Research Design

Article

PTSD Symptoms Mediate the Relationship Between Sexual Abuse and Substance Use Risk in Juvenile Justice–Involved Youth

Jasmyn Sanders 1 , Alexandra R. Hershberger

2 , Haley M. Kolp

3 , Miji Um

2 ,

Matthew Aalsma 4 , and Melissa A. Cyders

2

Abstract Juvenile justice–involved youth face disproportionate rates of sexual abuse, which increases the risk of post-traumatic stress disorder (PTSD) and substance use disorders (SUDs), both of which are associated with poor long-term outcomes. The present study tested two mediation and moderation models, controlling for age, race, and history of physical abuse, with gender as a moderator, to determine whether PTSD symptoms serve as a risk factor and/or mechanism in the relationship between sexual abuse and substance use. Data were examined for 197 juvenile justice–involved youth (mean age ¼ 15.45, 68.9% non-White, 78.4% male) that completed court-ordered psychological assessments. Results indicated that PTSD symptoms significantly mediated the relationship between sexual abuse and drug (b ¼ 3.44, confidence interval [CI] [0.26, 7.41]; test for indirect effect z ¼ 2.41, p ¼ .02) and alcohol use (b ¼ 1.42, CI [0.20, 3.46]; test for indirect effect z ¼ 2.23, p ¼ .03). PTSD symptoms and gender were not significant moderators. Overall, PTSD symptoms mediate the relationship between sexual abuse and SUDs in juvenile justice–involved youth, which suggests viability of targeting PTSD symptoms as a modifiable risk factor to reduce the effects of sexual abuse on substance use in this high-risk population.

Keywords sexual abuse, substance use, PTSD, youth, juvenile justice

Substance use disorders (SUDs) occur in approximately 60% of juvenile justice–involved youth (Substance Abuse and Mental

Health Services Administration, 2016; Teplin et al., 2005).

This is particularly problematic, as juvenile justice–involved

youth with SUDs face a host of negative outcomes, some of

which include increased likelihood of having a co-occurring

severe mental illness (e.g., manic episode and psychosis;

Teplin, Abram, McClelland, Dulcan, & Mericle, 2002),

increased likelihood of recidivism (Conrad, Tolou-Shams,

Rizzo, Placella, & Brown, 2014), and increased likelihood of

engagement in sexual risk-taking behaviors, compared to youth

in the general population (Teplin et al., 2005). Although there

are multiple potential explanations for the high prevalence of

SUDs in this population, such as genetic risk or social norms in

line with substance use (Kendler, Prescott, Myers, & Neale,

2003), emerging research suggests sexual abuse victimization

may be one risk factor for the development of SUDs in juvenile

justice–involved youth.

The prevalence of sexual abuse victimization in juvenile

justice–involved youth is high, with 31% of girls and 15% of boys (Baglivio et al., 2014; Dierkhising et al., 2013) in the

juvenile justice system reporting a history of sexual abuse.

Extensive research on adolescent and adult populations demon-

strate that a history of sexual abuse is associated with a host of

negative outcomes (Finkelhor, Cross, & Cantor, 2005; Mullers

& Dowling, 2008) including increased engagement in risky

sexual behaviors (Ruffolo, Sarri, & Goodkind, 2004; Saar,

Epstein, Rosenthal, & Vafa, 2015; Smith, Leve, & Chamber-

lain, 2006; Townsend, 2013), a greater vulnerability to revicti-

mization (Townsend, 2013), and increased rates of delinquency

and criminal behavior (Asscher, Van der Put, & Stams, 2015;

Baglivio et al., 2014; Townsend, 2013; Widom & Maxfield,

2001).

Importantly, sexual abuse is related to problematic sub-

stance use (Townsend, 2013). Adolescents with a history of

sexual abuse are 4 times more likely to have an SUD and nearly

1 Department of Psychology, University of Miami, Coral Gables, FL, USA

2 Department of Psychology, Indiana University—Purdue University,

Indianapolis, IN, USA 3 Department of Psychology, Ohio University, Athens, OH, USA

4 Section of Adolescent Medicine, Department of Pediatrics, Indiana University

School of Medicine, Indianapolis, IN, USA

Corresponding Author:

Alexandra R. Hershberger, Department of Psychology, Indiana University—

Purdue University, 402 North Blackford Street, Indianapolis, IN 46202, USA.

Email: [email protected]

Child Maltreatment 2018, Vol. 23(3) 226-233 ª The Author(s) 2017 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1077559517745154 journals.sagepub.com/home/cmx

3 times more likely to report problematic substance use in

adulthood in comparison to members of the general population

(Townsend, 2013). In addition, victims of sexual abuse begin

experimenting with drugs at a younger age (13.7 years old)

compared to those adolescents who are not victims of sexual

abuse (15.1 years old; Kingston & Raghavan, 2009; Townsend,

2013), and this earlier onset of substance use is associated with

a higher likelihood of developing an SUD (Moss, Chen, & Yi,

2014). Furthermore, substance use among sexual abuse victims

increases the likelihood of engaging in criminal activity

(Widom & White, 1997) and further exacerbates already

high recidivism rates among juvenile justice–involved youth

(Conrad et al., 2014).

One way in which sexual abuse may be related to substance

use in juvenile justice–involved youth is through the experience

of post-traumatic stress disorder (PTSD) symptoms. It is possible

that PTSD symptoms strengthen the relationship between sexual

abuse and substance use, as the experience of PTSD symptoms

could lead to substance use via self-medication (e.g., Wolitzky-

Taylor, Bobova, Zinbarg, Mineka, & Craske, 2012) in sexually

abused juvenile justice–involved youth. Thus, PTSD symptoms

in sexually abused juvenile justice–involved youth could serve

as a risk indicator for subsequent substance use. Additionally,

PTSD symptoms could have a mechanistic role between sexual

abuse and substance use in juvenile justice–involved youth, by

which decreasing PTSD symptoms could serve to mitigate sub-

stance use risk associated with sexual abuse. Regardless if PTSD

symptoms serve as a risk factor or mechanism in the relationship

between sexual abuse and substance use in juvenile justice–

involved youth, PTSD symptoms are a modifiable factor that

could be targeted through treatment, potentially decreasing the

relationship between sexual abuse and substance use and asso-

ciated negative outcomes.

Importantly, there is a burgeoning area of research that spe-

cifically aims to reduce PTSD symptoms in juvenile justice–

involved youth. Treatments, such as trauma affect regulation:

guide for education and therapy (TARGET; Ford, Chang,

Levine, & Zhang, 2013) and cognitive processing therapy

(CPT; Matulis, Resick, Rosner, & Steil, 2014), show initial

promise in reducing PTSD symptoms in juvenile justice–

involved youth, including those who have experienced sexual

abuse. An unexplored benefit of such treatments, however, is if

reductions in PTSD symptoms serve to mitigate substance use

risk associated with sexual abuse in juvenile justice–involved

youth. Determining whether PTSD symptoms serve to

strengthen or mediate the relationship between sexual abuse

and substance use in juvenile justice–involved youth is an

important step in clarifying treatment models for sexual abuse

victims in this high-risk population.

The goal of the current study was to test two alternative

models to better understand the way in which PTSD symptoms

might influence the relationship between sexual abuse and sub-

stance use in juvenile justice–involved youth. Specifically, we

examined the following research questions: (1) Is the relation-

ship between sexual abuse and substance use stronger in the

presence of PTSD symptoms (i.e., moderation) and (2) Is

sexual abuse related to substance use through PTSD symptoms

(i.e., mediation). Additionally, we examined each research

question with gender as a moderator. Differing base rates of

study variables by gender, such as sexual abuse (Baglivio et al.,

2014; Dierkhising et al., 2013), and varying mechanisms

thought to underlie posttraumatic stress responses by gender

(Norr, Albanese, Boffa, Short, & Schmidt, 2016), for example,

could result in varying relationships between sexual abuse,

PTSD symptoms, and substance use by gender. We also con-

trolled for history of physical abuse, as polyvictimization, par-

ticularly through multiple forms of abuse, is related to

psychological disorders and problems (e.g., substance use,

PTSD symptoms; Ford, Elhai, Connor, & Frueh, 2010).

This study is the first step in a program of research that seeks

to examine PTSD as a modifiable risk factor to reduce the

effects of sexual abuse on substance use in this high-risk pop-

ulation. We chose to examine PTSD as a moderator and med-

iator in the current sample, as opposed to alternative causal

models, given data that (1) sexual abuse typically temporally

precedes PTSD development (although PTSD does increase the

risk of revictimization) and (2) substance use is often a symp-

tom or result of PTSD.

Method

Participants

Data were taken from existing charts of 247 juvenile justice–

involved youth in a Midwestern city, who were court ordered to

complete a psychological assessment between 2009 and 2016.

Youth completed integrated assessments, and a subset of that

data are included in the current report. Data from the assess-

ments were deidentified, and analysis of this archival data was

approved by the institutional review board.

Materials

Demographics. Youth reported their age, gender (girl or boy), and race.

PTSD. First, PTSD symptoms were assessed through the Youth Self-Report (YSR; Achenbach & Rescorla, 2001). The YSR is

self-report assessment in which youth rate themselves on var-

ious behavioral and emotional problems. Response options

range from 0 (not true) to 2 (very often or often true), and

scores are converted to t scores. For the PTSD Scale, t scores

<65 are considered to fall in the “normal” range, and t scores

over 65 correspond with increases in symptom severity. The

YSR has shown good reliability (e.g., Ebesutani, Berstein,

Martinez, Chorpita, & Weisz, 2011). Further, the YSR has been

validated for use in samples of juvenile justice–involved youth

(Vreugdenhil, van den Brink, Ferdinand, Wouters, & Dorelei-

jers, 2006). PTSD symptoms were used as a variable in anal-

yses for the present study.

Second, PTSD diagnosis (distinct from the YSR PTSD

Scale score) was made by a licensed clinical psychologist con-

ducting or supervising each youth’s assessment. Diagnoses

Sanders et al. 227

were based on a structured clinical interview and YSR corre-

spondence with Diagnostic and Statistical Manual of Mental

Disorders, Version IV, Revised (DSM-IV-TR), (American Psy-

chiatric Association, 2000) or Diagnostic and Statistical Man-

ual of Mental Disorders, Version 5 (DSM-5), (American

Psychiatric Association, 2013). Due to differences in diagnos-

tic criteria between DSM versions and limited variability in

dichotomous diagnoses, our analyses focused on the YSR

PTSD score. PTSD diagnosis was used as a descriptive mea-

sure for the present study and not examined in study analyses.

Substance use. First, substance use (illicit drug use and alcohol use) was assessed through the Adolescent Substance Use

Subtle Screening Inventory (SASSI-A2; Miller & Lazowski,

2001). The SASSI-A2 is a self-report questionnaire for which

youth self-report various substance use–related behaviors. The

present study examined substance use using the Face Valid

Other Drugs (FVOD; e.g., “Taken drugs to improve your think-

ing and feeling,” “Taken drugs so you could enjoy sex more”;

Miller, Renn, & Lazowski, 2001) Scale and the Face Valid

Alcohol (FVA; e.g., “Tried to kill yourself while drunk,”

“Drank alcohol during the day”; Miller et al., 2001) Scale of

the SASSI-A2, with each scale assessing substance use–related

problems on a scale from 0 (never) to 3 (repeatedly). Results

are provided as t scores based on norms derived from an ado-

lescent sample (mean age ¼ 15, standard deviation [SD] ¼ 1.9) across addiction treatment centers, inpatient psychiatric hospi-

tals, outpatient behavioral health facilities, and juvenile correc-

tions programs. The SASSI-A2 FVA and FVOD Scales have

demonstrated acceptable to excellent reliability (a ¼ .61 and .95, respectively; Perera-Diltz & Perry, 2011) and test–retest

reliability (rs ¼ .71 and .92, respectively; Miller & Lazowski, 2001; Stein et al., 2005).

Second, an SUD diagnosis (distinct from the SASSI-A2

Scales) was made by a licensed clinical psychologist conduct-

ing or supervising each youth’s assessment. Diagnoses were

based on a structured clinical interview and SASSI-A2 corre-

spondence with DSM-IV-TR (American Psychiatric Associa-

tion, 2000) or DSM-5 (American Psychiatric Association,

2013). Due to differences in diagnostic criteria between DSM

versions and limited variability in dichotomous diagnoses, our

analyses focused on the SASSI-A2 scores. An SUD diagnosis

was used as a descriptive measure for the present study and not

examined in study analyses.

Sexual abuse. Youth self-reported their history of sexual abuse through a structured clinical interview. Sexual abuse was coded

as either reporting or not reporting sexual abuse. Additionally,

the relationship of alleged perpetrator to the victim of sexual

abuse was recorded (e.g., parent, paramour, relative).

Physical abuse. Youth self-reported their history of physical abuse through a structured clinical interview. Physical abuse

was coded as either reporting or not reporting physical abuse

and used as a covariate in analyses.

Procedure

Youth in the present sample were involved with the juvenile

court (e.g., through arrest, probation violation) and referred to

complete a court-ordered psychological assessment. Following

the referral, a licensed clinical psychologist or supervised doc-

toral student reported to the Juvenile Detention Center or the

youth’s current placement (e.g., group home, family home) to

complete the assessment. Clinicians conducted a structured

clinical interview and administered assessment measures,

including the YSR, SASSI-A2, and other measures unrelated

to the present study and not reported elsewhere. Clinicians used

assessment information to compile an integrated report for each

youth, which was submitted to the Marion County Juvenile

Court upon completion. Trained research assistants entered

YSR and SASSI-A2 data taken directly from the assessment

reports for each youth. Additionally, research assistants coded

whether or not the youth reported ever experiencing sexual

abuse (any form of illegal sex act conducted against the youth)

or physical abuse (any form of illegal physical act conducted

against the youth, not including physical altercations between

peers) based on the background information provided in the

assessment report. Twenty percent of the data were recoded

for interrater reliability. There were no discrepancies between

coders on study variables.

Analysis Plan

First, we examined sample characteristics stratified across sex-

ual abuse and gender. Second, we conducted moderated regres-

sion analyses using Hayes’s (2013) process macro, controlling

for age, race, and history of physical abuse (0 ¼ no abuse, 1 ¼ abuse), with sexual abuse (0 ¼ no abuse, 1 ¼ abuse) entered as the independent variable, PTSD symptoms (YSR PTSD Scale)

entered as the moderator, and gender (0 ¼ boys, 1 ¼ girls) entered as a second moderator (three-way interaction). Two

analyses were conducted with drug use and alcohol use as

dependent variables in separate models. Third, we conducted

moderated mediation analyses using Hayes’s (2013) process

macro, controlling for age, race, and history of physical abuse,

with sexual abuse entered as the independent variable, PTSD

symptoms entered as the mediator, and gender entered as a

moderator of the relationship between the independent and

dependent variable. Two analyses were conducted with drug

use and alcohol use as dependent variables in separate models.

Results

Descriptive Statistics

Of the 247 youth completing court-ordered psychological

assessments, 197 provided data for PTSD Scales, drug use, and

alcohol use, making the final sample N ¼ 197 (mean age ¼ 15.45, SD ¼ 1.31, range ¼ 13–18; 61.6% Black, 21.6% White, 3.2% Hispanic, 4.1% multiracial; 9.5% not specified; 78.4% male). Those youth who were not administered measures of

PTSD, drug use, or alcohol use did not differ significantly in

228 Child Maltreatment 23(3)

age or gender from youth who completed these measures.

Those completing the drug and alcohol use scales were more

likely to have a cannabis use disorder diagnosis compared to

those not completing (w2 ¼ 7.29, p ¼ .007). The majority of the sample was non-White (68.9%) and male (79.5%). The major- ity of assessments were conducted at the juvenile detention

center (82.1%) with youths having an average of six (SD ¼ 3.80) criminal referrals to the court. Of the 197 youth, 17.3% were diagnosed with PTSD, and 44.7% were diagnosed with a cannabis use disorder, with a small proportion meeting criteria

for alcohol use disorder (3%) and stimulant use disorder (1.5%).

Average drug use score was 56.32 (SD ¼ 12.67, range ¼ 6– 96), average alcohol use score was 45.75 (SD ¼ 7.10, range ¼ 39–95), and average PTSD score was 60.49 (SD ¼ 9.99, range ¼ 50–95). Alcohol use, drug use, and PTSD Scales were all significantly correlated (rs ¼ .31–.52, ps < .05; see Table 1).

A total of 13.2% of the sample reported any history of sexual abuse (7.8% of boys and 33.3% of girls in the sample) and 16% of the sample reported any history of physical abuse. A total of

2.9% of youth reported a history of both physical and sexual abuse. Girls were more likely than boys to report a history of

sexual abuse (w2 ¼ 17.73, p < .001) and physical abuse (w2 ¼ 5.97, p ¼ .02), and boys and girls were similarly likely to report a history of both physical and sexual abuse (w2 ¼ 2.57, p ¼ .11; see Table 2). Youth who reported a history of sexual abuse

were more likely to be diagnosed with PTSD compared to those

not reporting a history of sexual abuse (w2 ¼ 9.43, p < .002). Youth reporting only physical abuse (and not sexual abuse)

compared to youth reporting sexual abuse (and not physical

abuse) did not differ significantly in rate of PTSD diagnosis

(w2 ¼ 1.99, p ¼ .16), but youth reporting sexual abuse (and not physical abuse) scored higher on the PTSD Scale, t(41) ¼ �1.89, falling short of significance (p ¼ .07).

Table 1. Correlation Between Study Variables.

1 2 3 4 5 6 7 8 9

1. Age �.06 �.06 .09 .06 .0 .009 �.02 �.06 2. PTSD score — .18* .31** .10 .41** .06 �.06 .006 3. PTSD diagnosis — — �.03 �.08 .02 �.02 �.02 .09 4. FVA score — — .53** .52** �.007 �.04 .03 5. Alcohol use disorder diagnosis — — — — .27** �.002 �.02 �.001 6. FVOD score — — — — — .29** �.04 .06 7. Cannabis or other illicit drug use disorder diagnosis — — — — — — �.07 �.18* 8. Sexual abuse — — — — — — — �.03 9. Physical abuse

Note. PTSD ¼ post-traumatic stress disorder; FVA ¼ Face Valid Alcohol; FVOD ¼ Face Valid Other Drugs. *p < .05. **p < .001.

Table 2. Sample Characteristics by Sexual Abuse and Gender.

Boys (n ¼ 153) Girls (n ¼ 42) Test Statistic p Sexual Abuse No Sexual Abuse Test Statistic p

PTSD Scale 60.50 (10.30) 60.45 (8.88) 0.03 .98 65.27 (12.69) 59.77 (9.35) �2.66 .009 FVA (alcohol use) 45.06 (5.90) 48.29 (10.10) �2.65 .009 47.77 (7.83) 45.44 (6.96) �1.57 .12 FVOD (drug use) 56.60 (12.89) 55.31 (11.91) 0.59 .56 56.00 (11.56) 56.37 (12.86) 0.14 .89 PTSD diagnosis 13.5% 31% 7.01 .008 38.46% 14.03% 9.43 .002 Sexual abuse only 5.84% 26.32% 17.73 <.001 Physical abuse only 10.95% 21.05% 5.97 .02 Sexual and physical abuse 2.19% 5.26% 2.57 .11 Cannabis use disorder 46.45% 38.09% 0.93 .33 38% 45.6% 0.47 .49 Stimulant use disorder 1.3% 2.4% 0.26 .61 0 3.5% 0.46 .50 Alcohol use disorder 3.2% 2.4% 0.08 .78 0 3.5% 0.94 .33

Sexual Abuse Perpetrator Boys (n ¼ 14) Girls (n ¼ 17) Parent or paramour 21.4% 27.8% 0.17 .68 Adult relative 14.3% 5.6% 0.71 .40 Known adult nonrelative 28.6% 22.2% 0.17 .68 Adult stranger 7.1% 5.6% 0.03 .85 Child relative 28.6% 5.6% 3.16 .08 Known child nonrelative 14.3% 22.2% 0.33 .57

Note. Ranges for scale scores were as follows: PTSD YSR scale score ¼ 0–95; FVA ¼ 0–95; FVOD ¼ 6–96. Boys reports of sexual abuse n ¼ 14, girls reports of sexual abuse n ¼ 17. PTSD ¼ post-traumatic stress disorder; FVA ¼ Face Valid Alcohol; FVOD ¼ Face Valid Other Drugs; YSR ¼ Youth Self-Report. Significant p-values at the p<.05 level are bolded.

Sanders et al. 229

Moderation analyses: Is the relationship between sexual abuse and substance use stronger in the presence of PTSD symptoms and variable by gender?

Drug use. First, sexual abuse was not significantly related to drug use (b ¼ 10.69, p ¼ .60). Second, PTSD symptoms were significantly related to drug use (b ¼ 0.66, p < .001). Third, PTSD symptoms did not significantly moderate the relation-

ship between sexual abuse and drug use (b ¼�0.27, p ¼ .35). Fourth, gender did not significantly moderate the overall model

for drug use (b ¼ 0.64, p ¼ .24). Alcohol use. First, sexual abuse was not significantly related

to alcohol use (b ¼ 16.08, p ¼ .18). Second, PTSD symptoms were significantly related to alcohol use (b ¼ 0.23, p < .001). Third, PTSD symptoms did not significantly moderate the rela-

tionship between sexual abuse and alcohol use (b¼�0.22, p ¼ .17). Fourth, gender did not significantly moderate the overall

model for alcohol use (b ¼ 0.40, p ¼ .20).

Moderation mediation analyses: Is sexual abuse related to substance use through PTSD symptoms and does this vary by gender?

Drug use. First, sexual abuse was significantly related to PTSD symptoms (b ¼ 5.86, p ¼ .009). Second, PTSD symp- toms were significantly related to drug use (b¼ 0.59, p < .001). Third, sexual abuse was negatively related to drug use (b ¼ �7.52, p ¼ .05). Fourth, the relationship between sexual abuse and drug use was not moderated by gender (b ¼ 9.02, p ¼ .11).

Fifth, PTSD significantly mediated the relationship between

sexual abuse and drug use (b ¼ 3.44, 95% confidence interval [CI] ¼ [0.26, 7.41]; test of indirect effect z ¼ 2.41, p ¼ .02; see Figure 1).

Alcohol use. First, sexual abuse was significantly related to PTSD symptoms (b ¼ 5.86, p ¼ .009). Second, PTSD symp- toms were significantly related to alcohol use (b ¼ 0.24, p < .001). Third, sexual abuse was not significantly related to alco-

hol use (b ¼�0.57, p ¼ .80). Fourth, the relationship between sexual abuse and alcohol use was not moderated by gender

(b ¼ 0.68, p ¼ .80). Fifth, PTSD significantly mediated the relationship between sexual abuse and alcohol use (b ¼ 1.42, 95% CI [0.20, 3.46]; test of indirect effect z ¼ 2.23, p ¼ .03; see Figure 1).

Discussion

Juvenile justice–involved youth face a disproportionate amount

of negative outcomes compared to their same-age peers (e.g.,

Hershberger , Zapolski, & Aalsma, 2016; Fazel, Doll, & Lang-

strom, 2008) and exhibit high rates of sexual abuse, PTSD, and

substance use. Despite their high-risk nature, it is unclear how

best to reduce negative outcomes associated with sexual abuse

in juvenile justice–involved youth. There are viable treatments

targeting trauma, including sexual abuse, in this population

(e.g., Ford et al., 2013; Matulis et al., 2014), which reduce

PTSD symptoms; however, this study is unique in that it sug-

gests targeting PTSD symptoms as a means of reducing sexual

abuse–related substance use risk in juvenile justice–involved

youth. These findings provide support that not only may sexual

abuse serve as a risk indicator for PTSD and substance use in

juvenile justice–involved youth, but that PTSD may be a prime

modifiable risk factor to reduce the effects of sexual abuse on

subsequent substance use in this high-risk population. Once a

child experiences sexual abuse, it is no longer a preventable or

modifiable risk factor and intervening on substance use alone

underestimates the role PTSD symptomatology plays in the

onset and maintenance of substance use behaviors. PTSD

symptoms appear to mediate, rather than moderate, the rela-

tionship between sexual abuse and substance use in juvenile

justice–involved youth, which suggests that reducing PTSD

symptoms could protect against and reduce substance use in

this population. This is important, given the negative outcomes

associated with substance use in this high-risk group.

One likely explanation for the relationship between sexual

abuse and substance use through PTSD observed in the present

study is the self-medication hypothesis (Khantzian, 1987). The

self-medication hypothesis (Khantzian, 1987) indicates that

certain drugs are chosen deliberately by individuals who wish

to suppress or avoid their negative experiences and emotions.

Juvenile justice–involved youth often display high rates of

avoidant PTSD symptoms (Kerig & Becker, 2010), and sub-

stance use is often cited as one coping strategy used by sexually

victimized youth (Kilpatrick et al., 2003). Thus, juvenile jus-

tice–involved youth who are victims of sexual abuse may use

Figure 1. Results of mediation models run using Andrew Hayes process macro (Hayes, 2013), controlling for age, gender, and race. Top: Results indicated that the indirect effect of sexual abuse on drug use through post-traumatic stress disorder (PTSD) scores was sig- nificant (b ¼ 3.44, 95% confidence interval [CI] ¼ [0.26, 7.41]; test of indirect effect z ¼ 2.41, p ¼ .02). Bottom: Results indicated that the indirect effect of sexual abuse on alcohol use through PTSD scores was significant (b¼ 1.42, 95% CI [0.20, 3.46]; test of indirect effect z ¼ 2.23, p ¼ .03).

230 Child Maltreatment 23(3)

substances as a means to cope with or alleviate emotional dis-

tress. Interventions designed to provide more adaptive ways to

cope with or alleviate emotional distress have the potential to

reduce the reliance on substance use in this population.

Although previous research indicates mixed findings for the

relationship between PTSD and substance use in juvenile jus-

tice–involved youth (Abram et al., 2004; Danielson et al.,

2010; Dierkhising et al., 2013; Kingston & Raghavan, 2009;

Rosenberg et al., 2014), present findings support PTSD as a

proximal factor to substance use among youth who are victims

of sexual abuse.

Given that one way by which sexual abuse is related to

greater substance use is through the presence of greater PTSD

symptomology, it appears trauma-informed treatment could be

critical to reducing substance use in juvenile justice–involved

youth with sexual abuse. One well-studied model of trauma-

focused treatment for substance use is seeking safety (Najavits,

2002), and although this evidence-based treatment has not been

well-studied in juvenile justice–involved youth, it certainly

provides one framework for addressing issues highlighted in

the present study. Such therapies are easily implemented (Naja-

vits, Gallop, & Weiss, 2006) and provide psychoeducation on

the complex interplay between trauma and substance use. Risk

reduction through family therapy (Danielson et al., 2010;

Danielson et al., 2012) also demonstrates efficacy in reducing

substance use and PTSD symptoms in adolescent victims of

sexual assault, and such treatments could be expanded to exam-

ine efficacy in juvenile justice populations. Additionally, there

are trauma-related treatments for juvenile justice–involved

youth, such as TARGET (Ford et al., 2013) and CPT (Matulis

et al., 2014), which show initial evidence of reducing PTSD

symptomology in this population and thus could serve to miti-

gate sexual abuse–related substance use risk in this population.

It is imperative that trauma treatment research in juvenile jus-

tice–involved youth tracks not only changes in PTSD symp-

toms but also changes in substance use and related problems, to

best examine the mechanistic role of trauma-informed treat-

ment in reducing sexual abuse–related substance use risk.

It is important to note that the present findings are cross-

sectional; thus, there are other plausible reasons for the

observed relationships. For example, genetic factors may put

individuals at high risk for both PTSD and substance use

(Kendler et al., 2003). Although cross-sectional data are lim-

ited in making causal inferences, we view the current study as

the first step in a program of research examining these temporal

relationships in prospective designs and designing and testing

interventions to reduce substance use and risk in this high risk,

though relatively understudied, population. It is our aim that

this initial study will stimulate additional and much-needed

research and intervention design and testing in this group.

Additionally interesting, gender did not significantly

moderate the models examined in the present study. This

may indicate that there is no differential risk by gender in

juvenile justice–involved youth for the impact PTSD symp-

toms have on sexual abuse–related substance use or that the

relationship between sexual abuse and substance use does

not vary by gender in this population. First, although

research suggests varying mechanisms, such as anxiety sen-

sitivity, underlie posttraumatic responses by gender (Norr

et al., 2016), these gender differences may not extend to

reflect variable outcomes by gender, specifically sexual

abuse–related substance use.

Additionally, research indicates that juvenile justice–

involved girls experience sexual abuse at rates more than dou-

ble of boys (Baglivio et al., 2014; Dierkhising et al., 2013), but

it is possible that negative outcomes, particularly substance

use, associated with sexual abuse, cut across gender. It is also

possible that we were underpowered to detect an effect (N ¼ 42 girls) and there are gender differences in the impact that PTSD

symptoms have on sexual abuse–related substance use and the

direct relationship between sexual abuse and substance use.

Future research should aim to clarify the role of gender in

negative outcomes associated with sexual abuse in order to best

tailor treatment.

Although the present study is important in that it suggests

viability of a novel intervention approach to reduce substance

use in high-risk, juvenile justice–involved youth, there are

some limitations to discuss. First, as common in many studies,

data were self-report in nature and thus subject to self-report

bias. Additionally, the sample was limited in the number of

youth reporting sexual abuse; thus, null findings might reflect

failure to find a true effect. Present data are cross-sectional, and

no causal pathways can be determined from our findings,

although they can suggest viability of examining these relation-

ships in temporal models in the future. For the present study,

we were given access to juvenile justice–involved youth’s

assessment reports, which did not include individual item

scores on study measures, thus we could not compute reliability

for the measures; however, the YSR and SASSI-A2 demon-

strate good psychometric properties across studies (e, g. Miller

& Lazowski, 2001; Stein et al., 2005; Vreugdenhil et al., 2006),

increasing our confidence in these measures in the present

study. The present sample had a limited number of females

and non-Whites, thus results may not generalize beyond the

demographic makeup of the present sample, although this

makeup does reflect typical juvenile justice populations (Child

Trends Data Bank, 2015; Teplin et al., 2002). Last, although we

controlled for history of physical abuse, there are other traumas

juvenile justice–involved youth often experience, such as emo-

tional abuse, which may serve to amplify their trauma-related

risk of substance use. The present findings cannot determine

whether sexual abuse–related risk of substance use through

PTSD symptomology can be better accounted for by polyvic-

timization beyond sexual abuse and physical abuse; however,

as we detected the present findings after controlling for the

additional presence of physical abuse, this increases our confi-

dence that particular attention should be paid to sexual abuse–

related substance use. In light of these limitations, we hope our

findings will move the research field of juvenile justice–

involved youth forward, particularly in measuring and examin-

ing PTSD symptom change as a means of mitigating sexual

abuse–related substance use risk.

Sanders et al. 231

Conclusion

Juvenile justice–involved youth face disproportionate rates of

sexual abuse, which increases the risk of PTSD and SUDs, both

of which are associated with negative long-term outcomes.

However, how best to reduce the effects of sexual abuse in this

group is not yet clear. Overall, findings suggest viability that

targeting PTSD symptoms in juvenile justice–involved youth

has potential as a prime modifiable treatment target to reduce

the effects of sexual abuse on substance use in this high-risk

population. We view the current study as providing key initial

evidence, which should be replicated in temporal work, includ-

ing the design and testing of PTSD interventions to reduce risk

in this high-risk population.

Acknowledgments

The authors would like to acknowledge Katie Schwartz for her coor-

dination of the forensic psychology clinic at the Indiana University

School of Medicine.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to

the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for

the research, authorship, and/or publication of this article: The pre-

paration of this article was supported in part by an F31 grant to Alex-

andra Hershberger (F31 AA024682) and fellowship to Jasmyn N.

Sanders (R25 GM109432) both under the mentorship of Melissa A

Cyders.

References

Abram, K. M., Teplin, L. A., Charles, D. R., Longworth, S. L.,

McClelland, G. M., & Dulcan, M. K. (2004). Posttraumatic stress

disorder and trauma in youth in juvenile detention. Archives of

General Psychiatry, 61, 403–410. doi:10.1001/archpsyc.61.4.403

Achenbach, T. M., & Rescorla, L. (2001). ASEBA school-age forms

& profiles. Burlington, VT: University of Vermont, Research Cen-

ter for Children, Youth, & Families.

American Psychiatric Association. (2000). Diagnostic and statistical

manual of mental disorders: DSM-IV-TR. Washington, DC: Amer-

ican Psychiatric Association.

American Psychiatric Association. (2013). Diagnostic and statistical

manual of mental disorders : DSM-5(5th ed.). Arlington, VA:

American Psychiatric Association.

Asscher, J. J., Van der Put, C. E., & Stams, G. J. J. M. (2015). Gender

differences in the impact of abuse and neglect victimization on

adolescent offending behavior. Journal of Family Violence, 30,

215–225. doi:10.1007/s10896-014-9668-4

Baglivio, M. T., Epps, N., Swartz, K., Huq, M. S., Sheer, A., & Hardt,

N. S. (2014). The prevalence of adverse childhood experiences

(ACE) in the lives of juvenile offenders. Office of Juvenile Justice

and Delinquency Prevention Journal of Juvenile Justice, 3, 1–14.

Retrieved from http://www.journalofjuvjustice.org/JOJJ0302/ arti

cle01.htm

Child Trends Data Bank. (2015). Indicators on children and youth.

Juvenile Detention, 1–17. Retrieved from http://www.childtrends.

org/wp-content/uploads/2012/05/88_Juvenile_Detention.pdf

Conrad, S. M., Tolou-Shams, M., Rizzo, C., Placella, N., & Brown, L.

K. (2014). Gender differences in recidivism rates for juvenile jus-

tice youth: The impact of sexual abuse. Law and Human Behavior,

38, 305–314. doi:10.1037/lhb0000062

Danielson, C., Macdonald, A., Amstadter, A. B., Hanson, R., de Are-

llano, M. A., Saunders, B. E., . . . Kilpatrick, D. G. (2010). Risky

behaviors and depression in conjunction with—or in the absence

of—lifetime history of PTSD among sexually abused adolescents.

Child Maltreatment, 15, 101–107.

Danielson, C., McCart, M. R., de Arellano, M. A., Macdonald, A.,

Doherty, L. S., & Resnick, H. S. (2012). Risk reduction for sub-

stance use and trauma-related psychopathology in adolescent sex-

ual assault victims: Findings from an open trial. Child

Maltreatment, 15, 261–268.

Dierkhising, C. B., Ko, S. J., Woods-Jaeger, B., Briggs, E. C., Lee, R.,

& Pynoos, R. S. (2013). Trauma histories among justice-involved

youth: Findings from the National Child Traumatic Stress Net-

work. European Journal of Psychotraumatology, 4, 1–12. doi:10.

3402/ejpt.v4i0.20274

Ebesutani, C., Bernstein, A., Martinez, J. I., Chorpita, B. F., & Weisz,

J. R. (2011). The youth self report: Applicability and validity

across younger and older youths. Journal of Clinical Child &

Adolescent Psychology, 40, 338–346.

Fazel, S., Doll, H., & Långström, N. (2008). Mental disorders among

adolescents in juvenile detention and correctional facilities: A sys-

tematic review and metaregression analysis of 25 surveys. Journal

of the American Academy of Child & Adolescent Psychiatry, 47,

1010–1019.

Finkelhor, D., Cross, T. P., & Cantor, E. N. (2005). How the justice

system responds to juvenile victims: A comprehensive model of

case flow. Trauma, Violence, & Abuse, 6, 83–102. doi:10.1177/

1524838005275090

Ford, J. D., Chang, R., Levine, J., & Zhang, W. (2013). Randomized

clinical trial comparing affect regulation and supportive group

therapies for victimization-related PTSD with incarcerated

women. Behavior Therapy, 44, 262–276. doi:10.1016/j.beth.

2012.10.003

Ford, J. D., Elhai, J. D., Connor, D. F., & Frueh, B. C. (2010).

Poly-victimization and risk of posttraumatic, depressive, and

substance use disorders and involvement in delinquency in a

national sample of adolescents. Journal of Adolescent Health,

46, 545–552.

Hayes, A. F. (2013). Introduction to mediation, moderation, and con-

ditional process analysis: A regression-based approach. New

York, NY: Guilford Press. Hershberger, A., Zapolski, T., &

Aalsma, M. C. (2016). Social support as a buffer between discrim-

ination and cigarette use in juvenile offenders. Addictive beha-

viors, 59, 7-11.

Hershberger, A., Zapolski, T., & Aalsma, M. C. (2016). Social support

as a buffer between discrimination and cigarette use in juvenile

offenders. Addictive behaviors, 59, 7–11.

Kendler, K. S., Prescott, C. A., Myers, J., & Neale, M. C. (2003). The

structure of genetic and environmental risk factors for common

232 Child Maltreatment 23(3)

psychiatric and substance use disorders in men and women.

Archives of General Psychiatry, 60, 929–937.

Kerig, P. K., & Becker, S. P. (2010). From internalizing to externaliz-

ing: Theoretical models of the processes linking PTSD to juvenile

delinquency. In S. J. Egan (Ed.), Posttraumatic stress disorder

(PTSD): Causes, symptoms and treatment (pp. 33–78). Haup-

pauge, NY: Nova Science.

Khantzian, E. J. (1987). The self-medication hypothesis of addictive

disorders: Focus on heroin and cocaine dependence. In: D. F. Allen

(eds) The cocaine crisis (pp. 65–74). Boston, MA: Springer.

Kilpatrick, D. G., Ruggiero, K. J., Acierno, R., Saunders, B. E.,

Resnick, H. S., & Best, C. L. (2003). Violence and risk of PTSD,

major depression, substance abuse/dependence, and comorbidity:

Results from the National Survey of Adolescents. Journal of Con-

sulting & Clinical Psychology, 71, 602–700. doi:10.1037/0022-

006X.71.4.692

Kingston, S., & Raghavan, C. (2009). The relationship of sexual

abuse, early initiation of substance use, and adolescent trauma to

PTSD. Journal of Traumatic Stress, 22, 65–68. doi:10.1002/jts.

20381

Matulis, S., Resick, P. A., Rosner, R., & Steil, R. (2014). Develop-

mentally adapted cognitive processing therapy for adolescents suf-

fering from posttraumatic stress disorder after childhood sexual or

physical abuse: A pilot study. Clinical Child and Family Psychol-

ogy Review, 17, 173–190. doi:10.1007/s10567-013-0156-9

Moss, H. B., Chen, C. M., & Yi, H. Y. (2014). Early adolescent

patterns of alcohol, cigarettes, and marijuana polysubstance use

and young adult substance use outcomes in a nationally represen-

tative sample. Drug and Alcohol Dependence, 136, 51–62.

Miller, F. G., & Lazowski, L. E. (2001). The Adolescent Substance

Abuse Subtle Screening Inventory-A2 (SASSI-A2) manual. Spring-

ville, IN: SASSI Institute.

Miller, F. G., Renn, W. R., & Lazowski, L. E. (2001). Adolescent

SASSI-A2 user’s guide. Bloomington, IN: Baugh Enterprises.

Mullers, E. S., & Dowling, M. (2008). Mental health consequences of

child sexual abuse. British Journal of Nursing, 17, 1428–1433. doi:

10.12968/bjon.2008.17.22.31871

Najavits, L. M. (2002). Seeking safety: A treatment manual for PTSD

and substance abuse. New York, NY: Guilford Press.

Najavits, L. M., Gallop, R. J., & Weiss, R. D. (2006). Seeking safety

therapy for adolescent girls with PTSD and substance use disorder:

A randomized controlled trial. The Journal of Behavioral Health

Services & Research, 33, 453–463.

Norr, A. M., Albanese, B. J., Boffa, J. W., Short, N. A., & Schmidt, N.

B. (2016). The relationship between gender and PTSD symptoms:

Anxiety sensitivity as a mechanism. Personality and Individual

Differences, 90, 210–213. doi:10.1016/j.paid.2015.11.014

Perera-Diltz, D. M., & Perry, J. C. (2011). Screening for adolescent

substance-related disorders using the SASSI-A2: Implications for

nonreporting youth. Journal of Addictions & Offender Counseling,

31, 66–79.

Rosenberg, H. J., Vance, J. E., Rosenberg, S. D., Wolford, G. L.,

Ashley, S. W., & Howard, M. L. (2014). Trauma exposure, psy-

chiatric disorders, and resiliency in juvenile-justice-involved

youth. Psychological Trauma: Theory, Research, Practice, and

Policy, 6, 430.

Ruffolo, M. C., Sarri, R., & Goodkind, S. (2004). Study of delinquent,

diverted, and high-risk adolescent girls: Implications for mental

health intervention. Social Work Research, 28, 237–245. doi:10.

1093/swr/28.4.237

Saar, M. S., Epstein, R., Rosenthal, L., & Vafa, Y. (2015). The sexual

abuse to prison pipeline: The girls’ story. Georgetown Law Center

on Poverty & Inequality, 1–43. Retrieved from http://rights4girls.

org/wp-content/uploads/r4g/2015/02/2015_COP_sexual-abuse_

layout_web-1.pdf

Smith, D. K., Leve, L. D., & Chamberlain, P. (2006). Adolescent girls’

offending and health-risking sexual behavior: The predictive role

of trauma. Child Maltreatment, 11, 346–353. doi:10.1177/

1077559506291950

Stein, L. A. R., Lebeau-Craven, R., Martin, R., Colby, S. M., Barnett,

N. P., Golembeske, C., . . . Penn, J. V. (2005). Use of the adoles-

cent SASSI in a juvenile correctional setting. Assessment, 12,

384–394. doi:10.1177/1073191105279433

Substance Abuse and Mental Health Services Administration. (2016).

Criminal and juvenile justice: Overview. Retrieved from https://

www.samhsa.gov/criminal-juvenile-justice

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.

Teplin, L. A., Elkington, K. S., McClelland, G. M., Abram, K. M.,

Mericle, A. A., & Washburn, J. J. (2005). Major mental disorders,

substance use disorders, comorbidity, and HIV-AIDS risk beha-

viors in juvenile detainees. Psychiatric Services, 56, 823–828.

Townsend, C. (2013). Prevalence and consequences of child sexual

abuse compared with other childhood experiences. Darkness to

Light, 1–19. Retrieved from www.d2l.org

Vreugdenhil, C., van den Brink, W., Ferdinand, R., Wouters, L., &

Doreleijers, T. (2006). The ability of YSR scales to predict DSM/

DISC–C psychiatric disorders among incarcerated male adoles-

cents. European Child & Adolescent Psychiatry, 15, 88–96.

Widom, C. S., & Maxfield, M. G. (2001). An update on the “cycle of

violence.” National Institute of Justice Research in Brief, 1–8.

Retrieved from https://www.ncjrs.gov/pdffiles1/nij/184894.pdf

Widom, C. S., & White, H. R. (1997). Problem behaviors in abused

and neglected children grown up: Prevalence and co-occurrence of

substance abuse, crime, and violence. Criminal Behavior & Mental

Health, 7, 287–310. doi:10.1002/cbm.191

Wolitzky-Taylor, K., Bobova, L., Zinbarg, R. E., Mineka, S., &

Craske, M. G. (2012). Longitudinal investigation of the impact

of anxiety and mood disorders in adolescence on subsequent sub-

stance use disorder onset and vice versa. Addictive Behaviors,

37(8), 982–985. doi:10.1016/j.addbeh.2012.03.026

Sanders et al. 233