This article was downloaded by: [University of Glasgow]

On: 01 January 2015, At: 07:43

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Attachment & Human Development

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/rahd20

Attachment styles, traumatic

events, and PTSD: a cross-sectional
investigation of adult attachment and
trauma
a a
Maja O'Connor & Ask Elklit
a
Department of Psychology , University of Aarhus , Denmark
Published online: 19 Mar 2008.

To cite this article: Maja O'Connor & Ask Elklit (2008) Attachment styles, traumatic events, and
PTSD: a cross-sectional investigation of adult attachment and trauma, Attachment & Human
Development, 10:1, 59-71, DOI: 10.1080/14616730701868597

To link to this article: http://dx.doi.org/10.1080/14616730701868597

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 Attachment & Human Development
Vol. 10, No. 1, March 2008, 59–71

Attachment styles, traumatic events, and PTSD: a cross-sectional

investigation of adult attachment and trauma
Maja O’Connor* and Ask Elklit

Department of Psychology, University of Aarhus, Denmark

The aim of the present study was to examine the association between post-traumatic
Downloaded by [University of Glasgow] at 07:43 01 January 2015

stress disorder (PTSD) and adult attachment in a young adult population. A sample of
328 Danish students (mean age 29.2 years) from four different schools of intermediate
education level were studied by the Harvard Trauma Questionnaire (HTQ), the Revised
Adult Attachment Scale (RAAS), the Trauma Symptom Checklist (TSC), the Crisis
Support Scale (CSS), the Coping Style Questionnaire (CSQ), and the World
Assumption Scale (WAS). Attachment styles were associated with number of PTSD
symptoms, negative affectivity, somatization, emotional coping, attributions, and social
support. The distribution of attachment styles in relation to PTSD symptoms could be
conceived as uni-dimensional.
Keywords: adult attachment; traumatic events; posttraumatic stress symptoms;
protective factors; risk factors

Introduction
During the last few decades it has been recognized that theory originally designed to
describe attachment patterns between mother and child is also relevant in understanding
adult relationships and psychopathology (Sroufe, 2005). Attachment style in adulthood is
related to the adult experience of romantic love and mental models of self (Hazan & Shaver,
1987). The field of adult attachment has developed greatly over recent years (Fraley &
Shaver, 2000) and is currently influential in areas of personality and social psychology
including social cognition (Stein, Jacobs, Ferguson, Allen, & Fonagy, 1998). It has been
suggested that stressful attachment-related events such as the unresolved loss of a loved one
can lead to a number of symptoms, including those of Posttraumatic Stress Disorder
(PTSD; Fearon & Mansell, 2001). Bowlby (1953) identified a number of negative life events
that would be expected to influence the stability of attachment: the death of a parent, foster
care, parental divorce, chronic and severe illness of parent and child, single parent, parental
psychiatric disorder, drug and alcohol abuse, and child experience of physical or sexual
abuse. Weinfeld, Sroufe, and Egeland (2000), among others, in a study of a high-risk
sample provided evidence that attachment continuity was jeopardized by child maltreat-
ment, maternal depression, and family functioning in early adolescence. Quite similar
findings are reported as risk factors for PTSD in a meta-analysis by Brewin, Andrews, and

*Corresponding author. Email: maja@psy.au.dk

ISSN 1461-6734 print/ISSN 1469-2988 online

Ó 2008 Taylor & Francis
DOI: 10.1080/14616730701868597
http://www.informaworld.com
 60 M.O’Connor and A. Elklit

Valentine (2000) who concluded that general childhood adversity predicted PTSD more
consistently than e.g., did gender and age at trauma. Psychiatric history, reported
childhood abuse, and family psychiatric history had more uniform predictive effects.
A search in PsycINFO (August 2006) revealed only a few studies examining the
relationship between adult attachment and PTSD based on self-report inventories. Two
studies had male participants, such as former prisoners of war or victims of torture, who
had experienced extreme stress (Dieperink, Leskela, Thuras, & Engdahl, 2001; Kanninen,
Punamaki, & Qouta, 2003). Two studies had female participants investigating CSA survi-
vors and women who attended a clinic for early pregnancy termination (Allanson &
Astbury, 2001; Anderson & Alexander, 1996), with findings, in general, indicating a secure
attachment style to be associated with decreased number of PTSD symptoms and insecure
attachment styles to be associated with enhanced number of PTSD symptoms when expo-
sed to traumatic experiences. Two studies had both genders as participants. Miller (2000)
Downloaded by [University of Glasgow] at 07:43 01 January 2015

assessed 31 parents of childhood leukemia survivors and found that secure attachment was
associated with less posttraumatic symptoms and dysphoria, while fearful and dismissive
attachment styles were associated with an increased level of posttraumatic symptoms.
Waldinger, Schulz, Barsky, and Ahern (2006) tested whether insecure attachment mediated
the link between childhood trauma and adult somatization in a sample of 101 couples.
For women fearful attachment fully mediated this link, but for men both childhood trauma
and insecure attachment were independent predictors of adult somatization.
Lazarus and Folkman (1984) proposed that cognitive factors such as coping styles and
cognitive schemas or attributions mediate response to traumatic experiences in the direction
that problem-centered coping generally is more effective in dealing with traumatic stress that
emotional coping. A recent literature review gave evidence to the Lazarus and Folkman
theory (Agaibi & Wilson, 2005). Furthermore, it has been found in recent studies that high
levels of emotional coping style and low levels of problem-focused or rational coping style
preceding a traumatic event predicted PTSD (Gil, 2005; Nielson, 2003), and that chronic
PTSD in war veterans was associated with basic assumptions such as lower levels of self-
worth and beliefs about the benevolence of other people (Dekel, Solomon, Elklit, &
Ginzburg, 2004). In addition to basic assumptions (Janoff-Bulman, 1992), availability of
social support is likely to be perceived differently by individuals with various attachment
styles as these styles (Bartholomew, 1990) are defined by some authors as positive and
negative working models of self and others. In their empirical and theoretical review, Sarason
and Duck (2001) conclude that cognitive components, such as locus of control, schemata,
and working models, are some of the personality variables influencing social support.
The term negative affectivity refers to a mood dimension and reflects a general tendency
to react to and have a negative perspective on the surrounding world and oneself (Krog &
Duel, 2003). Negative affectivity can both be seen as a state and a personality factor or trait
similar to neuroticism (Krog & Duel, 2003). Schnurr and Vielhauer (2005) described
negative affectivity/neuroticism as a likely risk factor for PTSD because the personality
dimension reflects a sensitivity to negative stimuli. Shapinsky, Rapport, Henderson, and
Axelrod (2005) found a strong positive correlation between negative affectivity and PTSD
measures, and a substantial variance in PTSD was accounted for by trait characteristics
such as negative affectivity. Significant associations between PTSD and negative affectivity
were also found by e.g., Bennett, Owen, Koutsakis, and Bisson (2002), Feldner, Lewis,
Leen-Feldner, Schnurr, and Zvolensky (2006), and Monson, Price, Rodriguez, Ripley, and
Warner (2004). To the best of the authors’ knowledge, no research on the relationship
between attachment style and PTSD in a general population sample and no research on the
associations between attachment, PTSD, and coping styles has been published.
 Attachment & Human Development 61

Adult attachment
The adult attachment theory by Hazan and Shaver (1987) was further developed by
Bartholomew and colleagues (Bartholomew, 1990; Bartholomew, Cobb, & Poole, 1997;
Griffin & Bartholomew, 1994) by describing a connection between attachment and models
of self and others resulting in a two-dimensional model of adult attachment based on self-
report inventories. Collins and colleagues (Collins, Cooper, Albino, & Allard, 2002;
Collins & Feeney, 2000, 2004; Feeney & Collins, 2001, 2003) likewise based on self-report
data derived the same four attachment styles from two dimensions: (1) the combined
measures of closeness and dependency and (2) anxiety. These dimensions are comparable
to the model of self and others based on the dimensions of dependency and avoidance
originally described by Bartholomew (Bartholomew, 1990; Griffin & Bartholomew, 1994).
Downloaded by [University of Glasgow] at 07:43 01 January 2015

The current study

In the present study the definition of psychological trauma and the diagnostic criteria for
PTSD in the DSM IV (American Psychiatric Association, 1994) is used (First, Frances, &
Pincus, 2002). The aim of the study is to examine whether attachment security is related to
(1) post-traumatic symptomatology and (2) several other closely linked aspects of social
and psychological functioning believed to be implicated in the development of and/or
maintenance of PTSD, namely negative affectivity and somatization. A secondary aim is
to test whether attachment security is associated with post-traumatic symptomatology
independently of these PTSD-related social and psychological factors.

Method
Sample
A total of 328 students (65% female) with a mean age 29.2 years (SD ¼ 11.63;
range ¼ 15–61 years) participated in the study. The participants came from four different
schools of intermediate education level (nurses’ aids, two types of social workers, and
craftsmen). The schools were selected for their predominantly young student populations,
the majority of which have a non-university background. Traumatic experiences ad
modum Kessler, Chiu, Demler, and Walters (1995) were investigated.

Procedure
Data collection was based on questionnaires. The first author introduced the study to the
participants and was available for clarifying questions while the participants filled out the
questionnaire. Participation was voluntary. The response rate was 94%.

Measures
Harvard Trauma Questionnaire
The Harvard Trauma Questionnaire (HTQ) part IV (Mollica, Caspi-Yavin, Bollini, &
Truong, 1992) measures the severity of PTSD symptomatology and can be used for an
estimate of the PTSD diagnosis. The HTQ consists of 30 items (4-point Likert scale;
1 ¼ not at all, 4 ¼ very often). Sixteen items relate to the three core clusters in PTSD in
DSM-IV: avoidance (7 items), reexperiencing (4 items; one item contains two related
 62 M.O’Connor and A. Elklit

symptoms), and hypervigilance (5 items). The A2 criterion (the experience of horror or

helplessness) of the DSM-IV diagnosis (First, Frances, & Pincus, 2002) was assessed
separately. Only scale items 3 on HTQ were considered for a PTSD diagnosis. The scale
has been shown to be both reliable and valid (Bach, 2003). The HTQ self-report measure
of PTSD had 88% concordance with interview based estimates of PTSD (Mollica et al.,
1992).
In the present study, participants chose the most traumatic event experienced in their
lifetime. On the basis of the chosen event, each participant first answered HTQ
retrospectively in relation to their reaction in the time immediately after the stressful event
(Lifetime, LHTQ), then subsequently answered HTQ in relation to their current reaction
(CHTQ). The internal consistencies, as measured by Cronbach’s alpha, of the subscales in
this study were excellent for LHTQ, with avoidance (a ¼ .82), reexperiencing (a ¼ .84),
hypervigilance (a ¼ .76), LHTQ total (a ¼ .95), and good for CHTQ, with avoidance
Downloaded by [University of Glasgow] at 07:43 01 January 2015

(a ¼ .71), reexperiencing (a ¼ .74), hypervigilance (a ¼ .67), and CHTQ total (a ¼ .93).

Revised Adult Attachment Scale

The Revised Adult Attachment Scale (RAAS) is an 18-item self-report scale, on which
participants rate statements about how they function and feel in a relationship with a
partner, someone close, and people in general (5-point Likert scale; 1 ¼ not at all
characteristic, 5 ¼ very characteristic). The scale is two-dimensional; (1) items on
closeness and dependency are merged into one dimension (a ¼ .73) (Collins, 1995,
unpublished research note) and (2) an anxious attachment dimension (a ¼ .73).
Attachment styles corresponding to Ainsworth’s three original attachment patterns
(secure, anxious-ambivalent, and avoidant) can be generated based on the RAAS (Collins
& Read, 1990; Stein et al., 1998). A fourth attachment pattern corresponding to a fearful
style was also included (Collins & Feeney, 2004). The categorical attachment styles are
distributed by values on the dimensions of ‘‘close-dependency’’ and anxiety; e.g., secure
attached scores high on the close-dependency dimension and low on the anxious
dimension. High is defined as being above the midpoint on the 5-point scale, and low as
below the midpoint. Scores exactly on the midpoint are normally excluded from the
sample (Collins, 1995, unpublished research note).
In the present study, 12.8% (n ¼ 42) had missing values and 8.2% (n ¼ 27) had
midpoint scores and were therefore excluded from the sample leaving a new N of 259.
Sixty-six percent (n ¼ 188) had a secure attachment style, 8.2% (n ¼ 27) had a
preoccupied attachment style, 8.7% (n ¼ 25) had a dismissive attachment style, and
finally 6.6% (n ¼ 19) had a fearful attachment style.

Trauma Symptom Checklist — Revised

The Trauma Symptom Checklist — Revised (TSC-R) consists of 23 items and measures a
number of general distress symptoms on a 4-point Likert scale (1 ¼ never, 4 ¼ always).
The TSC-R yields specific scores for negative affectivity and somatization. Krog and Duel
(2003) identified the two subscales based on a factor analysis of 4152 cases from 16 studies
of trauma populations in which the original TSC (Briere & Runtz, 1989) was used. The
TSC-R subscales have good internal consistency reliability and factorial and criterion
validity (Krog & Duel, 2003). The internal consistencies of the subscales as measured by
Cronbach’s alpha were acceptable to very good (somatization a ¼ .65, negative affectivity
a ¼ .83; TSC total a ¼ .85).
 Attachment & Human Development 63

Crisis Support Scale

The Crisis Support Scale (CSS) measures social support after a traumatic event. The seven
items (7-point Likert scale; 1 ¼ never, 7 ¼ always) relate to perceived available support,
practical help, emotional support, contact with others in a similar situation, ability to
express thoughts and emotions related to the event, the degree to which one feels let down,
and general satisfaction with social support (Joseph, Andrews, Williams, & Yule, 1992).
The scale has good psychometric properties (Elklit, Pedersen, & Jind, 2001). The internal
consistencies of the subscales in the present study were good (Present CSS a ¼ .73; Past
CSS a ¼ .83).

Coping Style Questionnaire

Downloaded by [University of Glasgow] at 07:43 01 January 2015

The Coping Style Questionnaire (CSQ) originally consisted of 60 items (4-point Likert
scale; 1 ¼ never, 4 ¼ always) measuring four coping styles; rational coping, emotion-
focused coping, avoidance coping, and detached coping (Roger, Jarvis, & Najarian,
1993). In this study, 37 items loading significantly, both in the original study and in a
replication study (Elklit, 1996), on the four factors (cut-off point 0.30) were used. The
internal consistencies of the subscales in this study were good to acceptable with the
exception of detached coping (rational coping a ¼ .70; emotion-focused coping a ¼ .75;
avoidance coping a ¼ .65; detached coping a ¼ .43). The latter was excluded from this
study.

World Assumptions Scale

The World Assumptions Scale (WAS) is a 32-item self-report scale that examines the
participants’ cognitive schemes about self and the world. Items are scored on a 6-point
Likert scale (1 ¼ strongly disagree, 6 ¼ strongly agree). The internal consistency of the
subscales has previously been found to range from a ¼ .48 7 .82. (Elklit, Shevlin,
Solomon, & Dekel, 2007). In the present study the internal consistency ranged from
a ¼ .43 7 .70. Only subscales self-control (a ¼ .69) and benevolence of the world
(a ¼ .70) showed satisfactory internal consistency, therefore only these scales were used in
the present study.

Data analysis
The distribution of the data is presented as percentages, means, and standard deviations.
Nominal variables were compared using w2. Correlations were estimated with Pearson’s
correlation coefficient. Attachment styles were analysed in relation to all scales using a
one-way ANOVA. Multiple linear regression analysis was used to assess the effect of
independent variables on the LHTQ total scores. A 0/1 (no/yes) coding was used for
dichotomous variables. The analyses were performed by means of SPSS-PC, version 13.0.

Results
The average duration of the participants’ education was 12.7 years (SD ¼ 2.5; range
7–24). Forty-eight percent were single and lived alone, 20% were cohabiting, 26%
were married and lived together with a partner, and 6% were divorced. The average time
of living together with a partner/spouse was 11.3 years (SD ¼ 10.5). Thirty-nine percent
 64 M.O’Connor and A. Elklit

of the participants had children. The average number of children was 2 (SD ¼ 0.9;
range 1–5).
The average number of lifetime traumatic experiences was 1.5 (SD ¼ 1.4). Seventy-
seven percent had been exposed to at least one traumatic experience such as bereavement
(53%; n ¼ 174), threat of violence (15%; n ¼ 49), actual violence (10%; n ¼ 32), accident
(14%; n ¼ 47), shock because of significant others being exposed to a traumatic event
(14%; n ¼ 47), fire (4%; n ¼ 14), witnessing a person being killed or injured (9%; n ¼ 28),
rape (2%; n ¼ 8), childhood sexual abuse (3%; n ¼ 11), childhood physical abuse (4%;
n ¼ 14), or severe childhood neglect (7%; n ¼ 23).
Lifetime PTSD prevalence was estimated to 18.9% for all subjects (37.3% for all
exposed subjects who answered the HTQ); current PTSD rate was estimated to 4.6%
(11.9% for the exposed subjects who answered the HTQ). Missing data were not
imputed. There was an overrepresentation of young males who had missing data in the
Downloaded by [University of Glasgow] at 07:43 01 January 2015

various psychological scales particularly at the end, but not in the demographic or
event questions. As young males with a medium level of professional education are not
well represented in attachment and trauma research, we decided to keep them in the
study. Those not reporting any traumatic experiences were excluded in the following
analyses.

Associations between attachment styles and the psychosocial variables

The relationships between attachment styles and the independent variables were analysed
by means of one-way ANOVAs (Table 1). None of the attachment styles were associated
with level of trauma exposure (data not shown). Secure attachment was significantly
associated with low levels of lifetime and current number of PTSD symptoms, negative
affectivity, somatization, and emotional and avoidant coping, but with high levels of
perceived previous and present social support, rational coping, and a more favorable
attribution of benevolence of the world.
Preoccupied attachment was only associated with a high level of emotional and a low
level of rational coping. Dismissive attachment was associated with high levels of lifetime
and current number of PTSD symptoms, negative affectivity, and somatization and low
levels of benevolence of the world and perceived present support. Fearful attachment
presents almost the same pattern as dismissive attachment with the exception that fearful
attachment was not related to the intrusion subscales, a low level of perceived previous
social support, and high levels of emotional and avoidant coping.

Regression analysis
A stepwise linear regression analysis was carried out using the Lifetime HTQ (LHTQ)
total score as the dependent variable. The aim was to generally develop a model of the best
predictors of trauma symptoms and see whether attachment can be justifiably considered
to add unique information (i.e., whether it is independent of the other factors considered).
Demographic variables were entered first followed by trauma related variables, personality
related variables third, and social support variables last. The final model included six
variables that explained 54% of the total LHTQ score (Table 2).
The correlation between Lifetime HTQ and Current HTQ was r ¼ .76. Lifetime HTQ
was predicted by (female) gender (gender accounted for 16% of the variance), rape
experience, feeling of helplessness in relation to the traumatic event (helplessness was rated
independently of the HTQ), high degree of fearful attachment and of emotional coping,
 Attachment & Human Development 65

Table 1. One-way ANOVAs between attachment styles and dependent variables (F-ratio values;
df ¼ 1).

Secure Preoccupied Dismissive Fearful

CSS
Past Mean 34.74**** 0.12 3.35 21.43****
Mean (SD) 36.48 (8.6) 34.21 (10.5) 29.25 (6.7) 22.50 (9.4)
Present 59.89**** 0.02 18.95**** 32.43****
Mean (SD) 37.37 (5.4) 34.93 (8.5) 29.35 (6.2) 26.95 (6.6)
CSQ
Rational 7.24** 4.38* 3.80 1.81
Mean (SD) 26.93 (6.5) 23.45 (3.1) 23.64 (6.0) 28.11 (8.4)
Emotional 22.16**** 8.90*** 0.28 21.65****
Mean (SD) 15.97 (5.4) 20.46 (7.6) 17.83 (5.2) 23.28 (5.6)
Avoidant 7.25** 1.94 0.97 20.11****
Downloaded by [University of Glasgow] at 07:43 01 January 2015

Mean (SD) 19.84 (5.3) 21.92 (2.8) 19.43 (4.2) 26.06 (9.5)
TSC
Negative affectivity 42.58**** 1.64 6.33* 45.67****
Mean (SD) 17.61 (4.2) 20.38 (6.4) 21.90 (5.6) 26.72 (6.1)
Somatization 25.87**** 0.00 6.00* 22.54****
Mean (SD) 14.51 (3.6) 15.54 (3.7) 17.83 (6.1) 20.44 (6.1)
Total 43.21**** 0.55 6.94** 42.75****
Mean (SD) 32.11 (6.5) 35.92 (9.4) 39.75 (11.1) 47.17 (11.4)
WAS
Benevolence of the world 7.80** 0.12 4.19* 0.01
Mean (SD) 18.16 (4.0) 17.22 (4.6) 15.52 (7.5) 17.42 (8.8)
Control 0.36 0.09 1.50 3.11
Mean (SD) 12.36 (3.9) 12.46 (2.9) 11.38 (3.3) 10.78 (3.8)
HTQ
L-Intrusion 7.40** 0.62 8.29*** 2.70
Mean (SD) 9.27 (3.1) 10.39 (3.7) 12.06 (3.0) 11.33 (2.8)
L-Arousal 6.50* 0.13 5.03* 8.86***
Mean (SD) 11.17 (3.7) 12.05 (4.5) 13.81 (2.9) 14.92 (3.2)
L-Avoidance 14.52**** 0.47 6.70* 8.60***
Mean (SD) 12.27 (5.9) 14.50 (6.5) 17.25 (5.6) 18.42 (4.1)
L-Total 15.36**** 0.38 14.37**** 5.75*
Mean (SD) 57.24 (18.4) 66.73 (26.9) 89.63 (17.5) 81.71 (19.9)
C-Intrusion 9.46*** 0.63 10.82*** 1.91
Mean (SD) 6.68 (2.1) 7.6 (3.0) 9.25 (2.5) 8.22 (2.8)
C-Arousal 19.89**** 1.19 6.71* 8.24**
Mean (SD) 8.32 (2.8) 10.00 (3.0) 11.50 (2.6) 11.89 (2.6)
C-Avoidance 39.00**** 2.68 17.73**** 10.24***
Mean (SD) 9.41 (2.8) 13.23 (5.1) 16.73 (9.4) 16.00 (3.8)
C-Total 25.23**** 2.56 8.65*** 7.28**
Mean (SD) 45.07 (11.5) 57.82 (20.1) 68.33 (12.9) 67.00 (15.1)

Note: *p 5 .05; **p 5 .01; ***p 5 .005; ****p 5 .0005.

and lack of perceived sympathy and social support at the time after the traumatic
experience (Table 2).

Discussion
The present study attempted to clarify whether PTSD symptomatology and other
psychosocial factors were associated with adult attachment security. The results of the
 66 M.O’Connor and A. Elklit

Table 2. Hierarchical multiple regression analysis to predict present number of lifetime PTSD
symptoms (LHTQ).

Step 1 Step 2 Step 3 Step 4 Step 5 Step 6

b p b p b p b p b p b p
Gender .41 .000 .37 .000 .27 .000 .31 .000 .27 .000 .26 .000
Rape .28 .001 .23 .003 .20 .007 .15 .049 .13 .087
Helplessness/horror (A2) .36 .000 .36 .000 .34 .000 .34 .000
Fearful attachment .32 .000 .24 .002 .14 .067
Emotional coping .26 .002 .25 .002
Past sympathy -.26 .001

Note: Step 1: F ¼ 24.9; R2 ¼ .16; p 5 .005. Step 2: F ¼ 19.53; R2 ¼ .23; p 5 .0005. Step 3: F ¼ 21.05; R2 ¼ .34;
p 5 .0005. Step 4: F ¼ 25.24; R2 ¼ .48; p 5 .0005. Step 5: F ¼ 21.03; R2 ¼ .50; p 5 .0005. Step 6: F ¼ 19.9;
R2 ¼ .54; p 5 .0005.
Downloaded by [University of Glasgow] at 07:43 01 January 2015

study provide evidence that the number of PTSD symptoms was associated with certain
attachment styles. A negative correlation between secure attachment and all subscales of
current and lifetime HTQ was found, suggesting that secure attachment may have a
protective effect on the development of PTSD. The results are in accordance with a study
of former male political prisoners in which secure attachment was found to be protective in
relation to development of PTSD following physical torture (Kanninen et al., 2003).
No significant pattern appeared in the associations between preoccupied attachment
style and number of PTSD symptoms, whereas the fearful and the dismissive attachment
styles showed a positive relationship with number of PTSD symptoms on both current and
the lifetime measures of HTQ. The pattern of associations between number of PTSD
symptoms and secure attachment on the one hand and fearful and dismissive attachment
styles on the other was inverted. The level of associations indicates that secure attachment
may be a more salient factor than fearful and dismissive attachment. The regression
analysis showed that one particularly attachment style, the fearful, could explain part of the
variation of the PTSD symptomatology. This finding is similar to a finding by Fraley,
Waller, and Brennan (2000) that measures of the insecure attachment styles are more
reliable and therefore stronger predictors than measures of secure attachment. As expected,
fearful and dismissive attachment styles were linked to poor psychological adjustment on a
variety of parameters in the present study. The present findings showed significant
associations between fearful attachment and low levels of perceived social support and
emotional and avoidant coping styles. Dismissive attachment was unrelated to coping
and past social support and was less strongly associated with the negative affectivity and
somatization than fearful attachment. Dismissive attachment was associated with less
favorable assumptions about the benevolence of the world. The latter attribution has
previously been associated with poor psychological outcome (Janoff-Bulman, 1992).
The use of avoidant and especially emotional coping has been previously associated
with increased psychological distress (Bödvarsdottir & Elklit, 2004; Ireland, Boustead, &
Ireland, 2005). The results of the current study are in line with these findings. Further-
more, fearful attachment was found to correlate strongly and dismissive attachment
moderately with the TSC subscales of somatization and negative affectivity. A search on
PsycINFO (August 2006) revealed no publications on studies combining the TSC and
attachment style. The results are, however, comparable to results from related studies
finding a positive correlation between insecure attachment styles (especially the fearful
type) and somatization, hypochondria, or subjective body complaints (Ciechanowski,
 Attachment & Human Development 67

Walker, Katon, & Russo, 2002; Schmidt, Strauss, & Braehler, 2002; Wayment &
Vierthaler, 2002). In line with the present results, one study found a positive correlation
between fearful attachment and negative affectivity (Wearden, Lamberton, Crook, &
Walsh, 2005). Negative affectivity has previously been found to share the same underlying
structure as neuroticism (Meyer & Shack, 1989). The results from the present study are
therefore in line with related studies finding a positive correlation between neuroticism
and insecure attachment styles (Howell Rolston, 2003; Moreira et al., 1998; Wolfgang,
2005).
Demographic variables, with the exception of gender, were unrelated to traumatization
in the current study. The finding of a higher number of PTSD symptoms reported by
women remained significant after controlling for social support, emotional coping, fearful
attachment, and feeling of helplessness or horror in relation to traumatic events. This is in
line with previous research (Kessler, Chiu, Demler, & Walters, 2005; Kessler, Sonnega,
Downloaded by [University of Glasgow] at 07:43 01 January 2015

Bromet, & Hughes, 1995; Krog & Duel, 2003). Of interest is the fact that women, in spite
of the higher number of PTSD symptoms, reported more satisfying social support and
more positive attributions than men. The possibility of attachment security being
confounded with trauma exposure was not supported in the present study.
This study has several limitations. First, the study is cross-sectional and retrospective.
Furthermore, the number of persons with fearful attachment was relatively small. Self-
report measures may, in the light of different results found between self-reported adult
attachment versus adult attachment interviews, not fully capture the underlying
organization of attachment representation. The focus of this study has been limited to
identifying associations between adult attachment in romantic and close relationships and
PTSD. While this focus is appropriate when studying relationships between attachment
style and processing of traumatic events experienced in adulthood, it may not be the
most appropriate choice for studying relationships involving traumatic events experienced
in childhood. Finally, the population is only representative of a segment of the
Danish population. Generalizations to other populations should therefore be made with
caution.
Despite the limitations, one strength of the study was that it consisted of an
understudied population representative of a social stratum without a university
background. Hence the studied population represents a different population than that
found in many studies of young college populations. In addition, this study supports the
hypothesis that the number of PTSD symptoms is associated with attachment styles in a
younger adult population of intermediate education level not selected for being a survivor
of a traumatic event, and the results are consistent with results from traumatic populations
(Allen, Huntoon, & Evans, 1999; Dieperink et al., 2001; Kanninen et al., 2003). The
findings provide further evidence that fearful and dismissive attachment is linked to high
levels of psychological distress while secure attachment is linked to low levels of
psychological distress. This suggests that attachment style may be an important
interpersonal dimension in relation to traumatic experiences and reactions.
Sixty-six percent of the valid population was classified as securely attached. This figure
is fairly high compared to prior findings on other types of non-traumatized populations
(Bartholomew & Horowitz, 1991). Age, class, and cultural differences in the populations
investigated may be the reason for the various results, as could technicalities such as the
number of midpoint exclusions (8.7%; n ¼ 27).
The inverse relationship between secure versus fearful and dismissive attachment styles
found in this study opens an interesting discussion. The preoccupied attachment style was
hardly associated with the dependent variables. These results could suggest a type of linear
 68 M.O’Connor and A. Elklit

relationship between the four attachment styles and scores on most dependent variables; a
linear relationship with secure attachment at one end followed by preoccupied, then
dismissive, and lastly fearful attachment (Table 2). The results are in opposition to Collins’
(1996) findings that preoccupied persons report exaggerated levels of negative emotion,
compared to avoidant/dismissive persons.
Is this merely a coincidental result caused by a relatively small population of fearfully
attached individuals or is it possible that attachment could be conceived as uni-
dimensional rather than bi-dimensional? Fraley and Brumbaugh (2004) investigated
stability across time based on uni-dimensional attachment security in contrast to the bi-
dimensional attachment styles described by Bartholomew (Bartholomew & Horowitz,
1991; Griffin et al., 1994), and Collins (1996). Thus, in line with Fraley and Baumbaugh
(2004), this study may indicate that attachment could be conceived as a uni-dimensional,
high/low security dimension. Furthermore, the study indicates that attachment may be an
Downloaded by [University of Glasgow] at 07:43 01 January 2015

intermediate factor in relation to psychological trauma in the direction that a high degree
of secure attachment could be a protective factor, whereas preoccupied exhibits a neutral
middle ground, while fearful and dismissive attachment are risk factors in relation to
specific types of abuse and psychological distress.
A number of questions need to be addressed in future research. The current study gave
evidence that attachment is an important factor in relation to trauma. Attachment can be
viewed as a subjective, interpersonal dimension in relation to psychological trauma,
supplementing well-known psychological measures such as coping and personality styles,
when investigating the possible effects of personality in relation to traumatic experiences
and reactions. While secure attachment style seems to serve as a protective factor and
fearful and dismissive attachment styles represent risk factors in the development of
traumatization following a traumatic experience, new preventive interventions may be at
hand. As the study is correlational, we have no knowledge about causality. Attachment
security may as well be the result of traumatic events that cause destabilization of existing
attachment styles, as it may moderate or mediate the effects of traumatization. Inclusion
of attachment measures in prospective studies of PTSD and inclusion of PTSD measures
in prospective studies of attachment need to be taken into consideration in future research
to better understand the relationship between PTSD and adult attachment. Furthermore,
additional work is needed to understand the nature of dimensions of attachment across
gender, age, socioeconomic status, trauma type, and culture.

References
Agaibi, C.E., & Wilson, J.P. (2005). Trauma, PTSD, and resilience. A review of the literature.
Trauma, Violence, and Abuse, 6, 195–216.
Allanson, S., & Astbury, J. (2001). Attachment style and broken attachments: Violence, pregnancy,
and abortion. Australian Journal of Psychology, 53, 146–151.
Allen, J.G., Huntoon, J., & Evans, R.B. (1999). Complexities in complex posttraumatic stress
disorder in inpatient women: Evidence from cluster analysis of MCMI-III personality disorder
scales. Journal of Personality Assessment, 73, 449–471.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders
(4th ed.). Washington, DC: APA.
Anderson, C.L., & Alexander, P.C. (1996). The relationship between attachment and dissociation in
adult survivors of incest. Psychiatry: Interpersonal and Biological Processes, 59, 240–254.
Bach, M.E. (2003). En empirisk belysning og analyse af ‘‘Emotional Numbing’’ som eventuel
selvstændig faktor i PTSD. Psykologisk Studie Skriftserie, 6, 1–199.
Bartholomew, K. (1990). Avoidance of intimacy: An attachment perspective. Journal of Social &
Personal Relationships, 7, 147–178.
 Attachment & Human Development 69

Bartholomew, K., Cobb, R.J., & Poole, J.A. (1997). Adult attachment patterns and social support
processes. In G.R. Pierce & B. Lakey (Eds.), Sourcebook of social support and personality (pp.
359–378). New York: Plenum Press.
Bartholomew, K., & Horowitz, L.M. (1991). Attachment styles among young adults: A test of a
four-category model. Journal of Personality & Social Psychology, 61, 226–244.
Bennett, P., Owen, R.L., Koutsakis, S., & Bisson, J. (2002). Personality, social context and cognitive
predictors of post-traumatic stress disorder in myocardial infarction patients. Psychology &
Health, 17, 489–500.
Bowlby, J. (1953). Child care and the growth of love. London: Pelican Books.
Brewin, C., Andrews, B., & Valentine, J. (2000). Meta-analysis of risk factors for posttraumatic
stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology, 68, 748–
766.
Briere, J., & Runtz, M. (1989). The Trauma Symptom Checklist (TSC-33): Early data on a new scale.
Journal of Interpersonal Violence, 4(2), 151–163.
Bödvarsdottir, I., & Elklit, A. (2004). Psychological reactions in Icelandic earthquake survivors.
Scandinavian Journal of Psychology, 45, 3–13.
Downloaded by [University of Glasgow] at 07:43 01 January 2015

Ciechanowski, P., Walker, E.A., Katon, W.J., & Russo, J.E. (2002). Attachment theory: A model for
healthcare utilization and somatization. Psychosomatic Medicine, 64, 660–667.
Collins, N.L. (1996). Working models of attachment: Implications for explanation, emotion, and
behavior. Journal of Personality & Social Psychology, 71, 810–832.
Collins, N.L., Cooper, M.L., Albino, A., & Allard, L. (2002). Psychosocial vulnerability from
adolescence to adulthood: A prospective study of attachment style differences in relationship
functioning and partner choice. Journal of Personality, 70, 965–1008.
Collins, N.L., & Feeney, B.C. (2000). A safe haven: An attachment theory perspective on support
seeking and caregiving in intimate relationships. Journal of Personality & Social Psychology, 78,
1053–1073.
Collins, N.L., & Feeney, B.C. (2004). Working models of attachment shape perceptions of social
support: Evidence from experimental and observational studies. Journal of Personality & Social
Psychology, 87, 363–383.
Collins, N.L., & Read, S.J. (1990). Adult attachment, working models, and relationship quality in
dating couples. Journal of Personality & Social Psychology, 58, 644–663.
Dekel, R., Solomon, Z., Elklit, A., & Ginzburg, K. (2004). World assumptions and combat-related
posttraumatic stress disorder. Journal of Social Psychology, 144, 407–420.
Dieperink, M., Leskela, J., Thuras, P., & Engdahl, B. (2001). Attachment style classification and
posttraumatic stress disorder in former prisoners of war. American Journal of Orthopsychiatry,
71, 374–378.
Elklit, A. (1996). Coping styles questionnaire: A contribution to the validation of a scale for
measuring coping strategies. Personality & Individual Differences, 21, 809–812.
Elklit, A., Pedersen, S.S., & Jind, L. (2001). The crisis support scale: Psychometric qualities and
further validation. Personality & Individual Differences, 13, 1291–1302.
Elklit, A., Shevlin, M., Solomon, Z., & Dekel, R. (2007). Factor structure and current validity of the
world assumption scale. Journal of Traumatic Stress, 20, 303–312.
Fearon, R.M.P., & Mansell, W. (2001). Cognitive perspectives on unresolved loss: Insights from the
study of PTSD. Bulletin of the Menninger Clinic, 65, 380–396.
Feeney, B.C., & Collins, N.L. (2001). Predictors of caregiving in adult intimate relationships:
An attachment theoretical perspective. Journal of Personality & Social Psychology, 25, 972–
994.
Feeney, B.C., & Collins, N.L. (2003). Motivations for caregiving in adult intimate relationships:
Influences on caregiving behavior and relationship functioning. Personality & Social Psychology
Bulletin, 26, 950–968.
Feldner, M.T., Lewis, S.F., Leen-Feldner, E.W., Schnurr, P.P., & Zvolensky, M.J. (2006). Anxiety
sensitivity as a moderator of the relation between trauma exposure frequency and posttraumatic
stress symptomatology. Journal of Cognitive Psychotherapy, 20, 201–213.
First, M.B., Frances, A., & Pincus, H.A. (2002). DSM-IV-TR handbook of differential diagnosis.
Psychiatric Services, 53, 878–890.
Fraley, R.C., & Brumbaugh, C.C. (2004). A dynamical systems approach to conceptualizing and
studying stability and change in attachment security. In W.S. Rholes & J.A. Simpson (Eds.),
Adult attachment: Theory, research, and clinical implications (pp. 86–132). New York: Guilford.
 70 M.O’Connor and A. Elklit

Fraley, R.C., & Shaver, P.R. (2000). Adult romantic attachment: Theoretical developments,
emerging controversies, and unanswered questions. Review of General Psychology, 4, 132–154.
Fraley, R.C., Waller, N.G., & Brennan, K.A. (2000). An item response theory analysis of self-report
measures of adult attachment. Journal of Personality & Social Psychology, 78, 350–365.
Gil, S. (2005). Coping style in predicting posttraumatic stress disorder among Israeli students.
Anxiety, Stress and Coping: An International Journal, 18, 351–359.
Griffin, D.W., & Bartholomew, K. (1994). Models of the self and other: Fundamental dimensions
underlying measures of adult attachment. Journal of Personality & Social Psychology, 67, 430–
445.
Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of
Personality & Social Psychology, 52, 511–524.
Howell Rolston, C.D. (2003). Attachment styles and the concepts of self and of others. Dissertation
Abstracts International: Section B: The Sciences & Engineering, 63, 3977.
Ireland, J.L., Boustead, R., & Ireland, C.A. (2005). Coping style and psychological health among
adolescent prisoners: A study of young and juvenile offenders. Journal of Adolescence, 28, 411–423.
Janoff-Bulman, R. (1992). Shattered assumptions: Towards a new psychology of trauma. New York:
Downloaded by [University of Glasgow] at 07:43 01 January 2015

Free Press.
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 Clinical
Psychology, 31, 63–73.
Kanninen, K., Punamaki, R-L., & Qouta, S. (2003). Personality and trauma: Adult attachment and
posttraumatic distress among former political prisoners. Peace & Conflict: Journal of Peace
Psychology, 9, 97–126.
Kessler, R.C., Chiu, W.T., Demler, O., & Walters, E.E. (2005). Prevalence, severity, and comorbidity
of 12-month DSM-IV disorders in the national comorbidity survey replication. Archives of
General Psychiatry, 62, 617–627.
Kessler, R.C., Sonnega, A., Bromet, E., & Hughes, M. (1995). Posttraumatic stress disorder in the
National Comorbidity Survey. Archives of General Psychiatry, 52, 1048–1060.
Krog, T., & Duel, M. (2003). Traume symptom checkliste (TSC): en validering og revidering/ Tina
Krog og Mette Duel. Psykologisk Studie Skriftserie, 6, 1–164.
Lazarus, R., & Folkman, S. (1984). Stress, appraisal and coping. New York: Springer.
Meyer, G.J., & Shack, J.R. (1989). Structural converge of mood and personality: Evidence for old
and new directions. Journal of Personality & Social Psychology, 57, 691–706.
Miller, C.A. (2000). The relationship between posttraumatic stress disorder symptomatology in
parents of leukemia survivors and adult attachment classifications. Dissertation Abstracts
International: Section B: The Sciences & Engineering, 60, 4238.
Mollica, R.F., Caspi-Yavin, Y., Bollini, P., & Truong, T. (1992). The Harvard Trauma Ques-
tionnaire: Validating a cross-cultural instrument for measuring torture, trauma, and posttrau-
matic stress disorder in Indochinese refugees. Journal of Nervous & Mental Disease, 180, 111–116.
Monson, C.M., Price, J.L., Rodriguez, B.F., Ripley, M.P., & Warner, R.A. (2004). Emotional
deficits in military-related PTSD: An investigation of content and process disturbances. Journal
of Traumatic Stress, 17, 275–279.
Moreira, J.M., Bernardes, S.A., Andrez, M., Aguiar, P., Moleiro, C., & de Fátima Silva, M. (1998).
Social competence, personality and adult attachment style in a Portuguese sample. Personality &
Individual Differences, 24, 565–570.
Nielson, M.S. (2003). Crisis support and coping as mediators of well-being in persons with spinal
cord lesion. Journal of Clinical Psychology in Medical Settings, 10, 91–99.
Roger, D., Jarvis, G., & Najarian, B. (1993). Detachment and coping: The construction and
validation of a new scale for measuring coping strategies. Personality & Individual Differences,
15, 619–626.
Sarason, B.R., & Duck, S. (2001). Personal relationships. Implications for clinical and community
psychology. Chichester: Wiley.
Schmidt, S., Strauss, B., & Braehler, E. (2002). Subjective physical complaints and hypochondrial
features from an attachment theoretical perspective. Psychology and Psychotherapy: Theory,
Research, and Practice, 75, 313–332.
Schnurr, P.P., & Vielhauer, M.J. (1999). Personality as a risk factor for PTSD. In R. Yehuda (Ed.),
Risk factors for posttraumatic stress disorder. Washington, DC: American Psychiatric
Association.
 Attachment & Human Development 71

Shapinsky, A.C., Rapport, L.J., Henderson, M.J., & Axelrod, B.N. (2005). Civilian PTSD scales:
Relationships with trait characteristics and everyday distress. Assessment, 12, 220–230.
Sroufe, A.L. (2005). Attachment and development: A prospective, longitudinal study from birth to
adulthood. Attachment & Human Development, 7, 349–367.
Stein, H., Jacobs, N.J., Ferguson, K.S., Allen, J.G., & Fonagy, P. (1998). What do adult attachment
scales measure? Bulletin of the Menninger Clinic, 33–82.
Waldinger, R.J., Schulz, M.S., Barsky, A.J., & Ahern, D.K. (2006). Mapping the road from
childhood trauma to adult somatization: The role of attachment. Psychosomatic Medicine, 68,
129–135.
Wayment, H., & Vierthaler, J. (2002). Attachment style and bereavement reactions. Journal of Loss
and Trauma, 7, 129–149.
Wearden, A.J., Lamberton, N., Crook, N., & Walsh, V. (2005). Adult attachment, alexithymia, and
symptom reporting.An extension to the four category model of attachment. Journal of
Psychosomatic Research, 58, 279–288.
Weinfeld, N.S., Sroufe, A., & Egeland, B. (2000). Attachment from infancy to early adulthood in a
high-risk sample: Continuity, discontinuity, and their correlates. Child Development, 71, 695–
Downloaded by [University of Glasgow] at 07:43 01 January 2015

702.
Wolfgang, G. (2005). Relationships between the two dimensions of adult attachment theory and the
five factor theory of personality. Dissertation Abstracts International: Section B: The Sciences &
Engineering, 65, 3763.