Childhood Maltreatment and Difficulties in Emotion Regulation - Associations with Sexual and Relationship ...

[ Pobierz całość w formacie PDF ]
JOURNAL OF SEX RESEARCH, 49(5), 434–442, 2012
Copyright # The Society for the Scientific Study of Sexuality
ISSN: 0022-4499 print
=
1559-8519 online
DOI: 10.1080/00224499.2011.565430
Childhood Maltreatment and Difficulties in Emotion Regulation: Associations with
Sexual and Relationship Satisfaction among Young Adult Women
Alessandra H. Rellini
Department of Psychology, University of Vermont
Anka A. Vujanovic
National Center for PTSD–Behavioral Science Division, VA Boston Healthcare System,
Boston University School of Medicine
Myani Gilbert and Michael J. Zvolensky
Department of Psychology, University of Vermont
This study examined relations among childhood maltreatment, difficulties in emotion
regulation, and sexual and relationship satisfaction among young adult women reporting
current involvement in committed, romantic relationships. A sample of 192 women (ages
18–25) completed self-report questionnaires as part of an Internet-based survey. It was
hypothesized that severity of childhood maltreatment and difficulties in emotion regulation
would each independently and negatively predict (a) sexual satisfaction, (b) relationship
intimacy, and (c) expression of affection within the context of the relationship. Furthermore,
it was hypothesized that greater emotion regulation difficulties would moderate the effects of
childhood maltreatment on these sexual and relationship variables (i.e., sexual satisfaction,
relationship intimacy, and expression of affection). Findings suggest that difficulties in emo-
tion regulation demonstrated an incremental effect with regard to sexual satisfaction, but not
with intimacy and affection expression. In contrast to predictions, no significant interactive
effects were documented. Clinical implications and future directions related to this line of
inquiry are discussed.
There is a high prevalence of relational and sexual
problems among adult women exposed to childhood
maltreatment, defined as childhood sexual abuse,
physical abuse, or neglect (Davis, Petretic-Jackson, &
Ting, 2001; Fromuth, 1986; Loeb et al., 2002; Meston,
Heiman, Trapnell, & Carlin, 1999; Rellini & Meston,
2007; Scholerdt & Heiman, 2003). The most commonly
reported sexual and relationship problems for women
with a history of childhood maltreatment include
inhibited sexual desire, lower levels of sexual satisfac-
tion, difficulties becoming sexually aroused or reaching
orgasm, difficulties developing emotional intimacy with
a partner, and interpersonal aggression (DeSilva, 2001;
Lewis et al., 2010; Rellini & Meston, 2007).
As not all individuals with a history of childhood
maltreatment report sexual or relational problems in
adulthood (Leonard & Follette, 2002; Loeb et al.,
2002; Rellini & Meston, 2007), increasing our under-
standing of malleable (changeable) vulnerabilities
remains a fecund area of intellectual and clinical pursuit.
There is increasing evidence that emotion regulation
offers promise for advancing our understanding of
potential risk and maintenance factors for sexual and
relational problems (Rellini, Vujanovic, & Zvolensky,
2010). Emotion regulation is a promising construct in
this context; it encapsulates an individual’s understand-
ing and regulation of emotional responses (Cole,
Michel, & Teti, 1994; Mennin, 2004; Salovey, Mayer,
Goldman, Turvey, & Palfai, 1995). Gratz and Roemer
(2004) developed an empirically grounded assessment
instrument entitled the Difficulties in Emotion Regula-
tion Scale (DERS), which measures emotional
regulation from a multidimensional framework. The
DERS assesses several facets of emotion regulation,
including difficulties relevant to an individual’s (a)
acceptance of emotional responses, (b) ability to engage
This research was supported by the McNeil Prevention and
Community Psychology Fund and the Undergraduate Research
Endeavour Competitive Award. The views expressed here are those
of the authors and do not necessarily represent those of the Depart-
ment of Veterans Affairs or the McNeil Fund. We thank Sarah P.
Roberts for her substantial contribution with data collection.
Correspondence should be addressed to Alessandra H. Rellini,
Department of Psychology, University of Vermont, John Dewey Hall,
2 Colchester Ave., Burlington, VT 05405. E-mail: arellini@uvm.edu
EMOTION DYSREGULATION AND SEXUALITY
in goal-directed behavior under distress, (c) ability to
control impulsive behaviors when distressed, (d) aware-
ness of emotional experiences, (e) access to emotion
regulation strategies, and (f) emotional clarity.
It has been theorized that emotion regulation difficul-
ties are the product of the interaction between biological
vulnerabilities and environmental factors (Linehan,
1993; Thompson, 1994). As part of the environmental fac-
tors, an invalidating, abusive, or neglectful childhood
environment wherein children do not learn adaptive ways
of coping with aversive emotions has been hypothesized
as a risk factor for the development of emotion dysregula-
tion (Linehan, 1993). Such emotion regulatory difficulties
may be formed and
=
or maintained via social learning
contexts and relationships in adolescence and early adult-
hood (Kim, Pears, Capaldi, & Owen, 2009). It is possible
that greater levels of emotion regulation difficulties may
exacerbate the effects of childhood maltreatment on sex-
ual and relationship problems in adulthood (Rellini,
2008). For example, women with histories of child mal-
treatment may be more likely to manifest sexual and
relationship problems if they also experience high levels
of difficulties regulating emotions. In contrast, emotion
regulation skillsmay decrease the negative effects of child-
hood maltreatment in terms of sexual and relationship
problems (e.g., Rellini et al., 2010).
There is a dearth of empirical literature devoted to
the examination of associations between emotion regu-
lation and sexual and relationship functioning. In the
only published empirical study to date, Rellini and
colleagues (2010) documented significant (negative)
incremental associations between emotion regulation
difficulties, as indexed by the DERS, and sexual satis-
faction, even after controlling for theoretically relevant
variables (i.e., posttraumatic stress symptom severity,
negative affectivity, anxiety sensitivity, and daily ciga-
rette consumption). This investigation relied on a
relatively small sample of 43 men and women, with a
history of varied types of trauma exposure. The sample
also was limited in that it focused exclusively on smo-
kers. There is a need to replicate and extend such work
to a more diverse population in a number of specific
ways. First, it is important to replicate the previously
documented findings among larger samples of indivi-
duals who are currently engaged in committed roman-
tic relationships independently from their smoking
behavior, so as to increase the generalizability of
results and improve accuracy of relationship function-
ing reports. Second, since men and women have been
shown to manifest unique sexual and relationship func-
tioning problems (Lewis et al., 2010; Parish et al.,
2007), it is important to examine these relations using
a gender-specific within-subjects design so as to attain
a clearer understanding of variables among men and
women, independently. Third, Rellini et al. (2010)
documented varied types of trauma exposure among
participants, thus impeding specific interpretations of
the effects of certain types of trauma (e.g., childhood
maltreatment).
Together, this investigation sought to replicate and
extend extant findings among a sample of young adult
women, currently involved in committed romantic rela-
tionships and reporting varying degrees of childhood
maltreatment. First, it was hypothesized that childhood
maltreatment and difficulties in emotion regulation
would be significantly (negatively) correlated with sexual
and relationship satisfaction. Second, it was hypothe-
sized that severity of childhood maltreatment and diffi-
culties in emotion regulation would each incrementally
(independently) and negatively predict (a) sexual satis-
faction, (b) relationship intimacy, and (c) expression of
affection within the context of the relationship over
and above the variance accounted for by age and
relationship duration. Third, it was hypothesized that
greater emotion regulation difficulties would interact
with the effects of childhood maltreatment on these sex-
ual and relationship variables (i.e., sexual satisfaction,
relationship intimacy, and expression of affection). This
prediction was driven by the perspective that childhood
maltreatment may, theoretically, have the most deleteri-
ous interpersonal effects when co-occurring with greater
difficulties in regulating emotional states.
Method
Participants
Participants were 192 young adult women (M
age
¼
21.8, SD
¼
3.7; observed range
¼
18–25), recruited from
online advertisements posted on social networking
websites (see the Procedure section for details). Parti-
cipants were included in the study if they self-identified
as (a) female, (b) between the ages of 18 and 25, (c)
non-virgin, (d) U.S. citizen, and (e) English speaking. A
total of 596women completed the online screener between
June and October 2009, and 218 qualified and enrolled in
the study and started questionnaire completion. A total of
192 women completed all questionnaires. Therefore,
analyses were based on 88% (N
¼
192) of the women
who qualified for the study. No statistically significant
group differences were evident between women who com-
pleted the study and women who completed only the first
demographics questions (n
¼
26) in terms of age, edu-
cation level, sexual orientation, relationship length, stu-
dent status, or race and ethnicity.
As illustrated in Table 1, participants were predomi-
nantly Caucasian women (90%), who had completed at
least some college (89%). Based on the Kinsey scale
(Kinsey, Pomeroy, & Martin, 1948), the sample reported
large variability in terms of sexual orientation, with
approximately 26% not identifying as exclusively or pre-
dominantly heterosexual. With regard to relationship
status at the time of the study, approximately 60%
435
RELLINI, VUJANOVIC, GILBERT, AND ZVOLENSKY
Table 1. Demographic and Relationship Characteristics for
All Participants in the Sample
(n
¼
116) of participants reported being involved in a
committed relationship or married, whereas the remain-
ing 41% (n
¼
79) of participants reported either dating
one partner or being in a relationship that was not com-
mitted. A small minority of participants (9%; n
¼
17)
reported being in a relationship with a woman. The mean
relationship length reported was 3.7 years (SD
¼
2.3).
One hundred-three participants (54%) reported at least
one type of childhood maltreatment, and 10 (5%) scored
above the normative cutoff for all five types of childhood
maltreatment (i.e., emotional neglect, physical neglect,
emotional abuse, physical abuse, and sexual abuse), as
measured by the Childhood Trauma Questionnaire
(CTQ; see the Measures section for details; Bernstein,
Fink, Handelsman, & Foote, 1994). Twenty-four parti-
cipants (13%) scored above the normative cutoff for sex-
ual abuse. Table 2 illustrates more details on frequency of
childhood maltreatment in the sample. Co-occurrence of
types of abuse was common, with only eight participants
(4%) meeting criteria exclusively for sexual abuse, six
participants (3.1%) meeting criteria exclusively for physi-
cal abuse, and none meeting criteria exclusively for
emotional abuse or emotional neglect.
Variable
n
%
Education
High school graduate or less
25
13.0
Some college or more
171
89.0
Currently a student
Yes
102
53.1
Ethnicity
African American or Black
6
3.1
Hispanic or Latin American
11
5.7
Asian or Asian American
7
3.6
Native Hawaiian or Pacific Islander
1
0.5
American Indian or Alaska Native
5
2.6
Euro-American or White
173
90.1
Multiracial
11
5.7
Other
3
1.5
Sexual orientation (Kinsey scale)
0
93
48.4
1
49
25.5
2
20
10.4
3
10
5.2
4
6
3.1
5
5
2.6
6
6
3.1
Other
7
3.6
Relationship status
Single
=
dating
79
41.1
Measures
Married
=
committed relationship
116
60.4
Length of current relationship
0–6 months
Demographics. Questions about age, education (i.e.,
high school graduate or less vs. some college or more),
student status (i.e., currently a student: yes or no), and
ethnicity (i.e., African American or Black, Hispanic or
Latin American, Asian or Asian American, Native
Hawaiian or Pacific Islander, American Indian or Alaska
Native, Euro-American or White, multiracial, and
‘‘other’’) were asked to characterize the sample. Addition-
ally, household income was measured separately for
people self-identifying as dependent or non-dependent
for tax purposes. Finally, the Kinsey scale (Kinsey et al.,
1948) was used to assess sexual orientation. This is a
seven-point (0–6) Likert scale that conceptualizes sexual
orientation as ranging continuously from exclusively
heterosexual (0) to exclusively homosexual (6). Relation-
ship characteristics measured in the study include
relationship status (dating, in a non-committed rela-
tionship, or married
=
committed relationship), length of
relationship (months), and partner’s gender.
45
23.4
6–12 months
30
15.6
1–2 years
31
16.1
3–5 years
39
20.3
5–10 years
6
3.1
Gender of current partner
Male
151
78.6
Female
17
8.9
MSD
Relationship length (years)
3.7
2.3
Annual income
=
year for non-dependent
a
$34,700
$21,600
Family income
=
year for dependent or student
b
$105,000
$83,700
a
Participants self-identified as non-students and non-dependent based
on taxes filed the previous year.
b
Participants self-identified as either full-time students or dependent
based on taxes filed the previous year.
Table 2. Childhood Maltreatment Scores
n
>
cutoff; n (%)
a
Trauma Type
M
SD
Minimum–Maximum
Cutoff
CTQ-total
40.5
15.6
25–96
—
—
Emotional neglect
9.9
4.5
5–25
15
21 (10.9)
Emotional abuse
10.0
4.8
5–24
10
51 (26.6)
Physical neglect
6.7
2.7
5–19
8
27 (14.1)
Physical abuse
7.2
3.7
5–23
8
27 (14.1)
Sexual abuse
7.0
4.7
5–25
8
24 (12.5)
Note. CTQ
¼
Childhood Trauma Questionnaire (Bernstein, Fink, Handelsman, & Foote, 1994).
a
Cutoff based on specificity and sensitivity scores (Walker et al., 1999).
436
 EMOTION DYSREGULATION AND SEXUALITY
CTQ (Bernstein et al., 1994). The CTQ is a 28-item
self-report measure on which respondents indicate,
using a five-point Likert scale ranging from 1 (never
true)to5(very often true), their childhood experiences
of maltreatment—a construct indexed along five domains:
(a) Emotional Abuse, (b) Physical Abuse, (c) Sexual
Abuse, (d) Physical Neglect, and (e) Emotional Neglect
(Bernstein et al., 1994). The test–retest reliability of the
CTQ, over a two- to six-month period, has ranged from
r
¼
.79 to r
¼
.95; and convergent and discriminant val-
idity was determined to be adequate (Bernstein et al.,
1994). Overall internal consistency of the entire measure
was high (a
¼
.91; Scher, Stein, Asmundson, McCreary,
& Forder, 2001). A cutoff for each domain was used based
on findings from Walker et al. (1999), indicating that a
score of 8 for Physical Abuse, Physical Neglect, and Sex-
ual Abuse; a score of 10 for Emotional Abuse; and a score
of 15 for Emotional Neglect were sensitive and specific
scores to accurately identify individuals with these specific
types of abuse as assessed by experts in the field of trauma
(Walker et al., 1999). These cutoffs were utilized for a
description of the childhood maltreatment reported by
the sample. The continuous measure for severity of abuse
(CTQ) was the main variable in the analyses. In this study,
internal consistency was a
¼
.88, a
¼
.95, and a
¼
.90 for
Physical, Sexual, and Emotional Abuse, respectively;
and a
¼
.87 and a
¼
.77 for Emotional and Physical
Neglect, respectively. Consistent with past work (Rellini
et al., 2010), the CTQ-total score (logarithmic transform-
ation) was utilized as a measure of childhood maltreat-
ment severity in the analyses.
emotional issues (Communication); (b) compatibility
between partners in terms of sexual beliefs, preferences,
desires, and attraction (Compatibility); (c) contentment
with emotional and sexual aspects of the relationship (Con-
tentment), (d) personal distress concerning sexual prob-
lems (Personal Distress); and (e) distress regarding the
impact of sexual problems on partners and relationships
at large (Relational Distress). The five domains have
shown acceptable internal consistency (Cronbach’s
a
.74) and test–retest reliability after four to five weeks
(r
¼
.58
79). The SSS reliably differentiated between
women with and without sexual dysfunction on each of
the domains and total scores (Meston & Trapnell, 2005).
In this study, Cronbach’s alpha for the total score was .80.
Network of Relationships Inventory: Social Provisions
Version–Revised (NRI–SPV–R; Furman & Buhrmester,
2009). This version of the NRI–SPV–R comprises 36
items on which respondents indicated, using a five-point
Likert-type scale ranging from 1 (little or none)to5(the
most), their levels of relationship satisfaction as indexed
by 10 factors (Furman & Buhrmester, 2009). Based on
research suggesting that the association between sexual
function and emotion regulation may be mediated by dif-
ficulties in developing emotional connections with part-
ners (Fruzzetti & Iverson, 2004; Linehan, 1993; Rellini
et al., 2010), only the Intimacy and Affection factors were
considered for this study. The Intimacy factor was com-
prised of items relevant to how much the person shares
with her partner (e.g., ‘‘How much do you share your
secrets and private feelings with this romantic partner?’’).
The Affection factor included items related to how much
the person feels loved by her partner (e.g., ‘‘How much
does this romantic partner have a strong feeling of affec-
tion [loving or liking] toward you?’’). This scale can be
used to assess the quality of a romantic relationship.
The scale has shown high internal consistency for the
domains of Intimacy (a
¼
.86) and affection (a
¼
.91) with
romantic partners (Furman & Buhrmester, 2009). In this
investigation, internal consistency was similarly high,
with a
¼
.89 for Intimacy and a
¼
.95 for Affection.
DERS (Gratz & Roemer, 2004). The DERS is a
36-item self-report measure on which respondents indi-
cate, on a five-point Likert-style scale ranging from 1
(almost never)to5(almost always), how often each item
applies to them (Gratz & Roemer, 2004). The DERS is
multidimensional in that it is comprised of 6 factors in
addition to a total score. These factors include (a) Non-
acceptance of Emotional Responses, (b) Difficulties
Engaging in Goal-Directed Behavior, (c) Impulse Con-
trol Difficulties, (d) Lack of Emotional Awareness
(Aware), (e) Limited Access to Emotion Regulation Stra-
tegies, and (f) Lack of Emotional Clarity (Clarity). The
DERS has high levels of internal inconsistency (a
¼
.93;
Gratz & Roemer, 2004) and adequate test–retest
reliability over a four- to eight-week period (r
¼
.88;
Gratz & Roemer, 2004). The DERS-total sum score
represents a global emotional dysregulation composite
(Gratz & Roemer, 2004), and, in this study, it showed
adequate internal consistency (Cronbach’s a
¼
.89).
Procedure
Participants were recruited using online advertise-
ments posted on social networking websites, such as
Facebook.com
1
, Craigslist.org, and Backpage.com
1
,
indicating that (a) the online study was about sexual
health, and (b) people who completed the study would
be entered in a raffle to win $100 (with a 1 in 30 chance
of winning). An online site, PsychData
1
, was used for
anonymous and confidential collection of data from
recruited participants. Prior to beginning the online sur-
vey, interested individuals read a short explanation of
the study and clicked on a link leading to an online
screener. This eligibility screener assessed for age, gender
identity, virginity status,
Sexual Satisfaction Scale (SSS; Meston & Trapnell,
2005). The SSS is a 30-itemquestionnaire thatmeasures,
on a five-point Likert-type scale ranging from 1 (strongly
disagree)to5(strongly agree), five separate domains of sex-
ual satisfaction: (a) ease and comfort discussing sexual and
citizenship and language
437
RELLINI, VUJANOVIC, GILBERT, AND ZVOLENSKY
fluency. After completing the online screener, parti-
cipants received a more in-depth description of the study
and were informed that the questionnaires would take
between 30 and 45min to complete. Participants were
not permitted to stop data entry and continue at a later
time. At the conclusion of the study, participants
received a random participation identification code that
they were asked to send to the researcher to be entered
into the monetary raffle for possible compensation.
Age and relationship length were not significantly
correlated with the DERS subscales or the CTQ-total
score; however, they were significantly correlated with
the dependent variables used in the subsequent analy-
ses: SSS, NRI–SPV–R Affection, and NRI–SPV–R
Intimacy. Overall, both age and relationship length were
negatively correlated with SSS (r
¼
.18 and
.27,
respectively) and NRI–SPV–R Affection (r
¼
.26 and
.21, respectively). Age was significantly and negatively
correlated with NRI–SPV–R Intimacy (r
¼
.17). These
results indicate that older participants reported less sex-
ual satisfaction, less affection, and less intimacy, while
individuals reporting a longer relationship scored lower
in sexual satisfaction and affection.
As illustrated in Table 3, CTQ-total was signifi-
cantly and negatively associated with SSS, r
¼
.18,
p
<
.05; but not NRI–SPV–R Intimacy and NRI–
SPV–R Affection, suggesting that individuals reporting
more severe forms of maltreatment reported lower
sexual satisfaction. Also, CTQ-total was significantly
and positively associated with DERS-total (r
¼
.28,
p
<
.01), as well as all DERS subscales (range
r
¼
.15
27, p
<
.05), with the exception of DERS-
Aware and DERS-Clarity (ps
>
.05). These findings
indicate that individuals with more severe forms of
childhood maltreatment reported more difficulties in
emotion regulation.
Consistent with our hypotheses of greater sexual
and relationship difficulties for individuals who have
more emotion dysregulation, DERS-total was signifi-
cantly and negatively associated with SSS (r
¼
.29,
p
<
.01). Yet, in contrast to predictions, DERS-total
was not significantly associated with the NRI–SPV–R
Intimacy or Affection. DERS-Aware and DERS-
Clarity subfactors demonstrated significant correlations
with SSS (r
¼
.22 and
.36, respectively; p
<
.01),
NRI–SPV–R Intimacy (r
¼
.21 and
.29; respectively,
p
<
.01), and NRI–SPV–R Affection (r
¼
.19 and
.25, p
<
.01).
Data Analytic Plan
First, zero-order correlations were computed to evaluate
associations among theoretically relevant variables.
Second, a series of three-step hierarchical regressions were
conducted to assess themain and interactive effects of child-
hood trauma severity (CTQ-total) and difficulties in emo-
tion regulation (DERS-total) with regard to the following
criterion variables: (a) sexual satisfaction (SSS), (b)
relationship intimacy satisfaction (NRI–SPV–R Intimacy
subscale), and (c) relationship affection satisfaction
(NRI–SPV–R Affection subscale). Age and relationship
duration were entered as covariates at Step 1 of the
regressionmodels in an effort to account for developmental
differences in the sample, the main effects of childhood
trauma severity (CTQ-total) and difficulties in emotion
regulation (DERS-total) were entered at Step 2, and the
interactive effect of childhood trauma severity by difficult-
ies in emotion regulation (CTQ
DERS-total) was entered
at Step 3. DERS-total and CTQ-total were mean-centered.
Since CTQ scores were skewed, logarithmic transforma-
tions were applied before mean centering.
Results
Correlations between Emotion Dysregulation,
Relationship Variables, and Childhood Trauma
Table 3 presents a summary of zero-order correlations
and descriptive data for theoretically relevant variables.
Table 3. Zero-Order Correlations and Descriptive Characteristics for All Targeted Variables
Variable
CTQ-Total
Age
Relationship Length
SSS
Intimacy
Affection
M (SD)
DERS-Non-acceptance
.26
.09
.20
.02
.03
14.2 (5.8)
DERS-Goals
.15
.11
.03
.13
.06
15.3 (4.8)
DERS-Impulse
.27
.07
.17
.07
.09
12.4 (5.2)
DERS-Aware
.08
02
.22
.21
.19
13.4 (4.6)
DERS-Strategies
.22
.11
25
.01
05
18.7 (7.0)
.36
.29
.25
DERS-Clarity
.10
.05
11.1 (3.8)
.28
.29
DERS-total
.10
.07
.13
85.0 (22.2)
.18
CTQ-total
—
<
.01
.12
.07
1.58 (0.15)
.18
.17
.26
Age
<
.01
—
21.7 (3.3)
.15
.27
.21
Relationship length
.05
.13
3.7 (2.3)
M
1.58
21.7
3.7
89.3
11.3
12.4
—
SD
0.15
3.3
2.3
19.9
3.3
3.1
Note. Intimacy and affection are subscales of the Network of Relationships Inventory: Social Provisions Version–Revised. CTQ-total is log
transformed. CTQ
¼
Childhood Trauma Questionnaire; SSS
¼
Sexual Satisfaction Scale; DERS
¼
Difficulties in Emotion Regulation Scale.
p
<
.05.
p
<
.01.
438
  [ Pobierz całość w formacie PDF ]

  • zanotowane.pl
  • doc.pisz.pl
  • pdf.pisz.pl
  • trzonowiec.htw.pl

    • Odnośniki