mental disorder profiles in justice-involved youth by shiming … · 2016-12-07 · mental disorder...
TRANSCRIPT
Mental Disorder Profiles in Justice-Involved Youth
by
Shiming Huang
A thesis submitted in conformity with the requirements
for the degree of Master of Arts
Department of Applied Psychology & Human Development
Ontario Institute for Studies in Education
University of Toronto
© Copyright by Shiming Huang (2016)
ii
Mental Disorder Profiles in Justice-Involved Youth
Shiming Huang
Master of Arts
Department of Applied Psychology & Human Development
University of Toronto
2016
Abstract
As a population, justice-involved youth are characterized by high rates of comorbid mental
disorders. However, past research has generally considered disorders singly. It is unclear whether
disorders cluster together in a way that has implications for intervention and case management of
justice-involved youth. To begin to explore this issue, cluster analysis was used to identify
common diagnostic profiles in a sample of 195 youth who received court-ordered assessments.
Six mental disorder clusters emerged, which were then examined in relation to youths’ risk to
reoffend, criminogenic needs, interventions received during probation, and recidivism. Two
clusters were characterized by relatively higher reoffense rates: youth with disruptive behavior
disorders alone, and youth with the profile of disruptive behavior disorders, learning disability,
and ADHD. These clusters also significantly predicted odds of reoffense even after accounting
for youths’ static risk and criminogenic needs treatment. Implications for risk assessment and
rehabilitation of justice-involved youth are discussed.
iii
Acknowledgments
Thank you to the youth justice lab for the constant support and encouragement. I am fortunate to
be working in a wonderfully loving and passionate team. Thank you to Michele, for your
unwavering support and guidance throughout this journey. I look forward to learning from you
as you remain a constant source of inspiration for me. Thank you to Tracey, for your insight and
supervision, to teach me to think more conceptually and grow as a researcher. Lastly, thank you
to my SCCP colleagues, for your friendship and inspiration.
iv
Table of Contents
Abstract ........................................................................................................................................... ii
List of Tables ...................................................................................................................................v
Introduction ......................................................................................................................................1
Mental disorders in justice-involved youth .................................................................................1
Mental health and offending .......................................................................................................2
The present study ........................................................................................................................5
Method .............................................................................................................................................6
Participants ..................................................................................................................................6
Procedure.....................................................................................................................................7
Measures .....................................................................................................................................8
Mental disorder diagnoses .....................................................................................................8
Risk to reoffend and criminogenic needs ..............................................................................8
Matching of identified criminogenic needs ...........................................................................9
Recidivism ...........................................................................................................................10
Statistical analyses ....................................................................................................................10
Mental disorder profiles .......................................................................................................10
Treatment match across clusters ..........................................................................................10
Predicting recidivism ...........................................................................................................10
Results ............................................................................................................................................11
Preliminary analyses .................................................................................................................11
Mental illness profiles ...............................................................................................................13
Are identified criminogenic needs being met at a similar rate across the mental disorder
clusters?..............................................................................................................................14
Does matching services to identified criminogenic needs predict recidivism equally well
across the mental disorder clusters? ...................................................................................15
Discussion ......................................................................................................................................22
Mental disorder profiles ............................................................................................................22
Matching services to criminogenic needs .................................................................................25
Mental disorder profiles and recidivism ...................................................................................27
Implications for risk, need, and responsivity ............................................................................29
Strengths and limitations ...........................................................................................................30
References ......................................................................................................................................32
v
List of Tables
Table 1: Participants’ demographic, offense, and psychiatric information ....................................2
Table 2: Percentage treatment match scores within each criminogenic need and recidivism
rates across mental disorder clusters .......................................................................................15
Table 3: Hierarchical regression analyses predicting recidivism from criminal history, match,
and mental disorder profiles .....................................................................................................19
Table 4: Disruptive behavior disorders related YLS/CMI items by mental disorder clusters .......21
1
INTRODUCTION
Youth with mental illnesses are significantly overrepresented in the criminal justice system
(Teplin et al., 2007). Research results suggest that more than 65% of justice-system-involved
youth have at least one psychiatric disorder (e.g., Shufelt & Cocozza, 2006; Teplin et al., 2007;
Wasserman, McReynolds, Lucas, Fisher, & Santos, 2002), far exceeding rates found in the
general population (Waddell, Shepherd, Schwartz, & Barican, 2014). In addition, comorbidity
(i.e. the co-occurrence or two or more mental disorders) is highly prevalent in this population,
with 58% to 79% of justice-involved youth meeting diagnostic criteria for two or more mental
disorders (Abram, Teplin, McClelland, & Dulcan, 2003; Abrantes, Hoffmann, Anton, & Estroff,
2004; Shufelt & Cocozza, 2006; Ulzen & Hamilton, 1998). Justice-involved youth with
comorbid psychiatric conditions are a particularly vulnerable group, as individuals with multiple
diagnoses often exhibit poorer psychosocial outcomes compared to individuals with single
diagnoses (Drake, Mueser, Clark, & Wallach, 1996; Kessler et al., 1994). Thus, an increased
understanding of the relationship of comorbidity patterns to justice-involved youth’s outcomes is
needed. This entails elucidating the interrelationships among mental health variables, criminal
risk factors, intervention, and recidivism outcomes.
Mental disorders in justice-involved youth
Despite the high prevalence rates of comorbid mental disorders in justice-involved youth,
past research has focused mainly on single diagnoses and their influence on youth’s
rehabilitation and other life outcomes. Consistently across studies, disruptive behavior disorders
(DBD), such as conduct disorder (CD) and oppositional defiant disorder (ODD), have emerged
as the most prevalent mental disorders in justice involved youth (Fazel, Doll, & Langstrom, 2008;
Teplin, Abram, McClelland, Dulcan, & Mericle, 2002). Implications for justice-involved youth
2
with disruptive behavior disorders include increased likelihood of persistent adult dysfunction
(Zoccolillo, Pickles, Quinton, & Rutter, 1992) and continued engagement in antisocial behaviors
in adulthood (Loeber, Burke, Lahey, Winters, & Zera, 2000). Other common psychopathologies
in this population include substance use disorders (SUD), affective disorders, learning
disabilities (LD), and ADHD (Brink, 2005; Fazel et al., 2008; Murphy, 1986; Teplin et al., 2002).
The outcomes of these mental disorders in justice-involved youth have also been investigated.
For example, adolescents with learning disabilities display higher levels of antisocial behaviors
and arrest rates than non-learning-disabled peers (Keilitz & Dunivant, 1986; Waldie & Spreen,
1993), and justice-involved youth with ADHD were more likely to recidivate compared to non-
ADHD youth (Gordon, Diehl, & Anderson, 2012).
In contrast to the number of studies examining the prevalence rates of psychiatric
comorbidities in justice-involved youth (Abram et al., 2003; Abrantes et al., 2004), limited
research has examined the significance of such comorbidity patterns on outcomes in this
population. Some of the most common comorbidity patterns found in community, clinical, and
forensic studies include the co-occurrence of disruptive behavior disorders with ADHD,
disruptive behavior disorders with SUD, and SUD with ADHD (Abram et al., 2003; Armstrong
& Costelo, 2002; Chan, Dennis, & Funk, 2008; Milin, Halikas, Meller, & Morse, 1991;
Vermeiren, 2003). Despite their high prevalence, the specific relationships between these
comorbidity patterns and outcomes for justice-involved youth have yet to be elucidated.
Mental health and offending
The role of mental illness in criminal behavior and justice system involvement has been
examined in the literature, with research producing mixed results. Some studies have reported
that untreated mental disorders are risk factors for criminal justice system involvement and
recidivism (Hoeve, McReynolds, & Wasserman, 2013; Vermeiren, Schwab-Stone, Ruchkin, De
3
Clippele, & Deboutte, 2002) while others have found no meaningful correlations between mental
illness and antisocial behaviors (Skeem, Manchak, & Peterson, 2011). However, most of the
existing studies on mental health and offending have focused on single diagnoses and few have
examined links between particular patterns of mental disorders and offending. Within this small
literature, there is evidence that the co-occurrence of particular mental disorders is associated
with antisocial behaviors and recidivism outcomes. For example, some research shows that
individuals with substance abuse disorders alongside other psychopathologies are at increased
risk for recidivism compared to those with substance use disorders alone (Rezansoff,
Moniruzzaman, Gress, & Somers, 2013; Wallace, Mullen, & Burgess, 2004). The co-occurrence
of ADHD and CD is associated with earlier age of onset of antisocial behaviors (Loeber et al.,
2000), and greater variety and severity of antisocial behaviors than CD alone (Walker, Lahey,
Hynd, & Frame, 1987). However, differing definitions of mental disorders and differences in
which disorders are included in studies, as well as varied samples, hinder our understanding of
the relationship of comorbidities to antisocial outcomes.
Furthermore, few studies have examined justice-involved youth with mental disorders in
the context of the risk-need-responsivity framework, an assessment and case management
framework that is widely used to identify targets of intervention to reduce reoffending in justice
system-involved youth and adults (Andrews & Bonta, 2010; Andrews, Bonta, & Hoge, 1990).
Three principles form the core of the RNR framework. The risk principle states that the intensity
of service should match an offender’s level of risk level for recidivism. The need principle states
that each offender’s criminogenic needs – defined as risk factors that are strongly and directly
related to criminal behavior – should be the targets of rehabilitative interventions. Criminogenic
needs that have been well-validated in empirical studies include: history of antisocial behavior,
antisocial personality pattern, antisocial cognition/attitudes, antisocial peers, family dysfunction,
4
school/work obstacles, misguided use of leisure time, and substance abuse (Andrews & Bonta,
2010). With the exception of history of antisocial behavior, a static risk factor, criminogenic
needs represent dynamic risk factors in that they are amenable to change and therefore constitute
targets of intervention. Substantial research supports the utility of targeting criminogenic needs
in reducing reoffending (Andrews & Bonta, 2010; Andrews et al., 1990; Vieira, Skilling, &
Peterson-Badali, 2009). Finally, the responsivity principle states that treatment should entail
cognitive social learning interventions that are tailored to an individual’s unique characteristics
(Andrews et al. 1990; Bonta & Andrews, 2006).
In contrast to the dominance of the RNR framework in understanding and addressing risk
factors for continued offending and in the rehabilitation of justice-involved individuals, there is a
paucity of research examining mental health within this framework. Most research has focused
on the adult justice population and conceptualized mental health as singular disorders as defined
within the DSM-IV-TR (American Psychiatric Association, 2000). In studies where mental health
was defined as singular disorders and investigated within the context of the RNR framework,
criminogenic needs predicted recidivism equally well in individuals with and without mental
disorders and no incremental validity was added by the inclusion of clinical variables (e.g.,
mental disorder diagnoses, treatment history) to the prediction of reoffending (Bonta, Law, &
Hanson, 1998; Skeem, Winter, Kennealy, Louden, & Tatar, 2014). To date, only a handful of
studies have attempted to examine this issue in the youth justice population (McCormick,
Peterson-Badali, & Skilling, 2016; Shubert, Mulvey & Glasheen, 2011), and even fewer have
conceptualized mental health in terms of comorbid mental disorders or clusters rather than
discrete, singular diagnoses (Guebert & Olver, 2014). The few studies available relevant to
examining the links between mental health, criminogenic needs, and recidivism in justice-
involved youth has produced conflicting results. Some studies reported that mental health status
5
was unrelated to recidivism (McCormick et al., 2016) while others found that specific mental
disorder patterns, such as CD and substance abuse, were related to more serious criminogenic
need profiles and to higher rates of reoffending (Guebert & Olver, 2014). Given such
inconsistencies in the literature, further research regarding the relationships between mental
health variables, criminogenic needs, and recidivism outcomes among justice-involved youth is
warranted.
The present study
The existing literature has primarily investigated mental disorders as single diagnoses.
However, given the common occurrence of psychiatric comorbidities, conceptualizing mental
health solely as singular diagnoses may not be representative of the realities of the justice
populations. The high prevalence rates and the consistency of certain comorbidity profiles in the
justice populations may represent meaningful clusters that differentiate between subsets of
justice-involved individuals and have implications for treatment and/or recidivism outcomes.
Thus, the current study attempted to identify common mental disorder profiles in a sample of
justice-involved youth, and understand their relationships to criminogenic needs, treatments
received during probation, and recidivism outcomes.
Three research questions were examined: (1) What are the common mental disorder
profiles in justice-involved youth? It was expected that common profiles observed in community
and clinical studies would also be found in a sample of justice-involved youth. For example, we
anticipated the co-occurrence of disruptive behavior disorders with ADHD (Vermeiren, 2003),
and substance use/abuse disorders with disruptive behavior disorders (Armstrong & Castello,
2002; Chan et al., 2008). (2) Are criminogenic needs being addressed at similar rates across the
mental disorder profiles? The RNR framework stipulates that interventions should target
individuals’ identified criminogenic needs to reduce risk of reoffending (Andrews et al. 1990;
6
Bonta & Andrews, 2006). Thus, it is important to establish whether differences exist across the
mental disorders in the rate of addressing individuals’ identified criminogenic needs. (3) Does
addressing youths’ identified criminogenic needs via case management and treatment services
have an impact on re-offense rates across the mental disorder profiles? RNR principles and
previous research indicate that case management targeting individuals’ identified criminogenic
needs is associated with reduced reoffending (Peterson-Badali, Skilling, & Haqanee, 2015;
Vieira, Skilling, & Peterson-Badali, 2009). It was expected that this finding would also hold in
youth regardless of their specific mental disorder profile.
To answer the above research questions, information including mental disorder diagnoses,
services received during probation, and recidivism data were collected from a sample of
community-sentenced youth who received comprehensive assessments that included information
related to mental health, criminogenic needs, and risk to reoffend.
METHOD
Participants
The sample consisted of 195 youth (154 males and 41 females), ranging from 12 to 20
years of age (M = 16 years, SD = 1.56), referred for court-ordered assessments between 2003 and
2010 to a mental health agency in Toronto, Canada in order to assist the court in disposition
decisions; assessments were conducted by members of a multidisciplinary team including a
psychiatrist and a psychologist. Youth in this study sample met diagnostic criteria for at least one
or more mental disorders. Participants were included through consecutive admission and consent
was obtained from all current study participants for their clinical information to be used for
research purposes. The overall research consent rate in the clinic was 79%.
7
As shown in table 1, 32% of the participants reported their ethnicity as Black, 27% as
White, and 7% as Asian. The offenses leading to participants’ referrals for assessment included
non-violent (e.g., theft, drug-related, break and enter), violent but not sexual (e.g., assault,
robbery), and sexual offenses (e.g., sexual assault, invitation to touching). The majority of the
youth (63%) were charged with offenses in the violent but not sexual category, 23% in the non-
violent category, and 14% in the sexual category.
Procedure
Participants’ data were coded from their assessment files, probation case notes, and
criminal records according to a previously-developed coding scheme (Peterson-Badali, Skilling,
& Haqanee, 2015). Assessment files were reviewed to extract information on demographics,
mental health, offense history, charges leading to referral for assessment, scores on the YLS/CMI
(the Youth Level of Service/Case Management Inventory; Hoge & Andrews, 2002), and
clinician recommendations targeting identified criminogenic needs. At the time of assessment, a
psychologist, a psychiatrist, or a social worker with expertise in forensic assessment completed
the YLS/CMI and the court-ordered assessments using information from youth’s legal history,
standardized questionnaires and psychological tests, collateral reports from other sources (e.g.,
parents, probation officers, youth workers), and youth self-report (Peterson-Badali et al., 2015).
Probation case notes were reviewed to determine the details of youth’ sentence, case
management received during probation, and whether services youth received matched
recommendations made in clinical files. Recidivism information was obtained from criminal
records for youth with at least a two year follow-up period following their assessments.
8
Measures
(1) Mental disorder diagnoses
Mental disorder diagnoses, extracted from the clinical files, were made by either a
psychiatrist or a licensed psychologist who was part of the multidisciplinary team that conducted
the forensic assessments. There were six mental disorder variables in this study: disruptive
behavior disorders (DBD), Attention Deficit Hyperactivity Disorder (ADHD), Substance Use
Disorders (SUD), internalizing disorders, Learning Disorders (LD), and cognitive disabilities.
Disruptive behavior disorders include Oppositional Defiant Disorder (ODD) and Conduct
Disorder (CD). SUD include substance abuse disorder and substance dependence disorder.
Internalizing disorders include mood and/or anxiety disorders such as major depressive disorder,
dysthymia, generalized anxiety disorder, and post-traumatic stress disorder. LD includes specific
learning disorders and communication disorders. Cognitive disabilities include intellectual
disabilities and global developmental delay. The diagnostic criteria used by the clinicians for
these mental disorders was drawn from the Diagnostic and Statistical Manual of Mental
Disorders-IV-TR edition (American Psychiatric Association, 2000) and Learning Disabilities
Association of Ontario (LDAO, 2001). Learning disabilities and cognitive disabilities have not
always been included in studies examining mental health issues in justice-involved youth;
however, these two groups of disorders have established links with justice-system involvement
(Mauser 1974; Simpson & Hogg, 2001) and further, given the high rates of learning and
cognitive disabilities within this population of youth, any discussion of mental health in justice-
involved youth would be incomplete without them.
(2) Risk to reoffend and criminogenic needs
The seven dynamic criminogenic needs were coded into seven binary variables based on
clinician recommendations from the assessment reports, which were informed by the YLS/CMI
9
(Hoge & Andrews, 2002, 2011), a standardized forensic tool used to assess 12- to 18-year-old
youth’ risk to reoffend, criminogenic needs, and responsivity factors. The first section is a 42
item checklist divided into eight domains to assess youth’ risk to reoffend and criminogenic
needs. The eight domains map onto eight major risk factors relevant to criminal behavior as
identified in the RNR literature (Andrews & Bonta, 1990, 2010). These domains include: prior
and current criminal offenses (a major but static risk factor) and seven dynamic criminogenic
needs domains (family circumstances/parenting, education/employment, peer relations,
substance abuse, leisure/recreation, personality/behavior, and attitudes/orientation). Items within
each domain are summed into domain scores, which are also assigned categorical descriptors
(low, medium, high). An overall score is determined by summing all items; total scores also
categorize youth as low, moderate, high, or very high risk to reoffend.
The YLS/CMI has been reported to possess moderate to strong internal consistency for
most subscales (Catchpole & Gretton, 2003; Schmidt, Hoge, & Gomes, 2005; Vitopoulos,
Peterson-Badali, & Skilling, 2012). Moderate to strong concurrent validity has also been
reported between YLS/CMI total score and broad and narrow-band scores on the Child Behavior
Checklist (Schmidt et al., 2005) and Structured Assessment of Violence Risk in Youth
(Catchpole & Gretton, 2003). Predictive validity is reported to be moderate to high, with
YLS/CMI total scores significantly correlated with general and violent recidivism (Catchpole &
Gretton, 2003; Olver, Stockdale, & Wormith, 2009; Onifade et al., 2008; Rennie & Dolan, 2010).
In the current sample, inter-rater reliabilities for the YLS/CMI total score among the primary
clinicians ranged from .80 to .98 (average r = .93).
(3) Matching of identified criminogenic needs
Following procedures from previous studies (Peterson-Badali et al., 2015; Vitopoulos et
al., 2012), the extent to which youths’ identified criminogenic needs were targeted through
10
intervention and/or case management was examined using the concept of “treatment match”. For
each criminogenic need domain in which a youth had an identified need (i.e., a clinician
recommendation), match was coded as a binary variable (i.e., matched or not matched) by
reviewing the youth’s probation records to determine whether the need had been addressed via
service provision and/or case management activities. For example, a match in the Substance Use
domain meant that a youth successfully completed most or all of a substance abuse counseling
program recommended by a clinician. Other examples of match include completion of evidence-
based programs targeting clinician-identified Attitude and/or Personality needs, enrolment and
completion of school following recommendations in the Education domain, and association with
pro-social peers following recommendations in the Peer Relations domain.
For each youth an ‘overall match’ score was also calculated by dividing the number of
matched needs by the number of identified needs. For example, if a youth had two out of four
identified criminogenic needs addressed during probation, the youth’s overall match score was
50%.
(4) Recidivism
Recidivism was defined as whether a youth was convicted of one or more offenses within
a two year follow-up period following referral for assessment. Information was obtained from a
national police criminal record database. Reoffenses were only included for new offenses
occurring at least 3 months after forensic assessment completion in order to allow sufficient time
for probation service implementation.
Statistical analyses
(1)Mental disorder profiles
A two-step cluster analysis (SPSS Statistics/IBM Version 22, 2013) – which handles
categorical data and automatically tests multiple cluster solutions, outputting the optimal solution
11
– was used to determine patterns of mental disorder. Two-step cluster analysis employs a two-
stage process: In the first pre-clustering stage, each record in the dataset is scanned to determine
whether it is to merge with previously constructed dense regions or to form a new cluster by
itself based on a Log-likelihood distance measure. In the second stage, a hierarchical clustering
algorithm is applied to combine data records sequentially to form clusters (Chiu, Fang, Chen,
Wang, & Jeris, 2001; Mooi & Sarstedt, 2011). The six mental disorder variables described above
were used in the analysis: disruptive behavior disorders, ADHD, SUD, internalizing disorders,
LD, and cognitive disabilities. After the cluster analysis was completed, individuals within each
cluster were assigned a membership category, and subsequent analyses were conducted based on
the cluster group memberships.
(2) Treatment match across clusters
Overall treatment match scores were compared across the cluster groups using one-way
ANOVA tests. Post-hoc test Tukey was used to evaluate significant ANOVA results. Chi-square
tests were used to compare clusters within each criminogenic need domain, and Fisher’s exact
test was used in cases with small sample sizes.
(3) Predicting recidivism
A hierarchical binary logistic regression was conducted with recidivism as the outcome
variable. The predictors included individuals’ YLS/CMI domain scores for Criminal History,
overall match scores, and mental disorder cluster membership.
RESULTS
Preliminary analyses
Table 1 presents the descriptive statistics for participants’ demographic information,
offense information, and rates of psychiatric diagnoses. In the current sample, 67.2% of the
participants met diagnostic criteria for two or more mental disorder diagnoses. The most
12
common diagnoses included disruptive behavior disorders (74.9%), which included individuals
with CD and/or ODD diagnoses. This is followed by ADHD (43.1%), substance use/abuse
disorders (37.4%), LD (25.6%) , cognitive disabilities (10.3%), and internalizing disorders
(23.1%).The overall recidivism rate was 63%, and mean time to first reoffense conviction was
approximately 17 months for youth who reoffended.
Table 1: Participants’ demographic, offense, and psychiatric information
Variable Total (N = 195)
% Ethnicity (n)
Black 32.2 (63)
White 27.2 (53)
Asian 7.2 (14)
Other 23.6 (46)
Missing 9.7 (19)
% Index offense (n)
Non-violent 22.6 (43)
Violent (non-sexual) 63.1 (123)
Sexual 13.8 (27)
Psychiatric diagnoses (n)
Any two or more diagnoses 67.2 (131)
Any disruptive behavior disorders 74.9 (146)
Conduct disorder 72.8 (142)
Oppositional defiant disorder 10.8 (21)
Attention deficit hyperactivity disorder 43.1 (84)
Substance use/abuse diagnosis 37.4 (73)
13
Learning disabilities 25.6 (50)
Cognitive disabilities 10.3 (20)
Internalizing disorder 23.1 (45)
Any mood disorder 19.0 (37)
Any anxiety disorder 10.8 (21)
Recidivism-yes (n) 63.1 (123)
Time to recidivism conviction in days for youth who
reoffended(s.d.)
508 (300.4)
Mental illness profiles
The cluster analysis resulted in a six cluster solution. The first cluster included individuals with
primarily LD diagnoses (n = 33, 16.9% of the total sample). The second cluster included
individuals with disruptive behavior disorders and internalizing disorders (DBD & Internalizing,
n = 38, 19.5% of the total sample). The third cluster included individuals with ADHD and
disruptive behavior disorders diagnoses (DBD & ADHD, n = 24, 12.3% of the total sample). The
fourth cluster included individuals with a disruptive behavior disorders, LD, and ADHD
diagnoses (DBD, LD, ADHD, n = 28, 14.4% of the total sample). The fifth cluster included
individuals with a disruptive behavior disorders, SUD, and ADHD diagnoses (DBD, SUD,
ADHD, n = 36, 18.5% of the total sample). The sixth cluster included individuals with primarily
disruptive behavior disorders diagnoses (n = 36, 18.5% of the total sample). Table 2 presents the
recidivism rates for the mental disorder clusters; a chi-square analysis resulted in a significant
14
model, 2 (5) = 11.03, p = .05. It appears that clusters 4 (DBD, LD, ADHD) and 6 (DBD) have
significantly higher rates of offending than clusters 1 (LD) and 2 (DBD & Internalizing
Disorders).
Are identified criminogenic needs being met at a similar rate across the mental disorder
clusters?
Table 2 displays the percentage of youth for whom a criminogenic need was identified and
matched during probation within each criminogenic needs domain and across mental disorder
clusters. The overall proportion of treatment match varied from 17% (in the peer relations
domain) to 35% (in the family domain). When examined across the mental disorder clusters, peer
relations, leisure, and antisocial attitudes consistently displayed the lowest proportions of
treatment match. There were significant differences across the mental disorder clusters with
respect to overall treatment match scores (F(5,189) = 3.66, p = .004, η²= 0.09). Individuals in
cluster 2 (DBD, Internalizing) (M = .43, SD = .33) were more likely to have their identified
needs successfully addressed compared to individuals in cluster 5 (DBD, SUD, ADHD) (M = .17,
SD = .26) and 6 (DBD) (M = .22, SD = .25). There were no significant differences across the
other mental disorder clusters. When criminogenic needs were examined individually, Family
was the only domain with significant differences across the clusters, with clusters 5 (DBD, SUD,
ADHD) and 6 (DBD) less likely to have their need matched compared to the other clusters.
Table 2: Percentage treatment match scores within each criminogenic need and recidivism rates across mental disorder clusters
Variables
Cluster 1
(LD)
Cluster 2
(DBD &
Internalizing)
Cluster 3
(DBD &
ADHD)
Cluster 4
(DBD, LD,
& ADHD)
Cluster 5
(DBD,
SU/A/D, &
ADHD)
Cluster 6
(DBD)
Total
Family 37.0 (10) 44.1 (15) 52.4 (11) 50.0 (13) 18.8 (6) 12.9 (4) 34.5 (59)
Education/Employment 29.0 (9) 45.9 (17) 31.8 (7) 32.1 (9) 20.6 (7) 41.7 (15) 34.0 (64)
Peer Relations 23.8 (5) 4.8 (1) 26.9 (7) 25.0 (6) 11.4 (4) 11.8 (4) 16.8 (27)
Substance Use 41.7 (5) 7.7 (1) 36.4 (8) 25.0 (5) 17.1 (6) 5.3 (1) 21.5 (26)
Leisure 39.1 (9) 15.0 (3) 30.0 (9) 19.0 (4) 10.7 (3) 16.1 (5) 21.6 (33)
Personality 30.8 (8) 34.8 (8) 58.6 (17) 28.6 (8) 21.9 (7) 30.3 (10) 33.9 (58)
Attitude 25.0 (2) 11.8 (2) 31.6 (6) 20.0 (4) 16.7 (4) 12.5 (4) 18.3 (22)
Reoffended 48.5 (16) 50.0 (19) 66.7 (16) 78.6 (22) 63.9 (23) 75.0 (27) 63.1 (123)
Does matching services to identified criminogenic needs predict recidivism equally well
across the mental disorder clusters?
To address this question I first examined the predictive validity of the YLS/CMI as a risk
measure in this sample. A logistic regression analysis resulted in a significant model, 2 (1) =
12.97, p < .001; and its predictor, YLS/CMI total score, was also significant, B = .07, 2 (1) =
12.04, p = .001, odds ratio (OR) = 1.07, 95% confidence interval (CI) = [1.03, 1.11]. Therefore,
for each unit increase in YLS/CMI score, the odds of reoffense were 7% greater. Additionally, a
Receiver Operating Characteristic (ROC) analysis was conducted to examine the ability of the
model to correctly classify individuals. The area under the curve (AUC) statistic was significant
(0.66, p < .001), indicating that YLS/CMI scores classified the sample significantly better than
chance; at a 95% CI, there was a 66% probability that a randomly selected recidivist would
obtain a higher YLS/CMI score than a randomly selected non-recidivist (CI range = 0.57-0.74).
Next, a hierarchical logistic regression with recidivism as the outcome was conducted to
examine whether matching services to identified criminogenic needs predicted recidivism
equally well across the mental disorder clusters. YLS/CMI Criminal History domain scores were
first entered into the regression model in lieu of overall YLS/CMI scores to control for risk
because: 1) static factors such as offense history are strong predictors of reoffending (Bonta, Law,
& Hanson, 1998; Cottle, Lee & Heilburn, 2001), and 2) the overall treatment match scores
(which are entered in the second step of the hierarchical logistic regression) were calculated from
the total YLS/CMI risk scores, resulting in an overlap if YLS/CMI total scores were used
(Vitopoulos et al., 2012). Furthermore, Criminal History domain scores and YLS/CMI total risk
scores were highly correlated, r (236) = .65, p < .001, making Criminal History an appropriate
proxy for YLS/CMI total scores. In the second step, overall treatment match scores for each
17
participant were entered into the model. In the third and final step, mental disorder clusters were
entered into the model, with cluster 1 (LD) as a reference group because of its ‘low risk’ status.
The model was significant at every step (see Table 4). Criminal history was significant at
every step, with higher scores associated with an increase in recidivism. Matching services to
clinician identified needs was also significant at every step, with higher overall treatment match
scores associated with decreased odds of recidivism. With respect to the mental disorder clusters,
clusters 4 (DBD, LD, ADHD) and 6 (DBD) significantly predicted increased odds of recidivism
when compared to cluster 1 (LD).
In order to understand why clusters 4 (DBD, LD, ADHD) and 6 (DBD) predicted
increased odds of recidivism even after controlling for Criminal History scores and treatment
match, the compositions of the clusters were explored in more detail. In particular, given that the
‘disruptive behavior disorders only’ cluster significantly predicted recidivism (and was
associated with higher reoffence rates than most of the other clusters), it seemed prudent to
examine specific YLS/CMI items related to disruptive behavior disorders across the mental
disorder clusters. The items included Inadequate guilt feelings under the Antisocial Personality
criminogenic need domain, as well as Antisocial/procriminal attitudes, Not seeking help, Actively
rejecting help, and Callous/little concern for others under the Antisocial Attitude criminogenic
need domain. Chi-square analyses revealed significant differences across the mental disorder
clusters for each of these YLS/CMI items (see Table 4). Compared to clusters 1 (LD) and 2
(DBD & Internalizing Disorders), a higher percentage of individuals in cluster 6 (DBD) were
identified by clinicians as presenting Inadequate guilt feelings and Not seeking help. Compared
to cluster 1 (LD), a greater percentage of youth in clusters 4 (DBD, LD, ADHD) and 5 (DBD,
SUD, ADHD) had the Antisocial/procriminal attitudes item checked off by clinicians. Compared
to cluster 1 (LD), a greater percentage of youth in cluster 4 (DBD, LD, ADHD) had the Actively
18
rejecting help item endorsed by clinicians. Compared to cluster 1 (LD), a greater percentage of
youth in clusters 3 (DBD & ADHD), 4 (DBD, LD, ADHD), 5 (DBD, SUD, ADHD), and 6
(DBD) were identified as Callous/little concern for others item; cluster 6 (DBD) was also more
likely to have this item checked off than cluster 2 (DBD & Internalizing Disorders).
Table 3: Hierarchical regression analyses predicting recidivism from criminal history, match, and mental disorder profiles
Model Variables Β SEβ Wald’s χ2 Df p exp(B)
Lower
CI
(95%)
Upper
CI
(95%)
Model (Step) 1
Criminal history 0.47 0.09 28.26 1 <.001 1.60 1.35 1.90
Constant -0.43 0.19 4.94 1 .03 0.65
Overall model at Step 1
33.42 1 <.001
Model (Step) 2
Criminal history 0.40 0.09 18.64 1 <.001 1.48 1.24 1.78
Overall treatment match -1.77 0.49 13.15 1 <.001 0.17 0.07 0.44
Constant 0.27 0.27 0.97 1 .33 1.31
Overall model at Step 2
47.35 2 <.001
Model (Step) 3
Criminal history 0.44 0.10 17.89 1 <.001 1.55 1.27 1.91
Overall treatment match -1.75 0.53 11.04 1 .001 0.17 0.06 0.49
20
Cluster 2 0.31 0.46 0.47 1 .49 1.37 0.56 3.34
Cluster 3 0.49 0.57 0.73 1 .39 1.62 0.54 4.93
Cluster 4 1.20 0.55 4.79 1 .03 3.33 1.13 9.79
Cluster 5 -0.45 0.52 0.77 1 .38 0.64 0.23 1.75
Cluster 6 1.06 0.49 4.72 1 .03 2.88 1.11 7.47
Constant -0.13 0.34 0.16 1 .69 0.88
Overall model at Step 3 59.91 7 <.001
21
Table 4: Disruptive behavior disorders related YLS/CMI items by mental disorder clusters
Variables
Cluster 1
(LD)
%(n)
Cluster 2
(DBD &
Internalizing)
%(n)
Cluster 3
(DBD &
ADHD)
%(n)
Cluster 4
(DBD,
LD, &
ADHD)
%(n)
Cluster 5
(DBD,
SU/A/D, &
ADHD)
%(n)
Cluster 6
(DBD)
%(n)
X2 df p V
7f. Inadequate guilt feelings 37.5 (12) 40.5 (15) 68.2 (15) 71.4 (20) 65.7 (23) 72.2 (26) 17.04 5 .004 .30
8a. Antisocial/procriminal
attitudes
22.6 (7) 29.7 (11) 40.9 (9) 60.7 (17) 62.9 (22) 44.4 (16) 17.17 5 .004 .30
8b. Not seeking help 34.4 (11) 43.2 (16) 63.6 (14) 64.3 (18) 54.3 (19) 75.0 (27) 15.04 5 .010 .28
8c. Actively rejecting help 9.4 (3) 27.0 (10) 22.7 (5) 50.0 (14) 31.4 (11) 30.6 (11) 12.74 5 .026 .26
8e. Callous, little concern for
others
6.3 (2) 16.2 (6) 45.5 (10) 42.9 (12) 48.6 (17) 41.7 (15) 23.03 5 <.001 .35
DISCUSSION
The present study aimed to further understand the role of mental health in offending by
examining mental disorder profiles in justice-involved youth in the context of the risk-need-
responsivity framework. To this end, interrelationships among mental disorder clusters, risk
levels, criminogenic needs, case management, and recidivism were explored.
Consistent with previous findings of high comorbidity rates in justice-system involved
(Abram et al., 2003; Abrantes et al., 2004; Shufelt & Cocozza, 2006; Vermeiren, 2003), over 67%
of the youth in the current sample exhibited two or more concurrent diagnoses. Not surprisingly,
disruptive behavior disorders was the most frequently diagnosed mental disorder, followed by
ADHD, and substance use disorders. The base rates of mental disorder diagnoses found in the
current study were consistent with those found in the wider literature. For example, 43% of the
sample met diagnostic criteria for ADHD, which falls within the typical range observed in
justice-involved youth (Lederman, Dakof, Larrea, & Li, 2004; Shufelt & Cocozza, 2006; Teplin
et al., 2007; Vreugdenhil et al., 2004). Results from the current study confirm previous findings
that comorbidity of mental disorders is the norm, rather than the exception, in this sample of
justice-involved youth.
Mental disorder profiles
Five of the six mental disorder clusters included disruptive behavior disorders, an
unsurprising finding given that antisocial and/or criminal behaviors are features of diagnoses
such as conduct disorder. Disruptive behavior disorders and ADHD was the most prevalent of
the comorbid diagnoses, which was manifested in three of the six mental disorder clusters. This
finding is also congruent with the extant literature on comorbidities in the community and justice
populations, such that ADHD has been found to frequently overlap with CD (Angold et al., 1999;
Biederman, Newcorn, & Sprich, 1991; Vermeiren, 2003). This high comorbidity rate may reflect
23
similarities in the presentation of symptoms for these disorders, and/or synergies arising from the
interaction of these two disorders. For example, a youth with ADHD may, in a moment of
impulsivity, engage in reckless thrill seeking behaviors that introduce him to the justice system.
It may also be that the combination of symptoms represented in these two disorders puts
individuals at risk for engagement in antisocial acts (e.g., aggression, lying, impulsivity). For
example, ADHD has been more frequently found in children with early-onset, rather than late-
onset, CD (Loeber, Green, Keenan, & Lahey, 1995). Given that individuals with early-onset CD
often engage in persistent antisocial behaviors across the lifespan, concurrent diagnoses of
ADHD and disruptive behavior disorders may reflect intrinsic vulnerabilities in some youth that
explain their entries into the justice system.
One mental disorder profile included individuals with disruptive behavior disorders,
ADHD, and LD comorbidities. The overlap between LD and ADHD is a robust finding in the
general population (Biederman et al., 1991). Children and youth with both diagnoses have
similar difficulties related to learning and often exhibit poor academic achievement, leading to
disengagement from schools (Barry, Lyman, & Klinger, 2002; Marder & D’Amico, 1992;
Rapport, Scanlan, & Denney, 1999). In fact, the co-occurrence of both LD and ADHD is
associated with poorer outcomes such as more disruptions in cognitive abilities and poorer
performance in academic tasks compared to children with either ADHD or LD (Jakobson &
Kikas, 2007; Kamphaus & Frick, 2005; Mayes, Calhoun, & Crowell, 2000). Academic
underachievement and school disengagement are potent risk factors for delinquency and later
problematic behaviors (Henry, Knight, & Thomberry, 2012; Murray & Farrington, 2010).
Therefore, individuals with comorbid LD and ADHD may be at higher risk for academic
underachievement than those with either LD or ADHD diagnoses alone, and combined with
DBD, predispose them to engage in antisocial behaviors.
24
Another mental disorder cluster with disruptive behavior disorders and ADHD also
included substance use/abuse disorders. Substance use/abuse disorders are highly prevalent in
justice-involved youth (Abram et al., 2003; Abrantes et al., 2004) and their presence is related to
more severe antisocial behaviors and higher incidence of psychiatric comorbidity (Vermeiren,
2003). The aggregation of substance abuse, CD and ADHD may be an especially hazardous
mixture, as substance abusers tend to demonstrate more aggressive CD and more severe ADHD
than non-abusers (Milin et al., 1991), and they tend to start using substances at earlier ages
(Thompson, Riggs, Mikulich, & Crowley, 1996)). Therefore, the co-occurrence of these three
diagnoses may represent a complex developmental relationship that is especially relevant for
justice-involved youth.
The disruptive behavior disorders and internalizing disorders cluster primarily included
individuals with a disruptive behavior disorders and mood and/or anxiety disorders. The
internalizing disorders included a varied array of diagnoses, such as major depressive disorders,
dysthymia, generalized anxiety disorders, and/or post traumatic stress disorders. There is some
evidence from the literature indicating a tendency for CD to co-occur with both depressive and
anxiety disorders in children and adolescents in the general (Zoccolillo, 1992) and clinical
populations (Russo & Beidel, 1994). In fact, increasing severity of antisocial behaviors is
associated with increasing risk for affective disorders (Zoccolillo, 1992), and recently one study
has reported that justice-involved youth with both disruptive behavior disorders and internalizing
disorders are at an increased risk for recidivism compared to non-disordered justice-involved
youth (Hoeve et al., 2013). Therefore, this mental disorder profile may represent a group of
youth with more severe antisocial behaviors and poorer outcomes.
Two mental disorder clusters emerged consisting primarily of single diagnoses. The LD
cluster included youth with varied impairments in aspects of learning ranging from reading,
25
mathematics, to language. Individuals with LD are at a higher risk for involvement in the justice
system, and three hypotheses have been put forward to explain the link (Brier, 1989; Keilitz &
Dunivant, 1986). The susceptibility hypothesis proposes that individuals with LD possess certain
cognitive and neurological characteristics (e.g., inability to anticipate the consequences actions,
poor perception of social cues) that leave them vulnerable to engagement in antisocial activities
(Murray, 1976). The school failure hypothesis proposes that learning disability leads to academic
failure, which contributes to negative self-image, frustration, and school dropout, in turn leading
to antisocial behaviors (Murray, 1976). The differential treatment hypothesis proposes that the
justice system treats individuals with LD more harshly than individuals without LD (Dunivant,
1982). Current literature suggests that a multifactorial explanation – combining factors from the
three hypotheses – affect the relationship between learning disability and youth delinquency
(Brier, 1989).
The disruptive behavior disorders -only cluster consisted of individuals with CD and/or
ODD diagnoses, with a defining feature in antisocial behaviors. By virtue of their inclusion in
the justice system, all youth in the current sample have displayed antisocial behaviors. However,
given that most of the youth in the sample exhibited psychiatric comorbidities, the fact that youth
in this cluster have predominantly disruptive behavior disorders diagnoses without other
psychopathological complications makes them a unique group for comparison and study.
Despite limited research in psychiatric comorbidities in justice-involved youth, the mental
disorder profiles that emerged from the current study are consistent with the extant literature
from the justice-involved, as well as general child and adolescent, populations. Thus, these six
mental disorder profiles reflect robust findings that are relevant for the treatment and
rehabilitation of justice-involved youth (discussed below).
Matching services to criminogenic needs
26
In addition to identifying mental disorder profiles, it is also important to examine their
relationships to justice-involved youth. To our knowledge, no research has investigated mental
disorder clusters in the context of the RNR framework. The RNR need principle stipulates that
services must target identified criminogenic needs in order to reduce risk (Andrews et al., 1990).
One of my goals was to understand whether mental disorder profiles were associated with
delivery of such services, or “treatment matching” (i.e., addressing identified criminogenic needs
through probation-directed case management). I found that youth in cluster 2 (disruptive
behavior disorders & internalizing disorders) had higher proportion of needs matched compared
to youth in cluster 5 (disruptive behavior disorders, SUD, ADHD) and 6 (disruptive behavior
disorders). One explanation for this finding relates to variability in the availability of resources
for targeting different criminogenic needs. Few programs and services exist that target the
criminogenic need of Antisocial Attitude, which includes features coinciding with the diagnostic
criteria for disruptive behavior disorders. In the current study, Antisocial Attitudes displayed one
of the lowest proportion treatment match across the mental disorder clusters. These findings fall
in line with previous research, in which lack of programming was identified by probation
officers as a significant barrier to treatment (Haqanee, Peterson-Badali, & Skilling, 2015). In
contrast, resources that target the criminogenic need of Education are more abundant, including
special education programs, alternative schools, cooperative education programs, and vocational
programs. In this study, Education was one of the criminogenic needs with the highest proportion
of treatment match across the mental disorder clusters, which coincides with previous research
findings that probation officers may “over-program” the criminogenic need of Education (i.e.,
targeting Education when it was not identified as a relevant need) (Haqanee et al., 2015; Luong
& Wormith, 2011). Thus, successful matching of services to identified criminogenic needs may
be dependent on the nature of the needs. Individuals in clusters 5(disruptive behavior disorders,
27
SUD, ADHD) and 6 (disruptive behavior disorders) exhibited higher needs in Antisocial Attitude
domains, one of the criminogenic needs with limited programming, which may explain their
diminished proportion match.
Mental disorder profiles and recidivism
Consistent with previous research on similar samples, youths’ reoffending was
significantly predicted by their criminogenic risk (as measured by Criminal History) as well as
by the extent to which their identified needs were addressed during their probation terms overall
treatment match (Peterson-Badali et al., 2015; Vieira et al., 2009). In the current study – even
after accounting for risk and treatment matching – the specific nature of the mental disorder
clusters also differentiated individuals with respect to recidivism outcomes. Individuals in
clusters 4 (disruptive behavior disorders, LD, ADHD) and 6 (disruptive behavior disorders) had
higher odds of reoffense than those in cluster 1 (LD), which was chosen as the reference group
due to its consistently low risk scores. In addition, the observed rates of recidivism were also
elevated for clusters 4 (disruptive behavior disorders, LD, ADHD) and 6 (disruptive behavior
disorders) compared to other clusters, such as 1 (LD), and 2 (disruptive behavior disorders &
internalizing disorders).
The comorbidity profile of disruptive behavior disorders, LD, and ADHD (cluster 4) in
justice-involved youth is not surprising given that high comorbidity rates of disruptive behavior
disorders with ADHD (Vermeiren, 2003), and LD with ADHD (Biederman et al., 1991) have
been long established in the forensic and general literature. However, the particular profile of
disruptive behavior disorders, LD, and ADHD has hitherto not been emphasized as an important
mental disorder profile for the justice population. This profile’s inclusion of both LD and ADHD
suggests that individuals in this group likely demonstrate poor academic achievement (Barry et
al., 2002), and poor academic achievement has been linked to behavioral problems and higher
28
rates of recidivism in justice-involved adolescents (Katsiyannis, Ryan, Zhang, & Spann, 2008).
However, if the link between academic achievement and delinquency is robust, individuals in
cluster 1 (LD) should also reoffend at similar or higher rates compared to other clusters, but this
was not the case since cluster 1 demonstrated the lowest recidivism rate out of the mental
disorder clusters. Given the disruptive behavior disorders, LD, and ADHD cluster, in particular,
predicted recidivism, there must be characteristics at play other than academic underachievement.
In a study where justice-involved youth with LD were followed longitudinally, it was found that
specific personality characteristics of impulsivity and poor judgment discriminated between
persisting and non-persisting antisocial behaviors (Waldie & Spreen, 1993). Therefore, it appears
that the unique combination of ADHD and LD, in the presence of disruptive behavior disorders
increases an individual’s likelihood for reoffending, and this comorbidity pattern warrants further
investigation.
Cluster 6 (disruptive behavior disorders) also predicted an increased odds of reoffending in
the current study. In order to explore this finding further, YLS/CMI items related to disruptive
behavior disorders were examined. Clinicians identified the YLS/CMI antisocial attitude items
inadequate guilt feelings, not seeking help, and callous/little concern for others for a greater
percentage of individuals in cluster 6 than for individuals in clusters 1 (LD) and 2 (disruptive
behavior disorders & internalizing disorders). This differential rate of identification of YLS/CMI
items is particularly interesting given that individuals in cluster 2 and 6 both have disruptive
behavior disorders diagnoses. Clinicians also identified the items inadequate guilt feelings (e.g.,
toward victims), antisocial attitudes, not seeking help, actively rejecting help, and callous/little
concern for others for a greater percentage of youth in cluster 4 (disruptive behavior disorders,
LD, ADHD) than for youth in cluster 1 (LD). Clinician identification of these items suggests that
youth in clusters 4 and 6 would qualify for the callous-unemotional specifier (e.g., lack of guilt
29
or remorse, callous-lack of empathy) under the current Conduct Disorder diagnostic criteria of
the Diagnostic and Statistical Manual of Mental Disorders-5th
edition (American Psychiatric
Association, 2013). Studies have found that youth with callous-unemotional traits engage in
particularly severe, stable, and aggressive patterns of antisocial behaviors (Frick & White, 2008;
Moran, Ford, Butler, & Goodman, 2008), which may explain the increased rates of recidivism in
clusters 4 and 6. These findings suggest that clinicians should go beyond diagnosing mental
disorders to understanding the specific subtypes and specifiers related to individuals’ diagnoses.
In this case, understanding whether justice-involved youth endorse callous-unemotional traits
may improve treatment planning and case management outcomes.
Implications for risk, need, and responsivity
Consistent with the RNR principles and previous research (Catchpole & Gretton, 2003;
Olver et al., 2009; Peterson-Badali et al., 2015), risk (as assessed by the YLS/CMI) was
predictive of youths’ reoffending, and addressing identified criminogenic needs through
interventions was associated with reduced rates of recidivism. In addition to adding to the body
of literature supporting the risk and need principles, this study also identified commonly
occurring mental disorder profiles in a sample of justice-involved youth, in which two profiles
(clusters 4 and 6) emerged as significant predictors of youths’ recidivism even after controlling
for risk and treatment matching. The latter finding is noteworthy given that the central tenets of
the RNR framework rest on criminogenic factors as necessary and sufficient to predict
reoffending and critical to treatment planning. Differences between clusters in the frequency
with which clinicians endorsed YLS/CMI antisocial attitude items suggest that it may be
important for case managers and treatment providers to gain finer grained mental health
information (i.e., specifiers of mental disorder diagnoses) as well as attend to specific YLS/CMI
30
items in order to inform and fine-tune intervention planning and case management for justice-
involved youth.
Strengths and limitations
Data used in the current study were collected from comprehensive assessments, often
involving multiple parties such as justice-involved youth, their parents, and other collateral
sources of information (e.g., frontline workers). This is a methodological strength as it provided
multiple sources of information regarding youths’ mental health functioning, which was used to
arrive at psychiatric diagnoses. Past research on justice-involved youth’ psychopathology heavily
relied on self-report measures (Vermeiren, 2003), which alone is not a reliable way to arrive at
mental disorder diagnoses, especially with respect to disruptive behavior disorders. Furthermore,
past research mainly focused on understanding single psychiatric diagnoses and their impacts on
youth’ outcomes. When comorbidities were examined in the justice-involved population, the
focus was often placed on understanding their prevalence rates. This study is one of the few that
examined psychiatric profiles in justice-involved youth and attempted to understand the
significance of such profiles in relation to the RNR model and youth’ reoffending.
The current study also had several limitations. The justice-involved youth sample was
drawn from a single mental health care facility in an urban area in Canada. Gender ratio in this
sample was heavily skewed, with nearly three times as many male as female participants.
Disproportionate gender ratio precluded the examination of mental disorder profiles by gender,
which is unfortunate given the documented differences in the manifestation of psychopathologies
in male and female justice-involved individuals (Abram et al., 2003). Furthermore, the current
study’s modest sample size also prevented more detailed classification of mental disorders. For
example, the low numbers of anxiety and mood disorders in the current sample necessitated the
creation of an overall internalizing disorder category; such amalgamation may have concealed
31
important differences between specific internalizing disorders with respect to youth’ risk, needs,
and/or outcomes. Therefore, future research should aim to draw larger samples from
geographically diverse areas with more balanced gender ratios in order to replicate and extend
the findings from this study.
References
Abram, K. M., Teplin, L. A., McClelland, G. M., & Dulcan, M. K. (2003). Comorbid psychiatric
disorders in youth in juvenile detention. Archives of General Psychiatry, 60, 1097-1108.
Abrantes, A. M., Hoffmann, N. G., Anton, R. P., & Estroff, T. W. (2004). Identifying co-
occurring disorders in Juvenile Justice populations. Youth Violence and Juvenile Justice, 2,
329-341.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental
disorders (4th ed., text rev.). Washington, DC: American Psychiatric Association.
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental
disorders (5th ed.). Arlington, VA: American Psychiatric Association.
Andrews, D. A., & Bonta, J. (2010). Rehabilitating criminal justice policy and
practice. Psychology, Public Policy, and Law, 16(1), 39-55.
Andrews, D. A., Bonta, J., & Hoge, R. D. (1990). Classification for effective rehabilitation:
Rediscovering psychology. Criminal Justice and Behavior, 17(1), 19-52.
Andrews, D. A., Bonta, J., & Wormith, S. J. (2004). The Level of Service/Case Management
Inventory (LS/CMI). Toronto: Multi-Health Systems.
Angold, A., Costello, E. J., & Erkanli, A. (1999). Comorbidity. Journal of Child Psychology and
Psychiatry, 40(01), 57-87.
33
Armstrong, T. D., & Costello, E. J. (2002). Community studies on adolescent substance use,
abuse, or dependence and psychiatric comorbidity. Journal of Consulting and Clinical
Psychology, 70, 1224-1239..
Barry, T. D., Lyman, R. D., & Klinger, L. G. (2002). Academic underachievement and attention-
deficit/hyperactivity disorder: The negative impact of symptom severity on school
performance. Journal of School Psychology, 40(3), 259-283.
Biederman, J., Newcorn, J., & Sprich, S. (1991). Comorbidity of attention deficit hyperactivity
disorder. Am J Psychiatry, 148(5), 564-577.
Bonta, J. (1995, September). The responsivity principle and offender rehabilitation. Forum on
Corrections Research, 7(3), 34-37.
Bonta, J., & Andrews, D. A. (2007). Risk-need-responsivity model for offender assessment and
rehabilitation. Rehabilitation, 6, 1-22.
Bonta, J., Law, M., & Hanson, K. (1998). The prediction of criminal and violent recidivism
among mentally disordered offenders: a meta-analysis. Psychological Bulletin, 123(2),
123-142.
Brier, N. (1989). The relationship between learning disability and delinquency a review and
reappraisal. Journal of Learning Disabilities, 22, 546-553.
Burke, J. D., Loeber, R., & Birmaher, B. (2002). Oppositional defiant disorder and conduct
disorder: a review of the past 10 years, part II. Journal of the American Academy of Child
& Adolescent Psychiatry, 41, 1275-1293.
34
Carrellas, N., Wilens, T. E., & Anselmo, R. (2016). Treatment of comorbid substance use
disorders and adhd in youth. Current Treatment Options in Psychiatry, 3(1), 15-27.
Catchpole, R. E., & Gretton, H. M. (2003). The predictive validity of risk assessment with
violent young offenders a 1-year examination of criminal outcome. Criminal Justice and
Behavior, 30, 688-708.
Chan, Y. F., Dennis, M. L., & Funk, R. R. (2008). Prevalence and comorbidity of major
internalizing and externalizing problems among adolescents and adults presenting to
substance abuse treatment. Journal of Substance Abuse Treatment, 34(1), 14-24.
Chiu, T., Fang, D., Chen, J., Wang, Y., & Jeris, C. (2001). A robust and scalable clustering
algorithm for mixed type attributes in large database environment: Proceedings of the
Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data
Mining, San Francisco, 2001.
Colins, O., Vermeiren, R., Vahl, P., Markus, M., Broekaert, E., & Doreleijers, T. (2011).
Psychiatric disorder in detained male adolescents as risk factor for serious recidivism. The
Canadian Journal of Psychiatry, 56(1), 44-50.
Colins, O., Vermeiren, R. R. J. M., Vahl, P., Markus, M., Broekaert, E., & Doreleijers, T. (2012).
Parent-reported attention-deficit hyperactivity disorder and subtypes of conduct disorder as
risk factor of recidivism in detained male adolescents. European Psychiatry, 27, 329-334.
Cottle, C. C., Lee, R. J., & Heilbrun, K. (2001). The prediction of criminal recidivism in
juveniles a meta-analysis. Criminal Justice and Behavior, 28, 367-394.
35
Dembo, R., Schmeidler, J., Pacheco, K., Cooper, S., & Williams, L. W. (1998). The relationships
between youths' identified substance use, mental health or other problems at a juvenile
assessment center and their referrals to needed services. Journal of Child & Adolescent
Substance Abuse, 6(4), 23-54.
Dolnicar, S. (2002). A review of unquestioned standards in using cluster analysis for data-driven
market segmentation: Proceedings of the Australian and New Zealand Marketing Academy
Conference, Deakin University, Melbourne, 2002.
Drake, R. E., Mueser, K. T., Clark, R. E., & Wallach, M. A. (1996). The course, treatment, and
outcome of substance disorder in persons with severe mental illness. American Journal of
Orthopsychiatry, 66(1), 42-51.
Dunivant, N. (1982). The relationship between learning disabilities and juvenile delinquency:
Executive summary. Williamsburg, VA: National Center for State Courts.
Fazel, S., Doll, H., & Långström, N. (2008). Mental disorders among adolescents in juvenile
detention and correctional facilities: a systematic review and metaregression analysis of 25
surveys. Journal of the American Academy of Child & Adolescent Psychiatry, 47, 1010-
1019.
Frick, P.J., & White, S.F. (2008). The importance of callous-unemotional traits for
developmental models of aggressive and antisocial behavior. Journal of Child Psychology
and Psychiatry, 49, 359–375.
Gordon, J. A., Diehl, R. L., & Anderson, L. (2012). Does ADHD matter? Examining attention
deficit and hyperactivity disorder on the likelihood of recidivism among detained
youth. Journal of Offender Rehabilitation, 51, 497-518.
36
Grella, C. E., Hser, Y. I., Joshi, V., & Rounds-Bryant, J. (2001). Drug treatment outcomes for
adolescents with comorbid mental and substance use disorders. The Journal of Nervous
and Mental Disease, 189, 384-392.
Grieger, L., & Hosser, D. (2012). Attention deficit hyperactivity disorder does not predict
criminal recidivism in young adult offenders: Results from a prospective
study. International Journal of Law and Psychiatry, 35(1), 27-34.
Guebert, A. F., & Olver, M. E. (2014). An examination of criminogenic needs, mental health
concerns, and recidivism in a sample of violent young offenders: implications for risk,
need, and responsivity. International Journal of Forensic Mental Health, 13, 295-310.
Haqanee, Z., Peterson-Badali, M., & Skilling, T. (2015). Making “What works” work:
Examining probation officers’ experiences addressing the criminogenic needs of juvenile
offenders. Journal of Offender Rehabilitation, 54(1), 37-59.
Henry, K. L., Knight, K. E., & Thornberry, T. P. (2012). School disengagement as a predictor of
dropout, delinquency, and problem substance use during adolescence and early
adulthood. Journal of Youth and Adolescence, 41(2), 156-166.
Hinshaw, S. P. (1992). Academic underachievement, attention deficits, and aggression:
comorbidity and implications for intervention. Journal of Consulting and Clinical
Psychology, 60, 893-903.
Hoge, R. D. (2002). Standardized instruments for assessing risk and need in youthful
offenders. Criminal Justice and Behavior, 29, 380-396.
37
Hoge, R. D., & Andrews, D. A. (2002). Youth Level of Service/Case Management Inventory:
YLS/CMI interview guide. Toronto, Ontario, Canada: Multi-Health Systems.
Hoge, R. D., & Andrews, D. A. (2011). Youth Level of Service/case Management Inventory 2.0
(YLS/CMI 2.0): User's Manual. Toronto, Canada: Multi-Health Systems.
Jakobson, A., & Kikas, E. (2007). Cognitive functioning in children with and without attention-
deficit/hyperactivity disorder with and without comorbid learning disabilities. Journal of
Learning Disabilities, 40, 194-202.
Kamphaus, R. W., & Frick, P. J. (2005). Clinical assessment of child and adolescent personality
and behavior. Springer Science & Business Media.
Katsiyannis, A., Ryan, J. B., Zhang, D., & Spann, A. (2008). Juvenile delinquency and
recidivism: The impact of academic achievement. Reading & Writing Quarterly, 24, 177-
196.
Keilitz, I., & Dunivant, N. (1986). The relationship between learning disability and juvenile
delinquency current state of knowledge. Remedial and Special Education, 7(3), 18-26.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005).
Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National
Comorbidity Survey Replication. Archives of General Psychiatry, 62, 593-602.
Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., ... &
Kendler, K. S. (1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric
disorders in the United States: Results from the National Comorbidity Survey. Archives of
General Psychiatry, 51(1), 8-19.
38
Lederman, C. S., Dakof, G. A., Larrea, M. A., & Li, H. (2004). Characteristics of adolescent
females in juvenile detention. Int J Law Psychiatry, 27, 321–337.
Loeber, R., Burke, J. D., Lahey, B. B., Winters, A., & Zera, M. (2000). Oppositional defiant and
conduct disorder: a review of the past 10 years, part I. Journal of the American Academy of
Child & Adolescent Psychiatry, 39, 1468-1484.
Loeber, R., Green, S. M., Keenan, K., & Lahey, B. B. (1995). Which boys will fare worse? Early
predictors of the onset of conduct disorder in a six-year longitudinal study. Journal of the
American Academy of Child & Adolescent Psychiatry, 34, 499-509.
Luong, D., & Wormith, J. S. (2011). Applying risk/need assessment to probation practice and its
impact on the recidivism of young offenders. Criminal Justice and Behavior, 38, 1177-
1199.
Marder, C., & D'Amico, R. (1992). How Well Are Youth with Disabilities Really Doing? A
Comparison of Youth with Disabilities and Youth in General. A Report from the National
Longitudinal Transition Study of Special Education Students. Menlo Park, CA: SRI
International.
Mauser, A. J. (1974). Learning disabilities and delinquent youth. Academic Therapy. 9, 389-402.
Mayes, S. D., Calhoun, S. L., & Crowell, E. W. (2000). Learning disabilities and ADHD
overlapping spectrum disorders. Journal of Learning Disabilities, 33, 417-424.
McCormick, S., Peterson-Badali, M., & Skilling, T. A. (2016). Mental health and specific
responsivity: Its role in juvenile justice rehabilitation. In Press.
39
Milin, R., Halikas, J. A., Meller, J. E., & Morse, C. (1991). Psychopathology among substance
abusing juvenile offenders. Journal of the American Academy of Child & Adolescent
Psychiatry, 30, 569-574.
Mooi, E., & Sarstedt, M. (2011). A concise guide to market research. Berlin, Heidelberg:
Springer.
Moran, P., Ford, T., Butler, G., & Goodman, R. (2008). Callous and unemotional traits in
children and adolescents living in Great Britain. British Journal of Psychiatry, 192, 65–66.
Murphy, D. M. (1986). The prevalence of handicapping conditions among juvenile
delinquents. Remedial and Special Education, 7(3), 7-17.
Murray, C. A. (1976). The link between learning disabilities and juvenile delinquency: current
theory and knowledge, Washington, D.C: American Institutes for Research in the
Biological Sciences.
Murray, J., & Farrington, D. P. (2010). Risk factors for conduct disorder and delinquency: key
findings from longitudinal studies. The Canadian Journal of Psychiatry, 55, 633-642.
Olver, M. E., Stockdale, K. C., & Wormith, J. S. (2009). Risk assessment with young offenders a
meta-analysis of three assessment measures. Criminal Justice and Behavior, 36, 329-353.
Onifade, E., Davidson, W., Campbell, C., Turke, G., Malinowski, J., & Turner, K. (2008).
Predicting recidivism in probationers with the youth level of service case management
inventory (YLS/CMI). Criminal Justice and Behavior, 35, 474-483.
40
Peterson-Badali, M., Skilling, T., & Haqanee, Z. (2015). Examining implementation of risk
assessment in case management for youth in the justice system. Criminal Justice and
Behavior, 42, 304-320.
Rapport, M. D., Scanlan, S. W., & Denney, C. B. (1999). Attention-deficit/hyperactivity disorder
and scholastic achievement: a model of dual developmental pathways. Journal of Child
Psychology Psychiatry, 40, 1169–1183.
Rennie, C., & Dolan, M. (2010). Predictive validity of the Youth Level of Service/Case
Management Inventory in custody sample in England. The Journal of Forensic Psychiatry
& Psychology, 21, 407-425.
Rezansoff, S. N., Moniruzzaman, A., Gress, C., & Somers, J. M. (2013). Psychiatric diagnoses
and multiyear criminal recidivism in a Canadian provincial offender
population. Psychology, Public Policy, and Law, 19, 443-453.
Russo, M. F., & Beidel, D. C. (1994). Comorbidity of childhood anxiety and externalizing
disorders: Prevalence, associated characteristics, and validation issues. Clinical
Psychology Review, 14, 199-221.
Schmidt, F., Hoge, R. D., & Gomes, L. (2005). Reliability and validity analyses of the youth
level of service/case management inventory. Criminal Justice and Behavior, 32, 329-344.
Schwarz, G. (1978). Estimating the dimension of a model. The Annals of Statistics, 6, 461-464.
Shufelt, J. L., & Cocozza, J. J. (2006). Youth with mental health disorders in the juvenile justice
system: Results from a multi-state prevalence study (1-6). Delmar, NY: National Center
for Mental Health and Juvenile Justice.
41
Simpson, M. K., & Hogg, J. (2001). Patterns of offending among people with intellectual
disability: A systematic review. Part I: Methodology and prevalence data. Journal of
Intellectual Disability Research, 45, 384-396.
Skeem, J. L., Manchak, S., & Peterson, J. K. (2011). Correctional policy for offenders with
mental illness: creating a new paradigm for recidivism reduction. Law and Human
Behavior, 35, 110-126.
Skeem, J. L., Winter, E., Kennealy, P. J., Louden, J. E., & Tatar II, J. R. (2014). Offenders with
mental illness have criminogenic needs, too: Toward recidivism reduction. Law and
Human Behavior, 38(3), 212-224.
Skilling, T. A., & Sorge, G. B. (2014). Measuring antisocial values and attitudes in justice-
involved male youth: Evaluating the psychometric properties of the Pride in Delinquency
Scale and the Criminal Sentiments Scale–Modified. Criminal Justice and Behavior, 41,
992-1007.
Teplin, L. A. (1990). The prevalence of severe mental disorder among male urban jail detainees:
comparison with the Epidemiologic Catchment Area Program. American Journal of Public
Health, 80, 663-669.
Teplin, L. A., Abram, K. M., & McClelland, G. M. (1996). Prevalence of psychiatric disorders
among incarcerated women: I. Pretrial jail detainees. Archives of General Psychiatry, 53,
505-512.
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,
1133-1143.
42
Teplin, L. A., Abram, K. M., McClelland, G. M., Mericle, A. A., Dulcan, M. K., Washburn, J. J.,
& Butt. (2007). Psychiatric disorders of youth in detention. In C. L. Kessler & L. J. Kraus
(Eds.), The Mental Health Needs of Young Offenders: Forging Paths toward Reintegration
and Rehabilitation. (7-47). UK: Cambridge University Press.
Thompson, L. L., Riggs, P. D., Mikulich, S. K., & Crowley, T. J. (1996). Contribution of ADHD
symptoms to substance problems and delinquency in conduct-disordered
adolescents. Journal of Abnormal Child Psychology, 24, 325-347.
Ulzen, T. P., & Hamilton, H. (1998). The nature and characteristics of psychiatric comorbidity in
incarcerated adolescents. Canadian Journal of Psychiatry, 43(1), 57-63.
Vermeiren, R. (2003). Psychopathology and delinquency in adolescents: a descriptive and
developmental perspective. Clinical Psychology Review, 23, 277-318.
Vermeiren, R., Schwab-Stone, M., Ruchkin, V., De Clippele, A., & Deboutte, D. (2002).
Predicting recidivism in delinquent adolescents from psychological and psychiatric
assessment. Comprehensive Psychiatry, 43, 142-149.
Vieira, T., Skilling, T. & Peterson-Badali, M. (2009). Matching court-ordered services with
youths’ clinically identified treatment needs: Predicting treatment success for young
offenders Criminal Justice and Behavior, 36, 385-401.
Vitopoulos, N. A., Peterson-Badali, M., & Skilling, T. A. (2012). The relationship between
matching service to criminogenic need and recidivism in male and female youth
examining the RNR principles in practice. Criminal Justice and Behavior, 39, 1025-1041.
43
Vreugdenhil, C., Doreleijers, T. A., Vermeiren, R., Wouters, L. F., & Van Den Brink, W. (2004).
Psychiatric disorders in a representative sample of incarcerated boys in the
Netherlands. Journal of the American Academy of Child & Adolescent Psychiatry, 43(1),
97-104.
Waddell, C., Shepherd, C., Schwartz, C., & Barican, J. (2014). Child and youth mental disorders:
Prevalence and evidence-based interventions. Children’s Health Policy Centre, Simon
Fraser University: Vancouver.
Waldie, K., & Spreen, O. (1993). The relationship between learning disabilities and persisting
delinquency. Journal of Learning Disabilities, 26, 417-423.
Walker, J. L., Lahey, B. B., Hynd, G. W., & Frame, C. L. (1987). Comparison of specific
patterns of antisocial behavior in children with conduct disorder with or without coexisting
hyperactivity. Journal of Consulting and Clinical Psychology, 55, 910-913.
Wallace, C., Mullen, P. E., & Burgess, P. (2004). Criminal offending in schizophrenia over a 25-
year period marked by deinstitutionalization and increasing prevalence of comorbid
substance use disorders. American Journal of Psychiatry, 161, 716-727.
Wasserman, G. A., McReynolds, L. S., Lucas, C. P., Fisher, P., & Santos, L. (2002). The voice
DISC-IV with incarcerated male youths: prevalence of disorder. Journal of the American
Academy of Child & Adolescent Psychiatry, 41, 314-321.
Zoccolillo, M. (1992). Co-occurrence of conduct disorder and its adult outcomes with depressive
and anxiety disorders: A review. Journal of the American Academy of Child & Adolescent
Psychiatry, 31, 547-556.
44
Zoccolillo, M., Pickles, A., Quinton, D., & Rutter, M. (1992). The outcome of childhood conduct
disorder: implications for defining adult personality disorder and conduct
disorder. Psychological Medicine, 22, 971-971.