JOURNAL

10.1177/0886260505278720
Bal et al. / PREDICTORS
OF INTERPERSON
OF SYMPTOMA
AL VIOLENCE
TOLOGY
/ November 2005

Predictors of Trauma
Symptomatology in Sexually
Abused Adolescents
A 6-Month Follow-Up Study

SARAH BAL
ILSE DE BOURDEAUDHUIJ
GEERT CROMBEZ
PAULETTE VAN OOST
Ghent University

This study examines the natural course of trauma-specific symptoms 6 months after
disclosure. Furthermore, this study investigates whether severity and type of abuse
(intrafamilial or extrafamilial sexual abuse), negative appraisals, coping strategies,
and crisis support measured at time of disclosure can be predictive of trauma symp-
toms 6 months later. Sixty-five sexually abused Flemish adolescents are reassessed 6
months after disclosure. Information from the participants is obtained through self-
report questionnaires. Forty-six percent of the adolescents report clinically signifi-
cant trauma symptoms. Although internalizing symptoms significantly decreases
after 6 months, externalizing symptoms persist. Type or severity of the abuse does not
account for differences in symptomatology. Two predictors of ongoing trauma
symptomatology are identified: postdisclosure trauma symptomatology and a lack of
initial crisis support. Information on the victims’ postdisclosure symptomatology as
well as information on the initial received social support is critical in understanding
which abused adolescents are most at risk for poor outcomes in the long term.

Keywords: sexual abuse; adolescents; predictors; crisis support; follow-up

Sexual abuse of adolescents may induce a considerable variability in short-

term symptoms (Kendall-Tackett, Williams, & Finkelhor, 1993; Spaccarelli,
1994). Internalizing problems, such as anxiety, depression, dissociative
complaints and problems related to posttraumatic stress disorder, and
externalizing symptoms, such as sexual problems and anger, are among the
Authors’ Note: We want to thank Vanessa Gryspeerd and Jasmine Andries for their help in
data collection. This research was supported by a grant (01109600) from the Research Council of
the Ghent University.
JOURNAL OF INTERPERSONAL VIOLENCE, Vol. 20 No. 11, November 2005 1390-1405
DOI: 10.1177/0886260505278720
© 2005 Sage Publications
1390

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

 Bal et al. / PREDICTORS OF SYMPTOMATOLOGY 1391

most frequently reported symptoms (Bal, Crombez, Van Oost, & De

Bourdeaudhuij, 2003; Wolfe & Birt, 1997). Several cross-sectional studies
attribute symptom variability to differences in severity of abuse, appraisal
processes, coping strategies, and social support (Spaccarelli & Fuchs, 1997;
Tremblay, Hébert, & Piché, 1999). Much is already known about symptom
development shortly after disclosure of sexual abuse as well as about the
long-term effects of abuse in adults, but there is a scarcity of follow-up stud-
ies that cover the transition period between short- and long-term effects of
sexual abuse. The aim of this study is to explore the natural course of
symptomatology and influencing variables (negative appraisals, coping and
social support) 6 months after disclosure and to investigate which of the
above mentioned factors influence the outcomes 6 months later.
In their review of longitudinal and follow-up studies on child and adoles-
cent sexual abuse, Kendall-Tackett et al. (1993) concluded that for one half to
two thirds of all children and adolescents, postabuse symptoms decreased
with time, whereas 10% to 24% of symptoms intensified. However, this pat-
tern of recovery seemed to be different for different symptoms. In their
follow-up study with sexually abused children and adolescents, Gomes-
Schwartz, Horowitz, Cardarelli, and Sauzier (1990) found that anxiety prob-
lems tended to decrease, whereas problems of anger and sexual preoccupa-
tion seemed to persist or worsen. This is consistent with Mannarino, Cohen,
Smith, and Moore-Motily (1991), who found that at 6- and 12-month follow-
ups, sexually abused children improved significantly on internalizing prob-
lems but not on externalizing problems. Other studies, however, did not find
significant improvements in symptomatology with time. Oates, O’Toole,
Lynch, Stern, and Cooney (1994) assessed 84 sexually abused children and
adolescents until 18 months after intake. Results showed that self-esteem and
depression scores, which were significantly worse at intake than the levels
measured with control children, failed to show significant improvement. Five
years later, 68 of these children and adolescents were reassessed, and again,
no significant changes in depression, self-esteem, and behavior problems
were found (Tebbutt, Swanston, Oates, & O’Toole, 1997). The authors con-
cluded that children and adolescents who are sad or depressed at the time of
initial assessment are likely to have similar problems later on.
Several factors appear to influence symptom variability. Some cross-
sectional studies of sexually abused adolescents have revealed that character-
istics of the abusive experience as well as the type of abuse (intrafamilial or
extrafamilial) may account for variations in trauma symptomatology
(Browne & Finkelhor, 1986; Gomes-Schwartz, Horowitz, & Cardarelli,
1990). Other cross-sectional studies, however, have shown that variability in
symptoms cannot, or only partly, be explained by abuse characteristics (Bal,

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

1392 JOURNAL OF INTERPERSONAL VIOLENCE / November 2005

Crombez, De Bourdeaudhuij, & Van Oost, 2004; Chaffin, Wherry, & Dykman,
1997; Johnson & Kenkel, 1991; Spaccarelli, 1995). In addition, some studies
have suggested that older children, at initial assessment, displayed more
symptoms than younger children, whereas other studies did not find any rela-
tionship between age at initial assessment and symptom variability (Bal, Van
Oost, De Bourdeaudhuij, & Crombez, 2003; Einbender & Friedrich, 1989;
Kendall-Tackett et al., 1993; Wolfe, Gentile, & Wolfe, 1989). Recently, there
is a growing consistency among empirical cross-sectional studies that the
consequences of sexual abuse are dependent on factors such as appraisals,
coping, and social support (Spaccarelli & Fuchs, 1997; Tremblay et al.,
1999; Ullman, & Filipas, 2001). The observation by some researchers that
trauma symptomatology abates with time could be related to the influence of
these variables. However, we could find no study examining this idea.
Follow-up studies focused on some of these influencing variables. In the
majority of previous studies, abuse-related factors (e.g., proximity to the
abuser, frequency and duration of the abuse, severity of the acts) were not
found to be predictive of children’s or adolescents’adjustment following sex-
ual abuse (Dubowitz, Black, Harrington, & Verschoore, 1993; Manion et al.,
1998; Mannarino et al., 1991; Tebbutt et al., 1997). Some evidence was found
for the predictive value of negative appraisals, symptomatology at initial
assessment, and social support. In their study on 56 sexually abused children
and adolescents, Manion et al. (1998) found that negative appraisals of guilt
and blame predicted emotional functioning at 12 months postdisclosure.
Also, and consistent with the results of Freedman, Brandes, Peri, and Shalev
(1998), the authors show that internalizing symptoms, such as dissociation,
depression and anxiety, and 3-month postdisclosure, are associated with
more symptomatology at 12 months postdisclosure. They suggest that early
internalizing symptoms and their concomitant negative appraisals of the
traumatic event may reduce the victim’s ability to recover (Freedman et al.,
1998). Social support also has a predictive value in recovery after sexual
abuse (see Kendall-Tackett et al., 1993). Lynskey and Fergusson (1997) iden-
tified factors that discriminated sexually abused youngsters who developed a
psychiatric disorder or adjustment difficulties from sexually abused young-
sters who did not develop such problems. Results showed that parental sup-
port was an important factor protecting against the development of adjust-
ment difficulties. With increasing reports of support, affection, and nurture,
the occurrence of later adjustment difficulties decreased. This is consistent
with Joseph (1999) who argued that crisis support immediately after the trau-
matic event is influential on later functioning. In his study on adolescent
survivors of a ship disaster, greater direct crisis support was predictive of
fewer feelings of depression and anxiety 18 months later.

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

 Bal et al. / PREDICTORS OF SYMPTOMATOLOGY 1393

In a previous study, Bal et al. (2004) tested a theoretical model to explain

symptom variety after initial disclosure of sexual abuse in 100 Flemish ado-
lescents. Thirty-five percent reported clinically significant trauma-specific
symptoms, 47% reported clinically significant internalizing problems, and
28% reported clinically significant externalizing problems. Results showed
that more negative appraisals, use of avoidant coping strategies, and lack of
crisis support contributed to more internalizing as well as more externalizing
postdisclosure trauma symptomatology. The buffering role of crisis support
and the role of negative appraisals, which predominated on coping strategies,
appeared to contribute largely to symptom variability. As these factors have a
predictive value on postdisclosure trauma symptomatology (Time 1), the
question emerges whether these variables can predict symptomatology 6
months later (Time 2).
The present study is an attempt to further build on our earlier study. Until
now, we know of no study that has investigated the predictive value of all
these variables together on internalizing and externalizing trauma symptom-
atology 6 months after disclosure. The aim of this study is to focus on two
research questions: (a) What is the natural course of trauma-specific symp-
toms 6 months after disclosure? and (b) Are negative appraisals, coping strat-
egies, and crisis support predictive of internalizing and externalizing trauma
symptoms 6 months after disclosure?

METHOD

Participants and Procedures

At Time 1, 100 Flemish adolescents (87% girls) who initially disclosed
sexual abuse participated in the study. Age varied between 11 and 18 (M =
14.34, SD = 1.82). Sixty-three percent of the adolescents reported intra-
familial sexual abuse (the perpetrator was a family member), and 37%
reported extrafamilial sexual abuse (the perpetrator was an acquaintance or a
stranger). All adolescents were recruited in six Confidential Centers for
Child Abuse and Neglect and six Child Guidance Centers. The study was
approved by the ethical committees of Ghent University, each participating
Confidential Center on Child Abuse and Neglect and each participating
Child Guidance Center. Research at Time 1 took place at the center and was
carried out by the researcher. The adolescent was asked to fill in a standard-
ized test battery. The guiding therapist was present at the center during the
time of research. Inclusion and exclusion criteria in the study at Time 1 have
been described previously (Bal et al., 2004).

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

1394 JOURNAL OF INTERPERSONAL VIOLENCE / November 2005

At the end of the assessment process at Time 1, every adolescent was

asked to participate in a second part of the study 6 months later. No adoles-
cent refused and each filled out a second informed consent form. Six months
after the initial assessment, the researcher contacted the adolescent at home.
Researches took place at the adolescents’ home (16 participants) or at the
center where the initial assessment was carried out (in 49 cases). Similar to
Time 1, the adolescent was asked to fill out a standardized test battery.
Of the original sample of 100 sexually abused adolescents, 65 were reas-
sessed at 6 months after initial disclosure (Time 2). The group consisted of
85% girls, with a mean age of 14.38 (SD = 1.72). Fifty-one percent of them
were between 11 and 14 years old. Sixty-two percent were intrafamilially
abused, and 38% were extrafamilially abused. Thirty-five adolescents dropped
out. Of this group, 27 adolescents declined to participate and 8 adolescents
could not be found after the initial evaluation. Dropout analyses did not show
significant differences between these 35 adolescents and the remaining 65
adolescents on Time 1: internalizing trauma symptoms, t(98) = .92, p = .36,
externalizing trauma symptoms, t(98) = .04, p = 97, active coping, t(98) =
.79, p = .43, avoidance, t(98) = .64, p = .52, Negative Self-Evaluations
Related to One’s Character and Sexuality (NES), t(98) = 1.81, p = .07, Nega-
tive Evaluations and Critical Appraisals of the Character and Trustworthi-
ness of Others (NEO/C), t(98) = 1.00, p = .32, and harm, t(98) = 1.80, p = .07
(see Table 1).

Instruments
Three of the four questionnaires, namely the Negative Appraisals of Sexual
Abuse Scale (NASAS), the Crisis Support Scale (CSS), the How I Cope Under
Pressure Scale (HICUPS), and the Trauma Symptom Checklist for Children
were translated into Dutch according to a standardized back-translation pro-
tocol (Saito, Nomura, Noguchi, & Tezuka, 1996).

Severity of the abuse. Based on Draijer (1990), a short semistructured inter-

view was developed to assess duration and severity of the abusive acts. Accord-
ing to Draijer, the severity of abusive acts was divided into four levels: (a)
objectionable kissing and sexual contact on the clothes, (b) sexual touching
using exposed genitals and/or breasts, (c) attempts at sexual intercourse and/or
demand of masturbation, and (d) forms of penetration. This information was
obtained in a semistructured interview with the adolescent and/or parent(s).

NASAS. The NASAS (Spaccarelli, 1995) is a 56-item (eight subscales)

self-report questionnaire. Based on the theory and research of Spaccarelli

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

 TABLE 1: Means and Standard Deviations on Time 1 and Time 2

Characteristics
of Sample Characteristics Characteristics
Lost to of Follow-Up of Follow-Up
Follow-Up at Sample at Sample at
Time 1 (n = 35) Time 1 (n =65) Time 2 (n = 65)
M SD M SD M SD

Symptomatology Internalizing trauma symptoms 51.26 19.71 47.45 19.89 40.66 20.70
Externalizing trauma symptoms 14.06 7.72 13.98 9.12 13.95 8.55
Coping Active coping 55.20 8.00 53.45 11.72 53.66 9.31
Avoidance 33.74 4.31 33.02 5.92 32.55 5.63
Negative appraisals NES 33.23 9.09 29.75 9.21 27.69 8.13
NEO/C 28.26 9.41 26.48 7.92 26.70 9.28
HARM 52.00 15.25 46.78 14.28 45.86 14.19
Crisis support 35.69 6.82 34.83 7.15 35.05 7.11

NOTE: NES = Negative Self-Evaluations Related to One’s Character and Sexuality Scale; NEO/C = Negative Evaluations and Critical Appraisals of the Char-
acter and Trustworthiness of Others; HARM = Perceptions of Physical Harm or Damage to Self, Important Others, and Important Relationships or Resources.
*p < .05. **p < .01. ***p < .001.

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

1395
1396 JOURNAL OF INTERPERSONAL VIOLENCE / November 2005

(1994, 1995; Spaccarelli & Fuchs, 1997) and for the purpose of structural
equation modeling in this study, subscales were subdivided into three scales:
(a) NES (e.g., “it was your fault”), (b) NEO/C (e.g., “someone you cared
about was disappointed with you”), and (c) Perceptions of Physical Harm or
Damage to Self, Important Others, and Important Relationships or Resources
(HARM; e.g., “someone you care about got hurt”). Questions of the NASAS
are answered on a 4-point Likert-type scale (1 = not at all, 4 = a lot). In
research, the questionnaire has proven good internal consistency (α = .96)
and good convergent and discriminant validity (see Spaccarelli, 1995).
Cronbach’s alpha values for the subscales in this study were (a) .85, (b) .85,
and (c) .93 for Time 1 and (a) .57, (b) .71, and (c) .78 for Time 2.

CSS. The CSS (Joseph, Andrews, Williams, & Yule, 1992) measures
received social support after a traumatic experience. On a 7-point Likert-type
scale, ranging from 1 = never to 7 = always, the CSS assesses the availability
of others, contact with other survivors, confiding in others, emotional sup-
port, practical support, negative response, and satisfaction with support. The
CSS has a maximum score of 49, with high scores indicating high levels of
support. Research showed an adequate internal reliability (Cronbach’s
alpha = .80) of the scale and good validity (Joseph, Williams, & Yule, 1992).
Internal reliability in this study for Time 1 was .74 and for Time 2 was .71.

HICUPS. The HICUPS (Ayers, Sandler, West, & Roosa, 1996) assesses
coping strategies that children and adolescents use when they have a specific
problem. On a 4-point Likert-type scale (1 = never to 4 = usually), the adoles-
cent has to indicate how often he or she has used a particular strategy in the
past month. This study focused on two higher order dimensions, derived
from factor analysis, from this questionnaire that in prior research was
related to children’s mental health problems of active and avoidant coping
(Sandler, Tein, & West, 1994). Active Coping consists of 18 items that refer
to problem solving (e.g., “you did something to make things better”) and pos-
itive cognitive restructuring (e.g., “you reminded yourself that you are better
off than other adolescents”). Avoidant Coping consists of 8 items that refer to
avoidant actions (e.g., “you avoided people who made you feel bad”), repres-
sion (e.g., “you tried to put it out of your mind”), and wishful thinking (e.g.,
“you daydreamed that everything was OK”). There are no norms available.
Research has indicated a good to sufficient internal consistency of the
HICUPS respectively: active coping (.84), avoidance coping (.78), and suffi-
cient validity (Spaccarelli & Fuchs, 1997). In this study, the reliability analy-
sis of the subscales demonstrated sufficient to high internal consistency;

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

 Bal et al. / PREDICTORS OF SYMPTOMATOLOGY 1397

Cronbach alpha values were: Time 1: (1) α = .85, (2) α = .85, (3) α = .93;
Time 2: (1) α = .57, (2) α = .71, (3) α = .78.

The Trauma Symptom Checklist for Children (TSCC). The TSCC (Briere,
1996) assesses trauma symptoms among adolescents. On a 4-point Likert-
type scale (1 = never to 4 = almost all the time), the adolescent indicates how
often a thought, feeling, or behavior occurs. The scale consists of 54 items
and 6 subscales. For this study, the subscales of Anxiety, Depression,
Posttraumatic Stress, and Dissociation are taken together as internalizing
trauma symptoms; the subscales of Anger and Sexual Problems are taken
together as externalizing trauma symptoms. The questionnaire has been fre-
quently used in research on trauma among adolescents, which confirmed the
good psychometric qualities of this questionnaire (Nader, 1997). The reliabil-
ity of the subscales within this research was good to very good. Cronbach’s
alphas for internalizing problems were .90 for Time 1 and .90 for Time 2, and
for externalizing problems were .73 for Time 1 and .65 for Time 2.

DATA ANALYSES

To examine differences in trauma-related symptoms, negative appraisals,

and coping and crisis support between Time 1 and Time 2, repeated measures
were used.
Hierarchical regression analyses were performed to determine the poten-
tial value of demographic and abuse-related variables (age, type of abuse,
and severity and duration of the abuse), negative appraisals, and coping and
crisis support at Time 1 in predicting, internalizing, and externalizing trauma
symptoms at Time 2.

RESULTS

Differences Between Initial Assessment (Time 1)

and Follow-Up (Time 2) on Trauma-Specific Symptoms,
Negative Appraisals, and Coping and Crisis Support

Multivariate analysis of variance showed main effects of time for internal-

izing symptoms, F(1, 63) = 10.35, p < .01, and NES, F(1, 64) = 5.09, p < .05.
Results showed that compared to Time 1, adolescents reported fewer inter-
nalizing symptoms and less NES. No significant differences between Time 1
and Time 2 were found for Externalizing Symptoms, HARM, NEO/C,

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

1398 JOURNAL OF INTERPERSONAL VIOLENCE / November 2005

Active Coping, and Avoidance and CSS (all F values < 1). Means and stan-
dard deviations are presented in Table 1.

Influence of Negative Appraisals and Coping and Crisis

Support at Time 1 on Trauma-Specific Symptoms at Time 2
We first examined zero-order correlations between the hypothesized pre-
dictors and each of the criterion variables (see Table 2).
Second, we evaluated two models of hierarchical regression analyses. In
the first hierarchical regression analysis (Model 1), demographic and abuse-
related variables were first entered as a block. Then, the two coping variables
(active and avoidant coping), the three negative appraisals variables (NES,
NEO/C, and HARM), and the measure of crisis support, all measured on
Time 1, were entered using a stepwise inclusion method. The dependent vari-
ables were internalizing and externalizing symptoms, which were measured
on Time 2. The results of the regression analyses are presented in Table 3.
Variance inflation factors (VIFs) were small (ranging from 1.033 to 1.459),
indicating that there was no problem of colinearity in the regression analyses
(Tabachnick & Fidell, 1996). Results showed that internalizing symptoms at
Time 2 are predicted by high perceptions of harm and less crisis support,
resulting in an explained variance of 27%. Externalizing symptoms were
only predicted by high perceptions of harm. Total explained variance was
16%.
In the second hierarchical regression analyses (Model 2), a third step was
added: Internalizing and externalizing symptoms measured at Time 1 were
entered using a stepwise inclusion method. Steps 1 and 2, as well as the
dependent variables, remained the same. VIFs were small (ranging from
1.084 to 1.855), indicating that there was no problem of colinearity. Results
showed that internalizing symptoms at Time 2 are predicted by less crisis
support and more internalizing symptoms at Time 1, resulting in an explained
variance of 38%. Externalizing symptoms were predicted by externalizing
symptoms at Time 1. Total explained variance was 38%.
These results suggest that the relationship between negative appraisals
and internalizing as well as externalizing symptoms at Time 2 could be medi-
ated by symptoms at Time 1 (see Model 2). Results in our previous study (Bal
et al., 2004) showed that the independent variables affected the mediator
variable (internalizing and externalizing symptoms at Time 1). Model 1 in
this study shows that the independent variables affected the dependent vari-
ables (internalizing and externalizing symptoms at Time 2). Finally in Model
2, it was shown that the mediator affected the dependent variables (see Model
2). According to Baron and Kenny (1986), mediation holds if the independ-

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

 Bal et al. / PREDICTORS OF SYMPTOMATOLOGY 1399

TABLE 2: Zero-Order Correlations Between Predictor and Criterion Variables in the

Study

Criterion Variables
Internalizing Externalizing
Predictor Variable Symptoms Time 2 Symptoms Time 2

Age .02 .02

Type of abuse .07 .22
Severity of act –.00 –.01
Duration of abuse .01 –.09
Active coping .10 .09
Avoidance .19 .20
NES .47*** .46***
NEO/C .48*** .33***
HARM .50*** .45***
Crisis support –.35** –.22
Internalizing symptoms Time 1 .61*** .51***
Externalizing symptoms Time 1 .48*** .63***

NOTE: NES = Negative Self-Evaluations Related to One’s Character and Sexuality Scale; NEO/
C = Negative Evaluations and Critical Appraisals of the Character and Trustworthiness of Oth-
ers; HARM = Perceptions of Physical Harm or Damage to Self, Important Others, and Important
Relationships or Resources Scale.
**p < .01. ***p < .001.

ent variables have no significant effect when the mediator is controlled. Cri-
sis support has, in both models, a direct influence on internalizing symptoms
at Time 1 and Time 2.

DISCUSSION

The purpose of this study was to examine the natural course of internaliz-
ing and externalizing trauma-specific symptoms 6 months after disclosure of
sexual abuse and to investigate whether variables such as negative appraisals,
coping strategies, social support, and trauma symptomatology measured at
the time of disclosure were predictive of trauma symptomatology 6 months
later.
First, results showed that 46% of the adolescents, compared to 53% at dis-
closure, persisted to report clinically significant trauma symptoms 6 months
later. Twenty-eight percent of the adolescents reported clinically significant
internalizing symptoms, and 45% reported clinically significant externalizing
symptoms on Time 2. Conforming to the results of other studies, these find-

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

 1400
TABLE 3: Results of Hierarchical Regression Analyses of Internalizing and Externalizing Symptoms

Model 1 Model 2
2 2
Criterion Variable Step Predictor Beta ∆r Adjusted r Beta ∆r2 Adjusted r2

Internalizing symptoms 1 Age .02 .01 –.07 .03 .01 –.07

Type of abuse .10 .12
Severity of act –.10 –.07
Duration of abuse .04 .12
2 Harm .45*** .26 .20 .16 .26 .20
Crisis support –.31** .08 .27 –.26* .08 .27
3 Internalizing symptoms — — — .45** .11 .38
Externalizing symptoms 1 Age .01 .04 –.03 .10 .04 –.03
Type of abuse .17 .14
Severity of act –.01 –.02
Duration of abuse –.04 –.02
2 Harm .44*** .19 .16 .20 .19 .16
3 Externalizing symptoms — — — .52*** .21 .38

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

*p < .05. **p < .01. ***p < .001.
 Bal et al. / PREDICTORS OF SYMPTOMATOLOGY 1401

ings show that sexual abuse is an important problem with serious long-term
sequelae (Beitchman et al., 1992; Lynskey & Fergusson, 1997).
Notwithstanding the fact that nearly half of all adolescents still reported
clinically significant trauma symptoms, results showed a significant decrease
of internalizing symptoms, such as anxiety, depression, posttraumatic stress
complaints, and dissociative problems at a 6 month follow-up. These results
are consistent with the results of previous follow-up studies that observed
how trauma symptoms abated with time (Gomes-Schwartz et al., 1990;
Mannarino et al., 1991; Oates et al., 1994). For the total group, externalizing
trauma symptoms did not significantly change with time.
To explain these variations in adolescents’ trauma symptomatology, the
second aim of our study was to examine the predictive value of severity and
type of abuse, negative appraisals, coping strategies, and crisis support on
trauma symptomatology 6 months after disclosure. Results showed that
demographic and abuse-related variables did not account for differences in
internalizing or externalizing symptomatology. Apart from the influence of
initial trauma symptomatology, results of our first model suggested that high
perceptions of harm and a lack of social support predicted more internalizing
symptoms 6 months after disclosure. More externalizing symptoms, 6 months
later, were predicted by high perceptions of harm. Previous cross-sectional
studies with sexually abused adolescents found a direct relationship between
these influencing variables and postdisclosure trauma symptomatology (Bal
et al., 2004; Spaccarelli & Fuchs, 1997). Follow-up studies focused more on
the relationship between initial and follow-up symptomatology. Freedman
et al. (1998), Manion et al. (1998), and Tebbutt et al. (1997) found that
depression and anxiety at initial assessment were predictive of posttraumatic
stress symptoms 1 to 5 years later. However, these studies did not look at
other influencing variables nor control postdisclosure symptomatology.
Therefore, in our second model, we added postdisclosure symptomatology
to our regression analyses. Results showed that internalizing symptoms at
disclosure, as well as a lack of crisis support, were directly predictive of inter-
nalizing symptoms 6 months later and that externalizing symptoms at Time 1
were predictive of externalizing symptoms at Time 2. Thus, the direct influ-
ence of perceptions of harm disappeared. Hence, we suggest that postdisclo-
sure symptomatology and crisis support are predictive of trauma symptom-
atology at 6 months follow-up. Moreover, our study suggests that initial
trauma symptomatology could mediate the relationship between perceptions
of harm and outcomes in symptomatology 6 months later. The fact that
results of our study showed a decrease in negative self-evaluations could be
related to a decrease in internalizing symptoms at follow-up.

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

1402 JOURNAL OF INTERPERSONAL VIOLENCE / November 2005

Social support remained a direct predictor for internalizing symptomatology

6 months later. These results are in line with Lynskey and Fergusson (1997),
who found that with increasing reports of support, rates of later adjustment
difficulties decreased among sexually abused youngsters. Oates et al. (1994)
indicated that the environmental system around the victim influences the way
a child responds to a trauma. In his study, children’s outcomes after 4 months
were related to the extent to which the family was able to respond to the
trauma. Consistent with our findings, this may suggest that supportive rela-
tionships could mitigate the impact of sexual abuse.
There are a number of limitations that must be considered. First, the sam-
ple was obtained through a respondent’s contacts with mental health profes-
sionals, and as a result of this, all the participating adolescents received treat-
ment of some type. Prevalence rates of sexual abuse in nonclinical groups
showed that among adolescents, nearly 15% of girls and 6% of boys reported
sexual abuse (Finkelhor & Dziuba-Leatherman, 1994. This indicates that
there are many sexually abused adolescents who do not come in contact with
professional help. Hence, our findings could not be generalized to these sex-
ually abused adolescents. Second, the results are limited by the relatively
short time period between disclosure and follow-up. Longitudinal studies are
needed to further explore the influence of specific variables.
Most of the studies on long-term adverse effects of sexual abuse focused
mainly on the course of symptomatology. In contrast, our study also investi-
gated whether variables that influenced internalizing and externalizing
symptomatology postdisclosure still had a predictive value 6 months later.
One of the strengths of this study was that it was carried out with a relatively
large group of sexually abused adolescents who, postdisclosure as well as 6
months later, were examined with a standardized assessment procedure.
In conclusion, our study showed that 46% of the adolescents persisted to
report clinically significant trauma symptoms 6 months after disclosure.
Although improvements in internalizing symptoms and in negative evalua-
tions of oneself were noted, externalizing symptoms remained the same. The
type or severity of the abuse did not account for differences in symptomatology,
suggesting a high risk for all sexually abused adolescents. Predictors of
ongoing trauma symptomatology occurred to be postdisclosure trauma
symptomatology and a lack of crisis support following disclosure of the
abuse. Regardless of the severity or type of abuse, thorough assessment of
trauma symptomatology, crisis support, and negative appraisals at the moment
of disclosure are critical in understanding which abused adolescents are most
at risk for poor outcomes.

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

 Bal et al. / PREDICTORS OF SYMPTOMATOLOGY 1403

REFERENCES

Ayers, T. S., Sandler, I. N., West, S. G., & Roosa, M. W. (1996). A dispositional and situational
assessment of children’s coping: Testing alternative models of coping. Journal of Personal-
ity, 64, 923-958.
Bal, S., Crombez, G., De Bourdeaudhuij, I., & Van Oost, P. (2004). Symptomatology in adoles-
cents following initial disclosure of sexual abuse: The roles of crisis support, appraisals and
coping. Manuscript submitted for publication.
Bal, S., Crombez, G., Van Oost, P., & De Bourdeaudhuij, I. (2003). The role of social support in
well-being and coping with self-reported stressful events in adolescents. Child Abuse &
Neglect, 27, 1377-1395.
Bal, S., Van Oost, P., De Bourdeaudhuij, I., & Crombez, G. (2003). Avoidant coping as a media-
tor between self-reported sexual abuse and stress-related symptoms in adolescence. Child
Abuse & Neglect, 27, 883-897.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psy-
chological research: Conceptual, strategic, and statistical considerations. Journal of Person-
ality and Social Psychology, 51, 1173-1182.
Beitchman, J. H., Zucker, K. J., Hood, J. E., DaCosta, G. A., Akman, D., & Cassavia, E. (1992). A
review of the long-term effects of child sexual abuse. Child Abuse & Neglect, 16, 101-118.
Briere, J. (1996). Trauma Symptom Checklist for Children (TSCC). Odessa, FL: Psychological
Assessment Resources.
Browne, A., & Finkelhor, D. (1986). Impact of child sexual abuse: Review of the research. Psy-
chological Bulletin, 99, 66-77.
Chaffin, M., Wherry, J. N., & Dykman, R. (1997). School age children’s coping with sexual
abuse: Abuse stresses and symptoms associated with four coping strategies. Child Abuse &
Neglect, 21, 227-240.
Draijer, N. (1990). Seksuele traumatisering in de jeugd: Gevolgen op lange termijn van seksueel
misbruik van meisjes door verwanten [Sexual trauma in youth: Long term effects of sexual
abuse in girls by kins]. Amsterdam: SUA (Academisch Proefschrift).
Dubowitz, H., Black, M., Harrington, D., & Verschoore, A. (1993). A follow-up study of behav-
ior problems associated with child sexual abuse. Child Abuse & Neglect, 17, 743-754.
Einbender, A. J., & Friedrich, W. N. (1989). Psychological functioning and behavior of sexually
abused girls. Journal of Consulting and Clinical Psychology, 57, 155-157.
Finkelhor, D. & Dziuba-Leatherman, J. (1994). Children as victims of violence: A national sur-
vey. Pediatrics, 94, 413-420.
Freedman, S. A., Brandes, D., Peri, T., & Shalev, A. (1998). Predictors of post traumatic stress
disorder. British Journal of Psychiatry, 174, 353-359.
Gomes-Schwartz, B., Horowitz, J. M., & Cardarelli, A. P. (1990). Child sexual abuse: The initial
effects. Sage library of social research, 179.
Gomes-Schwartz, B., Horowitz, J. M., Cardarelli, A. P., & Sauzier, M. (1990). The aftermath of
child sexual abuse: 18 months later. In B. Gomes-Schwartz, J. M. Horowitz, & A. P.
Cardarelli (Eds.), Child sexual abuse: The initial effects (pp. 132-152). Newbury Park, CA:
Sage.
Johnson, B. K., & Kenkel, M. B. (1991). Stress, coping, and adjustment in female adolescent
incest victims. Child Abuse & Neglect, 15, 293-305.
Joseph, S. (1999). Social support and mental health following trauma. In W. Yule (Ed.),
Posttraumatic stress disorders, concepts and therapy (pp. 71-91). New York: John Wiley &
Sons.

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

1404 JOURNAL OF INTERPERSONAL VIOLENCE / November 2005

Joseph, S., Andrews, B., Williams, R., & Yule, W. (1992). Crisis support and psychiatric
symptomatology in adult survivors of the Jupiter cruise ship disaster. British Journal of Clin-
ical Psychology, 31, 63-73.
Joseph, S., Williams, R., & Yule, W. (1992). Crisis support, attributional style, coping style, and
post-traumatic symptoms. Personality and Individual Differences, 13, 1249-1251.
Kendall-Tackett, K. A., Williams, L. M., & Finkelhor, D. (1993). Impact of sexual abuse on chil-
dren: A review and synthesis of recent empirical studies. Psychological Bulletin, 113, 164-
180.
Lynskey, M. T., & Fergusson, D. M. (1997). Factors protecting against the development of
adjustment difficulties in young adults exposed to childhood sexual abuse. Child Abuse &
Neglect, 21, 1177-1190.
Manion, I., Firestone, P., Cloutier, P., Ligezinska, M., McIntyre, J., & Ensom, R. (1998). Child
extrafamilial sexual abuse: Predicting parent and child functioning. Child Abuse & Neglect,
22, 1285-1304.
Mannarino, A. P., Cohen, J. A., Smith, J. A., & Moore-Motily, S. (1991). Six- and twelve-month
follow-up of sexually abused girls. Journal of Interpersonal Violence, 6, 494-511.
Nader, K. O. (1997). Assessing traumatic experiences in children. In J. P. Wilson & T. M. Keane
(Eds.), Assessing psychological trauma and PTSD (pp. 291-348). New York: Guilford.
Oates, R. K., O’Toole, B. I., Lynch, D. L., Stern, A., & Cooney, G. (1994). Stability and change in
outcomes for sexually abused children. Journal of the American Academy of Child and Ado-
lescent Psychiatry, 33, 945-953.
Saito, S., Nomura, N., Noguchi, Y., & Tezuka, I. (1996). Translatability of family concepts into
the Japanese culture: Using the Family Environment Scale. Family Process, 35, 239-257.
Sandler, I. N., Tein, J. Y., & West, S. G. (1994). Coping, stress, and psychological symptoms of
divorce: A cross-sectional and longitudinal study. Child Development, 65, 1744-1763.
Spaccarelli, S. (1994). Stress, appraisal, and coping in child sexual abuse: A theoretical and
empirical review. Psychological Bulletin, 116, 340-362.
Spaccarelli, S. (1995). Measuring abuse stress and negative cognitive appraisals in child sexual
abuse: Validity data on two new scales. Journal of Abnormal Child Psychology, 23, 703-727.
Spaccarelli, S., & Fuchs, C. (1997). Variability in symptom expression among sexually abused
girls: Developing multivariate models. Journal of Clinical Child Psychology, 26, 24-35.
Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York:
HarperCollins College.
Tebbuttt, J., Swanston, H., Oates, R. K., & O’Toole, B. I. (1997). Five years after child sexual
abuse: Persisting dysfunction and problems of prediction. Journal of the American Academy
of Child and Adolescent Psychiatry, 36, 330-339.
Tremblay, C., Hébert, M., & Piché, C. (1999). Coping strategies and social support as mediators
of consequences in child sexual abuse victims. Child Abuse & Neglect, 23, 929-945.
Ullman, S. E., & Filipas, H. H. (2001). Predictors of PTSD symptom severity and social reactions
in sexual assault victims. Journal of Traumatic Stress, 14, 369-389.
Wolfe, V. V., & Birt, J. (1997). Child sexual abuse. In E. J. Mash & L. G. Terdal (Eds.), Assess-
ment of childhood disorders (pp. 596-605). New York: Guilford.
Wolfe, V. V., Gentile, C., & Wolfe, D. A. (1989). The impact of sexual abuse on children: A PTSD
formulation. Behavior Therapy, 20, 215-228.

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016

 Bal et al. / PREDICTORS OF SYMPTOMATOLOGY 1405

Sarah Bal, Ph.D., received her M.S. and Ph.D. in psychology from the Ghent University.
She is currently working as a clinical child psychologist and as a researcher at the child
psychiatry department of the University Hospital, Ghent. Her research interests include
fear and anxiety disorders in children and adolescents.

Ilse De Bourdeaudhuij, Ph.D., received her M.S. and Ph.D. in psychology from the Ghent
University. She is currently associate professor of health psychology at the Faculty of
Medicine and Health Sciences, Ghent University. Her research interests are health pro-
motion, determinant studies, and intervention evaluation related to nutrition and physi-
cal activity.

Geert Crombez, Ph.D., received his M.S. and Ph.D. in psychology from the Catholic Uni-
versity of Leuven. He is currently full professor of health psychology at Ghent University.
His research interests include the psychology of pain, fatigue, obesity, fear, and anxiety
disorders.

Paulette Van Oost, Ph.D., received her M.S. and Ph.D. in psychology from the Ghent Uni-
versity. She is currently full professor of clinical psychology at Ghent University. Her
research interests include the psychology of peer aggression and violent behavior.

Downloaded from jiv.sagepub.com at PENNSYLVANIA STATE UNIV on September 20, 2016