up” over 3 years

2.31.3Number of prior petitions to

court

14.3 years14.9 yearsAge at first petition

to court

MALEFEMALE

Prevalence of Placements Over 3 Years

0

10

20

30

40

50

60

Percent

Contracted Res Contracted Res

MH

YDC/AYDC Detention Jail/Prison Other

Male Female

Youths may have experienced multiple placements over the 3 year follow-up

Gender Differences in Mental Health and Substance Abuse

0

5

10

15

20

25

30

35

Percent Diagnosis in Past Year

Affective

Disorder

PTSD Alcohol Abuse

/Dependence

Drug Abuse

/Dependence

Male Female

Gender Differences in Historyof Victimization

0

10

20

30

40

50

60

70

Victim of

Violence

Beaten Up Attacked with

Weapon

Shot at Raped

Male Female

Percent at Baseline

The “Boyfriend” Factor….

� Number of romantic partners:

� Females: 8 partners

� Males: 14 partners

� Age of 1st sexual encounter

� Females: 13.7 years

� Males: 12.9 years

� Number of sexual partners

� Females: 5 partners

� Males: 10 partners

� Average age of current partner

� Females: 18.5 years

� Males: 16.3 years

0

5

10

15

20

25

30

35

40

Per

cent

0 1~2 3~4 5~6

Gender Differences in Self-Reported Offending

0.0

0.5

1.0

1.5

2.0

2.5

Mean Number of Self-

Reported Acts

6 12 18 24 30 36

Months After Initial Interview

Male Female

Are there chronic female

offenders?

Number of Time Points Reporting Antisocial Acts

Are Chronic Female Offenders Different From Girls Who Do Not Offend Chronically?

� There is variability among these girls…..

� Chronic Female Offenders are more likely to be….

� Less psychosocially mature

� Less future oriented

� Less responsible

� Less able to consider others

� More impulsive

� More likely to be a victim of violence

� More likely to have anxiety and a diagnosis of PTSD

� More likely to have a drug dependence disorder

Conclusion Conclusion

� Preliminary results from a large, comprehensive, operationally successful, ongoing study

� Considerable variability

� Offending patterns over time

� Substance use over time

� Parenting

� Gender

� Substance use matters

� Strongly related to offending

� Treatment for substance use problems could have a short-term, positive effect

� Interventions matter

� Police behavior relates to attitudes

� State facilities do a good job of targeting services

� Involvement during the aftercare period matters