Psychiatric Disorders and Future Violent Arrests: A Prospective Study Among Detained Girls

• Author: Olivier F. Colins,Lore Van Damme
• DOI: 10.1177/1541204020916826
• Published date: 01 October 2020
• Date: 01 October 2020
• Subject Matter: Articles

Article

Psychiatric Disorders

and Future Violent Arrests:

A Prospective Study Among

Detained Girls

Olivier F. Colins

1,2

and Lore Van Damme

1

Abstract

This study scrutinizes if detained girls with psychiatric disorders were at risk for future violent

arrests during adolescence. A structured diagnostic interview was performed to determine the

presence of various psychiatric disorders in 313 detained girls. Official juvenile arrest records were

collected. With three exceptions that are in need of replication, psychiatric disorders (e.g., post-

traumatic stress disorder, subtypes of conduct disorder), psychiatric disorder categories (e.g.,

anxiety disorders), and psychiatric comorbidity patterns were not prospectively related to future

violent arrests. Our findings suggest that detained girls with psychiatric disorders should not be

considered more dangerous than their counterparts without disorders, at least not during

adolescence.

Keywords

violence, criminal justice–involved females, mental disorder, limited prosocial emotions, recidivism

Worldwide, detained girls remain an understudied population of criminal justice–involved youth,

especially in comparison to their male counterparts. Burgeoning evidence, nevertheless, shows that

these girls have a myriad of mental health problems (Krabbendam et al., 2015; Vahl et al., 2016; Van

Damme et al., 2014) and are at risk of becoming violent offenders (Colins, Van Damme, Andershed

et al., 2017; Wolff et al., 2017). Being responsive to recent calls to bolster what is known about

juvenile female violent offending and associated risk factors (Lim et al., 2019; Van Damme, Hoeve

et al., 2016; Wolff et al., 2017), this article examines whether psychiatric (co)morbidity prospec-

tively predicts violent offending in a sample of detained girls.

Psychiatric disorders defined in the Diagnostic and Statistical Manual of Mental Disorders

(DSM; American Psychiatric Publishing [APA], 2013) allow a clinically meaningful approach to

study the link between mental health and violent offending (Elkington et al., 2015). However, few

1

Department of Special Needs Education, Faculty of Psychology and Educational Sciences, Ghent University, Belgium

2

Center for Criminological and Psychosocial Research, O

¨rebro University, Sweden

Corresponding Author:

Olivier F. Colins, Department of Special Needs Education, Faculty of Psychology and Educational Sciences, Ghent University,

H. Dunantlaan 2, B-9000 Gent, Belgium.

Email: olivier.colins@ugent.be

Youth Violence and JuvenileJustice

2020, Vol. 18(4) 365-380

ªThe Author(s) 2020

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1541204020916826

journals.sagepub.com/home/yvj

studies among detained youth examined the association between psychiatric disorders and future

self-reported or officially registered offending (Colins et al., 2011, 2012; Colins & Vermeiren, 2013;

Elkington et al., 2015; Harrington et al., 2005; Hoeve, McReynolds, & Wasserman, 2013; Hoeve,

McReynolds, Wasserman, & McMillan, 2013; Plattner et al., 2009; Wierson & Forehand, 1995).

Taken together, these studies y ielded a mixed and inconsistent pattern of findings in terms of

specific psychiatric disorders (e.g., conduct disorder [CD]) and disorder categories (e.g., substance

use disorders [SUDs]) that predicted future offending. These mixed and inconsistent findings might

be explained by methodological differences across studies, including the use of different samples,

the focus on different disorders, and the use of different operationalization and measures of disorders

and recidivism.

Crucially, the overwhelming majority of prospective studies among detained youth only recruited

males (Colins et al., 2011, 2012; Colins & Vermeiren, 2013; Harrington et al., 2005; Wierson &

Forehand, 1995) or merely recruited a s mall number (Ns¼56–138) of detained girls (H oeve,

McReynolds, & Wasserman, 2013; Hoeve, McReynolds, Wasserman, et al., 2013; Plattner et al.,

2009). Also, most studies did not include indices of future self-reported or officially registered

violent offenses (e.g., violent arrests) as an outcome but instead used othe r, broader indices of

offending, such as arrests for any type of offending and repeated incarcerations (Hoeve, McRey-

nolds, & Wasserman, 2013; Hoeve, McReynolds, Wasserman, et al., 2013; Plattner et al., 2009;

Wierson & Forehand, 1995). Contemporary research, therefore, lacked the possibility to thoroughly

scrutinize if psychiatric disorders in detained girls increase the risk for future violent offending, with

one notable exception (Elkington et al., 2015). This study showed that clusters of psychiatric

disorders, such as disruptive behavior disorders (DBDs), were generally not predictive of self-

reported future violence in detained boys (N¼945) and girls (N¼542). Yet other drug use disorders

were positively related to future self-reported violent offenses in boys, whereas alcohol and mar-

ijuana use disorders were positively related to future self-reported violent offenses in girls (Elk-

ington et al., 2015). This latter finding converges with prior work showing that female (vs. male)

offenders’ drug use was linked to more risky lifestyles (McCuish, 2017) and underscores the

importance of testing the disorder–recidivism link separately across gender.

Regardless of its many strengths, the study by Elkington et al. (2015) is hallmarked by some

notable limitations. First, various specific disorders were collapsed into clusters. Consequently, it is

unclear if specific psychiatric disorders that are particularly salient when working with detained girls

(e.g., post-traumatic stress disorder [PTSD]) and have been linked with violence (Kerig, 2019) are

useful markers of later violent offending. Second, review studies suggest that subtyping schemes of

CD that are based on age of onset and/or on the lack of prosocial emotions are relevant for prognostic

purposes, including the identification of youth who are at a heightened risk for future violence (Frick

et al., 2014; Moffitt et al., 2008). Unfortunately, no study to date considered the utility of these CD

subtyping schemes for predicting future violent offending in detained girls.

Clearly, the prospective link between psychiatric disorders and violent offending in detained girls

is still ill-researched. This is unfortunate for various reasons. First, violence is considered the most

severe type of offending and has the greatest impact on the physical integrity of the victim and public

safety (e.g., Cohen, 1986). Therefore, it is of indisputable relevance to focus on violent offending

when testing the prognostic usefulness of psychiatric disorders. Second, there is some evidence to

suggest that treatment for psychiatric disorders in criminal justice–involved adolescents reduces

recidivism (Evans Cuellar et al., 2006). As such, it bears relevance to start exploring if there is at

least also a link between psychiatric disorders and later violence. Third, mounting evidence, nev-

ertheless, also suggests that mental health problems and psychiatric disorders, overall, are not

predictive of future violent offend ing among criminal justice–involve d boys (Colins & Grisso,

2019; Colins et al., 2011; Colins et al., 2013; Elkington et al., 2015). If this is the case among

detained girls, then the presence of psychiatric disorders should not be used as reason to prolong

366 Youth Violence and Juvenile Justice 18(4)