The Massachusetts Transition Youth Arrest Study MATAYA Maryann Davis, Ph.D. Principal Investigator Center for Mental Health Services Research Department of Psychiatry University of Massachusetts Medical School Research Team Maryann Davis, Ph.D., Psychologist Steven Banks, Ph.D. Statistician/Methodologist William Fisher, Ph.D. Sociologist Albert Grudzinskas, Jr., J.D. Attorney Bernice Gershenson, M.P.H. Project Coordinator Michael Pullmann, M.A. Psychologist Edward Mulvey, Ph.D. Psychologist Daniel Nagin, Ph.D. Statistician/Methodologist Robert Jones, Ph.D. Statistician/Methodologist Justice System Involvement of Youths in Mental Health Systems High Arrest Rates in adolescence and young adulthood in general population High rates of mental health conditions in those in juvenile and criminal justice system High arrest rates at these ages in adolescent and adult mental health system users and special education students Mental health system has an opportunity to prevent and reduce juvenile and criminal justice system involvement Gender Differences in Offending Abound GOAL: Develop Knowledge to Help Prevent or Minimize Adolescent and Young Justice System Involvement of Youths in MH Services Used Two Massachusetts Statewide Administrative Databases Offending rates are lower in females Females may require more aversive experience to engage in antisocial behavior Members of the opposite, not same sex encourage female criminal behavior Incarcerated juvenile girls are almost times more likely than boys to have a non conduct psychiatric disorder with impairment 1. What are the various within-individual longitudinal arrest patterns among intensive public adolescent mental health system users? 2. What is the relationship between mental health services and arrest patterns at these ages? 3. How do offending patterns differ between the mental health and general offender population? ALL QUESTIONS ASKED WITHIN GENDER Criminal Offender Record Information CORI Records Name Birthdate All Arraignments; Date Specific Charge Department of Mental Health DMH Records Name Birthdate (e.g. Silver, Smith, & Banks, 2000; Herrera, & McCloskey, 2001; Pettiway, 87; Teplin, et al. 2002) Funded by National Institute of Mental Health R01-MH782 03 CORI records pulled as of July 05 DMH records pulled as of March 0 1
21st Annual RTC Conference DMH DATA All individuals were born between 7-79 DMH individuals received adolescent case management services between 94-9 Not arrested DMH individuals have no CORI record Non-DMH arrestees have no DMH record Gender Differences in Transition Arrests Maryann Davis Steven Banks William Fisher Bernice Gershenson Albert Grudzinskas Center for Mental Health Services Research Department of Psychiatry University of Massachusetts Medical School Cumulative Proportion Arrested by age 25 0.30 MALES Females FEMALES 0.35 0.30 0.25 0.20 0.15 0.10 5 0 0.25 0% Males 0.15 0.10 Females 5 14 15 1 17 18 20 21 22 23 24 in 5 40% 31% 2 20% 15% 10% 0% 0 Males 5 50% 0.20 % Population 5 0 0.45 0.40 Proportion of Subjects Proportion of Subjects 0.70 Frequency of Arrest s by 25 Gender Differences Proportion Arrested at Each 14 15 1 17 18 20 21 22 23 24 No Arrests Single Arrests Multi Arrests in From Davis et al., 2007 From Davis et al., 2007 χ2 (df=2)=3.3, p<.001 2
Developmental Trajectory Modeling Nagin & Land, 93 Identifies groups (clusters) of individuals with similar longitudinal patterns over time Describes those arraignment patterns Statistical analysis optimizes explanation of the greatest amount of variance using the fewest number of clusters Trajectory Dataset DMH individuals whose first charge was at age 9. Deleted one outlier with 4 charges at age 24. Multiple Charge s Zero Charges Population DMH Males 421 243 DMH Females 204 39 Comparative Trajectory Approach 1. Analyze males and females separately 2. Determine ideal number of groups and pattern within those groups 3. Combined the populations, used starting values 4. Determined ideal number of groups and pattern within those groups.5.0 9 10 11 14 15 1 17 18 20 21 22 23 24 Trajectory Groups; none 51% low mid low late 11%.5.0 9 10 11 14 15 1 17 18 20 21 22 23 24 Trajectory Groups; moderate mid 5% moderate mid peaked 5% moderate early 3%.5.0 9 10 11 14 15 1 17 18 20 21 22 23 24 Trajectory Groups; high rising high mid high early 1% moderately high mid 5% 3
Trajectory Groups;.5.0 9 10 11 14 15 1 17 18 20 21 22 23 24 % Trajectory Cluster 100% 90% 80% 70% 0% 50% 40% 20% 10% 0% None Proportion of Males and Females in each Trajectory Group Low Mid % Males in multiage arrestees Low Late Moderate Early % Males in sample Moderate Mid Moderate Mid Peaked Moderately High Mid High Early High Mid High Rising There are no trajectory clusters that are unique to either gender Conclusions The biggest differences between DMH males and females are in the proportion never arrested and with multiple arrest Males tend to be overrepresented in moderate and high frequency groups Different trajectories have different prevention timing implications; take advantage of their presence in treatment and prevent future offending Several clusters had peaks and high charge rates at ages 18 and older; adult mental health systems need to help Future question; how to identify high risk before it happens Criminological Theory Mental Health and Criminal Justice The Relationship between Transition-d Mental Health Services and the Risk of Arrest Michael Pullmann Vanderbilt University Maryann Davis, Ph.D. University of Massachusetts Medical School Aberrant behavior is socially determined Criminology is the study of particular aberrant behavior in violation of law Laws are codified proscriptions sanctioning particular aberrant behavior Integrative models of delinquency propose a variety of factors that contribute to longitudinal criminal behavior, including biological (individual), social, and structural factors Even the best integrative models predict only 25-49% of the variance in long-term offence patterns 1 The Mental Health and Criminal Justice systems are structures designed to address aberrant behavior High rate of youth with MH problems in the justice system 2-4 50-80% of youth in detention are estimated to have a diagnosable disorder 20% of youth in detention are estimated to have a serious MH disorder Increased probability of justice contact for youth receiving MH services One study found that 20% of youth receiving public mental health services were arrested over the course of 38 months 5 4
Transition to hood The transition to adulthood is a particularly sensitive time, especially for people involved in institutions such as mental health and the justice system Changing individuals (brain maturation, hormones, moral development) Changing social networks (peers, mentors, authority) Changing structures: differing MH eligibility requirements, differing laws and sanctions Hypotheses People who offend in adolescence will be more likely to offend in adulthood People who are diagnosed with externalizing mental health disorders during adolescence will be more likely to offend in adulthood Mental health system contact during the transition to adulthood will be related to justice system contact in adulthood, such that those who received restrictive MH services (inpatient hospitalization and residential treatment) will be more likely to offend in adulthood. Methods Subsample of previous presentation; people born in 78 or 79 N = 80 For this analysis: Juvenile : ages 7-18 : ages -25 Transition : ages 1-18 Male Female Race/ethnicity White African American Hispanic Asian Other Missing/unknown 1 through 18 old* Outpatient services Residential tx Inpatient hospitalization through 25 old* Outpatient services Residential tx Inpatient hospitalization * Categories are not mutually exclusive n 4 393 584 72 3 11 0 1 3 202 2 223 140 114 Valid % 51% 49% 73% 9% 1% 45% 25% 29% 1 1 Any MH diagnosis Of those with a diagnosis: Mood Disorder PTSD Thought Disorder Conduct Disorder/ODD ADD/ADHD Substance Abuse Disorder Impulse Control Disorder Personality Disorder Anxiety Disorder Developmental Disorder Adjustment Disorder Learning Disorder Eating Disorder 7-18 n Valid % 37 25 70% 3 8 4 1 1 1 49 28 25 20 5% 1 9 3% 8-25 n Valid % 11 20% 102 3% 55 3 2 39% 1 4 29% 2 1% 41 14 9% 8 5% 1 1% 3 Any MH diagnosis before old Of those with a diagnosis: Mood Disorder PTSD Thought Disorder Conduct Disorder/ODD ADD/ADHD Substance Abuse Disorder Impulse Control Disorder Personality Disorder Anxiety Disorder Developmental Disorder Adjustment Disorder Learning Disorder Eating Disorder n 37 25 8 4 1 49 28 25 20 1 9 8 Valid % 70% 3 1 1 5% 3% 5
Charges -25 old charges Before old Any charge Any serious property or person charge to 25 old Any charge Any serious property or person charge n 372 210 340 15 Valid % 4 2 42 Charges 7-18 old None No None crime 347 (43%) 3 5 1 () (15%) 0 () (3%) 33 () 58 253 (31%) 9 () (1) Juvenile charges None No None crime 347 (43%) 3 5 1 () (15%) 0 () (3%) 33 () 58 253 (31%) 9 () (1) of arrested 7-18 were also arrested as -25, compared to 20% of 7-18 not arrested. OR = 8.5 (95% CI =.1 11.7) Male** % within gender No crime 2 3 3 1 (15%) 1 51% 28 3 253172 (31%) 4 Note: this is the subgroup of people with a diagnosis prior to old ODD/CD** Substance use* PTSD* Anxiety* No crime 149 (41%) 2 9% 5 41% 3 18 7 1 57 (1) 8 1 1 33% 1 45 () 9% 3 % 25 1 5% 2 11 (3) 34 53% 29% 25 51% 3 2 31% 4 3% Summary of Diagnosis Criminal Justice Involvement People diagnosed with ODD/CD or substance use disorders were more likely to initiate criminal behavior in childhood and persist into adulthood People diagnosed with PTSD were more likely to initiate criminal behavior in adulthood People diagnosed with anxiety disorders were less likely to engage in criminal behaviors at any time Surprisingly, people diagnosed with impulse control disorders were not significantly more likely to engage in criminal behaviors No other class of disorders was significant, including thought disorders, learning disorders, adjustment disorders, or ADD/ADHD Impulse control (NS) 3 11% 11% 11% Personality* 10 9 2 3 33% 2 ** p <.01; * p <.05 ** p <.01; * p <.05
Community based services 1-* % within services Residential treatment, 1- (NS) % within services Inpatient hospitalization (NS) % within services No crime 14 45% 4 74 3 88 4 25% 1 (15%) 57 4 3 1 33 2 47 5 28 1 3 31% 253 (31%) 98 2 39% 4 3 35% 4 25% Community based treatment 1-18 Residential treatment 1-18 Inpatient hospitalization 1-18 *p <.05 No Yes No Yes No Yes Any charge before 49%* OR=1.33 (0-1.75) 4* 45% 50% Any violent charge before 31%* OR=1.7 (1.28-2.44) 20%* 2 2 2 2 Any charge -25 4 40% 41% 4 43% Any violent charge - 25 1 20% Overall Summary The strongest predictor of adult offense is offending as a juvenile. Juveniles who were arrested were 8.5 times more likely to be arrested than juveniles who were not arrested. Males were more likely to offend at any time. Females were more likely to initiate offending in adulthood People diagnosed with ODD/CD or substance use disorders were more likely to initiate criminal behavior in childhood and persist into adulthood Specific forms of treatment during the transition, including residential treatment and inpatient hospitalization, do not seem to be related to offending in this group of mental health service receivers Specific forms of treatment during the transition to adulthood do not seem to be related to adult criminal behavior This is also true when examining only those people with substance use disorders or CD/ODD in adolescence (data not shown). ** p <.01; * p <.05 References 1. Kelley, T. M. (94). Crime and psychology of mind: A neo-cognitive view of delinquency. In G. Barak (Ed.), Varieties of criminology: Readings from a dynamic discipline (pp. 15-28). Westport: Praeger. 2. Cocozza, J. J., & Skowyra, K. R. (2000). Youth with mental health disorders: Issues and emerging responses. Juvenile Justice, 7(1), 3-. 3. Teplin, L. A., Abram, K. M., McClelland, G. M., Dulcan, M. K., & Mericle, A. A. (2002). Psychiatric disorders in youth in juvenile detention. Archives of General Psychiatry, 59(), 13-1143. 4. Goldstrom, I., Jaiquan, F., Henderson, M., Male, A., & Mandersheid, R. W. (2000). The availability of mental health services to young people in juvenile justice facilities: A national survey. In R. W. Mandersheid & M. J. Henderson (Eds.), Mental health, United States, 2000. Washington, DC: Substance Abuse and Mental Health Services Administration. 5. Rosenblatt, J. A., Rosenblatt, A., & Biggs, E. E. (2000). Criminal behavior and emotional disorder: Comparing youth served by the mental health and juvenile justice systems. Journal of Behavioral Health Services and Research, (2), 2-237.. Davis, M. (2003). Addressing the needs of youth in transition to adulthood. Administration and Policy in Mental Health, 30(), 495-509. 7