Gender Differences in Emotion-Mediated Pathways from Childhood Sexual Abuse to Problem Drinking in Adolescents in the Child Welfare System
It is well-established that childhood sexual abuse (CSA) increases risk for later substance abuse. However, less is known about the mechanisms by which CSA increases risk of substance abuse, including problem drinking. Research points to negative emotions as potential explanatory links between CSA and problem drinking, and suggests that pathways may differ for men and women. In the current study, we examined three potential mechanisms (anger, anxiety, and depression symptoms) by which CSA increases vulnerability for alcohol problems in a sample of youth receiving child welfare services (N = 301). Path analyses revealed that relations between CSA and problem drinking were partially mediated by anger symptoms in male adolescents. In female adolescents, links between CSA and problem drinking were fully mediated by anxiety and anger symptoms. These findings highlight similarities and differences in how CSA increases vulnerability for alcohol problems in male and female adolescents. Clinical and research implications are discussed.
Gender Differences in Emotion-Mediated Pathways
from Childhood Sexual Abuse to Problem Drinking
in Adolescents in the Child Welfare System
&Abby L. Goldstein
&Sherry H. Stewart
#Springer International Publishing 2016
Abstract It is well-established that childhood sexual abuse
(CSA) increases risk for later substance abuse. However, less
is known about the mechanisms by which CSA increases risk
of substance abuse, including problem drinking. Research
points to negative emotions as potential explanatory links be-
tween CSA and problem drinking, and suggests that pathways
may differ for men and women. In the current study, we ex-
amined three potential mechanisms (anger, anxiety, and de-
pression symptoms) by which CSA increases vulnerability for
alcohol problems in a sample of youth receiving child welfare
services (N = 301). Path analyses revealed that relations
between CSA and problem drinking were partially mediated
by anger symptoms in male adolescents. In female adoles-
cents, links between CSA and problem drinking were fully
mediated by anxiety and anger symptoms. These findings
highlight similarities and differences in how CSA increases
vulnerability for alcohol problems in male and female adoles-
cents. Clinical and research implications are discussed.
Keywords Sexual abuse .Problem drinking .Anger .
Child maltreatment is a global rights, public health, mental
health, and premature morbidity and mortality issue. Among
maltreatment forms, child sexual abuse can be argued to be the
most taboo and the most stigmatizing for the victim(Finkelhor
et al. 2014). The World Health Organization broadly defines
childhood sexual abuse (CSA) as involving Bunwanted and
inappropriate sexual solicitation of, or exposure to, a child by
an older person; genital touching or fondling; or oral, anal or
vaginal intercourse or attempted intercourse^(Andrews et al.
2004). Prevalence rates of CSA have been estimated to be
between 8 and 31% for girls and 3 and 17% for boys (Barth
et al. 2013; MacMillan et al. 1997). The rate of CSA reported
in the Canadian Incidence Study was approximately 10% in
1998 (Trocmé et al. 2003). Although rates are consistently
reported to be higher in girls than boys, there are a number
of factors that may skew our estimates of CSA in boys. Boys
may be reluctant to disclose CSA due to beliefs that victimi-
zation threatens masculinity or for fear of not being believed
(Collin-Vézina et al. 2013). Moreover, available information
on CSA in boys suggests that victimization most commonly
occurs in the preschool years. The rudimentary language skills
of this stage of development may limit disclosure by boys,
The research was supported by funding from: The Canadian Institutes of
Health Research; Ontario Mental Health Foundation; Health Canada;
Public Health Agency of Canada; Ontario Ministry of Children &
Youth Services, and the Provincial Centre of Excellence for Child and
Youth Mental Health at the Children’s Hospital of Eastern Ontario.
Secondary analyses for this report were supported by a CIHR Team
grant in boys’and men’s health (TE3 138302) awarded to Chris Wekerle.
Department of Psychology and Neuroscience, Dalhousie University,
Halifax, NS, Canada
Department of Pediatrics, McMaster University, Hamilton, ON,
Department of Applied Psychology and Human Development,
University of Toronto, Toronto, ON, Canada
School of Social Work, Memorial University, St. John’s, NL, Canada
Department of Bioethics, St. George’sUniversity,West
Present address: St. Francis Xavier University, Antigonish, NS,
Department of Psychiatry, Dalhousie University, Halifax, NS,
Journ Child Adol Trauma
further contributing to the underestimate of CSA among boys
(Wekerle et al. 2016).
Childhood sexual abuse (CSA) has been associated with a
number of enduring consequences that can negatively affect
development and persist into adulthood. In terms of substance
use and misuse, CSA is a robust risk factor, predicting out-
comes from early use of illicit drugs to polysubstance use to
risk of fatal overdose (Dube et al. 2003;Harrisonetal.1989,
1997; Cutajar et al. 2010). With regard to alcohol misuse,
CSA has consistently been identified as a risk factor for heavy
drinking and alcohol-related problems (Miller et al. 1993;
Simpson and Miller 2002; Wilsnack et al. 1997).
Although links between CSA and subsequent alcohol mis-
use have been well-established, the mechanisms by which
CSA influences later drinking outcomes merit further investi-
gation (Hovdestad et al. 2011). In fact, the small body of
research on pathways from CSA to alcohol problems has been
conducted almost exclusively with women (Grayson and
Nolen-Hoeksema 2005;SchuckandWidom2001; White
and Widom 2008), despite significant rates of CSA across
both genders (Briere and Elliott 2003; Finkelhor et al. 2014).
In general, sexual abuse (SA) in boys and men is
underreported and understudied (Collin-Vézina et al. 2013),
emphasizing a need to isolate pathways from SA to outcomes
in boys and men, and to compare these with pathways for
girls and women.
In theory, links between CSA and later alcohol misuse
might be driven, at least in part, by emotion symptoms. A
large proportion of survivors of childhood abuse experience
negative emotion symptoms, including anxiety, depression,
and anger (Briere and Runtz 1988; Kiser et al. 1991;
Faulkner et al. 2014; Wolfe et al. 2004). In fact, trauma-
related emotion symptoms have been described as key medi-
ators explaining links between childhood maltreatment and a
range of behavioral and emotional issues (Flett et al. 2012;
Weke r le et al. 2001,2009a,b). Previous research has indicated
that anger is predictive of alcohol and illicit drug use among
adolescents involved with child welfare (Goldstein et al.
2011). Indeed, in testing the mediating roles of several
emotion symptom clusters, Faulkner et al. (2014)found
that anger was the only significant mediator of the effects
of childhood maltreatment on problem drinking in youth
in child protective services. Specific symptoms associated
with anxiety and depression (e.g., sleep problems and
rumination) have also been identified as risk factors for
heightened alcohol consumption following other poten-
tially traumatic events, such as military deployment
(Trautmann et al. 2015). Likewise, research with women
victims of CSA has emphasized the significance of
post-traumatic stress disorder (PTSD), which resembles
anxiety disorders in some ways, in predicting alcohol-
related problems in adulthood by way of encouraging
drinking to regulate emotions (Hannan et al. 2015).
Although this body of research on maltreatment and other
stressful events is informative, only one study gives insight
into the effects of CSA (Hannan et al. 2015). Moreover, al-
though existing work elucidates affect-related mechanisms
linking traumatic experiences to alcohol use and alcohol prob-
lems, it does not answer whether different pathways exist for
males and females. Clinical evidence consistently reveals gen-
der differences in rates of affective disorders. Men outnumber
women in terms of externalizing problems, such as aggres-
sion, and masculinity is associated with the expression of an-
ger (Campbell 2007; Kopper and Epperson 1991), whereas
women experience a greater burden of internalizing disorders,
such as anxiety and mood disorders (Weissman et al. 1991).
With respect to emotional outcomes of childhood abuse, male
victims are at greater risk of anger-fueled aggression and vio-
lence (LaPorte et al. 2011), whereas anxiety and depression
are more prevalent amongst female victims (Briere and Elliott
2003; Weiss et al. 1999).
Gender differences in emotional symptoms suggest that
different symptoms may characterize pathways from victimi-
zation to alcohol problems for men and women (Leyton and
Stewart 2014). Indeed, Goldstein et al. (2010) found that path-
ways from child maltreatment to drinking problems were me-
diated by different motives for young adult men and women.
For women, the relationship between child maltreatment and
drinking problems was mediated by drinking to cope with
negative affect, whereas this relationship was mediated by
drinking to enhance positive affect in men. These findings,
combined with the relative paucity of research on pathways
from CSA to alcohol problems, provide support for the need
to take a gendered approach to examining pathways from
child maltreatment to alcohol problems, and to extend this
work to address links between CSA, emotion symptoms,
and alcohol problems more explicitly.
Given that adolescence isthe time when alcohol use is most
commonly initiated (Kandel 1998), and considering the im-
portance of early interventions (Stewart et al. 2005), studying
these questions in a sample of at-risk youth was deemed to be
of particular relevance. Adolescence has been described as a
high-risk period for individuals involved with child protective
services (Flynn and Vincent 2008) and child welfare popula-
tions show higher rates of substance abuse than the general
population (Wekerle et al. 2009a,b), emphasizing a need to
focus research attention on this population. Moreover, as ad-
olescence is generally characterized by increased risk-taking,
it represents a key window of opportunity to intervene before
poor behavioral health patterns are established (Wekerle et al.
In the current study, we aimed to provide a more detailed
account of links between CSA and alcohol problems in a
vulnerable population –specifically, youth in the child welfare
system –by exploring potential emotion-focused mecha-
nisms. In light of different rates of specific emotional
Journ Child Adol Trauma
symptoms in men and women (Briere and Elliott 2003;
Campbell 2007; Weiss et al. 1999), and based on gender-
specific links between childhood maltreatment, reasons for
drinking, and drinking problems (Goldstein et al. 2010), we
examined gender as a potential moderator. We expected inter-
nalizing emotional symptoms (anxiety; depression) to be im-
portant mediators for female adolescents and the externalizing
emotional symptom of anger to be an important mediator for
male adolescents, in explaining the link between CSA and
alcohol problems in adolescence.
The current study includes data from a subset of participants
from the Maltreatment and Adolescent Pathways (MAP) pro-
ject –a research program aimed at understanding predictive
factors and outcomes of youth receiving child welfare services
in Ontario, Canada (Wekerle et al. 2009a,b). Participants were
randomly selected through child welfare agencies (Children’s
Aid Societies; CAS). Mostadolescents who participated in the
MAP project were in child welfare services for an extended
period (6 months or more) and about 60% were Crown wards
(receiving the highest level of services, in which parental care
was terminated and the government took legal responsibility).
At initial testing, 561 youth had agreed to participate and had
completed an initial questionnaire package. Only MAP study
participants with complete data on questionnaire items
assessing CSA, negative emotion symptoms, and problem
drinking were included in the current analyses (N= 301;
56% female). CSA was self-reported by 37% of participants.
Participants over 16 years of age provided consent to partici-
pate themselves, and those less than 16 years of age assented
to participate and had their parents or guardians provide con-
sent. On average, participants were 15.9 years old (SD =1.1)
and ethnicities included White (28%), Black (21%), Biracial
(31%), Asian (4%), Latin American (3%), Aboriginal (2%),
and other (11%). All procedures were approved by the partic-
ipating CAS agencies and university Research Ethics Boards.
Rutgers Alcohol Problem Index (RAPI; White and
Labouvie 1989)The RAPI was used to measure problem
drinking. It lists 23 negative consequences adolescents may
experience as a result of their alcohol use. Participants indicat-
ed the frequency with which they had experienced each nega-
tive outcome in the past 12 months. Response options included
0 (never), 1 (1–2 times), 2 (3–5 times), 3 (6–10 times), and 4
(more than 10 times). Scores were created by summing re-
sponses across the 23 items. The RAPI has demonstrated good
face, convergent, and discriminant validity in clinical and com-
munity samples, and has high internal consistency (Noel et al.
2010; White and Labouvie 1989; Winters 1999). The internal
reliability in the present sample was high, α=.92.
The Childhood Experiences of Violence Questionnaire
(CEVQ; Walsh et al. 2008)Self-reports of CSAwere collect-
ed using the CEVQ, which consists of 18 items on violence
and maltreatment that are appropriate for administration with
youth. The CEVQ has good psychometric properties, includ-
ing test–retest reliability and intra-class correlations (ICCs),
and demonstrates strong face and convergent validity (Walsh
et al. 2008). The current study uses data only from the sexual
abuse subscale (six items), which inquires about the occur-
rence and frequency of sexual abuse (0 = never, 1 = 1–2times,
2=3–5times,3=6–10 times, 4 = more than 10 times). CSA
scores were created by summing scores across these six items.
A sample item assessing sexual abuse is BDid anyone ever
threaten to have sex with you when you did not want them
to?^In the current sample, α=.85.
The Trauma Symptom Checklist for Children (TSCC;
Briere 1996)The TSCC assesses trauma symptoms in chil-
dren and adolescents, and can be applied to assess the effects
of CSA (Briere 1996). The TSCC provides an evaluation of
six symptom clusters; however, only anger, anxiety, and de-
pression scales were relevant to the analyses in the current
study. For each item, participants report the frequency with
which they experience the listed concern on a 4-point scale
ranging from 0 (never) to 3 (almost all of the time). Subscale
scores are calculated by summing scores for all items within
the given subscale. Internal consistencies range from accept-
able to good for the various subscales (Ohan, Myers, and
Collett 2002), and research supports its construct validity in
clinical and nonclinical samples (Elliott and Briere 1994).
Internal reliabilities were high for all three scales in the current
Independent sample t-tests were conducted to examine gender
differences in CSA frequency, negative emotion symptoms,
and problem drinking. Pearson correlational analyses were
conducted to examine bivariate correlations among variables,
for male and female adolescents separately.
Using maximum likelihood estimations in MPlus 7.11
(Muthén and Muthén 2010), path models were used to exam-
ine whether anxiety, depression, and anger symptoms
mediated the association between CSA and problem
drinking. Indirect effects were estimated in MPlus using the
INDIRECT command, which uses Sobel (1982)standarder-
rors. One thousand bootstrap samples and bias-corrected 95%
confidence intervals were also used to determine the
Journ Child Adol Trauma
significance of the hypothesized indirect effects consistent
with MacKinnon et al. (2002) recommendations. Analyses
included age as covariates in the model, and mediating vari-
ables were allowed to correlate.
Gender differences in the models were tested using
multiple-group analysis in MPlus (Bollen and Curran 2006).
We first tested an unconstrained model in which the path co-
efficients for males and female adolescents were not
constrained to be equal. We then tested the model constraining
all paths to be equal for males and females. A chi-square
difference test between the two models was used to determine
whether the unconstrained model fit better than the
constrained model, which would indicate that gender moder-
ated the relationships among the predictor, potential media-
tors, and outcome variables.
Bivariate correlations between CSA, anxiety symptoms, de-
pression symptoms, anger symptoms, and problem drinking
are presented separately for male and female adolescents in
Tab le 1. All measures were significantly interrelated, for both
male and female adolescents.
Independent samples t-tests were conducted to establish
whether male and female adolescents differed in terms of
CSA, emotion symptoms (anxiety, depression, anger), and
problem drinking (as assessed by RAPI scores). Results re-
vealed that female adolescents reported higher frequencies of
CSA than male adolescents, had more anxiety and depression
symptoms than males, but showed comparable levels of the
emotion symptom of anger. Alcohol problems on the RAPI
did not differ by gender. Descriptive statistics for all measures
are presented, by gender, in Table 2.
Our hypothesized model was just identified with all paths
estimated (0 degrees of freedom); thus, standard fit indices are
not available. We tested for gender moderation using a
multiple-group model. Results showed that the constrained
model fit significantly worse than the unconstrained model
=75.52,Δdf = 22, p< .001), indicating significant mod-
eration by gender. Therefore, the final path models are pre-
sented separately for male and female adolescents.
For female adolescents, results from the path models
showed that CSA was positively associated with higher levels
of anxiety, depression, and anger (Fig. 1a). As expected, there
was a significant indirect effect of CSA on problem drinking
through anxiety (.15, p= .02), but also, unexpectedly, through
anger (.13, p= .005). Higher levels of CSA predicted higher
levels of both anxiety and anger symptoms, which in turn pre-
dicted higher levels of problem drinking. The expected media-
tional pathway through depression for female adolescents was
not significant and, in fact, the relationship was in the opposite
direction to that hypothesized (−.10, p= .10). A summary of
the indirect effects is presented in Table 3.Overall,forfemale
adolescents, the association between CSA and problem drink-
ing was fully mediated by negative emotion symptoms (total
indirect effect = .89, p< .001) and the model accounted for
28% of the variance in problem drinking.
Similarly, for male adolescents, the results of the path
models showed that CSA was positively associated with
higher levels of all three potential mediators (Fig. 1b). As
expected, there was an indirect effect of CSA on problem
drinking via anger (.08, p= .03; Table 2) and, as expected,
the indirect effects of CSA on problem drinking via the
Tabl e 1 Bivariate correlations
for childhood sexual abuse scores
(CEVQ), anxiety, depression, and
anger subscale scores (TSCC),
and problem drinking scores
(RAPI) for female and male
Depression (TSCC) Anger(TSCC) Problem Drinking
Childhood Sexual Abuse (CEVQ)
Females .437* .410* .384* .277*
Males .469* .374* .242* .318*
Females .863* .757* .451*
Males .749* .579* .355*
Females .733* .358*
Males .612* .263*
CEVQ childhood experiences of violence questionnaire, TSCC trauma symptom checklist for children, RAPI
rutgers alcohol problem index
*All correlations are significant at p<.01
Journ Child Adol Trauma
emotion symptoms of anxiety and depression were not signif-
icant for male adolescents (Table 2). However, for males, the
direct association between CSA and problem drinking
remained statistically significant after controlling for indirect
effects. Thus, in contrast to females, the results suggest only
partial mediation of the association between CSA and
problem drinking via emotion symptoms for males. The full
model accounted for 23% of the variance in problem drinking.
In the present study, we sought to examine potential pathways
from CSA to emotion symptoms to problem drinking in a
sample of youth involved with the child welfare system.
Based on differential rates of internalizing versus externaliz-
ing problems in men and women (Campbell 2007; Weissman
et al. 1991), and based on documented gender differences in
links between childhood maltreatment, reasons for drinking,
and drinking problems (Goldstein et al. 2010), we expected
different mechanisms to explain links between CSA and prob-
lem drinking for male and female adolescents. Specifically,
we hypothesized that internalizing emotion symptoms (anxi-
ety; depression) would be significant mediators for female
adolescents and externalizing emotion symptoms (anger)
would be a significant mediator for males.
Tabl e 2 Means (and standard deviations) for childhood sexual abuse
scores (CEVQ), anxiety, depression, and anger subscale scores (TSCC),
and problem drinking scores (RAPI) for female and male adolescents
Females Males t-
Childhood Sexual Abuse (CEVQ) 1.68 (2.62) 0.41 (1.13) 5.24*
Anxiety (TSCC) 5.9 (6.31) 3.27 (3.78) 4.25*
Depression (TSCC) 6.85 (6.82) 3.80 (4.58) 4.43*
Anger (TSCC) 7.42 (7.57) 6.24 (6.06) 1.46
Problem Drinking (RAPI) 9.41 (12.47) 10.28 (12.39) −.60
CEVQ childhood experiences of violence questionnaire, TSCC trauma
symptom checklist for children, RAPI rutgers alcohol problem index
Fig. 1 Path Models Showing
Mediation of the Effects of
Childhood Sexual Abuse on
Problem Drinking Separately for
aFemales and bMales.
Standardized coefficients are
presented for each path. Solid
lines represent significant paths;
dotted lines represent non-
significant paths. †p=.08,*
Journ Child Adol Trauma
Preliminary analyses comparing males and females on
CSA, emotion symptoms, and problem drinking revealed that
females reported a higher frequency of SA incidences in child-
hood, and experienced higher levels of anxiety and depression
in adolescence. These findings corroborate previous research,
which reliably finds that girls experience higher prevalence
rates of CSA than boys (Finkelhor et al. 2014), and that wom-
en, in both general and victimized populations, are more prone
to internalizing disorders, like anxiety and depression (Briere
and Elliott 2003; Hankin and Abramson 2002; Weiss et al.
1999; Weissman et al. 1991).
Unexpectedly, no gender differences emerged for anger
symptoms, or problem drinking. Research in the general pop-
ulation tends to report gender differences in anger and related
behaviors, such as aggression (Campbell 2007;Kopperand
Epperson 1991). The lack of gender differences in anger in the
current study may be attributable to the particular population.
Indeed, research by Wolfe et al. (2001) found that girls who
were victims of maltreatment reported significantly more an-
ger symptoms than girls without histories of childhood mal-
treatment. As youth involved in the Child Welfare System are
more likely to have experienced childhood maltreatment or
neglect, this is one potential explanation for the comparable
levels of anger among female and male adolescents in our
sample. Another potential explanation is that, because females
experienced more frequent CSA than males, they reported
greater emotional symptoms, in general, such that typical gen-
der differences in anger were eliminated. Related research has
failed to find gender differences in externalizing symptoms in
youth who were exposed to violence (Foster et al. 2004;
Spilsbury et al. 2007). Furthermore, Jones et al. (2013)found
that externalizing pathways from CSA to adolescent risk-
taking behavior existed for both boys and girls. It is possible
that the lack of gender differences in drinking problems may
also be a feature of this population. Conversely, recent re-
search suggests that female youth, in general, are increasingly
showing similar rates of alcohol use as male adolescents, in a
process known as gender convergence (Boak et al. 2015;
Keyes et al. 2010). Therefore, this finding may more accurate-
ly reflect a generational trend seen across different
In line with expectations, path analyses showed that CSA
had an indirect effect on problem drinking in female adoles-
cents, wherein higher levels of CSA predicted higher levels of
anxiety symptoms, in turn, explaining increases in problem
drinking. Somewhat surprisingly, anger also emerged as a
significant mediator in our model from CSA to problem drink-
ing in females. In fact, together, anxiety and anger fully me-
diated links between CSA and problem drinking. In contrast,
there was a significant direct association between CSA and
problem drinking for male adolescents, in which higher levels
of CSA were associated with more problematic drinking.
There was also an indirect effect of CSA on problem drinking,
mediated solely by anger symptoms. Thus, of the three emo-
tion symptoms studied here, anger appears to be related to
drinking-related outcomes of CSA in both male and female
Our finding that anxiety was uniquely associated with
problem drinking in females with more frequent experiences
of sexual victimization is in accord with work by Goldstein
et al. (2010), who found that childhood maltreatment put
women at risk for drinking problems due to their coping-
based motives for drinking. Furthermore, the current study
expands on findings by Goldstein et al. by illustrating which
negative emotion symptoms women are attempting to cope
with through their alcohol use, namely anxiety symptoms.
The observation that anxiety symptoms were associated with
problem drinking in female adolescents, but not males, cor-
roborates clinical findings that anxiety and alcohol use disor-
ders are more commonly linked in women than men (Stewart
et al. 2009). It should be noted that anxiety, and not depres-
sion, mediated pathways from CSA to problem drinking. At
first glance, this is surprising considering high comorbidity
rates of these negative emotion states, and given the strong
correlations between these symptoms in the current study.
However, upon closer inspection, it seems likely that the mar-
ginal negative correlation between depression and drinking
problems emerged due to shared variance between anxiety
and depression symptoms that was accounted for in the
anxiety-mediated pathway. In turn, aspects of depressive emo-
tions that are distinct from anxiety (i.e., anhedonia, apathy)
might show a slight negative relation with problem drinking
because such symptoms discourage individuals from seeking
out contexts where drinking is common (i.e., social gather-
ings, parties, etc.). In other words, depression, in general, ap-
pears to be associated with problem drinking (according to the
observed bivariate correlations and previous literature; e.g.,
Dixit and Crum 2000), but unique aspects of depression might
be protective. It should also be noted that, although depression
was not a significant positive mediator, it was significantly
associated with CSA, such that more frequent CSA predicted
Tabl e 3 Indirect effects of childhood sexual abuse on problem drinking
via anger, anxiety, and depression
Indirect effect 95% CI Indirect effect 95% CI
Childhood sexual abuse via
Anger .13** .21–1.18 .08* .25–2.30
Anxiety .15** .02–1.64 .07 −.47–2.38
Depression −.10†−1.15–.05 −.05 −1.66–.15
CI confidence interval ** p<.01, * p<.01,†p<.10
Anger, anxiety, and depression were assessed via the TSCC
Journ Child Adol Trauma
higher levels of depression in both genders. This dovetails
with research on relations between CSA and depression in
adulthood (Lindert et al. 2014).
The observation that anger explained indirect effects of
CSA on problem drinking for both males and females in our
sample fits with broad research on pathways from
maltreatment/violence exposure to problem drinking
(Faulkner et al. 2014; Goldstein et al. 2011), and builds on
this work by illustrating that anger-based mechanisms can be
applied to understanding outcomes of CSA, as a specific form
of childhood victimization. Moreover, we established that
anger-mediated paths from CSA to problem drinking were
significant for males and females, whereas previous research
has controlled for gender in similar models (Faulkner et al.
2014; Goldstein et al. 2011), but has not studied gender as a
potential moderator. Future work should clarify how anger
increases risk for alcohol problems in victims of CSA. One
possibility is that early victimization results in a general exter-
nalizing tendency, which includes a variety of acting out be-
haviors, such as anger-related behaviors like aggression and
violence, aswell as alcohol misuse. Longitudinal research will
need to determine whether anger symptoms precede alcohol
misuse, or whether anger and drinking problems emerge con-
comitantly in youth who were victims of CSA.
Results should be understood in the context of several po-
tential study limitations. There are potential issues with our
use of self-report data on CSA in that such reports are often
influenced by unwillingness to disclose (Fergusson et al.
2000) and problems in remembering or interpreting inci-
dences (e.g., memory is unclear, incident is not understood
as assault). The issue of reluctance to report CSA is particu-
larly noteworthy in studies on gender differences, as barriers
to reporting might be greater in males (e.g., threatened mas-
culinity, fears of skepticism or disbelief; Collin-Vézina et al.
2013). Limitations associated with studying gender effects
also exist with regard to emotional symptom measures, in that
females might be more inclined to endorse items associated
with depression or anxiety (e.g., crying), whereas males might
preferentially endorse items assessing anger (e.g., wanting to
yell and break things). Indeed, Feingold (1994)suggeststhat
gender differences in reports of certain symptoms and traits
might, in part, reflect tendencies to respond in a way that is
consistent with gender norms (i.e., favoring masculine or fem-
inine attributes). However, gender differences in emotional
symptoms are replicated when using measures other than
self-report, such as ethological observation of behavior
(Troisi and Moles 1999). Potential limitations also exist when
using self-reports of alcohol use given that youth might feel
hesitant to give accurate information regarding their drinking
habits. Nonetheless, substantial research suggests that, when
confidentiality is ensured, respondents’self-reports of drink-
ing are highly accurate (Sobell and Sobell 1990). It should
also be noted that our measure of CSA reflected occurrence
and frequency of incidences, but did not consider other as-
pects of CSA, such as the duration or severity of abuse or
the age when abuse occurred. In addition, our assessment of
emotions associated with CSA did not include shame or guilt
–two self-conscious emotions that are common consequences
of child abuse (Feiring and Taska 2005; Stuewig and
McCloskey 2005). Follow-up work should assess whether
inclusion of these emotions broadens our understanding of
the chain of mechanisms that link CSA to problem drinking.
A body of work indicates links between shame and rage, and
guilt and anxiety (Briere and Elliott 1994;Lisak1995).
Therefore, it may be informative to examine serial mediation
models, whereby CSA may lead to problem drinking by
instilling feelings of shame and guilt, which in turn increase
anger and anxiety. Finally, because our data are of a cross-
sectional nature, causal inferences cannot be made. Future
research should replicate the findings presented here, but
using a longitudinal study design.
Despite these limitations, our findings have a number of
clinical implications for clinicians who treat adolescent pa-
tients with alcohol use disorder or problem drinking. First,
our results add to a substantial literature suggesting that
CSA is a risk factor for alcohol use problems in adolescence
(Champion et al. 2004;Milleretal.1993; Simpson and Miller
2002; Wilsnack et al. 1997). Therefore, individuals presenting
for treatment for alcohol use disorders should be screened for
histories of sexual victimization, and vice versa. Second, our
findings pinpoint different mechanisms by which CSA leads
to problem drinking in males and females. Both anxiety and
anger may be important emotion symptoms to address in
treating drinking problems in women with histories of sexual
abuse. In men, anger may merit increased attention in a clin-
ical setting, whereas internalizing symptoms are seemingly
less important, at least in relation to alcohol use outcomes. A
growing body of research suggests that targeted treatments are
effective in reducing substance use problems in youth and
adults (Conrod et al. 2000a,b,2006). These targeted treat-
ments adapt aspects of the treatment or intervention to meet
individual needs, such as motivational or personality risk fac-
tors (Conrod et al. 2000a,b,2006). Based on our findings,
tailored treatments for alcohol problems would do well to
focus on emotional symptoms that arise following trauma
(i.e., after sexual victimization). Moreover, such treatments
should take a gendered approach, acknowledging that distinct
emotion symptoms explain problem drinking in men and
women CSA survivors. Although recommendations based
on our results may pertain to child welfare populations, in
particular, they may also be informative for clinicians working
with other vulnerable populations. Future work will need to
examine these models with more diverse populations before
concrete recommendations can be made. Our findings also
suggest that early interventions for affective problems, which
impart healthy ways of dealing with anger and anxiety, might
Journ Child Adol Trauma
prevent the occurrence of alcohol problems later in life. Such
strategies are liable to confer substantial savings in the areas of
health and social services.
The current study’s results also have notable implications
for future research within this area. One finding that warrants
further consideration is the observation that negative emotion
symptoms (anxiety, anger) fully mediated links between CSA
and problem drinking in female adolescents, and anger symp-
toms only partially mediated this association in males. This
suggests that, while connections between CSA and problem
drinking in females can fully be accounted for by emotional
symptoms, relations may be more complex and multi-faceted
in males. Other contributors to problem drinking that merit
investigation, alongside emotion symptoms, in male victims
of CSA include motivational factors (reasons for drinking),
cognitive factors (beliefs, attitudes, expected outcomes), and
social factors (susceptibility to peer pressure and peer model-
ling). Future research will need to evaluate these possibilities.
Acknowledgements We thank the Child Welfare adolescents who par-
ticipated in the Maltreatment Adolescent Pathways (MAP) study, and
who remained involved over the 2 to 3 years of follow-up. We also thank
our partners in the Child Welfare agencies and Associations who support-
ed the study, as well as the MAP Advisory Board. We thank the many
funding agencies who supported the MAPstudy: The Canadian Institutes
of Health Research; Ontario Mental Health Foundation; Health Canada;
Public Health Agency of Canada; Ontario Ministry of Children & Youth
Services, and the Provincial Centre of Excellence for Child and Youth
Mental Health at the Children’s Hospital of Eastern Ontario. Secondary
analyses for this report were supported by a CIHR Team grant in boys’
and men’s health (TE3 138302). Finally, we thank all of the MAP study
investigators: Christine Wekerle, Harriet MacMillan, Michael Boyle,
Nico Trocmé, Eman Leung, Randall Waechter, Deb Goodman, Brenda
Moody, and Bruce Leslie.
Compliance with ethical standards
Ethical Standards and Informed Consent All participants provided
informed consent. Participants over 16 years of age provided consent to
participate themselves, and those less than 16 years of age provided
verbal assent to participate and had their parents or guardians provide
Research was conducted in accordance with institutional ethical
guidelines and APA ethical standards, and was approved by the institu-
tional research ethics board, as well as participating CAS agencies.
Conflict of Interest The authors declare that they have no conflict of
Andrews, G., Corry, J., Slade, T., Issakidis, C., & Swanston, H. (2004).
Child sexual abuse. Comparative Quantification of Health Risks, 2,
1851–1940. WHO publication.
Barth, J., Bermetz, L., Heim, E., Trelle, S., & Tonia, T. (2013). The
current prevalence of child sexual abuse worldwide: a systematic
review and meta-analysis. International Journal of Public Health,
Boak, A., Hamilton, H.A., Adlaf, E.M., & Mann, R.E. (2015). Drug Use
Among Ontario Students, 1977–2015: Detailed OSDUS Findings.
Centre for Addiction and Mental Health Report.
Bollen, K. A., & Curran, P. J. (2006). Latent curve models: A structural
equation perspective. Hoboken: Wiley-Interscience.
Briere, J. (1996). Trauma symptom checklist for children. Odessa:
Briere, J. N., & Elliott, D. M. (1994).Immediate and long-term impactsof
child sexual abuse. The Future of Children, 4,54–69. doi:10.2307
Briere, J.,& Elliott, D.M. (2003). Prevalence and psychological sequelae
of self-reported childhood physical and sexual abuse in a general
population sample of men and women. Child Abuse & Neglect, 27,
Briere, J., & Runtz, M. (1988). Symptomatology associated with child-
hood sexual victimization in a nonclinical adult sample. Child Abuse
& Neglect, 12,51–59. doi:10.1016/0145-2134(88)90007-5.
Campbell, A. (2007). Sex differences in aggression. In R. I. M. Dunbar &
L. Barrett (Eds.), Oxford handbook of evolutionary psychology (pp.
365–382). New York: Oxford University Press.
Champion, H. L., Foley, K. L., Durant, R. H., Hensberry, R., Altman, D.,
& Wolfson, M. (2004). Adolescent sexual victimization, use of alcohol
and other substances, and other health risk behaviors. Journal of
Adolescent Health, 35,321–328. doi:10.1016/j.jadohealth.2003.09.023.
Collin-Vézina, D., Daigneault, I., & Hébert, M. (2013). Lessons learned
from child sexual abuse research: prevalence, outcomes, and pre-
ventive strategies. Child and Adolescent Psychiatry and Mental
Health, 7,22–30. doi:10.1186/1753-2000-7-22.
Conrod, P. J., Pihl, R. O., Stewart, S. H., & Dongier, M. (2000a). Validation
of a system of classifying female substance abusers on the basis of
personality and motivational risk factors for substance abuse.
Psychology of Addictive Behaviors, 14, 243–256. doi:10.1037
Conrod, P. J., Stewart, S. H., Pihl, R. O., Côté, S., Fontaine, V., &
Dongier, M. (2000b). Efficacy of brief coping skills interventions
that match different personality profiles of female substance abusers.
Psychology of Addictive Behaviors, 14,231–242. doi:10.1037
Conrod, P. J., Stewart, S. H., Comeau, N., & Maclean, A. M. (2006).
Efficacy of cognitive–behavioral interventions targeting personality
risk factors for youth alcohol misuse. Journal of Clinical Child and
Adolescent Psychology, 35,550–563. doi:10.1207/s15374424
Cutajar, M. C., Mullen, P. E., Ogloff, J.R., Thomas, S. D., Wells,D. L., &
Spataro, J. (2010). Suicide and fatal drug overdose in child sexual
abuse victims: a historical cohort study. Medical Journal of
Dixit, A. R., & Crum, R. M. (2000). Prospective study of depression and
the risk of heavy alcohol use in women. American Journal of
Psychiatry, 157,751–758. doi:10.1176/appi.ajp.157.5.751.
Dube, S. R., Felitti, V. J., Dong, M., Chapman, D. P., Giles, W. H., &
Anda, R. F. (2003). Childhood abuse, neglect, and household dys-
function and the risk of illicit drug use: the adverse childhood expe-
riences study. Pe dia t ri c s, 111,564–572.
Elliott, D. M., & Briere, J. (1994). Forensic sexual abuse evaluations of
older children: disclosures and symptomatology. Behavioral
Sciences and the Law, 12,261–277. doi:10.1002/bsl.2370120306.
Faulkner, B., Goldstein, A. L., & Wekerle, C. (2014). Pathways from
childhood maltreatment to emerging adulthood: investigating
trauma-mediated substance use and dating violence outcomes
among child protective services-involved youth. Child
Maltreatment, 19,219–232. doi:10.7755/9514551944.
Journ Child Adol Trauma
Feingold, A. (1994). Gender differences in personality: a meta-analysis.
Psychological Bulletin, 116, 429–456. doi:10.1037/0033-
Feiring, C., & Taska, L. S. (2005). The persistence of shame following
sexual abuse: a longitudinal look at risk and recovery. Child
Maltreatment, 10,337–349. doi:10.1177/1077559505276686.
Fergusson, D. M., Horwood, L. J., & Woodward, L. J. (2000). The sta-
bility of child abuse reports: a longitudinal study of the reporting
behaviour of young adults. Psychological Medicine, 30,529–544.
Finkelhor, D., Shattuck, A., Turner, H. A., & Hamby, S. L. (2014). The
lifetime prevalence of child sexual abuse and sexual assault assessed
in late adolescence. Journal of Adolescent Health, 55, 329–333.
Flett, G. L., Goldstein, A. L., Hewitt, P. L., & Wekerle, C. (2012).
Predictors of deliberate self-harm behaviour among emerging ado-
lescents: an initial test of a self-punitiveness model. Current
Psychology, 31,49–64. doi:10.1007/s12144-012-9130-9.
Flynn, R. J., & Vincent, D. (2008). Canada. In M. S. Stein & E.R. Munro
(Eds.), Youn g p e ople ’s transitions from care to adulthood:
International research and practice. London: Jessica Kingsley
Foster, J. D., Kuperminc, G. P., & Price, A. W. (2004). Gender differences
in posttraumatic stress and related symptoms among inner-city mi-
nority youth exposed to community violence. Journal of Youth and
Adolescence, 33,59–69. doi:10.1023/A:1027386430859.
Goldstein, A. L., Flett, G. L., & Wekerle, C. (2010). Child maltreatment,
alcohol use and drinking consequences among male and female
college students: an examination of drinking motives as mediators.
Addictive Behaviors, 35,636–639. doi:10.1016/j.
Goldstein, A., Wekerle, C., Tonmyr, L., Thornton, T., Waechter, R.,
Pereira, J.,…Chung, R. (2011). The relationship between post-
traumatic stress symptoms and substance use among adolescents
involved with child welfare: Implications for emerging adulthood.
International Journal of Mental Health and Addiction, 9,507–524.
Grayson, C. E., & Nolen-Hoeksema, S. (2005). Motives to drink as me-
diators between childhood sexual assault and alcohol problems in
adult women. Journal of Traumatic Stress, 18,137–145.
Hankin, B. L., & Abramson, L. Y. (2002). Measuring cognitive vulnera-
bility to depression in adolescence: reliability, validity, and gender
differences. Journal of Clinical Child and Adolescent Psychology,
Hannan, S. M., Orcutt, H. K., Miron, L. R., & Thompson, K. L. (2015).
Childhood sexual abuse and later alcohol-related problems investi-
gating the roles of revictimization, PTSD, and drinking motivations
among college women. Journal of Interpersonal Violence.
Harrison, P. A., Hoffmann, N. G., & Edwall, G. E. (1989). Differential
drug use patterns among sexually abused adolescent girls in treat-
ment for chemical dependency. International Journal of the
Addictions, 24,499–514. doi:10.3109/10826088909081832.
Harrison, P. A., Fulkerson, J. A., & Beebe, T. J. (1997). Multiple sub-
stance use among adolescent physical and sexual abuse victims.
Child Abuse & Neglect, 21,529–539. doi:10.1016/S0145-2134(97
Hovdestad, W. E., Tonmyr, L., Wekerle, C., & Thornton, T. (2011). Why
is childhood maltreatment associated with adolescent substance
abuse? A critical review of explanatory models. International
Journal of Mental Health and Addiction, 9,525–542. doi:10.1007
Jones, D. J., Lewis, T., Litrownik, A., Thompson, R., Proctor, L. J., Isbell,
P. , …& Runyan, D. (2013). Linking childhood sexual abuse and
early adolescent risk behavior: The intervening role of internalizing
and externalizing problems. Journal of Abnormal Child Psychology,
41,139–150. doi: 10.1007/s10802-012-9656-1.
Kandel, D. B. (1998). Persistent themes and new perspectives on adoles-
cent substance use: A lifespan perspective. In J. Richard (Ed.), New
perspectives on adolescent risk behavior (pp. 43–89). New York:
Cambridge University Press.
Keyes, K., Martins, S., Blanco, C., & Hasin, D. (2010). Telescoping and
gender differences in alcohol dependence: new evidence from two
national surveys. American Journal of Psychiatry, 167,969–976.
Kiser, L. J., Heston, J., Millsap, P. A., & Pruitt, D. B. (1991). Physical and
sexual abuse in childhood: relationship with post-traumatic stress
disorder. Journal of the American Academy of Child & Adolescent
Psychiatry, 30,776–783. doi:10.1016/S0890-8567(10)80015-2.
Kopper, B. A., & Epperson, D. L. (1991). Sex and sex‐role comparisons
in the expression of anger. Psychology of Women Quarterly, 15,7–
Laporte, L., Jiang, D., Pepler, D. J., & Chamberland, C. (2011). The
relationship between adolescents’experience of family violence
and dating violence. Youth & Society, 43,3–27. doi:10.1177
Leyton, M., & Stewart, S. (Eds.). (2014). Substance abuse in Canada:
Childhood and adolescent pathways to substance use disorders.
Canadian Centre on Substance Abuse Report.
Lindert, J., von Ehrenstein, O. S., Grashow, R., Gal, G., Braehler, E., &
Weisskopf, M. G. (2014). Sexual and physical abuse in childhood is
associated with depression and anxiety over the life course: system-
atic review and meta-analysis. International Journal of Public
Health, 59,359–372. doi:10.1007/s00038-013-0519-5.
Lisak, D. (1995). Integrating a critique of gender in the treatment of male
survivors of childhood abuse. Psychotherapy: Theory, Research,
Practice, Training, 32,258.doi:10.1037/0033-318.104.22.1688.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., &
Sheets, V. (2002). A comparison of methods to test mediation and
other intervening variable effects. Psychological Methods, 7,83–
MacMillan, H. L., Fleming, J. E., Trocmé, N., Boyle, M. H., Wong, M.,
Racine, Y. A., …& Offord, D. R. (1997). Prevalence of child phys-
ical and sexual abuse in the community: Results from the Ontario
Health Supplement. The Journal of the American Medical
Association,278, 131–135. doi:10.1001
Miller, B. A., Downs, W. R., & Testa, M. (1993). Interrelationships be-
tween victimization experiences and women’s alcohol use. Journal
of Studies on Alcohol. Supplement, 11, 109–117. doi:10.15288
Muthén, B., & Muthén, L. (2010). Mplus user’s guide (6th ed.). Los
Angeles: Muthén & Muthén.
Noel, M., O’connor, R. M., Boudreau, B., Mushquash, C. J., Comeau, M.
N., Stevens, D., & Stewart, S. H. (2010). The Rutgers Alcohol
Problem Index (RAPI): a comparison of cut-points in First Nations
Mi’kmaq and non-Aboriginal adolescents in rural Nova Scotia.
International Journal of Mental Health and Addiction, 8, 336–
Ohan, J. L., Myers, K., & Collett, B. R. (2002). Ten-year review of rating
scales, IV: scales assessing trauma and its effects. Journal of the
American Academy of Child & Adolescent Psychiatry, 41,1401–
Schuck, A. M., & Widom, C. S. (2001). Childhood victimization and
alcohol symptoms in females: causal inferences and hypothesized
mediators. Child Abuse and Neglect, 25,1069–1092. doi:10.1016
Simpson, T. L., & Miller, W. R. (2002). Concomitance between child-
hood sexual and physical abuse and substance use problems: a re-
view. Clinical Psychology Review, 22,27–77. doi:10.1016/S0272-
Journ Child Adol Trauma
Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects
in structural equation models. Sociological Methodology, 13,290–
Sobell, L. C., & Sobell, M. C. (1990). Self-report issues in addictive
behaviors [Special issue]. Behavioral Assessment, 12,77–90.
Spilsbury, J.C., Belliston, L., Drotar, D., Drinkard, A., Kretschmar, J.,
Creeden, R., …& Friedman, S. (2007). Clinically significant trauma
symptoms and behavioral problems in a community-based sample
of children exposed to domestic violence. JournalofFamily
Vio l enc e ,22,487–499. doi: 10.1007/s10896-007-9113-z.
Stewart, S. H., Conrod, P. J., Marlatt, G. A., Comeau, M. N., Thush, C., &
Krank, M. (2005). New developments in prevention and early inter-
vention for alcohol abuse in youths. Alcoholism: Clinical and
Experimental Research, 29, 278–286. doi:10.1097/01.
Stewart, S. H., Gavric, D., & Collins, P. (2009). Women, girls, and alco-
hol. In K. T. Brady, S. E. Back, & S. F. Greenfield (Eds.), Wom e n
and addiction: A comprehensive handbook (pp. 341–359). New
York: Guilford Press.
Stuewig, J., & McCloskey, L. A. (2005). The relation of child maltreat-
ment to shame and guilt among adolescents: psychological routes to
depression and delinquency. Child Maltreatment, 10, 324–336.
Trautmann, S., Schönfeld, S., Behrendt, S., Heinrich, A., Höfler, M.,
Siegel, S., …& Wittchen, H. U. (2015). Predictors of changes in
daily alcohol consumption in the aftermath of military deployment.
Drug and Alcohol Dependence,147, 175–182. doi:10.1016/j.
Trocmé, N. M., Tourigny, M., MacLaurin, B., & Fallon, B.(2003). Major
findings from the Canadian incidence study of reported child abuse
and neglect. Child Abuse & Neglect, 27,1427–1439.
Troisi, A., & Moles, A. (1999). Gender differences in depression: an
ethological study of nonverbal behavior during interviews. Journal
of Psychiatric Research, 33,243–250. doi:10.1016/S0022-3956(98
Walsh, C. A., MacMillan, H. L., Trocmé, N., Jamieson, E., & Boyle, M.
H. (2008). Measurement of victimization in adolescence: develop-
ment and validation of the childhood experiences of violence ques-
tionnaire. Child Abuse & Neglect, 32,1037–1057. doi:10.1016/j.
Weiss, E. L., Longhurst, J. G., & Mazure, C. M. (1999). Childhood sexual
abuse as a risk factor for depression in women: psychosocial and
neurobiological correlates. American Journal of Psychiatry, 156,
Weissman, M. M., Bruce, M. L., Leaf, P. J., Florio, L. P., & Holzer, C.
(1991). Affective disorders. In L. N. Robins & D. A. Regier (Eds.),
Psychiatric disorders in America: the epidemiological catchment
area study (pp. 53–80). New York: Free Press.
Wekerle, C., Wolfe, D. A., Hawkins, D. L., Pittman, A.-L., Glickman, A.,
& Lovald, B. E. (2001). The value and contribution of youth self-
reported maltreatment history to adolescent dating violence: testing
a trauma mediational model. Development and Psychopathology,
Wekerle, C., Leung, E., Goldstein, A. L., Thornton, T., & Tonmyr, L.
(2009a). Substance use among adolescents in child welfare versus
adolescents in the general population: A comparison of the mal-
treatment and adolescent pathways (MAP) longitudinal study and
Ontario student drug use survey (OSDUS) datasets. Ottawa: Report
prepared for Health Canada.
Wekerle, C., Leung, E., MacMillan, H. L., Boyle, M., Trocmé, N., &
Waechter, R. (2009b). The impact of childhood emotional maltreat-
ment on teen dating violence. Child Abuse & Neglect: The
International Journal, 33,45–58. doi:10.1016/j.
Wekerle, C., Wolfe, D. A., Cohen, J., Bromberg, D., & Murray, L. (2016).
Child maltreatment (2nd ed.). Toronto: Hogrefe/Huber.
White, H. R., & Labouvie, E. W. (1989). Towards the assessment of
adolescent problem drinking. Journal of Studies on Alcohol, 50,
White, H. R., & Widom, C. S. (2008). Three potential mediators of the
effects ofchild abuse and neglect on adulthood substance use among
women. Journal of Studies on Alcohol and Drugs, 69,337–347.
Wilsnack, S. C., Vogeltanz, N. D., Klassen, A. D., & Harris, T. R. (1997).
Childhood sexual abuse and women’s substance abuse: national
survey findings. Journal of Studies on Alcohol, 58,264–271.
Winters, K.C. (1999). Screening and assessing adolescents for substance
use disorders. National Clearinghouse for Alcohol and Drug
Information (NCADI). DHHS Publication No. (SMA) 99–3282.
Wolfe, D. A., Scott, K., Wekerle, C., & Pittman, A. L. (2001). Child
maltreatment: risk of adjustment problems and dating violence in
adolescence. Journal of the American Academy of Child &
Adolescent Psychiatry, 40,282–289. doi:10.1097/00004583-
Wolfe, D. A., Wekerle, C., Scott, K., Straatman, A.-L., & Grasley, C.
(2004). Predicting abuse in adolescent dating relationships over
1 year: the role of child maltreatment and trauma. Journal of
Abnormal Child Psychology, 113,406–415. doi:10.1037/0021-843
Journ Child Adol Trauma