Assessment;
Analysis
Assessment type;
Treatment Plan
Word limit/length; 2000 words
Overview
Theory is useful as an underpinning to care for consumers, but it’s important to be able to move the theory into action. In the following case study you will have the opportunity to demonstrate that you can take what you have learned and apply it to real-world scenarios
Learning Outcomes
This assessment task is aligned to the following learning outcomes:
1. Appraise a range of complex acute care conditions engagement and interventions
2. Design a package of Trauma Informed Care applicable to acute mental health settings.
Assignment detail ***VERY IMPOTANT FOR YOU TO FOLLOW
Review and analyse the following scenario and create a holistic wellbeing assessment and treatment plan for the individual based on the principles of ‘Trauma Informed Care’. The plan you develop must be supported by critiqued evidenced-based research, with links between physical and mental wellbeing created and justified. Implementation processes of the package/plan must also be designed with specific reference to overcoming inhibiting factors in the consumer’s experience.
The Scenario
Mark was brought into the Emergency Department by a friend who said Mark told him that he was feeling unsafe. His friend advised that he had to leave and didn’t have time to stay. Mark is a 23- year-old Caucasian male. As you greet Mark, you note that he is around 6ft, has dark shoulderlength hair, and has a stocky build with weight held in his abdominal region. Mark presents as unkempt, wearing track pants, torn plain t-shirt and thongs. His feet appear to be unwashed, and his fingernails and toenails are dirty and unmanicured. Mark has limited eye contact and does not respond to your offer to shake his hand. You explain to Mark that your role is to offer support and help him. You further explain that this is done by listening to his concerns and working with him to identify helpful interventions whilst supporting him in keeping safe.
You ask Mark to share some information about himself. Mark explains in a regular rate and softly spoken voice that he has recently separated from his partner Annie and doesn’t know how to live without her. They had been together since he was 17. He explained that they have a four-year-old son together who now lives with Annie. Mark advised that Annie kicked him out of the house because she was sick of him drinking and passing out on the lounge. When she was packing up his stuff, she also found a few bottles of vodka around the house, about which she got upset. He advised that he is now staying with his mates or sleeping in his car. He also reported that he had been picked up by police and charged for high range drink driving a few days ago. Mark states that he hasn’t slept properly in a week.
Mark stated that he works as a labourer doing whatever needs to be done on the job site; however, that job is off and on again. He advised it was the only job he knew and that he didn’t like working out in the heat and that his back had started to hurt. Mark said the only thing he likes about work is going to the pub with the boys for drinks after a day’s work. Even though he knows he shouldn’t, Mark said he sometimes needs a drink before he finishes work to help him get through the day. He doesn’t get as much work now as he did when he was younger and most of his money goes on alcohol and to Annie to help with their son.
His friendship group is limited to the guys he met at the pub and on the worksite. He lost touch with his old school friends when he moved away from home. His parents live in a small rural town, and he doesn’t have anything to do with them. He described his upbringing as really rough and that his father used to drink heavily and would hit his mother when drunk. Mark recalled not having enough money for food or new clothes because his Dad was repeatedly fired from several jobs. He recalled his Dad ended up in the local lock-up for a few days for assaulting a man who confronted him about how he spoke to Mum at the pub. Mark shared how he tried to stop his Dad hitting his Mum but stated that his Dad ended up just hitting him too. Mark believes he was a disappointment to his parents, he feels that his Mum loves him but knows his Dad can’t stand the sight of him and nothing he did was ever good enough. Mark said he missed his Mum but knew he couldn’t go back.
The only thing he liked about growing up in the country was that it taught him how to handle guns. He has a rifle that he hasn’t used in a while as he has no one to go ‘roo shooting’ with.
Mark explained that he hated to leave his Mum but needed to move away from his Dad. He met Annie online when he was in year 11 at school and moved in with her. He didn’t finish school. Mark said that they got along well at the start, and Annie loved him so much. He and Annie did everything together, and then he started work as a labourer. Mark began to drink more, and Annie got upset with this. Mark said that Annie didn’t understand that that’s how he prefers to wind down after a hard day at work. They argued even more, when their son was born. Mark said that everything changed when he came along. He said that Annie told him that he had a problem with drinking and needed to stop.
Mark described his mood as really low, and this is congruent with the way he presented. He added that he felt hopeless and couldn’t do anything right. Mark stated, “I don’t know how I am going to live without Annie; she was my best friend.” Upon assessing the risk of self-harm, Mark stated he had thought about suicide when he was younger, but his uncle talked him out of it. He had a friend who died due to suicide when he was 15, and Mark saw what that did to his family “tore them up”. He further said that he had thought recently about suicide after breaking up with Annie and that the feelings come in waves. He said he had strong feelings again when the police charged him. Mark said he always talked to his uncle about these things as his uncle ‘really understood me’. His uncle passed away from cancer when Mark was 18.
Mark denied having experienced any hallucinations or other altered sensory perceptions. Mark believes he is ‘a loser’ and stated he was afraid he would turn out just like his old man (father). Mark didn’t appear to have any long-term memory impairment; however, he does describe blackouts when he can’t remember how he gets home from the pub. This has been increasing in recent months. He is orientated to place but cannot recall the day, date or time.
You ask Mark what he might need and what might help. His response includes that he hit rock bottom when he got caught drink driving and knows he needs to stop drinking. He would like to get back with Annie and have more time with his son. He also wants to increase his energy – he stated, “look at me, I have no energy, I can’t even be bothered having a shower”. He said his friend brought him here as he told his friend that he couldn’t take much more and wanted to end it all. However, he did add that he doesn’t feel like that all the time and that he currently feels safe talking to you. Mark said he would like a new job and always wanted to work as a mechanic. He also needs somewhere to stay while he works it out with Annie
TASK
*****VERY IMPORTANT TO HELP IN STRUCTURE THIS ASSIGNMENT****
Part A – Approx. 1500 words
Concerning the assessment of Mark provided in the scenario, outline what additional information you would obtain and why to complete a comprehensive and holistic assessment. Ensure you include information for an evaluation of risk.
Part B – Approx. 500 words
Write a letter to Marks GP explaining that you assessed Mark, summarise your findings, what your brief interventions and recommendations were and why, and include any further followup Mark may require from the GP. Keep in mind that Mark will be provided with a copy of the letter, and he should be able to easily read and understand what you consider the problems are and what needs to happen.
Letter Requirements:
1) Outline the critical biological, psychological and social health issues that need to be addressed in this episode of care — paying attention to past trauma/experiences and how they may have contributed to the Marks current presentation.
2) Highlight anything that needs to be addressed to keep Mark safe.
3) Outline your plan of care of brief therapeutic interventions to holistically address/resolve the health problems, referencing the latest evidence and authoritative guidelines.
4) Provide any recommendations that the GP may address to assist Mark in further improving his health, again with reference to the latest evidence.
Assignment Rubric
1-
Criteria; Part A – Additional information & why
Mark 50%
Comprehensive and nuanced outline of the additional information, and why, in order to complete a comprehensive and holistic assessment. Integrates with high quality and relevant evidence and guidelines.
2-
Criteria; Part B – Letter
Mark; 40%;
Comprehensive overview of critical health issues.
Clear and succinct plan of care and recommendations to the GP
3-
Criteria; Writing, grammar and referencing
Mark; 10%
Clear, concise and logically structured paper with a succinct, clear introduction and cogent conclusion.
Very well organised.
Free of grammar and spelling errors.
All citations follow required style.
DEBATE Open Access
Borderline personality disorder and
childhood trauma: exploring the affected
biological systems and mechanisms
Nadia Cattane1, Roberta Rossi2, Mariangela Lanfredi2 and Annamaria Cattaneo1,3,4*
: According to several studies, the onset of the Borderline Personality Disorder (BPD) depends on the
combination between genetic and environmental factors (GxE), in particular between biological vulnerabilities and
the exposure to traumatic experiences during childhood. We have searched for studies reporting possible
alterations in several biological processes and brain morphological features in relation to childhood trauma
experiences and to BPD. We have also looked for epigenetic mechanisms as they could be mediators of the effects
of childhood trauma in BPD vulnerability.
: We prove the role of alterations in Hypothalamic-Pituitary-Adrenal (HPA) axis, in neurotrasmission, in
the endogenous opioid system and in neuroplasticity in the childhood trauma-associated vulnerability to develop
BPD; we also confirm the presence of morphological changes in several BPD brain areas and in particular in those
involved in stress response.
Summary: Not so many studies are available on epigenetic changes in BPD patients, although these mechanisms
are widely investigated in relation to stress-related disorders. A better comprehension of the biological and
epigenetic mechanisms, affected by childhood trauma and altered in BPD patients, could allow to identify “at high
risk” subjects and to prevent or minimize the development of the disease later in life.
Keywords: Borderline personality disorder, Childhood trauma, HPA axis, Endogenous opioid system,
Neurotransmission, Neuroplasticity, Neuroimaging studies, Epigenetic mechanisms
Background
Borderline Personality Disorder (BPD) is a pervasive pat-
tern of emotional dysregulation, impulsiveness, unstable
sense of identity and difficult interpersonal relationships
[1]. The prevalence rates of BPD are between 0.2–1.8%
in the general community, 15–25% among psychiatric
inpatients and 10% of all psychiatric outpatients [2, 3].
Among the different aetiopathological theories that have
been proposed over years, the most supported is the one
proposed by Linehan in 1993 [4], which suggests that
BPD can be the result of the interactions between
biological and psychosocial factors [2], in particular be-
tween biologically based temperamental vulnerabilities
and adverse and traumatic experiences during childhood.
BPD is a disorder primarily characterized by emotion
dysregulation and indeed, patients with BPD show
heightened emotional sensitivity, inability to regulate in-
tense emotional responses, and a slow return to emo-
tional baseline. Linehan proposed also that the
development of BPD occurs within an invalidating devel-
opmental context characterized by intolerance toward
the expression of private emotional experiences during
childhood [4]. As a consequence, children exposed to
this adverse environment show inability to learn how to
understand, label, regulate, or tolerate emotional
responses and, conversely, they vacillate between emo-
tional inhibition and extreme emotional lability.
Recently, Hughes and colleagues [5] have proposed an
integration to the aethiopathogenetic model of BPD,
* Correspondence: annamaria.cattaneo@kcl.ac.uk;
acattaneo@fatebenefratelli.eu
1Biological Psychiatry Unit, IRCCS Istituto Centro San Giovanni di Dio –
Fatebenefratelli, via Pilastroni 4, Brescia, Italy
3Stress, Psychiatry and Immunology Laboratory, Department of Psychological
Medicine, Institute of Psychiatry, King’s College London, 125 Coldharbour
Lane, London SE5 9NU, UK
Full list of author information is available at the end of the article
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Cattane et al. BMC Psychiatry (2017) 17:221
DOI 10.1186/s12888-017-1383-2
http://crossmark.crossref.org/dialog/?doi=10.1186/s12888-017-1383-2&domain=pdf
mailto:annamaria.cattaneo@kcl.ac.uk
mailto:acattaneo@fatebenefratelli.eu
http://creativecommons.org/licenses/by/4.0/
http://creativecommons.org/publicdomain/zero/1.0/
emphasizing the role played by a lack of social proximity
or responsiveness from relevant caregivers in the
development of BPD symptoms, which in turn impairs
the individual’s emotion regulation. Affect regulation dif-
ficulties have been also proposed as key mediators in the
relationship between childhood trauma and BPD [6].
Several studies have shown that a diagnosis of BPD is
associated with child abuse and neglect more than any
other personality disorders [7, 8], with a range between
30 and 90% in BPD patients [7, 9].
Adverse childhood experiences are also related to BPD
symptom severity [9–11]. In support to this, Widom and
collaborators [12] followed 500 children who had suf-
fered physical and sexual abuse and neglect and 396
matched controls, and they observed that significantly
more abused/neglected children met criteria for BPD in
adulthood in comparison to controls. However, the pres-
ence of a risk factor, such as adverse childhood events,
was not necessary or sufficient to explain the reason
why some individuals developed BPD symptoms in
adulthood, whereas others did not.
In a recent study, Martin-Blanco and collaborators
[10] have hypothesized that the interaction of childhood
trauma and temperamental traits could be associated
with the severity of BPD. In this regard, they have evalu-
ated the self-reported history of trauma, the psychobio-
logical temperamental traits and the severity of the BPD
symptoms in a cohort of 130 BPD patients. Data showed
a correlation only between childhood maltreatment and
sociability and no other correlation was observed. More-
over, the interaction between high neuroticism-anxiety
traits and the presence of severe emotional abuse was
associated with the severity of the disorder.
Symptom overlap has been reported between BPD
diagnosis and other disorders including Post-Traumatic
Stress Disorder (PTSD) and other axis I disorders [13].
Moreover, in recent decades, different nosographic de-
scriptions have been suggested to characterize the differ-
ent symptoms associated with trauma, like complex
Post-Traumatic Stress Disorder (cPTSD) [14], also
known as Disorders of Extreme Stress Not Otherwise
Specified (DESNOS) [15], which describes a clinical syn-
drome following an experience of interpersonal trau-
matic victimization and shares many similarities with
BPD, including pathological dissociation, somatizations,
dysregulation of emotions, altered central self and rela-
tional schemas. The definition of cPTSD therefore refers
to the experience of severe and/or prolonged traumatic
situations, and does not merely identify the effects of
devastating traumatic events (like violence or chronic
maltreatment), which fall under the category of PTSD or
Acute Stress Disorder. Indeed, exposure to particular
types of traumatic experiences may result in far more in-
sidious and crippling psychopathogenic disorders than
PTSD, compromising the sound development of attach-
ment behavior related systems and of the ability to
modulate emotions [16]. Recent research is currently try-
ing to determine whether cPTSD and BPD diagnosis in
comorbidity with PTSD are distinct or should both be
considered and named as trauma-related disorders [17]. A
recent review [18] has explored the mechanisms through
which childhood trauma is related to the development of
BPD in adulthood, and has discussed how interrelated fac-
tors (such as heritable personality traits, affect regulation
and dissociation, trauma symptoms) could be mediators
in the relationship between childhood trauma and BPD.
Based on all these findings, in the following para-
graphs we will discuss alterations in several neurobio-
logical systems and in brain morphology that can be
induced by exposure to early life adverse experiences
and that are also associated with BPD (see Table 1). We
will examine the impact of early stressful events on dif-
ferent biological systems and mechanisms, possibly iden-
tifying biomarkers that could be involved in BPD
vulnerability. This might allow to identify at high risk
BPD subjects earlier, and to develop intervention strat-
egies and programs.
Discussion
Neurobiological mechanisms involved in BPD
BPD and the hypothalamic-pituitary-adrenal axis
The Hypothalamic-Pituitary-Adrenal (HPA) axis is one
of the neuroendocrine systems which mediate the
response of the body to stress. Although the stress re-
sponse mechanism is meant to maintain stability or
homeostasis, its long-term activation, as consequence of
chronic stress exposure, may have deleterious effects on
the body, increasing the risk for developing different
kinds of illnesses, including stress-related psychiatric
disorders.
In stress conditions, corticotropin-releasing factor
(CRF) and arginine vasopressin (AVP) are released from
the paraventricular nucleus (PVN) located in the hypo-
thalamus. These peptides travel through the pituitary
portal system and act synergistically to stimulate the re-
lease of the adrenocorticotropic hormone (ACTH) from
the corticotroph cells. Then, ACTH is transported
throughout the systemic circulation and binds to recep-
tors in the adrenal cortex of the adrenal gland, resulting
in the biosynthesis and release of cortisol [19]. Cortisol
can affect multiple organs and biological processes, such
as metabolism, growth, inflammation, cardiovascular
function, cognition, and behavior [20, 21], by binding to
specific receptors in the body and in several brain re-
gions, as the hypothalamus, anterior pituitary and medial
prefrontal cortex. The central and peripheral effects of
cortisol are mediated by two intracellular glucocorticoid
receptor subtypes: the high-affinity type I receptor or
Cattane et al. BMC Psychiatry (2017) 17:221 Page 2 of 14
Table 1 Summary of the papers cited in the review and showing alterations in different biological systems in BPD
Biological systems Authors Sample size Date of study Main Results
HPA axis Southwick et al. [26] 37 subjects with PTSD comorbid
with BPD; 18 subjects only with
PTSD
2003 Higher 24 h urinary cortisol levels in
patients with PTSD compared to
patients with PTSD and comorbid BPD.
Wingenfeld et al. [27] 21 female patients with BPD; 24
healthy female controls.
2007 Higher overnight urinary cortisol levels
in BPD patients compared to controls.
Very high cortisol levels were found
only in BPD patients with a low number
of PTSD symptoms.
Rinne et al. [28] 39 BPD patients (24 with and 15
without sustained childhood
abuse and comorbid PTSD
(n = 12) or MDD (n = 11));
11 control subjects
2002 Higher ACTH and cortisol levels in the
blood of BPD females who had
experienced childhood abuse during
the DEX/CRH test.
Carvalho Fernando et al. [29] 32 female BPD patients; 32
healthy female
2013 Acute cortisol levels decreased the
reaction time to target stimuli in both
BPD patients and controls.
Martin-Blanco et al. [30] 481 subjects with BPD; 442
controls
2016 Case-control study focusing on 47 SNPs
in 10 HPA axis genes. An association
between polymorphic variants within
the FKPB5 and the CRHR genes with
the diagnosis of BPD was shown. Two
FKBP5 SNPs were more frequently
represented in patients with a history
of childhood trauma.
Neurotransmission Wagner et al. [42] 159 BPD patients 2009 Association between stressful events
and low impulsivity in BPD patients.
5-HTTLPR S-allele carriers showed
higher impulsivity scores when exposed
to stressful events than LL omozygotes.
Wagner et al. [47] 112 female BPD patients 2010 COMT Val158Met SNP was associated
with early life stressful events and
impulsive aggression in female BPD
patients
Wagner et al. [48] 159 BPD patients 2010 The effect of COMT Val158Met SNP on
the association between stressful life
events and impulsivity was not confirmed.
Tadic et al. [49] 161 Caucasian BPD patients;
156 healthy controls.
2009 The COMT Met158Met SNP was
over-represented in BPD patients compared
to controls. No differences in 5-HTTLPR
genotype were found. An interaction
between the COMT Met158 and the
5-HTTLPR s-allele was observed.
Martin-Blanco et al. [50] 481 BPD subjects; 442 controls 2015 Genetic variants within COMT, DBH and
SLC6A2 genes were associated with an
enhanced risk to develop BPD
Endogenous Opioid
System
Kalin et al. [57] 8 infant rhesus monkeys
(4 males and 4 females)
1988 The endogenous opioid system mediates
separate-induced vocalizations and
influences the HPA axis activation in
rhesus monkeys using the mother-infant
separation paradigm.
Prossin et al. [61] 18 un-medicated female BPD
patients; 14 female controls
2010 BPD patients had greater regional μ-opioid
availability at baseline in the left necleus
accumbens, the hypothalamus and the
right hippocampus/parahippocampus
relative to controls, showing an
endogenous opioid system activation.
Neuroimaging
studies
Driessen et al. [36] 21 female BPD patients;
21 female controls
2000 Volume reduction in the hippocampus
and in the amygdala in BPD patients
compared to controls.
Schmahl et al. [38] 25 unmedicated female patients
with BPD (10 with and 15 without
comorbid PTSD);
25 female controls
2009 Hippocampal volume reduction in
patients with BPD and comorbid PTSD
as compared to controls.
Cattane et al. BMC Psychiatry (2017) 17:221 Page 3 of 14
mineralcorticoid receptor (MR) and the low-affinity type
receptor or glucocorticoid receptor (GR). It has been
suggested that MRs have a high affinity for both cortisol
and aldosterone; they bind cortisol when it is detectable
at low concentrations. The GRs have a relatively low af-
finity for cortisol, but high affinity for dexamethasone
(DEX) [22]; moreover, they bind cortisol at high concen-
tration, reflecting what occurs in stress conditions.
The HPA axis is regulated by an auto-regulatory
mechanism mediated by cortisol itself, that is crucial in
the maintenance of the homeostatic functions of the
HPA axis. Indeed, when cortisol levels rise, as in
Table 1 Summary of the papers cited in the review and showing alterations in different biological systems in BPD (Continued)
Kreisel et al. [70] 39 BPD patients; 39 controls 2014 Smaller hippocampal volume in BPD
patients with a lifetime history than
those without comorbid PTSD.
Boen et al. [71] 18 women with BPD; 21 controls 2014 Two hippocampal structures (DG-CA4
and CA2–3 subfields) were significantly
smaller in patients with BPD than controls.
Kuhlmann et al. [73] 30 BPD patients; 33 controls 2013 Patients with BPD showed lower
hippocampal volumes than controls, but
higher volumes in the hypothalamus.
Rodrigues et al. [63] 124 BPD patients; 147 controls 2011 Both the left and the right sides of the
hippocampus were reduced in BPD patients
with PTSD when compared to controls.
Ruocco et al. [37] 205 BPD patients; 222 controls 2012 Bilateral volume reductions of the amygdala
and hippocampus were not related to
comorbid MDD, PTSD or substance use
disorders.
Epigenetics Martin-Blanco et al. [88] 281 subjects with BPD 2014 An association between NR3C1 methylation
levels and childhood trauma was found in
blood samples of BPD patients.
Dammann et al. [89] 26 BPD patients; 11 controls 2011 An increase in the methylation levels of
HTR2A,NR3C1,MAOA,MAOB and COMT
was found in BPD patients as compared
to controls.
Perroud et al. [91] 346 BD, BPD, and ADHD patients 2016 Differential 5-HT3AR methylation levels
were associated with the severity of
childhood trauma, mainly found in BPD
patients.
Teschler et al. [93] 24 female BPD patients;
11 female controls
2013 Genome-wide methylation analyses revealed
increased methylation levels of several genes
(APBA2,APBA3,GATA4,KCNQ1,MCF2,NINJ2,
TAAR5) in blood of BPD female patients
and controls.
Prados et al. [94] 96 BPD subjects suffering from a
high level of child adversity; 93
subjects suffering from MDD and
reporting a low rate of child
maltreatment
2015 Several CpGs within or near genes involved
in inflammation and in neuronal excitability
were differentially methylated in BPD patients
compared to MDD patients or in relation to
the severity of childhood trauma.
Teschler et al. [95] 24 female BPD patients;
11 female controls
2016 A significant aberrant methylation of rDNA and
PRIMA1 was revealed for BPD patients using
pyrosequencing. For the promoter of PRIMA1, the
average methylation of six CpG sites was higher in
BPD patients compared to controls. In contrast, the
methylation levels of the rDNA promoter region and
the 5′ETS were significantly lower in patients with BPD
compared to controls.
Neuroplasticity Koenigsberg et al. [109] 24 medication-free BPD patients;
18 healthy control subjects
2012 Decrease of PKC and BDNF protein levels in the blood
of BPD patients.
Tadic et al. [49] 161 Caucasian BPD patients;
156 healthy controls.
2009 Association between HTR1B A-161 variant and the
functional BDNF 196A allele in BPD patients.
Perroud et al. [90] 115 subjects with BPD;
52 controls
2013 Higher methylation levels in BDNF CpG exons I and IV
in BPD patients than in controls. Higher BDNF protein
levels in plasma of BPD patients than in controls.
Thaler et al. [92] 64 women with bulimia nervosa
and comorbid BPD; 32 controls
2014 Hypermethylation within BDNF promoter region sites
in women with bulimia nervosa and with a history of
BPD and/or trauma events.
Cattane et al. BMC Psychiatry (2017) 17:221 Page 4 of 14
response to stress, the MRs are saturated and, conse-
quently, cortisol binds the GRs, promoting a cascades of
events that represent the main transduction signals of
glucocorticoids in stress conditions.
So far, the HPA axis activity has been widely investi-
gated in the context of childhood trauma experiences
and findings support alterations in HPA axis in subjects
exposed to stress early in life. Indeed, several studies
have reported alterations in the cortisol circadian
rhythm and levels, indicating a deregulation of the HPA
axis responsiveness, due to childhood trauma experi-
ences, upon stress conditions [23–25].
Despite the large amount of data on the HPA axis
functionality as consequence of exposure to stress early
in life, only a few studies have investigated possible alter-
ations of this axis in BPD patients. For example, higher
urinary cortisol levels have been found in BPD patients
compared to controls [26, 27].
Southwick and colleagues [26] found higher 24 h urinary
cortisol levels in patients with PTSD compared to patients
with PTSD and comorbid BPD, suggesting that these alter-
ations might reflect differences in the severity of PTSD
symptoms rather than factors related to BPD per se.
Another study [27] explored overnight urinary free
cortisol levels showing higher cortisol levels in BPD pa-
tients than in controls. A negative association between
cortisol and PTSD symptoms was also observed. More-
over, when BPD patients were divided according to the
presence of high or low number of PTSD symptoms,
very high cortisol levels were found only in BPD patients
with a low number of PTSD symptoms. Rinne and col-
laborators [28] found an exaggerated ACTH and cortisol
response during the DEX/CRH test in the blood of BPD
female subjects who had experienced childhood abuse.
Carvalho Fernando and colleagues [29] investigated the
effects of cortisol administration on response inhibition
of emotional stimuli in patients with BPD compared to
controls. They found that acute cortisol elevations
decreased the reaction time to target stimuli in both
BPD patients and controls, but they did not differ in task
performance.
Also genetic association studies support alterations in
HPA axis functionality in association with childhood
trauma exposure. Martin-Blanco and collaborators [30]
have investigated the contribution of genetic variants
within genes in the HPA axis, also in the context of child-
hood trauma exposures, in a sample of BPD patients and
controls. The authors performed a case-control study fo-
cusing on 47 SNPs in 10 HPA axis genes. Data showed an
association between polymorphic variants within the
FK506 Binding Protein 5 (FKBP5) and Corticotropin Re-
leasing Hormone Receptor (CRHR) genes with the diag-
nosis of BPD. In particular, two FKBP5 polymorphisms,
rs4713902 and rs9470079, showed significant association
with BPD. Stronger associations were found in patients
exposed to childhood trauma where the risk alleles of
other two FKBP5 polymorphisms, rs3798347-T and
rs10947563-A, were more frequently represented in
patients with a history of childhood physical abuse and
emotional neglect than in patients who had never experi-
enced these trauma and controls.
All these findings suggest an association between a
deregulated functionality of the HPA axis and childhood
trauma and highlight the involvement of this biological
system in the development of BPD.
BPD and neurotransmission
In addition to the presence of HPA axis dysfunction,
several studies have also proposed that childhood
trauma can affect glutamatergic, serotonergic, dopamin-
ergic and noradrenergic transmission, suggesting that
BPD is the result of alterations in several interacting
neurotransmitter systems [31, 32].
Glutamatergic and N-methyl-D-aspartate (NMDA)
neurotransmissions play a critical role in neurodevelop-
ment, synaptic plasticity, learning and memory [33, 34]
and alterations in all these processes have been involved
also in the vulnerability and pathophysiology of BPD
[35]. For example, neuroimaging studies in BPD patients
as compared to controls have consistently demonstrated
the presence of decreased synaptic density and volume
in several brain regions involved in spatial or autobio-
graphical memory and in the modulation of vigilance
and negative emotional states, such as hippocampus and
amygdala, which are also enriched in NMDA receptors
[36] (see also paragraph “BPD and neuroimaging stud-
ies”). Moreover, early chronic stress and mistreatments
experienced during life by BPD patients have been found
able to impact dendritic arborization, thus contributing
to the development of morphological alterations associ-
ated with BPD symptoms [37, 38].
The serotonin transporter gene (5-HTTLPR) and its
related signaling in neurotransmission represent another
system involved in the pathogenesis of BPD [39–42]. In
particular, a functional single nucleotide polymorphism
(SNP) within this gene (the 5-HTTLPR S/L SNP) has
been widely reported to be a modulator of early life
stressful events by several studies [43–45]; interestingly,
it has been also associated with BPD symptoms [42, 46].
For example, Wagner and collaborators [42] investigated
the effects of 5-HTTLPR S/L SNP and of early life
stressful events on impulsivity, assessed by the Barratt
Impulsiveness Scale (BIS), in BPD patients. The authors
reported an association between the presence of stressful
events with lower BIS impulsivity scores, suggesting that
subjects who have experienced trauma, in particular sex-
ual abuse, may show a reduced impulsivity as a conse-
quence of the activation of coping mechanisms that
Cattane et al. BMC Psychiatry (2017) 17:221 Page 5 of 14
control behavior and social interaction. Further analyses
conducted by the same authors indicated that S-allele
carriers showed higher impulsivity scores when exposed
to early life stressful events as compared to LL omozy-
gotes, suggesting that patients with 5-HTTLPR S-allele
are more vulnerable to early life stress. These data high-
light the contribution of the serotonergic system on im-
pulsivity in BPD [42].
Another gene suggested to be a genetic risk factor for
BPD is represented by Catechol-O-methyltransferase
(COMT), an enzyme catalyzing the degradation of cate-
cholamines, including the neurotransmitters dopamine,
epinephrine, and norepinephrine; however, literature
data on the role of this gene are contrasting. In a first
study conducted by Wagner and collaborators [47], the
COMT Val158Met SNP has been found associated with
early life stressful events and impulsive aggression,
assessed by the Buss-Durkee-Hostility Inventory (BDHI)
sum score, in female BPD patients. In particular, the au-
thors identified that in COMT Val158Val carriers, but
not in Val/Met and Met/Met carriers, childhood sexual
abuse and the cumulative number of stressful events
were associated with lower BDHI impulsive aggression
scores. However, in another study conducted by the
same authors, the effect of the COMT Val158Met SNP
on the association between stressful life events and im-
pulsivity was not confirmed [48], probably due to the
small sample size. The same authors [49] also investi-
gated, in a group of BPD patients and controls, the role
of (i) the COMT Val158Met SNP, (ii) the 5-HTTLPR S/L
variant and (iii) their interaction as genetic vulnerability
factors for BPD. Data showed that the genotype COMT
Met158Met was over-represented in BPD patients than
in controls, whereas no differences in 5-HTTLPR geno-
type between BPD and controls were reported. In
addition, the COMT Met158Met genotype was signifi-
cantly over-represented in BPD patients carrying at least
one 5-HTTLPR S-allele and, interestingly, an interaction
between the COMT Met158 and the 5-HTTLPR S-allele
was also observed. These results suggest an interactive
effect of COMT and 5-HTTLPR gene variants on the
vulnerability to develop BPD and, according to the au-
thors, highlight again the key role of the serotonergic
and dopaminergic system in the pathogenesis of BPD.
Martin-Blanco and collaborators [50] investigated the
possible involvement of the noradrenergic system in BDP
pathogenesis, by evaluating genetic variants within 4 nor-
adrenergic genes. In addition to COMT, the authors se-
lected Dopamine Beta-Hydroxylase (DBH), that acts
transforming dopamine into noradrenaline, Solute Carrier
Family 6 Member 2 (SLC6A2), a transporter responsible
for the reuptake of extracellular neurotransmitters, and
Adrenoceptor Beta 2 (ADRB2), that mediates the
catecholamine-induced activation of adenylate cyclase
through the action of G proteins. The authors’ findings in-
dicated that only genetic variants within 3 genes (COMT,
DBH and SLC6A2) were associated with an enhanced risk
to develop BPD.
These studies, taken together, show that alterations in
several neurotransmitter systems could be involved in
BPD pathogenesis; however, due to the small number of
available studies, further investigations are needed.
BPD and the endogenous opioid system
According to Bandelow and Schmahl’s theory, a reduc-
tion in the sensitivity of the opioid receptors or in the
availability of endogenous opioids might constitute part
of the underlying pathophysiology of BPD [51].
Endogenous opioids mainly include three classes (en-
dorphins, enkephalins and dynorphins), which activate
three types of G protein-coupled receptors (μ, δ, and κ
opioid receptors [52]). One of the most important en-
dogenous opioid is β-endorphin which is synthesized in
part in the arcuate nucleus of the hypothalamus and is
released into the blood, the spinal cord and in various
brain regions, including reward-related areas [53]. β-
endorphin is activated by a variety of stressors [54] and
induce euphoria and analgesic effects (for example dur-
ing childbirth and during positive experiences [55]).
The μ-opioid receptors appear to be more relevant for
the social and affective regulation associated with BPD,
suggesting that this system can contribute to the interper-
sonal vulnerabilities and intrapersonal pain of BPD. These
receptors are widely distributed throughout the human
Central Nervous System (CNS), with a particular density
in the basal ganglia, cortical structures, thalamic nuclei,
spinal cord, and specific nuclei in the brainstem [56].
The endogenous opioid system modulates responses
to acute and chronic stressful and noxious stimuli that
induce physical, emotional, or social pain. In animal
models, the endogenous opioid system has been
implicated in affiliative responses, emotion and stress
regulation, including stress-induced analgesia and
impulsive-like behavior [57]. Using the mother-infant
separation paradigm in rhesus monkeys, Kalin and col-
laborators [57] studied for the first time the role of the
opioid system in modulating the behavioural and neuro-
endocrine consequences of a brief occurring stressor.
The authors conducted several experiments where ani-
mals received morphine, an opioid agonist, naloxone, an
opioid antagonist or both to test the increase in
vocalization and the activation of the HPA axis in infant
primates separated or not from their mothers. The re-
sults showed that morphine significantly decreased
separation-induced vocalizations and locomotion with-
out affecting activity levels, whereas naloxone increased
separation-induced vocalizations and environmental ex-
ploration. When the two drugs were co-administered,
Cattane et al. BMC Psychiatry (2017) 17:221 Page 6 of 14
the effect of morphine was reversed only with the
0.1 mg/kg dose of naloxone. The authors also assessed
the effects of separation on neuroendocrine function
and tested whether activation of the opioid system may
attenuate these effects by measuring plasma concentra-
tions of ACTH and cortisol in infant rhesus monkeys
separated or not separated from their mothers, treated
with morphine or naloxone or co-treated with the two
drugs. Plasma ACTH and cortisol levels were higher in
infant rhesus monkeys separated from their mothers
compared to not separated ones, confirming the involve-
ment of the HPA axis during stress exposure. However,
only ACTH plasma levels were modulated by morphine
and by naloxone and by their interaction in the group of
infant separated by their mothers. These findings suggest
that the endogenous opioid system is involved in medi-
ating separation-induced vocalizations and influences
the HPA axis activation following a stress condition.
In humans, regional endogenous opioid system activa-
tion has been associated with suppression of both sen-
sory and affective qualities of stressors and with trait
impulsivity [58–60] whereas its regional deactivation has
been related to hyperalgesic responses and increases in
negative affect during stress [61]. The hypothesis is that
the activation of the μ-opioid receptors could have a
suppressive effect during emotional or physical chal-
lenges that threaten organism homeostasis.
Research has described regional alterations in the
function of the endogenous opioid system and μ-opioid
receptors in brain regions involved in emotion and stress
processing, decision making, and pain and neuroendo-
crine regulation. However, to date, there is only limited
evidence of alterations of endogenous opioid levels in
BPD patients. In an interesting study Prossin and collab-
orators [61] investigated the role of the endogenous opi-
oid system and μ-opioid receptors in emotion regulation
in un-medicated female BPD patients compared to fe-
male controls by using positron emission tomography
(PET) (see paragraph “BPD and neuroimaging studies”
for details).
Comparing BPD patients to their matched controls,
the authors found significant differences in baseline re-
gional μ-opioid receptor concentrations in vivo, as well
as in this neurotransmitter system’s response to a nega-
tive emotional challenge that can be related to some of
the clinical characteristics of BPD.
BPD and neuroimaging studies
Volumetric alterations in brain areas involved in stress
response
To date, several functional and structural in vivo neuro-
imaging studies have been performed in BPD patients,
detecting alterations mainly localized in the limbic cir-
cuit and in frontal cortex. These regions are related to
the distinctive clinical features of the disorder (i.e impul-
sivity, aggression, and emotional reactivity). The most
replicated result, confirmed in recent meta-analyses [37,
62, 63], is represented by the reduction in the volumes
of the hippocampus and the amygdala of BPD patients
compared to controls [36, 64–69]. The robustness of this
finding seems to suggest that volumetric decreases in
these two brain areas could be specific for BPD and thus
useful as possible endophenotypes of illness. In 2000
Driessen and collaborators [36] performed the first mag-
netic resonance imaging volumetric measurement of the
hippocampus, amygdala, temporal lobes, and prosen-
cephalon in 21 female BPD patients and female controls,
reporting in BPD patients a volume reduction of the
16% in the hippocampus and of the 8% in the amygdala.
Moreover, hippocampal volumes were negatively corre-
lated with the extent and the duration of self-reported
early trauma, but only in the entire sample of BPD pa-
tients and controls.
The role of PTSD and trauma as comorbidity with BPD
on hippocampus and amygdala volumes has been object
of investigation but the results are still controversial.
Schmahl and colleagues [38] compared two groups of un-
medicated BPD female patients with and without comor-
bid PTSD and 25 female controls. They found reduced
hippocampal volumes only in patients with BPD and co-
morbid PTSD but not in BPD patients without a history
of PTSD as compared to controls. Similarly, Kreisel and
collaborators [70] investigated in details the hippocampal
structural volumes comparing 39 BPD patients with 39
matched controls, and, although no volume differences
were found between the two groups, patients with a life-
time history of PTSD had a smaller hippocampal volume
(−10,5%) than those without comorbid PTSD. Boen and
collaborators [71] investigated the volumes of the Cornu
Ammonis (CA) and the Dentate Gyrus (DG), two hippo-
campal structures prone to morphological changes [72] in
response to adverse environmental changes in a group of
18 women with BPD and 21 controls. The authors found
that the stress-vulnerable DG-CA4 and CA2–3 subfields
were significantly smaller in patients with BPD than in
controls. However, they did not identify any significant
association between subfield volumes and reported child-
hood trauma.
In another interesting study, Kuhlmann and collabora-
tors [73] investigated alterations in the grey matter of
central stress-regulating structures, including hippocam-
pus, amygdala, anterior cingulate cortex and hypothal-
amus, in female patients with BPD and controls. The
authors also explored whether grey matter volume of
these four brain structures was associated with child-
hood trauma, reporting that patients with BPD showed
lower hippocampal volumes than healthy controls, but
higher volumes in the hypothalamus. Interestingly,
Cattane et al. BMC Psychiatry (2017) 17:221 Page 7 of 14
hypothalamic volume correlated positively with a history
of trauma in patients with BPD.
Two recent meta-analyses [37, 63] evaluated whether
the magnitude of hippocampus and amygdala volume
reductions may be associated with state-of-illness factors
and psychiatric disorders (i.e. PTSD) which often co-
occured with BPD. In the Rodrigues’ meta-analysis, the
authors included 7 articles with a total number of 124
patients and 147 controls. They showed that both the
left and the right sides of hippocampal volumes were re-
duced in BPD patients with PTSD when compared to
controls. The left hippocampal volume was not signifi-
cantly smaller in BPD patients without PTSD relative to
healthy controls and the right hippocampal volume was
reduced in patients with BPD without comorbid PTSD,
but to a lesser degree than in BPD patients with PTSD.
In contrast, the results reported by Ruocco’s meta-
analysis [37] which included 11 studies with a total num-
ber of 205 BPD patients and 222 controls, revealed that
bilateral volume reductions of the amygdala and hippo-
campus were unrelated to comorbid Major Depressive
Disorder (MDD), PTSD, or substance use disorders.
Taken together, all these studies show that the main
brain regions involved in BPD are those associated to
stress and highlight the importance of classifying sub-
groups of patients with BPD, especially taking into ac-
count the presence of comorbidity with PTSD or of a
history of childhood trauma. Notwithstanding, the asso-
ciation between the volume reduction and the degree to
which childhood trauma could be responsible for these
changes remains unclear.
Endogenous opiod system alterations in brain regions
involved in stress response
Despite a large amount of data referred to volumetric
and morphological alterations in brain regions associated
to specific clinical features of BPD, not many neuroim-
aging studies have been conducted to investigate the role
of the endogenous opioid system in BPD. As previously
mentioned, Prossin and collaborators [61] measured the
in vivo availability of the μ-opioid receptors (non-dis-
placeable binding potential (BPND)) in a group of un-
medicated female BPD patients compared to female con-
trols by using PET and the selective radiotracer [11C]
carfentanil at baseline and during sustained sadness
states. Patients had greater regional μ-opioid BPND than
controls at baseline (neutral state) in the left nucleus ac-
cumbens, the hypothalamus, and the right hippocam-
pus/parahippocampus relative to comparison subjects,
showing an endogenous opioid system activation. As
suggested by the authors, differences between BPD pa-
tients and controls in baseline in vivo μ-opioid receptor
concentrations and in the endogenous opioid system re-
sponse to a negative emotional challenge can be related
to some of the clinical characteristics of BPD patients.
These findings show alterations in the function of the
endogenous opioid system and μ-opioid receptors in
brain regions involved in emotion and stress processing,
decision making, and pain and neuroendocrine regula-
tion, features also associated with BPD.
BPD and epigenetic mechanisms
The influence of environmental factors, such as child-
hood trauma, has been suggested to occur through
epigenetic mechanisms, which may underlie gene-
environment associated vulnerability to develop stress-
related disorders [74] including BPD where childhood
trauma history occurs in most of the patients (with a
range between 30 and 90%) [7, 9].
Among the most investigated epigenetic mechanisms
there are: (i) DNA methylation, which occurs at CG
dinucleotides (CpG) and can influence the spatial struc-
ture of the DNA and the binding or the repression of
specific DNA-binding proteins to the DNA [75], (ii) his-
tone modifications, which influence the condensation of
the DNA around histone proteins and regulate the ac-
cessibility of functional regions to transcriptional factors
[76] and (iii) post-transcriptional regulation by non-
coding RNAs such as microRNAs (miRNAs) [77].
All these epigenetic processes and, in particular,
changes in DNA methylation have been widely investi-
gated in the context of long-term negative effects of
early life stressful events. In non-human primates and in
rodents, several paradigms of stress early in life, includ-
ing maternal separation or prenatal stress have been as-
sociated with epigenetic alterations via DNA
methylation [78, 79]. For example, non-stressed dams
during pregnancy showed increased frequency of licking
and grooming in the first week of the puppies’ life that
were associated with changes in DNA methylation
within the promoter of genes, such as glucocorticoid re-
ceptor gene (NR3C1), known to be involved in behavior
and neurodevelopment.
The hypothesis is that the quality of maternal care, af-
fected by stress or depression in pregnancy and post-
partum [80, 81] could impact, through epigenetic mech-
anisms, on gene expression and behavioral traits that are
maintained throughout life [78].
Recently, McGowan and colleagues [79] examined
DNA methylation, histone acetylation and gene expres-
sion in a 7 million base pair region of chromosome 18
containing the NR3C1 gene in the hippocampus of adult
rat offspring, whose mothers differed in the frequency of
maternal care. The authors found that the adult off-
spring of high compared to low maternal care showed a
pattern of regions spanning the NR3C1 gene which were
differentially methylated and acetylated, highlighting the
idea that epigenetic changes, in the context of early life
Cattane et al. BMC Psychiatry (2017) 17:221 Page 8 of 14
stress, involve alterations in gene-networks rather than
in a single or few genes.
Similarly, studies in humans reported similar results as
those found in rodents, including the increased methyla-
tion levels within the NR3C1 promoter region in sub-
jects who reported a history of early life adverse events
[82–84]. For example, in another interesting study,
McGowan and collaborators [82] found that in humans
the cytosine methylation levels of the NR3C1 promoter
were significantly increased in the postmortem hippo-
campus obtained from suicide victims with a history of
childhood abuse as compared with those from suicide
victims with no childhood abuse or with control sam-
ples. Decreased levels of NR3C1 mRNA were also identi-
fied, suggesting an effect of childhood abuse on NR3C1
methylation status and gene expression, independently
from suicide.
Several epigenetic studies have been also conducted in
control subjects characterized for a history of childhood
trauma compared to those with no childhood trauma. In
this context, Suderman and colleagues [85] have demon-
strated, by using a genome-wide promoter DNA methy-
lation approach, an abuse-associated hypermethylation
in 31 miRNAs in a sample of control adult males
exposed to childhood abuse. The hypermethylated state
for 6 of these miRNAs was consistent with an hypome-
thylation status of their target genes.
Reduced methylation levels of FKBP5 gene within
regions containing functional glucocorticoid responsive
elements (GRE) were also found in the blood of control
individuals exposed to childhood abuse when compared
to subjects without a history of trauma [86]. This de-
methylation was linked to increased stress-dependent
gene transcription followed by a long-term dysregulation
of the stress hormone system and a global effect on the
function of immune cells and brain areas associated with
stress regulation. Thus, according to the authors, the
changes in FKBP5 methylation levels might increase the
differential responsiveness of FKBP5 to GR activation
that can remain stable over time. Moreover, Labontè and
colleagues [87] have conducted a genome-wide study of
promoter methylation in the hippocampus of individuals
with a history of severe childhood abuse and control
subjects. Methylation profiles were then compared with
corresponding genome-wide gene expression profiles.
Among all the differentially methylated promoters, 248
showed hypermethylation whereas 114 demonstrated hy-
pomethylation and genes involved in cellular/neuronal
plasticity were among the most significantly differentially
methylated.
Despite the contribution of DNA methylation has been
extensively investigated in association with childhood
trauma in the context of pathologies related to stress,
studies on the possible involvement of epigenetic
mechanisms in BPD vulnerability are only at their birth.
Indeed, only few studies are available. In particular,
Martin-Blanco and colleagues, investigated the associ-
ation between NR3C1 methylation status, history of
childhood trauma and clinical severity in blood samples
of BPD subjects, showing an association between
NR3C1 methylation and childhood trauma, in the form
of physical abuse, and a trend towards significance for
emotional neglect [88]. Regarding NR3C1 methylation
and clinical severity, the authors also found a significant
association with self injurious behavior and previous
hospitalizations. All these findings support the hypoth-
esis that alterations in NR3C1 methylation can occur
early in life as consequence of stress exposure and can
persist up to adulthood where subjects with higher
NR3C1 methylation levels are also those with enhanced
vulnerability to develop BPD.
Above to DNA methylation changes within NR3C1,
hypo- or hyper-methylation within other genes have
been found to play a key role in mediating the impact of
early life stress on the development of stress-related dis-
orders, including BPD [89–92]. For example, in a study
conducted by Dammann and colleagues [89] DNA
methylation pattern of 14 genes, selected because previ-
ously associated with BPD and other psychiatric disor-
ders, (COMT, Dopamine Transporter 1 (DAT1),
Gamma-Aminobutyric Acid Type A Receptor Alpha1
Subunit (GABRA1), G Protein Subunit Beta 3 (GNB3),
Glutamate Ionotropic Receptor NMDA Type Subunit 2B
(GRIN2B), 5-Hydroxytryptamine Receptor 1B (HTR1B),
5-Hydroxytryptamine Receptor 2A (HTR2A), Serotonin
Transporter 1 (5-HTT), Monoamine Oxidase A
(MAOA), Monoamine Oxidase B (MAOB), Nitric Oxide
Synthase 1 (NOS1), NR3C1, Tryptophan Hydroxylase 1
(TPH1) and Tyrosine Hydroxylase (TH)), was analyzed
in the whole blood of BPD patients and controls. An in-
crease in the methylation levels of HTR2A, NR3C1,
MAOA, MAOB and COMT was observed in BPD
patients as compared to controls, suggesting that an in-
creased methylation of CpG sites within these genes
may contribute to BPD aetiopathogenesis. Recently,
Perroud and colleagues [91] investigated the role of
childhood trauma on the methylation status of the
Serotonin 3A Receptor (5-HT3AR), including several
CpGs located within or upstream this gene. They ana-
lyzed its association with clinical severity outcomes, also
in relation with a functional genetic SNP (rs1062613)
within 5-HT3AR in adult patients with Bipolar Disorder,
BPD, and Attention Deficit Hyperactivity Disorder
(ADHD). The results showed that differential 5-HT3AR
methylation status was dependent on the history of
childhood maltreatment and the clinical severity of the
psychiatric disorder; this association was not specifically
restricted to one specific psychiatric disorders
Cattane et al. BMC Psychiatry (2017) 17:221 Page 9 of 14
investigated by the authors, but was found in patients
who reported the higher severity indexes of childhood
maltreatment, mainly represented by BPD patients. In
particular, childhood physical abuse was associated with
higher 5-HT3AR methylation levels, whereas childhood
emotional neglect was inversely correlated with CpG1 I
methylation levels. As suggested by the authors, these
results highlight the need to search for history of child-
hood maltreatment in patients suffering from psychiatric
disorders as these events could be associated with the
worse negative outcomes. Moreover, the authors found a
modulation of the 5HT3AR methylation status by
rs1062613 at CpG2 III, where patients carrying the risk
CC genotype showed the highest levels of methylation at
CpG2 III. Since C allele has been also associated with a
lower expression levels of 5HT3AR, the authors sug-
gested that increased methylation, due to exposure to
childhood maltreatment, could lead to a further decrease
in the expression of 5HT3AR mRNA.
Aiming to identify novel genes that may exhibit aber-
rant DNA methylation frequencies in BPD patients,
Teschler and collaborators [93] performed a genome-
wide methylation analysis in the blood of BPD female
patients and female controls. The authors reported in-
creased methylation levels of several genes, including
neuronal adaptor proteins (Amyloid Beta Precursor
Protein Binding Family A Member 2 (APBA2) and
Amyloid Beta Precursor Protein Binding Family A Mem-
ber 3 (APBA3)), zinc-finger transcription factors (GATA
Binding Protein 4 (GATA4)), voltage-gated potassium
channel gene (Potassium Voltage-Gated Channel Sub-
family Q Member 1 (KCNQ1)), guanine nucleotide ex-
change factors (Proto-Oncogene MCF-2 (MCF2)),
adhesion molecules (Ninjurin 2 (NINJ2)) and G protein-
coupled receptors (Trace Amine Associated Receptor 5
(TAAR5)) in BPD samples compared to controls. Simi-
larly, using a whole-genome methylation approach, Pra-
dos and colleagues [94] analyzed the global DNA
methylation status in the peripheral blood leukocytes of
BPD patients with a history of childhood adversity and
also in patients with MDD characterized by a low rate of
childhood maltreatment. Contrary to Teschler [93], who
used control subjects as reference group, in this study
the authors used MDD subjects, most of them suicide
attempters, thus controlling not only for MDD but also
for a history of suicide. The authors also assessed pos-
sible correlations between methylation signatures and
the severity of childhood maltreatment. Data showed
that several CpGs within or near genes involved in in-
flammatory processes (Interleukin 17 Receptor A
(IL17RA)), regulation of gene expression (miR124–3)
and neuronal excitability and development/maintenance
of the nervous system (Potassium Voltage-Gated Chan-
nel Subfamily Q Member 2 (KCNQ2)) were differentially
methylated, either in BPD compared with MDD or in re-
lation to the severity of childhood maltreatment.
In a more recent study, Teschler and collaborators
[95] have analyzed also DNA methylation patterns of the
ribosomal RNA gene (rDNA promoter region and 5′-ex-
ternal transcribed spacer/5′ETS) and the promoter of
the proline rich membrane anchor 1 gene (PRIMA1) in
peripheral blood samples of female BPD patients and
controls. The authors have identified a significant aber-
rant methylation of rDNA and PRIMA1 in the group of
BPD patients. Specifically, the average methylation of 6
CpG sites in the promoter of PRIMA1 was 1.6-fold
higher in BPD patients compared to controls. In con-
trast, the methylation levels of the rDNA promoter re-
gion and the 5′ETS were significantly lower (0.9-fold) in
patients with BPD compared to controls. Furthermore,
decreased methylation levels were found for nine CpGs
located in the rDNA promoter region and for 4 CpGs at
the 5′ETS in peripheral blood of patients compared to
controls. These results suggest that aberrant methylation
of rDNA and PRIMA1 could be associated with the
pathogenesis of BPD.
Taken together, all these studies reveal a complex
interplay between BPD, early-life stressful adversities and
epigenetic signatures.
BPD and neuroplasticity (the role of BDNF)
Neuroplasticity refers to brain-related mechanisms
associated with the ability of the brain to perceive, adapt
and respond to a variety of internal and external stimuli
[96, 97], including stress.
The exposure to acute stressful challenges can induce
several beneficial and protective effects for the body,
which responds to almost any sudden, unexpected events
by releasing chemical mediators – i.e. catecholamines that
increase heart rate and blood pressure – and help the in-
dividual to cope with the situation [20, 98–101]. However,
a chronic exposure to stress and thus a chronic exposure
to glucocorticoids can have negative and persistent effects
on the body, including altered metabolism, altered im-
munity, enhanced inflammation, cognitive deficits, and
also an enhanced vulnerability for psychiatric disorders
and for medical conditions such as cardiovascular disease,
metabolic disorders and cancer [102, 103].
Neurotrophic factors, and in particular the neurotro-
phin Brain-Derived Neurotrophic Factor (BDNF), have
been identified as key mediators of stress on neuronal
connectivity, dendritic arborization, synaptic plasticity
and neurogenesis [104–107]. Since its crucial role in
brain development and brain plasticity, BDNF has been
widely investigated also in several psychiatric diseases,
including BPD [108].
For example, Koenigsberg and colleagues [109] found
a decrease of Protein Kinase C (PKC) isoenzyme, which
Cattane et al. BMC Psychiatry (2017) 17:221 Page 10 of 14
is a molecule downstream the activation of BDNF, and
BDNF protein levels in the blood of BPD patients, sug-
gesting an alteration of BDNF signaling and conse-
quently of neuroplasticity-related mechanisms in BPD.
In another study, Tadic and collaborators [49] investi-
gated the association between BPD and genetic variants
within HTR1B and BDNF genes. Although data showed
no significant differences in genotype or haplotype dis-
tribution for both HTR1B and BDNF variants between
BPD patients and controls, logistic regression analyses
revealed an association between the HTR1B A-161 vari-
ant and the functional BDNF 196A allele in BPD.
Importantly, several findings have also documented epi-
genetic modifications on BDNF gene in patients with
BPD, suggesting that childhood maltreatment in BPD pa-
tients can cause long term epigenetic alterations of genes
crucially involved in brain functions and neurodevelop-
ment, including BDNF, and that these alterations may
contribute to enhanced vulnerability to develop BPD path-
ology. In this regard, Perroud and collaborators [90] mea-
sured the percentage of methylation at BDNF CpG exons
I and IV and also plasma BDNF protein levels in subjects
with BPD and controls. The authors reported significantly
higher methylation status in both CpG regions in patients
than in controls, with the number of childhood trauma
exposures associated with the high levels of BDNF methy-
lation. Moreover, BPD patients had significantly higher
BDNF plasma protein levels than controls, but this in-
crease was not associated with changes in BDNF methyla-
tion status. More recently, Thaler and collaborators [92]
analyzed DNA methylation patterns in the promoter re-
gion of BDNF gene in women with bulimia nervosa and
with history of BPD and/or trauma events. They reported
that bulimia nervosa was associated per se with an hyper-
methylation within BDNF promoter region sites. This was
particularly evident when co-occurring with childhood
abuse or BPD.
Overall, these studies support the hypothesis that child-
hood trauma could be associated with changes in BDNF
epigenetic signature, that in turn could contribute to alter
cognitive functions in BPD patients. Indeed, higher levels
of gene methylation are commonly accompanied by a re-
duced gene expression. Thus higher BDNF methylation
levels should determine reduced expression of BDNF gene
and reduced BDNF mRNA levels are widely observed in
patients with psychiatric diseases [110–112].
Up to now, neither a specific gene variant or biological
mechanism has been exclusively associated with BPD,
but the onset of this disorder has been suggested to
depend on the combination of a vulnerable genetic back-
ground with adverse environmental factors during
childhood.
Among the biological systems found involved in BPD
pathogenesis and particularly affected by childhood
trauma events, there are: the HPA axis, the neurotrans-
mission mechanisms, the endogenous opioid system and
the neuroplasticity. In line with the involvement of these
processes, neuroimaging studies in BPD patients have
shown volume reductions in the hippocampus and amyg-
dala, both brain regions mainly involved in stress
responses, cognition, memory and emotion regulation and
an increase in the μ-opioid receptors in the same areas.
Among the environmental factors, early life stressful
events, in particular childhood trauma, have been pro-
posed to negatively impact brain development through
epigenetic mechanisms. Although a complex interplay
between BPD, early-life stressful adversities and epigen-
etic signatures has been suggested, further investigations
are needed in order to better understand the role of gen-
etic background and traumatic events during childhood
in the onset of BPD. A better comprehension of these
interactions could allow to identify at risk subjects, who
could be treated with preventive therapies, such as psy-
chotherapy, and to prevent or minimize the develop-
ment of the disease later in life.
5-HT3AR: Serotonin 3A Receptor; 5-HTT: Serotonin Transporter 1; 5-HTTLPR: Serotonin
transporter gene; ACTH: Adrenocorticotropic Hormone; ADHD: Attention Deficit
Hyperactivity Disorder; ADRB2: Adrenoceptor Beta 2; APBA2: Amyloid Beta Precursor
Protein Binding Family A Member 2; APBA3: Amyloid Beta Precursor Protein Binding
Family A Member 3; AVP: Arginine Vasopressin; BDHI: Buss-Durkee-Hostility Inventory;
BDNF: Brain-Derived Neurotrophic Factor; BIS: Barratt Impulsiveness Scale;
BPD: Borderline Personality Disorder; CA: Cornu Ammonis; CNS: Central Nervous
System; COMT: Catechol-O-methyltransferase; CpG: CG dinucleotides;
cPTSD: complex Post-Traumatic Stress Disorder; CRF: Corticotropin-Releasing Factor;
CRHR: Corticotropin Releasing Hormone Receptor; DAT1: Dopamine Transporter 1;
DBH: Dopamine Beta-Hydroxylase; DESNOS: Disorders of Extreme Stress Not
Otherwise Specified; DEX: Dexamethasone; DG: Dentate Gyrus; FKBP5: FK506 Binding
Protein 5; GABRA1: Gamma-Aminobutyric Acid Type A Receptor Alpha1 Subunit;
GATA4: GATA Binding Protein 4; GNB3: G Protein Subunit Beta 3; GR: Glucocorticoid
Receptor; GRE: Glucocorticoid Responsive Elements; GRIN2B: Glutamate Ionotropic
Receptor NMDA Type Subunit 2B; HPA axis: Hypothalamic-Pituitary-Adrenal axis;
HTR1B: 5-Hydroxytryptamine Receptor 1B; HTR2A: 5-Hydroxytryptamine Receptor 2A;
IL17RA: Interleukin 17 Receptor A; KCNQ1: Potassium Voltage-Gated Channel Sub-
family Q Member 1; KCNQ2: Potassium Voltage-Gated Channel Subfamily Q Member
2; MAOA: Monoamine Oxidase A; MAOB: Monoamine Oxidase B; MCF2: Proto-
Oncogene MCF-2; MDD: Major Depressive Disorder; miRNAs: microRNAs;
MR: Mineralcorticoid Receptor; NINJ2: Ninjurin 2; NMDA: N-methyl-D-aspartate;
NOS1: Nitric Oxide Synthase 1; NR3C1: Glucocorticoid receptor gene; PET: Positron
Emission Tomography; PKC: Protein Kinase C; PRIMA1: Prolin Rich Membrane Anchor
1; PTSD: Post-Traumatic Stress Disorder; PVN: Paraventricular Nucleus; SLC6A2: Solute
Carrier Family 6 Member 2; SNP: Single nucleotide polymorphism; TAAR5: Trace
Amine Associated Receptor 5; TH: Tyrosine Hydroxylase; TPH1: Tryptophan
Hydroxylase 1
Not applicable.
This work was supported by an Eranet-Neuron Grant to A.C. (Inflame-D
project) and by funding from the Italian Ministry of Health (MoH) to A.C.
The data supporting the conclusions of this article are included within the
article.
Cattane et al. BMC Psychiatry (2017) 17:221 Page 11 of 14
N.C. managed the literature searches and wrote the first draft of the manuscript.
R.R. and M.L. managed the literature searches and completed the manuscript.
A.C. revised and approved the final version of the manuscript. All authors gave
their scientific contribution and have approved the final manuscript.
All the authors declare that they have no conflicts of interest.
All the authors certify that the submission is an original work and it is not
under review at any other journal.
Not applicable- as the submitted manuscript is a review.
Not applicable- as the submitted manuscript is a review.
1Biological Psychiatry Unit, IRCCS Istituto Centro San Giovanni di Dio –
Fatebenefratelli, via Pilastroni 4, Brescia, Italy. 2Psychiatry Unit, IRCCS Istituto
Centro San Giovanni di Dio – Fatebenefratelli, via Pilastroni 4, Brescia, Italy.
3Stress, Psychiatry and Immunology Laboratory, Department of Psychological
Medicine, Institute of Psychiatry, King’s College London, 125 Coldharbour
Lane, London SE5 9NU, UK. 4Department of Psychological Medicine, Institute
of Psychiatry, Psychology and Neuroscience, King’s College London, 125
Coldharbour Lane, London SE5 9NU, UK.
Received: 7 February 2017 Accepted: 6 June 2017
1. Regier DA, Kuhl EA, Kupfer DJ. The DSM-5: classification and criteria
changes. World psychiatry : official journal of the World Psychiatric
Association. 2013;12(2):92–8. doi:10.1002/wps.20050.
2. Leichsenring F, Leibing E, Kruse J, New AS, Leweke F. Borderline personality
disorder. Lancet. 2011;377(9759):74–84. doi:10.1016/S0140-6736(10)61422-5.
3. Lieb K, Zanarini MC, Schmahl C, Linehan MM, Bohus M. Borderline personality
disorder. Lancet. 2004;364(9432):453–61. doi:10.1016/S0140-6736(04)16770-6.
4. Linehan MM. Dialectical behavior therapy for treatment of borderline
personality disorder: implications for the treatment of substance abuse.
NIDA Res Monogr. 1993;137:201–16.
5. Hughes AE, Crowell SE, Uyeji L, Coan JA. A developmental neuroscience of
borderline pathology: emotion dysregulation and social baseline theory. J
Abnorm Child Psychol. 2012;40(1):21–33. doi:10.1007/s10802-011-9555-x.
6. van Dijke A, Ford JD, van der Hart O, van Son M, van der Heijden P, Buhring
M. Affect dysregulation in borderline personality disorder and somatoform
disorder: differentiating under- and over-regulation. J Personal Disord. 2010;
24(3):296–311. doi:10.1521/pedi.2010.24.3.296.
7. Battle CL, Shea MT, Johnson DM, Yen S, Zlotnick C, Zanarini MC, et al.
Childhood maltreatment associated with adult personality disorders:
findings from the collaborative longitudinal personality disorders study. J
Personal Disord. 2004;18(2):193–211.
8. Yen S, Shea MT, Battle CL, Johnson DM, Zlotnick C, Dolan-Sewell R, et al.
Traumatic exposure and posttraumatic stress disorder in borderline,
schizotypal, avoidant, and obsessive-compulsive personality disorders:
findings from the collaborative longitudinal personality disorders study. J
Nerv Ment Dis. 2002;190(8):510–8. doi:10.1097/01.NMD.0000026620.66764.78.
9. Zanarini MC, Frankenburg FR, Hennen J, Reich DB, Silk KR. Prediction of the
10-year course of borderline personality disorder. Am J Psychiatry. 2006;
163(5):827–32. doi:10.1176/ajp.2006.163.5.827.
10. Martin-Blanco A, Soler J, Villalta L, Feliu-Soler A, Elices M, Perez V, et al.
Exploring the interaction between childhood maltreatment and
temperamental traits on the severity of borderline personality disorder.
Compr Psychiatry. 2014;55(2):311–8. doi:10.1016/j.comppsych.2013.08.026.
11. Gunderson JG, Weinberg I, Daversa MT, Kueppenbender KD, Zanarini MC,
Shea MT, et al. Descriptive and longitudinal observations on the relationship
of borderline personality disorder and bipolar disorder. Am J Psychiatry.
2006;163(7):1173–8. doi:10.1176/appi.ajp.163.7.1173.
12. Widom CS, Czaja SJ, Paris J. A prospective investigation of borderline
personality disorder in abused and neglected children followed up into
adulthood. J Personal Disord. 2009;23(5):433–46. doi:10.1521/pedi.2009.23.5.
433.
13. Pagura J, Stein MB, Bolton JM, Cox BJ, Grant B, Sareen J. Comorbidity of
borderline personality disorder and posttraumatic stress disorder in the U.S.
population. J Psychiatr Res. 2010;44(16):1190–8. doi:10.1016/j.jpsychires.2010.
04.016.
14. Herman JL. Complex PTSD: A syndrome in survivors of prolonged and
repeated trauma. J Trauma Stress 1992;5(3):377–391. doi:10.1002/jts.
2490050305.
15. Luxenberg T, Spinazzola, J., Hidalgo, J., Hunt, C., Van Der Kolk, B.A. Complex
trauma and disorders of extreme stress (DESNOS) diagnosis, Part One:
Assessment Directions in Psychiatry 2001;21:373–393.
16. D’Andrea W, Ford J, Stolbach B, Spinazzola J, van der Kolk BA.
Understanding interpersonal trauma in children: why we need a
developmentally appropriate trauma diagnosis. The American journal of
orthopsychiatry. 2012;82(2):187–200. doi:10.1111/j.1939-0025.2012.01154.x.
17. Cloitre M, Garvert DW, Weiss B, Carlson EB, Bryant RA. Distinguishing PTSD,
Complex PTSD, and borderline personality disorder: a latent class analysis.
Eur J Psychotraumatol 2014;5. doi:10.3402/ejpt.v5.25097.
18. MacIntosh HG, N.; Dubash, N.;. Borderline personality disorder: disorder of
trauma or personality, a review of the empirical literature. Can Psychol
2015;56:227–241.
19. Pompili M, Serafini G, Innamorati M, Moller-Leimkuhler AM, Giupponi G,
Girardi P, et al. The hypothalamic-pituitary-adrenal axis and serotonin
abnormalities: a selective overview for the implications of suicide
prevention. Eur Arch Psychiatry Clin Neurosci. 2010;260(8):583–600. doi:10.
1007/s00406-010-0108-z.
20. Lupien SJ, Maheu F, Tu M, Fiocco A, Schramek TE. The effects of stress and
stress hormones on human cognition: implications for the field of brain and
cognition. Brain Cogn. 2007;65(3):209–37. doi:10.1016/j.bandc.2007.02.007.
21. Harris BN, Carr JA. The role of the hypothalamus-pituitary-adrenal/interrenal
axis in mediating predator-avoidance trade-offs. General and comparative
endocrinology. 2016;230–231:110–42. doi:10.1016/j.ygcen.2016.04.006.
22. De Kloet ER. Why dexamethasone poorly penetrates in brain. Stress.
1997;2(1):13–20.
23. Carpenter LL, Carvalho JP, Tyrka AR, Wier LM, Mello AF, Mello MF, et al.
Decreased adrenocorticotropic hormone and cortisol responses to stress in
healthy adults reporting significant childhood maltreatment. Biol Psychiatry.
2007;62(10):1080–7. doi:10.1016/j.biopsych.2007.05.002.
24. Maniam J, Antoniadis C, Morris MJ. Early-life stress, HPA Axis adaptation, and
mechanisms contributing to later health outcomes. Front Endocrinol.
2014;5:73. doi:10.3389/fendo.2014.00073.
25. Papadopoulos AS, Cleare AJ. Hypothalamic-pituitary-adrenal axis dysfunction
in chronic fatigue syndrome. Nat Rev Endocrinol. 2012;8(1):22–32. doi:10.
1038/nrendo.2011.153.
26. Southwick SM, Axelrod SR, Wang S, Yehuda R, Morgan CA 3rd, Charney D,
et al. Twenty-four-hour urine cortisol in combat veterans with PTSD and
comorbid borderline personality disorder. J Nerv Ment Dis. 2003;191(4):261–
2. doi:10.1097/01.NMD.0000061140.93952.28.
27. Wingenfeld K, Driessen M, Adam B, Hill A. Overnight urinary cortisol release
in women with borderline personality disorder depends on comorbid PTSD
and depressive psychopathology. European psychiatry : the journal of the
Association of European Psychiatrists. 2007;22(5):309–12. doi:10.1016/j.
eurpsy.2006.09.002.
28. Rinne T, de Kloet ER, Wouters L, Goekoop JG, DeRijk RH, van den Brink W.
Hyperresponsiveness of hypothalamic-pituitary-adrenal axis to combined
dexamethasone/corticotropin-releasing hormone challenge in female
borderline personality disorder subjects with a history of sustained
childhood abuse. Biol Psychiatry. 2002;52(11):1102–12.
29. Carvalho Fernando S, Beblo T, Schlosser N, Terfehr K, Wolf OT, Otte C, et al.
Acute glucocorticoid effects on response inhibition in borderline personality
disorder. Psychoneuroendocrinology. 2013;38(11):2780–8. doi:10.1016/j.
psyneuen.2013.07.008.
30. Martin-Blanco A, Ferrer M, Soler J, Arranz MJ, Vega D, Calvo N, et al. The role
of hypothalamus-pituitary-adrenal genes and childhood trauma in
borderline personality disorder. Eur Arch Psychiatry Clin Neurosci.
2016;266(4):307–16. doi:10.1007/s00406-015-0612-2.
31. Friedel RO. Dopamine dysfunction in borderline personality disorder: a
hypothesis. Neuropsychopharmacology : official publication of the American
College of Neuropsychopharmacology. 2004;29(6):1029–39. doi:10.1038/sj.
npp.1300424.
Cattane et al. BMC Psychiatry (2017) 17:221 Page 12 of 14
http://dx.doi.org/10.1002/wps.20050
http://dx.doi.org/10.1016/S0140-6736(10)61422-5
http://dx.doi.org/10.1016/S0140-6736(04)16770-6
http://dx.doi.org/10.1007/s10802-011-9555-x
http://dx.doi.org/10.1521/pedi.2010.24.3.296
http://dx.doi.org/10.1097/01.NMD.0000026620.66764.78
http://dx.doi.org/10.1176/ajp.2006.163.5.827
http://dx.doi.org/10.1016/j.comppsych.2013.08.026
http://dx.doi.org/10.1176/appi.ajp.163.7.1173
http://dx.doi.org/10.1521/pedi.2009.23.5.433
http://dx.doi.org/10.1521/pedi.2009.23.5.433
http://dx.doi.org/10.1016/j.jpsychires.2010.04.016
http://dx.doi.org/10.1016/j.jpsychires.2010.04.016
http://dx.doi.org/10.1002/jts.2490050305
http://dx.doi.org/10.1002/jts.2490050305
http://dx.doi.org/10.1111/j.1939-0025.2012.01154.x
http://dx.doi.org/10.3402/ejpt.v5.25097
http://dx.doi.org/10.1007/s00406-010-0108-z
http://dx.doi.org/10.1007/s00406-010-0108-z
http://dx.doi.org/10.1016/j.bandc.2007.02.007
http://dx.doi.org/10.1016/j.ygcen.2016.04.006
http://dx.doi.org/10.1016/j.biopsych.2007.05.002
http://dx.doi.org/10.3389/fendo.2014.00073
http://dx.doi.org/10.1038/nrendo.2011.153
http://dx.doi.org/10.1038/nrendo.2011.153
http://dx.doi.org/10.1097/01.NMD.0000061140.93952.28
http://dx.doi.org/10.1016/j.eurpsy.2006.09.002
http://dx.doi.org/10.1016/j.eurpsy.2006.09.002
http://dx.doi.org/10.1016/j.psyneuen.2013.07.008
http://dx.doi.org/10.1016/j.psyneuen.2013.07.008
http://dx.doi.org/10.1007/s00406-015-0612-2
http://dx.doi.org/10.1038/sj.npp.1300424
http://dx.doi.org/10.1038/sj.npp.1300424
32. Figueroa E, Silk KR. Biological implications of childhood sexual abuse in
borderline personality disorder. J Personal Disord. 1997;11(1):71–92.
33. Snyder MA, Gao WJ. NMDA hypofunction as a convergence point for
progression and symptoms of schizophrenia. Front Cell Neurosci. 2013;7:31.
doi:10.3389/fncel.2013.00031.
34. Kahn RS, Sommer IE. The neurobiology and treatment of first-episode
schizophrenia. Mol Psychiatry. 2015;20(1):84–97. doi:10.1038/mp.2014.66.
35. Grosjean B, Tsai GE. NMDA neurotransmission as a critical mediator of
borderline personality disorder. Journal of psychiatry & neuroscience : JPN.
2007;32(2):103–15.
36. Driessen M, Herrmann J, Stahl K, Zwaan M, Meier S, Hill A, et al. Magnetic
resonance imaging volumes of the hippocampus and the amygdala in
women with borderline personality disorder and early traumatization. Arch
Gen Psychiatry. 2000;57(12):1115–22.
37. Ruocco AC, Amirthavasagam S, Zakzanis KK. Amygdala and hippocampal
volume reductions as candidate endophenotypes for borderline personality
disorder: a meta-analysis of magnetic resonance imaging studies. Psychiatry
Res. 2012;201(3):245–52. doi:10.1016/j.pscychresns.2012.02.012.
38. Schmahl C, Berne K, Krause A, Kleindienst N, Valerius G, Vermetten E, et al.
Hippocampus and amygdala volumes in patients with borderline
personality disorder with or without posttraumatic stress disorder. Journal
of psychiatry & neuroscience : JPN. 2009;34(4):289–95.
39. Ni X, Sicard T, Bulgin N, Bismil R, Chan K, McMain S, et al. Monoamine
oxidase a gene is associated with borderline personality disorder. Psychiatr
Genet. 2007;17(3):153–7. doi:10.1097/YPG.0b013e328016831c.
40. Pascual JC, Soler J, Barrachina J, Campins MJ, Alvarez E, Perez V, et al. Failure
to detect an association between the serotonin transporter gene and
borderline personality disorder. J Psychiatr Res. 2008;42(1):87–8. doi:10.1016/
j.jpsychires.2006.10.005.
41. Tadic A, Baskaya O, Victor A, Lieb K, Hoppner W, Dahmen N. Association
analysis of SCN9A gene variants with borderline personality disorder. J
Psychiatr Res. 2008;43(2):155–63. doi:10.1016/j.jpsychires.2008.03.006.
42. Wagner S, Baskaya O, Lieb K, Dahmen N, Tadic A. The 5-HTTLPR
polymorphism modulates the association of serious life events (SLE) and
impulsivity in patients with borderline personality disorder. J Psychiatr Res.
2009;43(13):1067–72. doi:10.1016/j.jpsychires.2009.03.004.
43. Harkness KL, Bagby RM, Stewart JG, Larocque CL, Mazurka R, Strauss JS, et al.
Childhood emotional and sexual maltreatment moderate the relation of the
serotonin transporter gene to stress generation. J Abnorm Psychol. 2015;
124(2):275–87. doi:10.1037/abn0000034.
44. Benedetti F, Riccaboni R, Poletti S, Radaelli D, Locatelli C, Lorenzi C, et al.
The serotonin transporter genotype modulates the relationship between
early stress and adult suicidality in bipolar disorder. Bipolar Disord. 2014;
16(8):857–66. doi:10.1111/bdi.12250.
45. Duman EA, Canli T. Influence of life stress, 5-HTTLPR genotype, and SLC6A4
methylation on gene expression and stress response in healthy Caucasian males.
Biology of mood & anxiety disorders. 2015;5:2. doi:10.1186/s13587-015-0017-x.
46. Paaver M, Nordquist N, Parik J, Harro M, Oreland L, Harro J. Platelet MAO
activity and the 5-HTT gene promoter polymorphism are associated with
impulsivity and cognitive style in visual information processing.
Psychopharmacology. 2007;194(4):545–54. doi:10.1007/s00213-007-0867-z.
47. Wagner S, Baskaya O, Anicker NJ, Dahmen N, Lieb K, Tadic A. The catechol
o-methyltransferase (COMT) val(158)met polymorphism modulates the
association of serious life events (SLE) and impulsive aggression in female
patients with borderline personality disorder (BPD). Acta Psychiatr Scand.
2010;122(2):110–7. doi:10.1111/j.1600-0447.2009.01501.x.
48. Wagner S, Baskaya O, Lieb K, Dahmen N, Tadic A. Lack of modulating effects
of the COMT Val(158)met polymorphism on the association of serious life
events (SLE) and impulsivity in patients with borderline personality disorder.
J Psychiatr Res. 2010;44(2):121–2. doi:10.1016/j.jpsychires.2009.06.008.
49. Tadic A, Elsasser A, Victor A, von Cube R, Baskaya O, Wagner S, et al.
Association analysis of serotonin receptor 1B (HTR1B) and brain-derived
neurotrophic factor gene polymorphisms in borderline personality disorder.
J Neural Transm. 2009;116(9):1185–8. doi:10.1007/s00702-009-0264-3.
50. Martin-Blanco A, Ferrer M, Soler J, Arranz MJ, Vega D, Bauza J, et al. An
exploratory association study of the influence of noradrenergic genes and
childhood trauma in borderline personality disorder. Psychiatry Res. 2015;
229(1–2):589–92. doi:10.1016/j.psychres.2015.07.046.
51. Bandelow B, Schmahl C, Falkai P, Wedekind D. Borderline personality
disorder: a dysregulation of the endogenous opioid system? Psychol Rev.
2010;117(2):623–36. doi:10.1037/a0018095.
52. Feng Y, He X, Yang Y, Chao D, Lazarus LH, Xia Y. Current research on opioid
receptor function. Curr Drug Targets. 2012;13(2):230–46.
53. Dikshtein Y, Barnea R, Kronfeld N, Lax E, Roth-Deri I, Friedman A, et al. Beta-
endorphin via the delta opioid receptor is a major factor in the incubation
of cocaine craving. Neuropsychopharmacology : official publication of the
American College of Neuropsychopharmacology. 2013;38(12):2508–14. doi:
10.1038/npp.2013.155.
54. Roth-Deri I, Green-Sadan T, Yadid G. Beta-endorphin and drug-induced
reward and reinforcement. Prog Neurobiol. 2008;86(1):1–21. doi:10.1016/j.
pneurobio.2008.06.003.
55. Esch T, Stefano GB. The neurobiology of Love. Neuro endocrinology letters.
2005;26(3):175–92.
56. Stanley B, Siever LJ. The interpersonal dimension of borderline personality
disorder: toward a neuropeptide model. Am J Psychiatry. 2010;167(1):24–39.
doi:10.1176/appi.ajp.2009.09050744.
57. Kalin NH, Shelton SE, Barksdale CM. Opiate modulation of separation-
induced distress in non-human primates. Brain Res. 1988;440(2):285–92.
58. Zubieta JK, Ketter TA, Bueller JA, Xu Y, Kilbourn MR, Young EA, et al.
Regulation of human affective responses by anterior cingulate and limbic
mu-opioid neurotransmission. Arch Gen Psychiatry. 2003;60(11):1145–53.
doi:10.1001/archpsyc.60.11.1145.
59. Zubieta JK, Smith YR, Bueller JA, Xu Y, Kilbourn MR, Jewett DM, et al.
Regional mu opioid receptor regulation of sensory and affective dimensions
of pain. Science. 2001;293(5528):311–5. doi:10.1126/science.1060952.
60. Love TM, Stohler CS, Zubieta JK. Positron emission tomography measures of
endogenous opioid neurotransmission and impulsiveness traits in humans. Arch
Gen Psychiatry. 2009;66(10):1124–34. doi:10.1001/archgenpsychiatry.2009.134.
61. Prossin AR, Love TM, Koeppe RA, Zubieta JK, Silk KR. Dysregulation of
regional endogenous opioid function in borderline personality disorder. Am
J Psychiatry. 2010;167(8):925–33. doi:10.1176/appi.ajp.2010.09091348.
62. Nunes PM, Wenzel A, Borges KT, Porto CR, Caminha RM, de Oliveira IR.
Volumes of the hippocampus and amygdala in patients with borderline
personality disorder: a meta-analysis. J Personal Disord. 2009;23(4):333–45.
doi:10.1521/pedi.2009.23.4.333.
63. Rodrigues E, Wenzel A, Ribeiro MP, Quarantini LC, Miranda-Scippa A, de
Sena EP, et al. Hippocampal volume in borderline personality disorder with
and without comorbid posttraumatic stress disorder: a meta-analysis.
European psychiatry : the journal of the Association of European
Psychiatrists. 2011;26(7):452–6. doi:10.1016/j.eurpsy.2010.07.005.
64. Irle E, Lange C, Sachsse U. Reduced size and abnormal asymmetry of
parietal cortex in women with borderline personality disorder. Biol
Psychiatry. 2005;57(2):173–82. doi:10.1016/j.biopsych.2004.10.004.
65. Brambilla P, Soloff PH, Sala M, Nicoletti MA, Keshavan MS, Soares JC.
Anatomical MRI study of borderline personality disorder patients. Psychiatry
Res. 2004;131(2):125–33. doi:10.1016/j.pscychresns.2004.04.003.
66. Tebartz van Elst L, Hesslinger B, Thiel T, Geiger E, Haegele K, Lemieux L,
et al. Frontolimbic brain abnormalities in patients with borderline
personality disorder: a volumetric magnetic resonance imaging study. Biol
Psychiatry. 2003;54(2):163–71.
67. Rossi R, Lanfredi M, Pievani M, Boccardi M, Beneduce R, Rillosi L, et al.
Volumetric and topographic differences in hippocampal subdivisions in
borderline personality and bipolar disorders. Psychiatry Res. 2012;203(2–3):
132–8. doi:10.1016/j.pscychresns.2011.12.004.
68. Rossi R, Pievani M, Lorenzi M, Boccardi M, Beneduce R, Bignotti S, et al.
Structural brain features of borderline personality and bipolar disorders.
Psychiatry Res. 2013;213(2):83–91. doi:10.1016/j.pscychresns.2012.07.002.
69. O’Neill A, D’Souza A, Carballedo A, Joseph S, Kerskens C, Frodl T. Magnetic
resonance imaging in patients with borderline personality disorder: a study
of volumetric abnormalities. Psychiatry Res. 2013;213(1):1–10. doi:10.1016/j.
pscychresns.2013.02.006.
70. Kreisel SH, Labudda K, Kurlandchikov O, Beblo T, Mertens M, Thomas C, et al.
Volume of hippocampal substructures in borderline personality disorder.
Psychiatry Res. 2015;231(3):218–26. doi:10.1016/j.pscychresns.2014.11.010.
71. Boen E, Westlye LT, Elvsashagen T, Hummelen B, Hol PK, Boye B, et al.
Smaller stress-sensitive hippocampal subfields in women with borderline
personality disorder without posttraumatic stress disorder. Journal of
psychiatry & neuroscience : JPN. 2014;39(2):127–34.
72. Teicher MH, Anderson CM, Polcari A. Childhood maltreatment is associated
with reduced volume in the hippocampal subfields CA3, dentate gyrus, and
subiculum. Proc Natl Acad Sci U S A. 2012;109(9):E563–72. doi:10.1073/pnas.
1115396109.
Cattane et al. BMC Psychiatry (2017) 17:221 Page 13 of 14
http://dx.doi.org/10.3389/fncel.2013.00031
http://dx.doi.org/10.1038/mp.2014.66
http://dx.doi.org/10.1016/j.pscychresns.2012.02.012
http://dx.doi.org/10.1097/YPG.0b013e328016831c
http://dx.doi.org/10.1016/j.jpsychires.2006.10.005
http://dx.doi.org/10.1016/j.jpsychires.2006.10.005
http://dx.doi.org/10.1016/j.jpsychires.2008.03.006
http://dx.doi.org/10.1016/j.jpsychires.2009.03.004
http://dx.doi.org/10.1037/abn0000034
http://dx.doi.org/10.1111/bdi.12250
http://dx.doi.org/10.1186/s13587-015-0017-x
http://dx.doi.org/10.1007/s00213-007-0867-z
http://dx.doi.org/10.1111/j.1600-0447.2009.01501.x
http://dx.doi.org/10.1016/j.jpsychires.2009.06.008
http://dx.doi.org/10.1007/s00702-009-0264-3
http://dx.doi.org/10.1016/j.psychres.2015.07.046
http://dx.doi.org/10.1037/a0018095
http://dx.doi.org/10.1038/npp.2013.155
http://dx.doi.org/10.1016/j.pneurobio.2008.06.003
http://dx.doi.org/10.1016/j.pneurobio.2008.06.003
http://dx.doi.org/10.1176/appi.ajp.2009.09050744
http://dx.doi.org/10.1001/archpsyc.60.11.1145
http://dx.doi.org/10.1126/science.1060952
http://dx.doi.org/10.1001/archgenpsychiatry.2009.134
http://dx.doi.org/10.1176/appi.ajp.2010.09091348
http://dx.doi.org/10.1521/pedi.2009.23.4.333
http://dx.doi.org/10.1016/j.eurpsy.2010.07.005
http://dx.doi.org/10.1016/j.biopsych.2004.10.004
http://dx.doi.org/10.1016/j.pscychresns.2004.04.003
http://dx.doi.org/10.1016/j.pscychresns.2011.12.004
http://dx.doi.org/10.1016/j.pscychresns.2012.07.002
http://dx.doi.org/10.1016/j.pscychresns.2013.02.006
http://dx.doi.org/10.1016/j.pscychresns.2013.02.006
http://dx.doi.org/10.1016/j.pscychresns.2014.11.010
http://dx.doi.org/10.1073/pnas.1115396109
http://dx.doi.org/10.1073/pnas.1115396109
73. Kuhlmann A, Bertsch K, Schmidinger I, Thomann PA, Herpertz SC.
Morphometric differences in central stress-regulating structures between
women with and without borderline personality disorder. Journal of
psychiatry & neuroscience : JPN. 2013;38(2):129–37. doi:10.1503/jpn.120039.
74. Klengel T, Binder EB. Epigenetics of stress-related psychiatric disorders and
Gene x environment interactions. Neuron. 2015;86(6):1343–57. doi:10.1016/j.
neuron.2015.05.036.
75. Slatkin M. Epigenetic inheritance and the missing heritability problem.
Genetics. 2009;182(3):845–50. doi:10.1534/genetics.109.102798.
76. Levine A, Worrell TR, Zimnisky R, Schmauss C. Early life stress triggers
sustained changes in histone deacetylase expression and histone H4
modifications that alter responsiveness to adolescent antidepressant
treatment. Neurobiol Dis. 2012;45(1):488–98. doi:10.1016/j.nbd.2011.09.005.
77. Issler O, Chen A. Determining the role of microRNAs in psychiatric disorders.
Nat Rev Neurosci. 2015;16(4):201–12. doi:10.1038/nrn3879.
78. Kaffman A, Meaney MJ. Neurodevelopmental sequelae of postnatal
maternal care in rodents: clinical and research implications of molecular
insights. Journal of child psychology and psychiatry, and allied disciplines.
2007;48(3–4):224–44. doi:10.1111/j.1469-7610.2007.01730.x.
79. McGowan PO, Suderman M, Sasaki A, Huang TC, Hallett M, Meaney MJ,
et al. Broad epigenetic signature of maternal care in the brain of adult rats.
PLoS One. 2011;6(2):e14739. doi:10.1371/journal.pone.0014739.
80. Kammerer M, Marks MN, Pinard C, Taylor A, von Castelberg B, Kunzli H, et al.
Symptoms associated with the DSM IV diagnosis of depression in
pregnancy and post partum. Archives of women’s mental health. 2009;12(3):
135–41. doi:10.1007/s00737-009-0062-9.
81. Plant DT, Pariante CM, Sharp D, Pawlby S. Maternal depression during
pregnancy and offspring depression in adulthood: role of child
maltreatment. The British journal of psychiatry : the journal of mental
science. 2015;207(3):213–20. doi:10.1192/bjp.bp.114.156620.
82. McGowan PO, Sasaki A, D’Alessio AC, Dymov S, Labonte B, Szyf M, et al.
Epigenetic regulation of the glucocorticoid receptor in human brain
associates with childhood abuse. Nat Neurosci. 2009;12(3):342–8. doi:10.
1038/nn.2270.
83. Oberlander TF, Weinberg J, Papsdorf M, Grunau R, Misri S, Devlin AM.
Prenatal exposure to maternal depression, neonatal methylation of human
glucocorticoid receptor gene (NR3C1) and infant cortisol stress responses.
Epigenetics. 2008;3(2):97–106.
84. Perroud N, Paoloni-Giacobino A, Prada P, Olie E, Salzmann A, Nicastro R,
et al. Increased methylation of glucocorticoid receptor gene (NR3C1) in
adults with a history of childhood maltreatment: a link with the severity and
type of trauma. Transl Psychiatry. 2011;1:e59. doi:10.1038/tp.2011.60.
85. Suderman M, Borghol N, Pappas JJ, Pinto Pereira SM, Pembrey M, Hertzman
C, et al. Childhood abuse is associated with methylation of multiple loci in
adult DNA. BMC Med Genet. 2014;7:13. doi:10.1186/1755-8794-7-13.
86. Klengel T, Mehta D, Anacker C, Rex-Haffner M, Pruessner JC, Pariante CM,
et al. Allele-specific FKBP5 DNA demethylation mediates gene-childhood
trauma interactions. Nat Neurosci. 2013;16(1):33–41. doi:10.1038/nn.3275.
87. Labonte B, Suderman M, Maussion G, Navaro L, Yerko V, Mahar I, et al.
Genome-wide epigenetic regulation by early-life trauma. Arch Gen
Psychiatry. 2012;69(7):722–31. doi:10.1001/archgenpsychiatry.2011.2287.
88. Martin-Blanco A, Ferrer M, Soler J, Salazar J, Vega D, Andion O, et al.
Association between methylation of the glucocorticoid receptor gene,
childhood maltreatment, and clinical severity in borderline personality
disorder. J Psychiatr Res. 2014;57:34–40. doi:10.1016/j.jpsychires.2014.06.011.
89. Dammann G, Teschler S, Haag T, Altmuller F, Tuczek F, Dammann RH.
Increased DNA methylation of neuropsychiatric genes occurs in borderline
personality disorder. Epigenetics. 2011;6(12):1454–62. doi:10.4161/epi.6.12.
18363.
90. Perroud N, Salzmann A, Prada P, Nicastro R, Hoeppli ME, Furrer S, et al.
Response to psychotherapy in borderline personality disorder and
methylation status of the BDNF gene. Transl Psychiatry. 2013;3:e207. doi:10.
1038/tp.2012.140.
91. Perroud N, Zewdie S, Stenz L, Adouan W, Bavamian S, Prada P, et al.
Methylation of serotonin receptor 3a in Adhd, borderline personality, and
bipolar disorders: link with severity of the disorders and childhood
maltreatment. Depression and anxiety. 2016;33(1):45–55. doi:10.1002/da.
22406.
92. Thaler L, Gauvin L, Joober R, Groleau P, de Guzman R, Ambalavanan A, et al.
Methylation of BDNF in women with bulimic eating syndromes:
associations with childhood abuse and borderline personality disorder. Prog
Neuro-Psychopharmacol Biol Psychiatry. 2014;54:43–9. doi:10.1016/j.pnpbp.
2014.04.010.
93. Teschler S, Bartkuhn M, Kunzel N, Schmidt C, Kiehl S, Dammann G, et al. Aberrant
methylation of gene associated CpG sites occurs in borderline personality
disorder. PLoS One. 2013;8(12):e84180. doi:10.1371/journal.pone.0084180.
94. Prados J, Stenz L, Courtet P, Prada P, Nicastro R, Adouan W, et al. Borderline
personality disorder and childhood maltreatment: a genome-wide methylation
analysis. Genes Brain Behav. 2015;14(2):177–88. doi:10.1111/gbb.12197.
95. Teschler S, Gotthardt J, Dammann G, Dammann RH. Aberrant DNA Methylation
of rDNA and PRIMA1 in Borderline Personality Disorder. International journal of
molecular sciences. 2016;17(1). doi:10.3390/ijms17010067.
96. Cattaneo A, Macchi F, Plazzotta G, Veronica B, Bocchio-Chiavetto L, Riva MA,
et al. Inflammation and neuronal plasticity: a link between childhood
trauma and depression pathogenesis. Front Cell Neurosci. 2015;9:40. doi:10.
3389/fncel.2015.00040.
97. Briggs JA, Wolvetang EJ, Mattick JS, Rinn JL, Barry G. Mechanisms of long
non-coding RNAs in mammalian nervous system development, plasticity,
disease, and evolution. Neuron. 2015;88(5):861–77. doi:10.1016/j.neuron.
2015.09.045.
98. McIntyre CK, McGaugh JL, Williams CL. Interacting brain systems modulate
memory consolidation. Neurosci Biobehav Rev. 2012;36(7):1750–62. doi:10.
1016/j.neubiorev.2011.11.001.
99. Dhabhar FS. Enhancing versus suppressive effects of stress on immune
function: implications for immunoprotection and immunopathology.
Neuroimmunomodulation. 2009;16(5):300–17. doi:10.1159/000216188.
100. McEwen BS. Physiology and neurobiology of stress and adaptation: central role
of the brain. Physiol Rev. 2007;87(3):873–904. doi:10.1152/physrev.00041.2006.
101. McEwen BS. Understanding the potency of stressful early life experiences
on brain and body function. Metab Clin Exp. 2008;57(Suppl 2):S11–5. doi:10.
1016/j.metabol.2008.07.006.
102. McEwen BS. Protection and damage from acute and chronic stress: allostasis
and allostatic overload and relevance to the pathophysiology of psychiatric
disorders. Ann N Y Acad Sci. 2004;1032:1–7. doi:10.1196/annals.1314.001.
103. Herbert J, Goodyer IM, Grossman AB, Hastings MH, de Kloet ER, Lightman
SL, et al. Do corticosteroids damage the brain? J Neuroendocrinol. 2006;
18(6):393–411. doi:10.1111/j.1365-2826.2006.01429.x.
104. Duman RS, Monteggia LM. A neurotrophic model for stress-related mood
disorders. Biol Psychiatry. 2006;59(12):1116–27. doi:10.1016/j.biopsych.2006.02.013.
105. Kapczinski F, Frey BN, Andreazza AC, Kauer-Sant’Anna M, Cunha AB, Post
RM. Increased oxidative stress as a mechanism for decreased BDNF levels in
acute manic episodes. Rev Bras Psiquiatr. 2008;30(3):243–5.
106. Waterhouse EG, Xu B. New insights into the role of brain-derived
neurotrophic factor in synaptic plasticity. Mol Cell Neurosci. 2009;42(2):81–9.
doi:10.1016/j.mcn.2009.06.009.
107. Calabrese F, Molteni R, Gabriel C, Mocaer E, Racagni G, Riva MA. Modulation
of neuroplastic molecules in selected brain regions after chronic
administration of the novel antidepressant agomelatine.
Psychopharmacology. 2011;215(2):267–75. doi:10.1007/s00213-010-2129-8.
108. Ansorge MS, Hen R, Gingrich JA. Neurodevelopmental origins of depressive
disorders. Curr Opin Pharmacol. 2007;7(1):8–17. doi:10.1016/j.coph.2006.11.
006.
109. Koenigsberg HW, Yuan P, Diaz GA, Guerreri S, Dorantes C, Mayson S, et al.
Platelet protein kinase C and brain-derived neurotrophic factor levels in
borderline personality disorder patients. Psychiatry Res. 2012;199(2):92–7.
doi:10.1016/j.psychres.2012.04.026.
110. Polyakova M, Stuke K, Schuemberg K, Mueller K, Schoenknecht P, Schroeter
ML. BDNF as a biomarker for successful treatment of mood disorders: a
systematic & quantitative meta-analysis. J Affect Disord. 2015;174:432–40.
doi:10.1016/j.jad.2014.11.044.
111. Cattaneo A, Bocchio-Chiavetto L, Zanardini R, Milanesi E, Placentino A,
Gennarelli M. Reduced peripheral brain-derived neurotrophic factor mRNA
levels are normalized by antidepressant treatment. The international journal
of neuropsychopharmacology / official scientific journal of the Collegium
Internationale Neuropsychopharmacologicum. 2010;13(1):103–8. doi:10.
1017/S1461145709990812.
112. Cattaneo A, Gennarelli M, Uher R, Breen G, Farmer A, Aitchison KJ, et al.
Candidate genes expression profile associated with antidepressants
response in the GENDEP study: differentiating between baseline ‘predictors’
and longitudinal ‘targets’. Neuropsychopharmacology : official publication of
the American College of Neuropsychopharmacology. 2013;38(3):377–85. doi:
10.1038/npp.2012.191.
Cattane et al. BMC Psychiatry (2017) 17:221 Page 14 of 14
http://dx.doi.org/10.1503/jpn.120039
http://dx.doi.org/10.1016/j.neuron.2015.05.036
http://dx.doi.org/10.1016/j.neuron.2015.05.036
http://dx.doi.org/10.1534/genetics.109.102798
http://dx.doi.org/10.1016/j.nbd.2011.09.005
http://dx.doi.org/10.1038/nrn3879
http://dx.doi.org/10.1111/j.1469-7610.2007.01730.x
http://dx.doi.org/10.1371/journal.pone.0014739
http://dx.doi.org/10.1007/s00737-009-0062-9
http://dx.doi.org/10.1192/bjp.bp.114.156620
http://dx.doi.org/10.1038/nn.2270
http://dx.doi.org/10.1038/nn.2270
http://dx.doi.org/10.1038/tp.2011.60
http://dx.doi.org/10.1186/1755-8794-7-13
http://dx.doi.org/10.1038/nn.3275
http://dx.doi.org/10.1001/archgenpsychiatry.2011.2287
http://dx.doi.org/10.1016/j.jpsychires.2014.06.011
http://dx.doi.org/10.4161/epi.6.12.18363
http://dx.doi.org/10.4161/epi.6.12.18363
http://dx.doi.org/10.1038/tp.2012.140
http://dx.doi.org/10.1038/tp.2012.140
http://dx.doi.org/10.1002/da.22406
http://dx.doi.org/10.1002/da.22406
http://dx.doi.org/10.1016/j.pnpbp.2014.04.010
http://dx.doi.org/10.1016/j.pnpbp.2014.04.010
http://dx.doi.org/10.1371/journal.pone.0084180
http://dx.doi.org/10.1111/gbb.12197
http://dx.doi.org/10.3390/ijms17010067
http://dx.doi.org/10.3389/fncel.2015.00040
http://dx.doi.org/10.3389/fncel.2015.00040
http://dx.doi.org/10.1016/j.neuron.2015.09.045
http://dx.doi.org/10.1016/j.neuron.2015.09.045
http://dx.doi.org/10.1016/j.neubiorev.2011.11.001
http://dx.doi.org/10.1016/j.neubiorev.2011.11.001
http://dx.doi.org/10.1159/000216188
http://dx.doi.org/10.1152/physrev.00041.2006
http://dx.doi.org/10.1016/j.metabol.2008.07.006
http://dx.doi.org/10.1016/j.metabol.2008.07.006
http://dx.doi.org/10.1196/annals.1314.001
http://dx.doi.org/10.1111/j.1365-2826.2006.01429.x
http://dx.doi.org/10.1016/j.biopsych.2006.02.013
http://dx.doi.org/10.1016/j.mcn.2009.06.009
http://dx.doi.org/10.1007/s00213-010-2129-8
http://dx.doi.org/10.1016/j.coph.2006.11.006
http://dx.doi.org/10.1016/j.coph.2006.11.006
http://dx.doi.org/10.1016/j.psychres.2012.04.026
http://dx.doi.org/10.1016/j.jad.2014.11.044
http://dx.doi.org/10.1017/S1461145709990812
http://dx.doi.org/10.1017/S1461145709990812
http://dx.doi.org/10.1038/npp.2012.191
Reproduced with permission of copyright owner.
Further reproduction prohibited without permission.
Background
Discussion
Summary
Background
Discussion
Neurobiological mechanisms involved in BPD
BPD and the hypothalamic-pituitary-adrenal axis
BPD and neurotransmission
BPD and the endogenous opioid system
BPD and neuroimaging studies
Volumetric alterations in brain areas involved in stress response
Endogenous opiod system alterations in brain regions involved in stress response
BPD and epigenetic mechanisms
BPD and neuroplasticity (the role of BDNF)
Conclusions
Abbreviations
Acknowledgements
Funding
Availability of data and materials
Authors’ contributions
Competing interests
Consent for publication
Ethics approval and consent to participate
Author details
References
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=zept20
European Journal of Psychotraumatology
ISSN: 2000-8198 (Print) 2000-8066 (Online) Journal homepage: https://www.tandfonline.com/loi/zept20
Distinguishing PTSD, Complex PTSD, and
Borderline Personality Disorder: A latent class
analysis
Marylène Cloitre, Donn W. Garvert, Brandon Weiss, Eve B. Carlson & Richard
A. Bryant
To cite this article: Marylène Cloitre, Donn W. Garvert, Brandon Weiss, Eve B. Carlson & Richard
A. Bryant (2014) Distinguishing PTSD, Complex PTSD, and Borderline Personality Disorder:
A latent class analysis, European Journal of Psychotraumatology, 5:1, 25097, DOI: 10.3402/
ejpt.v5.25097
To link to this article: https://doi.org/10.3402/ejpt.v5.25097
© 2014 Marylène Cloitre et al. View supplementary material
Published online: 15 Sep 2014. Submit your article to this journal
Article views: 35822 View related articles
View Crossmark data Citing articles: 94 View citing articles
https://www.tandfonline.com/action/journalInformation?journalCode=zept20
https://www.tandfonline.com/loi/zept20
https://www.tandfonline.com/action/showCitFormats?doi=10.3402/ejpt.v5.25097
https://www.tandfonline.com/action/showCitFormats?doi=10.3402/ejpt.v5.25097
https://doi.org/10.3402/ejpt.v5.25097
https://www.tandfonline.com/doi/suppl/10.3402/ejpt.v5.25097
https://www.tandfonline.com/doi/suppl/10.3402/ejpt.v5.25097
https://www.tandfonline.com/action/authorSubmission?journalCode=zept20&show=instructions
https://www.tandfonline.com/action/authorSubmission?journalCode=zept20&show=instructions
https://www.tandfonline.com/doi/mlt/10.3402/ejpt.v5.25097
https://www.tandfonline.com/doi/mlt/10.3402/ejpt.v5.25097
http://crossmark.crossref.org/dialog/?doi=10.3402/ejpt.v5.25097&domain=pdf&date_stamp=2014-09-15
http://crossmark.crossref.org/dialog/?doi=10.3402/ejpt.v5.25097&domain=pdf&date_stamp=2014-09-15
https://www.tandfonline.com/doi/citedby/10.3402/ejpt.v5.25097#tabModule
https://www.tandfonline.com/doi/citedby/10.3402/ejpt.v5.25097#tabModule
CLINICAL RESEARCH ARTICLE
Distinguishing PTSD, Complex PTSD, and Borderline
Personality Disorder: A latent class analysis
Marylène Cloitre1,2*, Donn W. Garvert1, Brandon Weiss1,3, Eve B. Carlson
1
and Richard A. Bryant4
1National Center for PTSD, Veterans Affairs Palo Alto Health Care System, Palo Alto, CA, USA;
2Department of Psychiatry and Child & Adolescent Psychiatry, New York University Medical Center,
New York, USA; 3Department of Psychiatry and Behavioral Sciences, Stanford University School
of Medicine, Palo Alto, CA, USA; 4School of Psychology, University of New South Wales, Sydney,
NSW, Australia
Background: There has been debate regarding whether Complex Posttraumatic Stress Disorder (Complex
PTSD) is distinct from Borderline Personality Disorder (BPD) when the latter is comorbid with PTSD.
Objective: To determine whether the patterns of symptoms endorsed by women seeking treatment for child-
hood abuse form classes that are consistent with
diagnostic criteria for PTSD, Complex PTSD, and BPD.
Method: A latent class analysis (LCA) was conducted on an archival dataset of 280 women with histories of
childhood abuse assessed for enrollment in a clinical trial for PTSD.
Results: The LCA revealed four distinct classes of individuals: a Low Symptom class characterized by low
endorsements on all symptoms; a PTSD class characterized by elevated symptoms of PTSD but low endorse-
ment of symptoms that define the Complex PTSD and BPD diagnoses; a Complex PTSD class characterized by
elevated symptoms of PTSD and self-organization symptoms that defined the Complex PTSD diagnosis but
low on the symptoms of BPD; and a BPD class characterized by symptoms of BPD. Four BPD symptoms were
found to greatly increase the odds of being in the BPD compared to the Complex PTSD class: frantic efforts to
avoid abandonment, unstable sense of self, unstable and intense interpersonal relationships, and impulsiveness.
Conclusions: Findings supported the construct validity of Complex PTSD as distinguishable from BPD. Key
symptoms that distinguished between the disorders were identified, which may aid in differential diagnosis
and treatment planning.
Keywords: Complex PTSD; posttraumatic stress disorder; Borderline Personality Disorder; WHO; ICD-11
*Correspondence to: Marylène Cloitre, National Center for PTSD Dissemination and Training Division,
VAPAHCS, 795 Willow Road, Menlo Park, CA 94025, USA, Email: Marylene.cloitre@nyumc.org
For the abstract or full text in other languages, please see Supplementary files under Article Tools online
Received: 3 June 2014; Revised: 22 July 2014; Accepted: 18 August 2014; Published: 15 September 2014
T
here has long been debate about whether Complex
Posttraumatic Stress Disorder (Complex PTSD)
is distinct from Borderline Personality Disorder
(BPD) comorbid with PTSD. Part of the difficulty in this
evaluation has been the lack of clear and consistent
characterization of Complex PTSD. The World Health
Organization (WHO) Working Group on the Classification
of Stress-Related Disorders has proposed the inclusion
of Complex PTSD as a new diagnosis related to but sepa-
rate from PTSD (Maercker et al., 2013). Both of these
disorders are viewed as distinct and separate from BPD.
An emerging and accumulating empirical literature is
demonstrating consistent and clear differences between
ICD-11 PTSD and Complex PTSD. In addition, it is
important to determine the construct validity of Complex
PTSD as empirically distinct from BPD particularly
among those with a trauma history. This investigation
evaluated whether ICD-11 Complex PTSD could be dis-
tinguished from DSM-IV BPD in a treatment-seeking
population of women with childhood abuse.
The WHO proposed that the development of ICD-11
be guided by the principle of clinical utility. Characteris-
tics of clinical utility include the organization of disorders
that are consistent with clinicians’ mental health taxo-
nomies, that contain a limited number of symptoms so
that they can be easily recalled and used in the field, and
that are based on distinctions important for manage-
ment and treatment (Reed, 2010). The distinction between
PSYCHOTRAUMATOLOGY
EUROPEAN JOURNAL OF
�
European Journal of Psychotraumatology 2014. # 2014 Marylène Cloitre et al. This is an Open Access article distributed under the terms of the Creative Commons
Attribution 4.0 Unported (CC-BY 4.0) License (http://creativecommons.org/licenses/by/4.0/), allowing third parties to copy and redistribute the material in any medium or
format, and to remix, transform, and build upon the material, for any purpose, even commercially, under the condition that appropriate credit is given, that a link to the license
is provided, and that you indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
1
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097
(page number not for citation purpose)
http://www.eurojnlofpsychotraumatol.net/index.php/ejpt/rt/suppFiles/25097/0
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
ICD-11 PTSD and Complex PTSD are consistent with
these guidelines (see Cloitre, Garvert, Brewin, Bryant, &
Maercker, 2013; Maercker et al., 2013). For example,
ICD-11 PTSD is construed as a fear-based disorder and
symptoms are limited to and consistent with fear reactions
and consequent avoidance and hypervigilence. In contrast,
Complex PTSD has been described as typically associated
with chronic and repeated traumas and includes not only
the symptoms of PTSD but also disturbances in self-
organization reflected in emotion regulation, self-concept
and relational difficulties (see Cloitre et al., 2013) a symp-
tom profile that has been demonstrated as associated
with prolonged trauma (Briere & Rickards, 2007).
Three studies have found evidence supporting the vali-
dity of the ICD-11 PTSD versus Complex PTSD distinction
(see Table 1 for description of the diagnoses). Recently, in
order to evaluate whether PTSD and Complex PTSD
could be empirically distinguished from each other, Cloitre
and colleagues (2013) performed a latent profile analysis
(LPA) on assessment data from 302 treatment-seeking
individuals with diverse trauma histories, ranging from
single events (e.g., 9/11 attacks) to sustained exposures
(e.g., childhood or adult physical and/or sexual abuse).
The results were consistent with the ICD-11 formulation
for Complex PTSD, with the best fitting LPA model
delineating three classes of individuals: (1) a Complex
PTSD class, with high levels of both PTSD symptoms as
well as disturbances in self-organization related to affect
regulation problems, negative self-concept, and relational
difficulties; (2) a PTSD class, with high levels of PTSD
symptoms but relatively low on the disturbances in self-
organization that define Complex PTSD; and (3) a class
Table 1. Symptom profile for each diagnosis and items used in the LCA analyses
Symptoms for each diagnoses
ICD-11 PTSD ICD-11 CPTSD DSM-IV BPD Items
Re-experiencing
Re-experiencing
Flashbacks Flashbacks CAPS 1: Unwanted memories of the event
Nightmares Nightmares CAPS 2: Recurrent distressing dreams of the event
Avoidance
Avoidance
Thoughts Thoughts CAPS 6: Avoid thoughts, feelings or conversations
People, places,
activities
People, places, activities CAPS 7: Avoid activities, places, or people
Sense of threat
Sense of threat
Hypervigilance Hypervigilance CAPS 16: Being especially alert constantly on guard
Startle Startle CAPS 17: Strong startle reactions
Emotion regulation
Anger BSI 13: Temper outburst that you could not control
Hurt Feelings BSI 20: Your feelings being easily hurt
Negative self-concept
Worthless BSI 50: Feeling of worthlessness
Guilty BSI 52: Feelings of guilt
Interpersonal problems
Not close BSI 44: Never feeling close to another person
Feel disconnected CAPS 10: Feeling distant or cut off from other people
Frantic SCID-II 90: Frantic efforts to avoid abandonment
Unstable relationships SCID-II 91: Unstable and intense relationships with alternating
extremes of idealization and devaluation
Unstable sense of self SCID-II 92: Markedly and persistently unstable sense of self
Impulsiveness SCID-II 96: Impulsiveness that is potentially self-damaging
Self-harm SCID-II 97: Recurrent suicidal behavior, gestures, or threats,
or self-mutilating behavior
Mood changes SCID-II 99: Affective instability due to reactivity to mood
Empty SCID-II 100: Chronic feelings of emptiness
Temper SCID-II 101: Frequent displays of anger, constant anger,
recurrent physical fights
Paranoid/dissociation SCID-II 104: Transient, stress-related paranoid ideation or
severe dissociative symptoms
Marylène Cloitre et al.
2
(page number not for citation purpose)
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
relatively low on symptoms of both PTSD and Complex
PTSD. Notably, these identified classes were identical
when including an additional 86 participants with BPD,
providing further support for the stability of the identified
classes. Cloitre et al. (2013) also found that chronic trauma
was more predictive of Complex PTSD than PTSD and
that Complex PTSD resulted in significantly greater func-
tional impairment than PTSD.
Elklit, Hyland, and Shevlin (2014) replicated the find-
ings of Cloitre and colleagues (2013), performing a latent
class analysis (LCA) on three separate samples of trauma-
exposed Danish individuals who experienced primary
traumas of bereavement, sexual assault, and physical
assault. The investigators found that the LCA model with
the best fit for each sample consisted of three classes of
individuals identical to those identified by Cloitre et al.
(2013). Lastly, Knefel and Lueger-Schuster (2013) per-
formed confirmatory factor analysis (CFA) on data from
226 Austrian adults who had experienced institutional
abuse, defined as physical, sexual or emotional abuse
by individuals representing institutions responsible for
the care of children (i.e., Catholic Church, foster care).
Results indicated that individuals diagnosed with Com-
plex PTSD had experienced significantly longer exposure
to traumatic situations and that the theoretically derived
CFA model demonstrated good model fit. Overall, these
studies provide substantial support for the construct vali-
dity of Complex PTSD across international samples of
individuals exposed to diverse traumatic events, demon-
strating that it is a diagnostic entity distinct from PTSD
and supporting the recommendations of the ICD-11
proposal.
The argument that Complex PTSD is an amalgam of
PTSD and BPD has been built on reports of the relatively
high comorbidity between PTSD and BPD. For example,
using data from the National Epidemiologic Survey on
Alcohol and Related Conditions (NESARC), a nationally
representative sample of United States population, Pagura
and colleagues (2010) found that 24% of individuals with
lifetime PTSD also met criteria for BPD, 30% of indi-
viduals with BPD also met criteria for lifetime PTSD,
and 2% had comorbid PTSD and BPD. Also utilizing
NESARC data, Grant and colleagues (2008) found that
29% of individuals who currently met criteria for PTSD
in the past 12 months also met criteria for lifetime BPD,
and 32% of individuals with lifetime BPD met criteria for
12-month PTSD. In clinical samples, the rates of comor-
bidity are higher. PTSD patients are reported to have
BPD comorbidity ranging from 37 to 68% (Heffernan &
Cloitre, 2000; Zlotnick, Franklin, & Zimmerman, 2002)
and conversely, among BPD patients 25�58% are diag-
nosed with comorbid PTSD (Golier et al., 2003; Harned,
Rizvi, & Linehan, 2010; Zanarini et al., 1998).
Despite these high rates of comorbidity, the key clinical
features of Complex PTSD and BPD differ and lead to
different treatment implications, a consequence of signifi-
cance when considering the clinical utility of diagnostic
formulation. Complex PTSD includes PTSD symptoms
and, accordingly, treatment highlights the amelioration
of the trauma memory as a key goal (Cloitre et al., 2011).
In contrast, the key impairing features of BPD are self-
injurious and suicidal behaviors, and treatment activities
focus on the resolution of these behaviors (e.g., Linehan,
1993). There are several other ways in which the profile of
Complex PTSD differs from that of BPD. First, it should
be noted that BPD does not require a traumatic stressor
for diagnosis and PTSD symptoms may or may not be
present. Rather, BPD is characterized by fear of abandon-
ment, shifting self-image or self-concept, shifting ideali-
zation and devaluation in relationships, and frequent
impulsive and suicidal behaviors (see Table 1). In Complex
PTSD, as proposed in ICD-11, the fear of abandonment
is not a requirement of the disorder, self-identify is
consistently negative rather than shifting and relational
disturbances highlight chronic avoidance of relationships
rather than sustained chaotic engagement. While emotion
regulation difficulties are central to both Complex PTSD
and BPD, their expression is quite different. In Complex
PTSD they predominantly include emotional sensitivity,
reactive anger and poor coping responses (e.g., use of
alcohol and substances). In BPD, some of the preceding
may be observed, but the criteria, perhaps the defining
characteristics of the disorder, include suicide attempts
and gestures as well as self-injurious behaviors, events
which occur much less frequently in complex forms of
PTSD than in BPD samples (e.g., Zlotnick et al., 2002).
Given these identified differences in diagnostic formula-
tion and their treatment implications, empirical evaluation
of Complex PTSD and BPD in a symptom-by-symptom
manner is important.
The purpose of the current study was to determine
whether the symptoms endorsed by women seeking treat-
ment for childhood abuse form classes that are consistent
with diagnostic criteria for PTSD, Complex PTSD, and
BPD (see Table 1). We hypothesized that analyses would
reveal at least three distinct classes of individuals with the
following symptom profiles: (1) high levels of ICD-11
PTSD symptoms but not symptoms of Complex PTSD or
BPD, (2) high levels of Complex PTSD symptoms (PTSD
plus emotion regulation, negative self-concept and inter-
personal problems) but not BPD symptoms; (3) high
levels of BPD symptoms with an admixture of trauma-
related symptoms (e.g., PTSD, CPTSD symptoms).
Methods
Participants and procedures
The data for these analyses were obtained from an archi-
val set of measures administered as part of an assessment
procedure for a randomized controlled trial for PTSD
PTSD, Complex PTSD, and BPD
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097 3
(page number not for citation purpose)
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
related to childhood abuse (n�310) (see Cloitre et al.,
2010). The data are a subset of individuals for whom com-
plete measures of PTSD, BPD, general psychopathology,
and functional impairment were available (n�280).
Participants had a mean age of 37.13 (SD �10.86)
years. The entire sample was female and 40% identified
as Caucasian (40.4%, n�113), followed by African-
American (26.4%, n�74), Hispanic (18.6%, n�52), Asian
(3.9%, n�11), other (8.6%, n�24), and unreported
(2.1%, n�6). Marital status of the sample was as follows:
54.3% (n�152) reported being single, married (11.1%,
n�31), divorced or separated (16.1%, n�45), living with
a significant other (15.7%, n�44), widowed (0.7%, n�2),
and unreported (2.1%, n�6). College graduation or com-
pletion of some college was reported by 64.3% (180),
postgraduate education was reported by 24.3% (68), high
school graduation or some high school was reported by
9.3% (26), and education level was unavailable for 2.1%
(6). The majority of participants reported some employ-
ment with 41.4% (116) employed full-time (35� hours
per week) and 23.9% (n�67) employed part-time (B35
hours per week).
Frequency of traumas were as follows: childhood sexual
abuse (CSA) (65.1%), childhood physical abuse (CPA)
(80.8%), neglect (46.4%), emotional abuse (80.4%), 35.9%
were not living with their mother before the age of 18,
adulthood sexual assault (ASA) (49.6%), and adult-
hood physical assault (APA) (26.0%). All individuals
had experienced either CPA or CSA.
Measures
Clinician Administered PTSD Scale
The Clinician Administered PTSD Scale (CAPS) for
DSM-IV PTSD is a well-validated clinician administered
interview (see Weathers, Keane, & Davidson, 2001) that
evaluates the presence and severity of the 17 DSM-IV
PTSD symptoms over the past month with separate five-
point scales for frequency (ranging from 0 � ‘‘never’’ to
4 � ‘‘almost daily’’) and intensity (ranging from 0 �
‘‘none’’ to 4 � ‘‘extreme’’). The CAPS items used for
current analyses are listed on Table 1. An item with a
frequency score of 1 (‘‘once or twice in the past month’’)
or higher and an intensity score of 2 (‘‘moderate’’) or
higher was considered positive for that symptom follow-
ing guidelines suggested by Weathers et al. (2001).
Brief Symptom Inventory
The Brief Symptom Inventory (BSI) is a 53-item self-
report psychological symptom inventory with nine pri-
mary symptom dimensions. The measure assesses how
much a problem bothered or distressed a person using a
5-point Likert scale ranging from 0 � ‘‘not at all’’ and
4 � ‘‘extremely’’. The BSI has shown high convergent
and construct validity (Derogatis & Melisaratos, 1983).
The BSI items used for the current analyses are listed on
Table 1. An item score of 2 (‘‘moderately’’) or higher was
considered positive for
that symptom.
Structured Clinical Interview-II Borderline Personality
Disorder module
The Structured Clinical Interview for Axis II Disorders
(SCID-II) DSM�IV BPD Module has nine items where
a score of 1�absent or false, 2�subthreshold and
3�threshold or true (First, Spitzer, Gibbon, & Williams,
1994). The items used for the current analyses are listed
on Table 1. An item score of 3 was considered positive for
that symptom.
Social Adjustment Scale-Self Report
The Social Adjustment Scale-Self Report (SAS-SR;
Weissman & Bothwell, 1976) was utilized to measure func-
tional impairment. The SAS-SR consists of 42 Likert-
type items, which assess the level of functioning over the
past 2 weeks for six domains: work, social and leisure
activities, relationships with extended family, role as a
marital partner, parental role, and role within the family
unit. A mean score can be calculated for each of the six
domains, as well as one overall mean score, based on the
total number of items relevant to the responder. Higher
scores indicate greater impairment. The SAS-SR has
demonstrated strong psychometric properties among com-
munity and clinical samples (e.g., Weissman & Bothell,
1976).
Statistical analyses
Latent class analysis
The model fit for the optimal number of classes that were
examined were the Lo-Mendell-Rubin adjusted likeli-
hood ratio test (LMR-A), the bootstrap likelihood ratio
test (BLRT), the Bayesian Information Criterion (BIC),
the Sample-Size Adjusted BIC (SSA-BIC), and the Akaike
Information Criterion (AIC). In a simulation study, the
BLRT was shown to outperform the LMR-A and the
BIC (among other measures of model fit) in selecting
the number of classes (Nylund, Asparouhov, & Muthen,
2007). Since there is not a clear-cut decision rule on how
to select the best fitting model, we ranked ordered the
importance of fit indices as follows: BLRT, BIC, SSA-
BIC, AIC, and then the LMR-A. The general practice of
LCA is to test the fit of a two-class model and system-
atically increase the number of classes until adding more
classes is no longer warranted. The LMR-A and the
BLRT compare the fit of the specified class solution to
models with one less class. A pB0.05 suggests that the
specified model provides a better fit to the data relative to
the model with one less class. A total of 21 symptoms
(coded dichotomously as present or not) were used in
the LCA, 6 representing the ICD-11 PTSD symptoms,
6 representing the self-organization symptoms unique
to Complex PTSD), and 9 representing the SCID-II1
Marylène Cloitre et al.
4
(page number not for citation purpose)
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
symptoms of BPD (see Table 1). The LCA models were
estimated using robust maximum likelihood method with
400 initial stage random starts and 80 final stage opti-
mizations to determine if the best log-likelihood value
was obtained and replicated. Finally, 50 bootstrap draws
were used in the BLRT.
Descriptive statistics
Chi-square tests and ANOVAs were performed to assess
differences in sociodemographic characteristics, trauma
history, and symptom severity across the classes identi-
fied in the LCA. Descriptive statistics were computed
based on valid (non-missing) data.
Results
Latent class analysis
The fit indices of the different class models examined are
shown in Table 2. The two-class model yielded a signi-
ficant LMR-A and BLRT result at pB0.05. The three-
and four-class models both yielded a significant BLRT
result at pB0.05, but not a significant LMR-A result.
A five-class model was examined, but the best log-
likelihood value was not replicated, and it was not con-
sidered for the final model, as the p-value may not be
trustworthy due to local maxima. The four-class model
did not have the lowest BIC value, but it was selected over
the two- and three-class models because it was the model
with the largest number of classes that had a significant
and trustworthy BLRT result, had the lowest SSA-BIC
value, and had the lowest AIC value of the models consi-
dered. The three- and four-class models were examined
closely, as they both could have legitimate arguments for
being selected. However, based on all of the fit indices
examined and on the interpretability of the symptom
profiles of the classes (consistent with study hypotheses),
the four-class model was selected.
The pattern of symptom endorsement for all four classes
is presented in Fig. 1. The four classes were compared
on the 21 symptoms that were used to determine class
membership in order to provide descriptive labels for
each class. The Low Symptom class had relatively low
levels of all symptoms across all domains. The PTSD class
had generally high levels of PTSD symptoms, but relati-
vely low levels of self-organization and BPD symptoms.
The CPTSD class had high levels of PTSD and self-
organization symptoms, but relatively low levels of BPD
symptoms. The BPD class had a high percentage of BPD
symptoms as well as self-organization disturbances and
PTSD symptoms.
The average probability of latent class membership in
the four-class model was acceptable at 0.91 for the Low
Symptom class, 0.92 for the PTSD class, 0.87 for the
CPTSD class, and 0.91 for the BPD class, which implies
acceptable discrimination between the classes. An accep-
table entropy value probability of 0.81 lends support to
this result by suggesting adequate latent class separation.
Overall, 20.4% (n�57) of participants were classified
into the Low Symptom class, 25.7% (n�72) into the
PTSD class, 27.5% (n�77) into the CPTSD class, and
26.4% (n�74) into the BPD class.
Sociodemographic and trauma history characteristics
ANOVA and Chi-square analyses were performed to assess
differences in sociodemographic characteristics, trauma
history, and symptom severity across the classes identi-
fied in the LCA. Results shown in Table 3 indicate that
the four classes did not differ by age, ethnicity, or employ-
ment status. The classes also did not differ in the rates of
types of childhood or adulthood interpersonal traumas,
with the exception that CSA was reported more fre-
quently in the CPTSD class than in the Low Symptom
and BPD classes. Total number of different types of
traumatic experiences did not differ across classes.
Symptom characteristics
The rates of probable disorders (ICD-11 PTSD, ICD-11
CPTSD and BPD) as well as the percent of endorsed
symptom characteristics for all 21 symptoms across the
four classes are presented in Table 4. Overall, 53.9%
(n�151) had a PTSD diagnosis, 38.2% (n�107) had a
CPTSD diagnosis, and 29.3% (n�82) had a BPD
diagnosis. Of those with a BPD diagnosis, majority also
had a PTSD diagnosis (54.9%, n�45) and 45.1% (n�37)
had a CPTSD diagnosis. In the Low Symptom class, no
one met criteria for either PTSD or CPTSD while 12%
Table 2. Latent class models and fit indices
Model Log-likelihood BIC SSA-BIC AIC Entropy LMR-A p-value BLRT p-value
2 classes �3523.010 7288.315 7151.965 7132.020 0.781 0.029 B0.001
3 classes �3433.024 7232.310 7026.199 6996.048 0.817 0.066 B0.001
4 classes �3382.025 7254.278 6978.406 6938.051 0.808 0.401 B0.001
5 classes �3338.211 7290.613 6944.981 6894.421 0.848 0.639 B0.001a
Note. BIC, Bayesian Information Criterion; SSA-BIC, Sample-Size Adjusted BIC; AIC, Akaike Information Criterion; LMRA-A, Lo-Mendell-
Rubin adjusted likelihood ratio test; BLRT, bootstrap likelihood ratio test.
aThe best log-likelihood value was not replicated in 32 out of 50 bootstrap draws. The p-value may not be trustworthy due to local
maxima.
PTSD, Complex PTSD, and BPD
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097 5
(page number not for citation purpose)
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
met criteria for BPD. The most common symptoms were
unstable relationships, mood changes and feeling empty.
Of individuals in the PTSD class, 68% met criteria for
PTSD, but only 19.4% met criteria for CPTSD and 1.4%
met criteria for BPD. Of individuals in the CPTSD class,
77.9% met criteria for CPTSD but only 7.8% met criteria
for BPD. Of those in the BPD class, 91.9% met DSM-IV
BPD diagnosis. Overall, the DSM-IV BPD diagnosis fit
very few of the individuals in the CPTSD (7.8%) class but
the large majority of the BPD class (91.9%).
A review of the individual items indicates that,
consistent with the graphic depiction provided in Fig. 1,
the BPD class had a lower rate of endorsement of the
ICD-11 PTSD symptoms across all items as compared to
the CPTSD class. The rates were significantly lower for
nightmares and avoidance of trauma-related thoughts.
Endorsement of the individual items reflecting disturban-
ces in self-organization by the BPD class members was
similar to that of the CPTSD class. However, only 44.6%
of the BPD class met criteria for CPTSD suggesting that
Fig. 1. Symptom endorsement of Complex PTSD and BPD items by class.
Table 3. Demographic and trauma characteristics of the classes
Characteristics
Class 1 Low Symptom
n�57
Class 2 PTSD
n�72
Class 3 CPTSD
n�77
Class 4 BPD
n�74
Significance
test
Age 37.91 (10.17) 36.21 (10.64) 36.95 (10.42) 37.63 (12.10) NS
Race (% white) 44.4% 42.9% 33.8% 45.2% NS
Employed (FT or PT) 66.0% 71.0% 68.4% 64.4% NS
CSA 54.5% 66.7% 80.5% 55.4% p�0.003
3�1, 4
CPA 80.0% 82.9% 81.8% 78.4% NS
Neglect 34.5% 47.1% 54.5% 45.9% NS
Emotional abuse 78.2% 75.7% 79.2% 87.8% NS
Any childhood abuse 98.2% 98.6% 97.4% 95.9% NS
ASA 38.9% 46.4% 57.1% 52.7% NS
APA 29.6% 18.8% 32.5% 23.3% NS
Any adult assaults 53.7% 55.9% 72.7% 65.8% NS
Both child and adult events 51.9% 55.9% 71.4% 63.5% NS
Note. All tests were Chi-square tests with 3 degrees of freedom.
Marylène Cloitre et al.
6
(page number not for citation purpose)
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
individuals did not consistently endorse the CPTSD symp-
toms across all three categories of disturbance (emotion
dysregulation, negative self-concept, interpersonal pro-
blems) sufficient to complete the CPTSD profile. Indeed,
individuals in the BPD class were more likely to meet
criteria for PTSD (54%) than CPTSD.
The CPTSD class had significantly lower endorsement
of all the BPD symptoms than the BPD with the excep-
tion of feelings of emptiness. The CPTSD class was more
similar to the PTSD class in regard to endorsement of
BPD symptoms. The two classes did not differ from each
other on the BPD symptoms in seven out of nine symp-
toms, with the exception of unstable relationships and
mood changes, which were both endorsed at higher rates
in the CPTSD class than the PTSD class. Notably, almost
half of the BPD class members endorsed self-harm/
Table 4. Frequencies of endorsement for ICD-11 PTSD, CPTSD, and DSM-IV BPD items
Symptoms
Class 1 Low symptoms
(n�57)
Class 2 PTSD
(n�72)
Class 3 CPTSD
(n�77)
Class 4 BPD
(n�74)
Significant pairwise
post-hoc comparisons
ICD-11 PTSD diagnosis 0.0% 68.1% 80.5% 54.1% 2, 3, 4�1
3�4
Re-experiencing
Flashbacks 15.8% 80.6% 80.5% 75.7% 2, 3, 4�1
Nightmares 3.5% 45.8% 70.1% 41.9% 2, 3, 4�1
3�2, 4
Avoidance
Thoughts 10.5% 90.3% 89.6% 58.1% 2, 3, 4�1
2, 3�4
People, places, or activities 10.5% 66.7% 67.5% 50.0% 2, 3, 4�1
Sense of threat
Hypervigilance 21.1% 70.8% 75.3% 63.5% 2, 3, 4�1
Startle 26.3% 51.4% 70.1% 60.8% 2, 3, 4�1
ICD-11 CPTSD diagnosis 0.0% 19.4% 77.9% 44.6% 3�1, 2,4
4�1, 2
Affect regulation problems
Angry 28.1% 23.6% 54.6% 51.4% 3, 4�1, 2
Hurt feelings 54.4% 51.4% 97.4% 87.8% 3, 4�1, 2
Negative self-concept
Worthless 40.4% 20.8% 93.5% 87.8% 3, 4�1, 2
Guilty 54.4% 43.1% 92.2% 81.1% 3, 4�1, 2
Interpersonal problems
Not close 36.8% 29.2% 83.1% 70.3% 3, 4�1, 2
Feel disconnected 59.7% 76.4% 98.7% 85.1% 3, 4�1
3�2, 4
DSM-IV BPD diagnosis 12.3% 1.4% 7.8% 91.9% 4�1, 2, 3
Frantic 19.3% 15.3% 11.7% 63.5% 4�1, 2, 3
Unstable relationships 43.9% 8.3% 33.8% 83.8% 4�1, 2, 3
1, 3�2
Unstable sense of self 21.1% 2.8% 14.3% 67.6% 4�1, 2, 3
1�2
Impulsiveness 26.3% 16.7% 22.1% 73.0% 4�1, 2, 3
Self-harm 0.0% 16.7% 14.3% 48.7% 4�1, 2, 3
2, 3�1
Mood changes 40.4% 8.3% 41.6% 73.0% 4�1, 2, 3
1, 3�2
Empty 31.6% 22.2% 81.8% 78.4% 3, 4�1, 2
Temper 22.8% 15.3% 15.6% 59.5% 4�1, 2, 3
Paranoia/dissociation 21.1% 16.7% 32.5% 73.0% 4�1, 2, 3
Note. All tests were Chi-square tests with 3 degrees of freedom, and the significance of all tests was pB0.01; reported significant
pairwise post-hoc comparisons used an adjusted p-value using the Bonferroni method.
PTSD, Complex PTSD, and BPD
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097 7
(page number not for citation purpose)
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
suicidal behaviors while this behavior was not endorsed
by anyone in the Low Symptoms class and by a relatively
low and equal proportion in the PTSD and CPTSD
classes (16.7 and 14.3%, respectively).
Functional impairment
Functional impairment was greatest in the BPD (M�
2.34, SD�0.43) and CPTSD class (M�2.31, SD�0.39)
relative to the PTSD (M�2.76, SD�0.48) and Low
Symptom (M�2.71, SD� 0.52) classes. The BPD and
CPTSD classes did not differ significantly from each other
in functional impairment (p�0.920). Similarly, the PTSD
and Low Symptom classes did not differ significantly
from each other in functional impairment (p�0.983).
BPD symptoms as indicators of risk for BPD versus
CPTSD diagnosis
The salience of each of the BPD symptoms as a ‘‘marker’’
of being in the BPD class compared to the CPTSD class
was evaluated. Relative risk (RR) was computed for each
symptom (see Table 5). RR provides the likelihood that a
person positive on a particular symptom will be in the
BPD class relative to the CPTSD class. Each of the BPD
symptoms was much more likely to be associated with the
BPD class versus the CPTSD class, except for emptiness.
The strongest symptom predictors of class were: frantic
about abandonment, unstable relationships, unstable sense
of self and impulsiveness.
Discussion
Overall, the findings showed that the patterns of symp-
toms endorsed formed classes that are consistent with
diagnostic criteria for PTSD, Complex PTSD, and BPD.
The LCA identified four distinct classes of individuals
within a treatment-seeking sample: a Low Symptom class
that was relatively low in all measured symptoms; a
PTSD class that was high in symptoms of PTSD but
relatively low in self-organization symptoms and symp-
toms of BPD; a Complex PTSD class that was high in
symptoms of PTSD and self-organization symptoms but
relatively low in symptoms of BPD; and a BPD that was
high in symptoms of BPD with additional symptoms of
PTSD and CPTSD. These distinct classes demonstrated
acceptable discrimination. Additionally, these classes did
not differ in demographic variables (e.g., age, ethnicity,
employment status) or total number of traumas experi-
enced. These findings provide empirical support that the
symptom profiles endorsed by individuals with Complex
PTSD and BPD result in distinguishable subgroups of
trauma-exposed individuals.
While the individuals in the BPD reported many of
the symptoms of PTSD and CPTSD, the BPD class was
clearly distinct in its endorsement of symptoms unique
to BPD. The RR ratios presented in Table 5 revealed that
the following symptoms were highly indicative of place-
ment in the BPD rather than the CPTSD class: (1) frantic
efforts to avoid real or imagined abandonment, (2) un-
stable and intense interpersonal relationships characterized
by alternating between extremes of idealization and
devaluation, (3) markedly and persistently unstable self-
image or sense of self, and (4) impulsiveness. Given the
gravity of suicidal and self-injurious behaviors, it is
important to note that there were also marked differences
in the presence of suicidal and self-injurious behaviors
with approximately 50% of individuals in the BPD class
reporting this symptom but much fewer and an equivalent
number doing so in the CPSD and PTSD classes (14.3 and
16.7%, respectively). The only BPD symptom that in-
dividuals in the BPD class did not differ from the CPTSD
class was chronic feelings of emptiness, suggesting that
in this sample, this symptom is not specific to either BPD
or CPTSD and does not discriminate between them.
It should be noted that the endorsement of the CPTSD
symptoms related to self-organization disturbances was
high among members of the BPD class. It may be that the
presence emotion regulation problems does not distin-
guish CPTSD and BPD, although the severity and type
might, i.e., suicidality, self-injurious behavior are char-
acteristic of BPD not CPTSD. Alternatively, it may be that
the BSI is not optimal as a measure of self-organization
disturbances to provide differential diagnosis of CPTSD
versus BPD. The BSI tracks symptoms only for the past 2
weeks, and thus chronicity of symptoms was not assessed.
Members of the BPD class may have some but not all of
the CPTSD symptoms and may vary in their endorsement
of symptoms across time while the symptoms as endorsed
by the CPTSD class would be expected to be chronic and
stable. This interpretation is consistent with the data from
the SCID-II questions where items highlighting lack of
stability were strongly endorsed by the BPD but not the
CPTSD and PTSD class members.
Table 5. Relative risk of SCID-II BPD items*comparing
BPD versus CPTSD classes
BPD symptoms Relative risk 95% CI
Frantic 2.95* 2.10, 4.15
Unstable relationships 3.70* 2.18, 6.26
Unstable sense of self 3.07* 2.14, 4.42
Impulsiveness 3.04* 2.04, 4.55
Self-harm 2.10* 1.56, 2.83
Mood changes 2.04* 1.37, 3.04
Empty 0.90 0.61, 1.32
Temper outbursts 2.49* 1.80, 3.45
Dissociation 2.46* 1.65, 3.68
CI�Confidence Interval.
*pB0.01.
Interpretation example: Individuals positive on the Frantic
symptom had a 2.95 times greater risk of being in the BPD
class than those without the Frantic symptom.
Marylène Cloitre et al.
8
(page number not for citation purpose)
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
Overall, the findings indicate that there are several
ways in which Complex PTSD and BPD differ, consistent
with the proposed diagnostic formulation of CPTSD.
BPD is characterized by fears of abandonment, unstable
sense of self, unstable relationships with others, and impul-
sive and self-harming behaviors. In contrast, in CPTSD
as in PTSD, there was little endorsement of items related
to instability in self-representation or relationships. Self-
concept is likely to be consistently negative and relational
difficulties concern mostly avoidance of relationships
and sense of alienation. Lastly, a comment on the Low
Symptom class is deserved. The class seems comprised
of individuals who have very low endorsement of PTSD
symptoms but somewhat higher endorsements on distur-
bances in self-organization. These symptoms may reflect
the presence of subsyndromal BPD or symptoms result-
ing from a mix other Axis I disorders (Bipolar Disorder,
Major Depression). Future studies, which evaluate Axis I
disorders and provide subsyndromal diagnoses, will help
decipher the nature of this class.
The distinct symptom profiles characterizing CPTSD
and BPD lead to different treatment considerations. The
focus of treatment for BPD concerns reduction of life-
interfering behaviors such as suicidality and self-injurious
behaviors, a reduction in dependency on others and an
increase in an internalized and stable sense of self
(e.g., Dialectical Behavior Therapy, Linehan, 1993). In
contrast, treatment programs for CPTSD focus on reduc-
tion of social and interpersonal avoidance, development of
a more positive self-concept and relatively rapid engage-
ment in the review and meaning of traumatic memories
(e.g., Cloitre et al., 2006). Duration of treatment for each
disorder and attention to the termination phase are
different as well. Experts in the treatment of BPD have
noted that the termination of treatment is a time of risk
for relapse and symptom exacerbation (see Harned &
Linehan, 2008). The end of therapy may provoke feelings
of abandonment, destabilize identity and lead to impul-
sive and self-injurious behaviors. The DSM guidelines
for BPD recommend treatment duration of at least 1 year
(American Psychiatric Association, 2013). A treatment
course of a year or more may allow for demonstrated
success in reduction of life-interfering behaviors, the rein-
forcement and routinization of effective emotion manage-
ment skills and a carefully planned end to treatment.
While the recommended duration of treatment for Com-
plex PTSD has not yet been established, it seems likely be
shorter than for BPD given the presence of a stable sense
of self and relative absence of substantial risk for self-
injurious behaviors and suicidality, but longer than that
for PTSD, given the greater number and diversity of
symptoms (see Cloitre et al., 2012).
Growing attention to patient-centered care, which em-
phasizes the patient’s specific symptoms, needs and
preferences will hopefully facilitate the development of
treatment programming that neither under-treats nor
over-treats the patient. The proposed spectrum of diag-
noses moving from PTSD to CPTSD and BPD may
provide a foundation for developing algorithms of type of
interventions and duration of care that meets the needs of
patients with symptom profiles that differ in clinically
significant ways.
A number of limitations of the current study are worth
noting. First, the sample consisted of a treatment-seeking
sample with a history of childhood interpersonal trauma.
Replication of results is necessary with samples that are
more representative of populations in clinical and com-
munity settings. Future studies should include samples
with greater diversity in types of trauma as well as
diversity in the exposure to traumatic stressors. Secondly,
the data used in the analyses are from a secondary source
and do not represent the ideal basis for evaluating ICD-11
PTSD and Complex PTSD symptoms. The Structured
Interview for Disorders of Extreme Stress (SIDES, Pelcovitz
et al., 1997), a structured clinician driven measure which
assesses many of the symptoms of Complex PTSD was
not available in this data set. Also, the time duration
for which the symptoms were assessed differed across
measures and thus did not allow consistency in the
assessment of the chronicity or variability of the symp-
toms endorsed. However, the study results, which provide
evidence of qualitative differences between the CPTSD
and BPD symptom profiles, suggest the importance of
developing empirically validated measures of ICD-11
PTSD and CPTSD and their comparison to BPD in a
variety of clinical and epidemiological samples.
Conclusion
This study identified four distinct classes of individuals
who have experienced trauma, supporting the proposed
distinction between Complex PTSD and BPD. Key symp-
toms that differentiate BPD from Complex PTSD were
identified. These findings conform to ICD-11’s proposed
distinction between the diagnoses. They also point to the
merits of pursuing the construct of CPTSD as a separate
clinical entity from PTSD and BPD. However, to achieve
this agenda it is important that empirically validated
measures of CPTSD be developed for standardized assess-
ment of the construct in relation to PTSD and BPD.
Given that that there are efficacious treatments for
CPTSD (Cloitre et al., 2010) and BPD (e.g., Linehan,
1993), and these approaches vary in important ways, it is
useful for clinicians to be able to differentiate between
these presentations.
Disclaimer
M Cloitre and R Bryant are members of the WHO of the
Working Group on the Classification of Stress-Related
Disorders. However, the views expressed reflect the opinions
PTSD, Complex PTSD, and BPD
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097 9
(page number not for citation purpose)
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
of the authors and not necessarily the Working Group
and the content of this manuscript does not represent
WHO policy.
Conflict of interest and funding
There is no conflict of interest in the present study for
any of the authors. This manuscript was supported by
a National Institute of Mental Health grant, RO1 MH-
062347 to the first author (M. Cloitre).
References
American Psychiatric Association. (2013). Diagnostic and statistical
manual of mental disorders (5th ed.). Arlington, VA: Author.
Briere, J., & Rickards, S. (2007). Self-awareness, affect regulation,
and relatedness: Differential sequels of childhood versus adult
victimization experiences. Journal of Nervous and Mental
Disease, 195, 497�503.
Cloitre, M., Cohen, L., & Koenan, K. (2006). Treating survivors of
childhood abuse: psychotherapy for the interrupted life. New
York: Guilford Press.
Cloitre, M., Courtois, C. A., Ford, J. D., Green, B. L., Alexander, P.,
Briere, J., et al. (2012). The ISTSS expert consensus treatment
guidelines for complex PTSD in adults. Retrieved May 13,
2014, from http://www.istss.org/
Cloitre, M., Courtois, C. C., Charuvastra, A., Carapezza, R.,
Stolbach, B. C., & Breen, B. L. (2011). Treatment of complex
PTSD: Results of the ISTSS expert clinician survey on best
practices. Journal of Traumatic Stress, 24, 616�627.
Cloitre, M., Garvert, D. W., Brewin, C. R., Bryant, R. A., &
Maercker, A. (2013). Evidence for proposed ICD-11 PTSD
and complex PTSD: A latent profile analysis. European Journal
of Psychotraumatology, 4, 20706, doi: http://dx.doi.org/10.3402/
ejpt.v4i0.20706
Cloitre, M., Stovall-McClough, C. K., Nooner, K., Zorbas, P.,
Cherry, S., Jackson, C. L., et al. (2010). Treatment of PTSD
related to childhood abuse: A randomized controlled trial.
American Journal of Psychiatry, 167, 915�24.
Derogatis, L. R., & Melisaratos, N. (1983). The brief symptom
inventory: An introductory report. Psychological Medicine, 13,
595�605.
Elklit, A., Hyland, P. & Shevlin, M. (2014). Evidence of symptom
profiles consistent with posttraumatic stress disorder and
complex posttraumatic stress disorder in different trauma
samples. European Journal of Psychotraumatology, 5, 24221,
doi: http://dx.doi.org/10.3402/ejpt.v5.24221
First, M. D., Spitzer, M. D., Gibbon, M., & Williams, J. W. (1994).
Structured Clinical Interview for DSM�IV, Patient Edition.
New York: Biometrics Research Department, New York State
Psychiatric Institute.
Golier, J. A., Yehuda, R., Bierer, L. M., Mitropoulou, V., New, A. S.,
Schmeidler, J., et al. (2003). The relationship of borderline per-
sonality disorder to posttraumatic stress disorder and trau-
matic events. American Journal of Psychiatry, 160, 2018�2024.
Grant, B. F., Chou, S. P., Goldstein, R. B., Huang, B., Stinson, F. S.,
Saha, T. D., et al. (2008). Prevalence, correlates, disability,
and comorbidity of DSM-IV borderline personality disorder:
Results from the wave 2 national epidemiologic survey on
alcohol and related conditions. Journal of Clinical Psychiatry,
69, 533�545.
Harned, M. S., & Linehan, M. M. (2008). Integrating dialectical
behavior therapy and prolonged exposure to treat co-occurring
borderline personality disorder and PTSD: Two case studies.
Cognitive and Behavioral Practice, 15, 263�276.
Harned, M. S., Rizvi, S. L., & Linehan, M. M. (2010). The impact of
co-occurring posttraumatic stress disorder on suicidal women
with borderline personality disorder. American Journal of
Psychiatry, 167, 1210�1217.
Heffernan, K., & Cloitre, M. (2000). A comparison of posttraumatic
stress disorder with and without borderline personality dis-
order among women with a history of childhood sexual abuse:
Etiological and clinical characteristics. Journal of Nervous and
Mental Disease, 188, 589�595.
Knefel, M., & Lueger-Schuster, B. (2013). An evaluation of ICD-11
PTSD and complex PTSD criteria in a sample of adult sur-
vivors of childhood institutional abuse. European Journal of
Psychotraumatology, 4, 22608, doi: http://dx.doi.org/10.3402/
ejpt.v4i0.22608
Linehan, M. (1993). Cognitive-behavioral treatment of borderline
personality disorder. New York: Guilford Press.
Maercker, A., Brewin, C. R., Bryant, R. A., Cloitre, M., Reed, G. M.,
Van Ommeren, M., et al. (2013). Proposals for mental dis-
orders specifically associated with stress in the ICD-11. Lancet,
381(9878), 1683�1685. doi: 10.1016/S0140-6736(12)62191-6.
Nylund, K. L., Asparouhov, T., & Muthen, B. O. (2007). Deciding
on the number of classes in latent class analysis and growth
mixture modelling: A Monte Carlo simulation study. Struc-
tural Equation Modeling, 14(4), 535�569.
Pagura, J., Stein, M. B., Bolton, J. M., Cox, B. J., Grant, B., &
Sareen, J. (2010). Comorbidity of borderline personality dis-
order and posttraumatic stress disorder in the U.S. population.
Journal of Psychiatric Research, 44, 1190�1198.
Pelcovitz, D., Van der Kolk, B., Roth, S., Mandel, F., Kaplan, S., &
Resick, P. (1997). Development of a criteria set and a struc-
tured interview for disorders of extreme stress (SIDES).
Journal of Traumatic Stress, 10, 3�17.
Reed, G. M. (2010). Toward ICD-11: Improving the clinical utility
of WHO’s international classification of mental disorders.
Professional Psychology: Research and Practice, 41, 457�464.
Weathers, F. W., Keane, T. M., & Davidson, J. R. T. (2001).
Clinician-Administered PTSD Scale: A review of the first ten
years of research. Depression and Anxiety, 13, 132�156.
Weissman, E., & Bothell, S. (1976). Assessment of patient social
adjustment by patient self-report. Archives of General Psychia-
try, 33, 1111�1115.
Zanarini, M. C., Frankenburg, F. R., Dubo, E. D., Sickel, A.,
Trikha, A., Levin, A., et al. (1998). Axis I comorbidity of border-
line personality disorder. American Journal of Psychiatry, 155,
1733�1739.
Zlotnick, C., Franklin, C. L., & Zimmerman, M. (2002). Is
comorbidity of posttraumatic stress disorder and borderline
personality disorder related to greater pathology and impair-
ment? American Journal of Psychiatry, 159, 1940�1043.
Marylène Cloitre et al.
10
(page number not for citation purpose)
Citation: European Journal of Psychotraumatology 2014, 5: 25097 – http://dx.doi.org/10.3402/ejpt.v5.25097
http://www.istss.org/
http://dx.doi.org/10.3402/ejpt.v4i0.20706
http://dx.doi.org/10.3402/ejpt.v4i0.20706
http://dx.doi.org/10.3402/ejpt.v5.24221
http://dx.doi.org/10.3402/ejpt.v4i0.22608
http://dx.doi.org/10.3402/ejpt.v4i0.22608
http://eurojnlofpsychotraumatol.net/index.php/ejpt/article/view/25097
http://dx.doi.org/10.3402/ejpt.v5.25097
<<
/ASCII85EncodePages false
/AllowTransparency false
/AutoPositionEPSFiles true
/AutoRotatePages /None
/Binding /Left
/CalGrayProfile (Dot Gain 30%)
/CalRGBProfile (None)
/CalCMYKProfile (U.S. Sheetfed Coated v2)
/sRGBProfile (sRGB IEC61966-2.1)
/CannotEmbedFontPolicy /Error
/CompatibilityLevel 1.4
/CompressObjects /Off
/CompressPages true
/ConvertImagesToIndexed false
/PassThroughJPEGImages true
/CreateJobTicket false
/DefaultRenderingIntent /Default
/DetectBlends true
/DetectCurves 0.1000
/ColorConversionStrategy /LeaveColorUnchanged
/DoThumbnails false
/EmbedAllFonts true
/EmbedOpenType false
/ParseICCProfilesInComments true
/EmbedJobOptions true
/DSCReportingLevel 0
/EmitDSCWarnings false
/EndPage -1
/ImageMemory 1048576
/LockDistillerParams false
/MaxSubsetPct 100
/Optimize false
/OPM 1
/ParseDSCComments true
/ParseDSCCommentsForDocInfo true
/PreserveCopyPage true
/PreserveDICMYKValues true
/PreserveEPSInfo true
/PreserveFlatness true
/PreserveHalftoneInfo false
/PreserveOPIComments false
/PreserveOverprintSettings true
/StartPage 1
/SubsetFonts true
/TransferFunctionInfo /Apply
/UCRandBGInfo /Remove
/UsePrologue false
/ColorSettingsFile (None)
/AlwaysEmbed [ true
]
/NeverEmbed [ true
]
/AntiAliasColorImages false
/CropColorImages true
/ColorImageMinResolution 150
/ColorImageMinResolutionPolicy /OK
/DownsampleColorImages false
/ColorImageDownsampleType /Average
/ColorImageResolution 300
/ColorImageDepth -1
/ColorImageMinDownsampleDepth 1
/ColorImageDownsampleThreshold 1.50000
/EncodeColorImages true
/ColorImageFilter /DCTEncode
/AutoFilterColorImages false
/ColorImageAutoFilterStrategy /JPEG
/ColorACSImageDict <<
/QFactor 0.15
/HSamples [1 1 1 1] /VSamples [1 1 1 1]
>>
/ColorImageDict <<
/QFactor 0.40
/HSamples [1 1 1 1] /VSamples [1 1 1 1]
>>
/JPEG2000ColorACSImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/JPEG2000ColorImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/AntiAliasGrayImages false
/CropGrayImages true
/GrayImageMinResolution 150
/GrayImageMinResolutionPolicy /OK
/DownsampleGrayImages false
/GrayImageDownsampleType /Average
/GrayImageResolution 300
/GrayImageDepth -1
/GrayImageMinDownsampleDepth 2
/GrayImageDownsampleThreshold 1.50000
/EncodeGrayImages true
/GrayImageFilter /DCTEncode
/AutoFilterGrayImages false
/GrayImageAutoFilterStrategy /JPEG
/GrayACSImageDict <<
/QFactor 0.15
/HSamples [1 1 1 1] /VSamples [1 1 1 1]
>>
/GrayImageDict <<
/QFactor 0.40
/HSamples [1 1 1 1] /VSamples [1 1 1 1]
>>
/JPEG2000GrayACSImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/JPEG2000GrayImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/AntiAliasMonoImages false
/CropMonoImages true
/MonoImageMinResolution 1200
/MonoImageMinResolutionPolicy /OK
/DownsampleMonoImages false
/MonoImageDownsampleType /Average
/MonoImageResolution 1200
/MonoImageDepth -1
/MonoImageDownsampleThreshold 1.50000
/EncodeMonoImages true
/MonoImageFilter /CCITTFaxEncode
/MonoImageDict <<
/K -1
>>
/AllowPSXObjects false
/CheckCompliance [
/None
]
/PDFX1aCheck false
/PDFX3Check false
/PDFXCompliantPDFOnly true
/PDFXNoTrimBoxError false
/PDFXTrimBoxToMediaBoxOffset [
0.00000
0.00000
0.00000
0.00000
]
/PDFXSetBleedBoxToMediaBox false
/PDFXBleedBoxToTrimBoxOffset [
0.00000
0.00000
0.00000
0.00000
]
/PDFXOutputIntentProfile (Euroscale Coated v2)
/PDFXOutputConditionIdentifier ()
/PDFXOutputCondition ()
/PDFXRegistryName (http://www.color.org)
/PDFXTrapped /False
/CreateJDFFile false
/SyntheticBoldness 1.000000
/Description <<
/DEU
/FRA
/JPN
/PTB
/DAN
/NLD
/ESP
/SUO
/ITA
/NOR
/SVE
/ENU
>>
>> setdistillerparams
<<
/HWResolution [2400 2400]
/PageSize [612.000 792.000]
>> setpagedevice
Research Article
Relationship of Childhood Abuse and
Household Dysfunction to Many of the
Leading Causes of Death in Adult
s
The Adverse Childhood Experiences (ACE) Stud
y
Vincent J. Felitti, MD, FACP, Robert F. Anda, MD, MS, Dale Nordenberg, MD, David F. Williamson, MS, PhD,
Alison M. Spitz, MS, MPH, Valerie Edwards, BA, Mary P. Koss, PhD, James S. Marks, MD, MPH
Background: The relationship of health risk behavior and disease in adulthood to the breadth of
exposure to childhood emotional, physical, or sexual abuse, and household dysfunction
during childhood has not previously been described.
Methods: A questionnaire about adverse childhood experiences was mailed to 13,494 adults who had
completed a standardized medical evaluation at a large HMO; 9,508 (70.5%) responded.
Seven categories of adverse childhood experiences were studied: psychological, physical, or
sexual abuse; violence against mother; or living with household members who were
substance abusers, mentally ill or suicidal, or ever imprisoned. The number of categories
of these adverse childhood experiences was then compared to measures of adult risk
behavior, health status, and disease. Logistic regression was used to adjust for effects of
demographic factors on the association between the cumulative number of categories of
childhood exposures (range: 0–7) and risk factors for the leading causes of death in adult
life.
Results: More than half of respondents reported at least one, and one-fourth reported $2
categories of childhood exposures. We found a graded relationship between the number
of categories of childhood exposure and each of the adult health risk behaviors and
diseases that were studied (P , .001). Persons who had experienced four or more
categories of childhood exposure, compared to those who had experienced none, had 4-
to 12-fold increased health risks for alcoholism, drug abuse, depression, and suicide
attempt; a 2- to 4-fold increase in smoking, poor self-rated health, $50 sexual intercourse
partners, and sexually transmitted disease; and a 1.4- to 1.6-fold increase in physical
inactivity and severe obesity. The number of categories of adverse childhood exposures
showed a graded relationship to the presence of adult diseases including ischemic heart
disease, cancer, chronic lung disease, skeletal fractures, and liver disease. The seven
categories of adverse childhood experiences were strongly interrelated and persons with
multiple categories of childhood exposure were likely to have multiple health risk factors
later in life.
Conclusions: We found a strong graded relationship between the breadth of exposure to abuse or
household dysfunction during childhood and multiple risk factors for several of the
leading causes of death in adults.
Medical Subject Headings (MeSH): child abuse, sexual, domestic violence, spouse abuse,
children of impaired parents, substance abuse, alcoholism, smoking, obesity, physical
activity, depression, suicide, sexual behavior, sexually transmitted diseases, chronic obstruc-
tive pulmonary disease, ischemic heart disease. (Am J Prev Med 1998;14:245–258) © 1998
American Journal of Preventive Medicine
Department of Preventive Medicine, Southern California Perma-
nente Medical Group (Kaiser Permanente), (Felitti) San Diego,
California 92111. National Center for Chronic Disease Prevention
and Health Promotion, Centers for Disease Control and Prevention,
(Anda, Williamson, Spitz, Edwards, Marks) Atlanta, Georgia 30333.
Department of Pediatrics, Emory University School Medicine, (Nor–
denberg) Atlanta, Georgia 30333. Department of Family and Com-
munity Medicine, University of Arizona Health Sciences Center,
(Koss) Tucson, Arizona 85727.
Address correspondence to: Vincent J. Felitti, MD, Kaiser Perma-
nente, Department of Preventive Medicine, 7060 Clairemont Mesa
Boulevard, San Diego, California 92111.
245Am J Prev Med 1998;14(4) 0749-3797/98/$19.00
© 1998 American Journal of Preventive Medicine PII S0749-3797(98)00017-8
Introduction
Only recently have medical investigators in pri-
mary care settings begun to examine associa-
tions between childhood abuse and adult
health risk behaviors and disease.1–5 These associations
are important because it is now clear that the leading
causes of morbidity and mortality in the United States6
are related to health behaviors and lifestyle factors;
these factors have been called the “actual” causes of
death.7 Insofar as abuse and other potentially
damaging childhood experiences contribute
to the development of these risk factors, then
these childhood exposures should be recog-
nized as the basic causes of morbidity and
mortality in adult life.
Although sociologists and psychologists
have published numerous articles about the
frequency8–12 and long-term consequenc-
es13–15 of childhood abuse, understanding their rele-
vance to adult medical problems is rudimentary. Fur-
thermore, medical research in this field has limited
relevance to most primary care physicians because it is
focused on adolescent health,16–20 mental health in
adults,20 or on symptoms among patients in specialty
clinics.22,23 Studies of the long-term effects of child-
hood abuse have usually examined single types of abuse,
particularly sexual abuse, and few have assessed the im-
pact of more than one type of abuse.5,24–28 Conditions
such as drug abuse, spousal violence, and criminal activity
in the household may co-occur with specific forms of
abuse that involve children. Without measuring these
household factors as well, long-term influence might be
wrongly attributed solely to single types of abuse and
the cumulative influence of multiple categories of
adverse childhood experiences would not be assessed.
To our knowledge, the relationship of adult health risk
behaviors, health status, and disease states to childhood
abuse and household dysfunction29–35 has not been
described.
We undertook the Adverse Childhood Experiences
(ACE) Study in a primary care setting to describe the
long-term relationship of childhood experiences to
important medical and public health problems. The
ACE Study is assessing, retrospectively and prospec-
tively, the long-term impact of abuse and household
dysfunction during childhood on the following out-
comes in adults: disease risk factors and incidence,
quality of life, health care utilization, and mortality. In
this initial paper we use baseline data from the study to
provide an overview of the prevalence and interrelation
of exposures to childhood abuse and household dys-
function. We then describe the relationship between
the number of categories of these deleterious child-
hood exposures and risk factors and those diseases that
underlie many of the leading causes of death in
adults.6,7,36,37
Methods
Study Setting
The ACE Study is based at Kaiser Permanente’s San
Diego Health Appraisal Clinic. More than 45,000 adults
undergo standardized examinations there each year,
making this clinic one of the nation’s largest free-
standing medical evaluation centers. All en-
rollees in the Kaiser Health Plan in San
Diego are advised through sales literature
about the services (free for members) at the
clinic; after enrollment, members are ad-
vised again of its availability through new-
member literature. Most members obtain
appointments by self-referral; 20% are re-
ferred by their health care provider. A recent review of
membership and utilization records among Kaiser
members in San Diego continuously enrolled between
1992 and 1995 showed that 81% of those 25 years and
older had been evaluated in the Health Appraisal
Clinic.
Health appraisals include completion of a standard-
ized medical questionnaire that requests demographic
and biopsychosocial information, review of organ sys-
tems, previous medical diagnoses, and family medical
history. A health care provider completes the medical
history, performs a physical examination, and reviews
the results of laboratory tests with the patient.
Survey Methods
The ACE Study protocol was approved by the Institu-
tional Review Boards of the Southern California Per-
manente Medical Group (Kaiser Permanente), the
Emory University School of Medicine, and the Office of
Protection from Research Risks, National Institutes of
Health. All 13,494 Kaiser Health Plan members who
completed standardized medical evaluations at the
Health Appraisal Clinic between August–November of
1995 and January–March of 1996 were eligible to
participate in the ACE Study. Those seen at the clinic
during December were not included because survey
response rates are known to be lower during the
holiday period.38
In the week after visiting the clinic, and hence
having their standardized medical history already
completed, members were mailed the ACE Study
questionnaire that included questions about child-
hood abuse and exposure to forms of household
dysfunction while growing up. After second mailings
of the questionnaire to persons who did not respond
to the first mailing, the response rate for the survey
was 70.5% (9,508/13,494).
See
related
Commentary
on pages 354,
356, 361.
246 American Journal of Preventive Medicine, Volume 14, Number 4
A second survey wave of approximately the same
number of patients as the first wave was conducted
between June and October of 1997. The data for the
second survey wave is currently being compiled for
analysis. The methods for the second mail survey wave
were identical to the first survey wave as described
above. The second wave was done to enhance the
precision of future detailed analyses on special topics
and to reduce the time necessary to obtain precise
statistics on follow-up health events. An overview of the
total ACE Study design is provided in Figure 1.
Comparison of
Respondents and Nonrespondents
We abstracted the completed medical evaluation for
every person eligible for the study; this included their
medical history, laboratory results, and physical find-
ings. Respondent (n 5 9,508) and nonrespondent
(n 5 3,986) groups were similar in their percentages
of women (53.7% and 51.0%, respectively) and in their
mean years of education (14.0 years and 13.6 years,
respectively). Respondents were older than nonrespon-
dents (means 56.1 years and 49.3 years) and more likely
to be white (83.9% vs. 75.3%) although the actual
magnitude of the differences was small.
Respondents and nonrespondents did not differ with
regard to their self-rated health, smoking, other sub-
stance abuse, or the presence of common medical
conditions such as a history of heart attack or stroke,
chronic obstructive lung disease, hypertension, or dia-
betes, or with regard to marital status or current family,
marital, or job-related problems (data not shown). The
health appraisal questionnaire used in the clinic con-
tains a single question about childhood sexual abuse
that reads “As a child were you ever raped or sexually
molested?” Respondents were slightly more likely to
answer affirmatively than nonrespondents (6.1% vs.
5.4%, respectively).
Questionnaire Design
We used questions from published surveys to construct
the ACE Study questionnaire. Questions from the Con-
flicts Tactics Scale39 were used to define psychological
and physical abuse during childhood and to define
violence against the respondent’s mother. We adapted
four questions from Wyatt40 to define contact sexual
abuse during childhood. Questions about exposure to
alcohol or drug abuse during childhood were adapted
from the 1988 National Health Interview Survey.41 All
of the questions we used in this study to determine
childhood experiences were introduced with the
phrase “While you were growing up during your first 18
years of life . . .”
Questions about health-related behaviors and health
problems were taken from health surveys such as the
Behavioral Risk Factor Surveys42 and the Third Na-
tional Health and Nutrition Examination Survey,43
both of which are directed by the Centers for Disease
Control and Prevention. Questions about depression
came from the Diagnostic Interview Schedule of the
National Institute of Mental Health (NIMH).44 Other
information for this analysis such as disease history was
obtained from the standardized questionnaire used in
the Health Appraisal Clinic. (A copy of the question-
naires used in this study may be found at www.elsevier.
com/locate/amepre.)
Figure 1. ACE Study design. *After exclusions, 59.7% of the original wave I sample (8,056/13,494) were included in this analysis.
Am J Prev Med 1998;14(4) 247
Defining Childhood Exposures
We used three categories of childhood abuse: psycho-
logical abuse (2 questions), physical abuse (2 ques-
tions), or contact sexual abuse (4 questions). There
were four categories of exposure to household dysfunc-
tion during childhood: exposure to substance abuse
(defined by 2 questions), mental illness (2 questions),
violent treatment of mother or stepmother (4 ques-
tions), and criminal behavior (1 question) in the house-
hold. Respondents were defined as exposed to a cate-
gory if they responded “yes” to 1 or more of the
questions in that category. The prevalence of positive
responses to the individual questions and the category
prevalences are shown in Table 1.
We used these 7 categories of childhood exposures to
abuse and household dysfunction for our analysis. The
measure of childhood exposure that we used was simply
the sum of the categories with an exposure; thus the
possible number of exposures ranged from 0 (unex-
posed) to 7 (exposed to all categories).
Risk Factors and Disease Conditions Assessed
Using information from both the study questionnaire
and the Health Appraisal Clinic’s questionnaire, we
chose 10 risk factors that contribute to the leading
causes of morbidity and mortality in the United
States.6,7,36,37 The risk factors included smoking, severe
obesity, physical inactivity, depressed mood, suicide
attempts, alcoholism, any drug abuse, parenteral drug
abuse, a high lifetime number of sexual partners
($50), and a history of having a sexually transmitted
disease.
We also assessed the relationship between childhood
exposures and disease conditions that are among the
leading causes of mortality in the United States.6 The
presence of these disease conditions was based upon
medical histories that patients provided in response to
the clinic questionnaire. We included a history of
ischemic heart disease (including heart attack or use of
nitroglycerin for exertional chest pain), any cancer,
stroke, chronic bronchitis, or emphysema (COPD),
Table 1. Prevalence of childhood exposure to abuse and household dysfunction
Category of childhood exposurea Prevalence (%) Prevalence (%)
Abuse by category
Psychological 11.1
(Did a parent or other adult in the household . . .)
Often or very often swear at, insult, or put you down? 10.0
Often or very often act in a way that made you afraid that
you would be physically hurt?
4.8
Physical 10.8
(Did a parent or other adult in the household . . .)
Often or very often push, grab, shove, or slap you? 4.9
Often or very often hit you so hard that you had marks or
were injured?
9.6
Sexual 22.0
(Did an adult or person at least 5 years older ever . . .)
Touch or fondle you in a sexual way? 19.3
Have you touch their body in a sexual way? 8.7
Attempt oral, anal, or vaginal intercourse with you? 8.9
Actually have oral, anal, or vaginal intercourse with you? 6.9
Household dysfunction by category
Substance abuse 25.6
Live with anyone who was a problem drinker or alcoholic? 23.5
Live with anyone who used street drugs? 4.9
Mental illness 18.8
Was a household member depressed or mentally ill? 17.5
Did a household member attempt suicide? 4.0
Mother treated violently 12.5
Was your mother (or stepmother)
Sometimes, often, or very often pushed, grabbed, slapped,
or had something thrown at her?
11.9
Sometimes, often, or very often kicked, bitten, hit with a
fist, or hit with something hard?
6.3
Ever repeatedly hit over at least a few minutes? 6.6
Ever threatened with, or hurt by, a knife or gun? 3.0
Criminal behavior in household
Did a household member go to prison? 3.4 3.4
Any category reported 52.1%
aAn exposure to one or more items listed under the set of questions for each category.
248 American Journal of Preventive Medicine, Volume 14, Number 4
diabetes, hepatitis or jaundice, and any skeletal frac-
tures (as a proxy for risk of unintentional injuries). We
also included responses to the following question about
self-rated health: “Do you consider your physical health
to be excellent, very good, good, fair, or poor?” because
it is strongly predictive of mortality.45
Definition of Risk Factors
We defined severe obesity as a body mass index (kg/
meter2) $35 based on measured height and weight;
physical inactivity as no participation in recreational
physical activity in the past month; and alcoholism as a
“Yes” response to the question “Have you ever consid-
ered yourself to be an alcoholic?” The other risk factors
that we assessed are self-explanatory.
Exclusions from Analysis
Of the 9,508 survey respondents, we excluded 51
(0.5%) whose race was unstated and 34 (0.4%) whose
educational attainment was not reported. We also ex-
cluded persons who did not respond to certain ques-
tions about adverse childhood experiences. This in-
volved the following exclusions: 125 (1.3%) for
household substance abuse, 181 (1.9%) for mental
illness in the home, 148 (1.6%) for violence against
mother, 7 (0.1%) for imprisonment of a household
member, 109 (1.1%) for childhood psychological
abuse, 44 (0.5%) for childhood physical abuse, and 753
(7.9%) for childhood sexual abuse. After these exclu-
sions, 8,056 of the original 9,508 survey respondents
(59.7% of the original sample of 13,494) remained and
were included in the analysis. Procedures for insuring
that the findings based on complete data were gener-
alizable to the entire sample are described below.
The mean age of the 8,506 persons included in this
analysis was 56.1 years (range: 19–92 years); 52.1% were
women; 79.4% were white. Forty-three percent had
graduated from college; only 6.0% had not graduated
from high school.
Statistical Analysis
We used the Statistical Analysis System (SAS)46 for our
analyses. We used the direct method to age-adjust the
prevalence estimates. Logistic regression analysis was
employed to adjust for the potential confounding ef-
fects of age, sex, race, and educational attainment on
the relationship between the number of childhood
exposures and health problems.
To test for a dose-response relationship to health
problems, we entered the number of childhood expo-
sures as a single ordinal variable (0, 1, 2, 3, 4, 5, 6, 7)
into a separate logistic regression model for each risk
factor or disease condition.
Assessing the Possible Influence of Exclusions
To determine whether our results were influenced by
excluding persons with incomplete information on any
of the categories of childhood exposure, we performed
a separate sensitivity analysis in which we included all
persons with complete demographic information but
assumed that persons with missing information for a
category of childhood exposure did not have an expo-
sure in that category.
Results
Adverse Childhood Exposures
The level of positive responses for the 17 questions
included in the seven categories of childhood exposure
ranged from 3.0% for a respondent’s mother (or
stepmother) having been threatened with or hurt by a
gun or knife to 23.5% for having lived with a problem
drinker or alcoholic (Table 1). The most prevalent of
the 7 categories of childhood exposure was substance
abuse in the household (25.6%); the least prevalent
exposure category was evidence of criminal behavior in
the household (3.4%). More than half of respondents
(52%) experienced $1 category of adverse childhood
exposure; 6.2% reported $4 exposures.
Relationships between
Categories of Childhood Exposure
The probability that persons who were exposed to any
single category of exposure were also exposed to an-
other category is shown in Table 2. The relationship
between single categories of exposure was significant
for all comparisons (P , .001; chi-square). For persons
reporting any single category of exposure, the proba-
bility of exposure to any additional category ranged
from 65%–93% (median: 80%); similarly, the probabil-
ity of $2 additional exposures ranged from 40%–74%
(median: 54.5%).
The number of categories of childhood exposures by
demographic characteristics is shown in Table 3. Statis-
tically, significantly fewer categories of exposure were
found among older persons, white or Asian persons,
and college graduates (P , .001). Because age is
associated with both the childhood exposures as well as
many of the health risk factors and disease outcomes,
all prevalence estimates in the tables are adjusted for
age.
Relationship between
Childhood Exposures and Health Risk Factors
Both the prevalence and risk (adjusted odds ratio)
increased for smoking, severe obesity, physical inactiv-
ity, depressed mood, and suicide attempts as the num-
ber of childhood exposures increased (Table 4). When
Am J Prev Med 1998;14(4) 249
persons with 4 categories of exposure were compared
to those with none, the odds ratios ranged from 1.3 for
physical inactivity to 12.2 for suicide attempts (Table 4).
Similarly, the prevalence and risk (adjusted odds
ratio) of alcoholism, use of illicit drugs, injection of
illicit drugs, $50 intercourse partners, and history of a
sexually transmitted disease increased as the number of
childhood exposures increased (Table 5). In compar-
ing persons with $4 childhood exposures to those with
none, odds ratios ranged from 2.5 for sexually trans-
mitted diseases to 7.4 for alcoholism and 10.3 for
injected drug use.
Childhood Exposures and
Clustering of Health Risk Factors
We found a strong relationship between the number of
childhood exposures and the number of health risk
factors for leading causes of death in adults (Table 6).
For example, among persons with no childhood expo-
sures, 56% had none of the 10 risk factors whereas only
14% of persons with $4 categories of childhood expo-
sure had no risk factors. By contrast, only 1% of persons
with no childhood exposures had four or more risk
factors, whereas 7% of persons with $4 childhood
exposures had four or more risk factors (Table 6).
Relationship between
Childhood Exposures and Disease Conditions
When persons with 4 or more categories of childhood
exposure were compared to those with none, the
odds ratios for the presence of studied disease con-
ditions ranged from 1.6 for diabetes to 3.9 for
chronic bronchitis or emphysema (Table 7). Simi-
larly, the odds ratios for skeletal fractures, hepatitis
or jaundice, and poor self-rated health were 1.6, 2.3,
and 2.2, respectively (Table 8).
Significance of Dose-Response Relationships
In logistic regression models (which included age,
gender, race, and educational attainment as covariates)
we found a strong, dose-response relationship between
the number of childhood exposures and each of the 10
risk factors for the leading causes of death that we
studied (P , .001). We also found a significant (P ,
.05) dose-response relationship between the number
of childhood exposures and the following disease con-
ditions: ischemic heart disease, cancer, chronic bron-
chitis or emphysema, history of hepatitis or jaundice,
skeletal fractures, and poor self-rated health. There was
no statistically significant dose-response relationship
for a history of stroke or diabetes.T
ab
le
2.
R
el
at
io
n
sh
ip
s
be
tw
ee
n
ca
te
go
ri
es
of
ad
ve
rs
e
ch
ild
h
oo
d
ex
po
su
re
P
er
ce
nt
(%
)
E
xp
os
ed
t
o
A
no
th
er
C
at
eg
or
y
Fi
rs
t
C
at
eg
or
y
of
C
hi
ld
ho
od
E
xp
os
ur
e
Sa
m
pl
e
Si
ze
*
P
sy
c
h
ol
og
ic
a
l
A
bu
se
P
hy
si
ca
l
A
bu
se
Se
xu
al
A
bu
se
Su
bs
ta
nc
e
A
bu
se
M
en
ta
l
Il
ln
es
s
T
re
at
ed
V
io
le
nt
ly
Im
pr
is
on
ed
M
em
be
r
A
ny
O
ne
A
dd
it
io
na
l
C
at
eg
or
y
A
ny
T
w
o
A
dd
it
io
na
l
C
at
eg
or
ie
s
C
h
ild
h
oo
d
A
bu
se
:
Ps
yc
h
ol
og
ic
al
89
8
—
52
*
47
51
50
39
9
93
74
Ph
ys
ic
al
ab
us
e
87
4
54
—
44
45
38
35
9
86
64
Se
xu
al
ab
us
e
17
70
24
22
—
39
31
23
6
65
41
H
ou
se
h
ol
d
dy
sf
un
ct
io
n
:
Su
bs
ta
n
ce
ab
us
e
20
64
22
19
34
—
34
29
8
69
40
M
en
ta
l
ill
n
es
s
15
12
30
22
37
46
—
26
7
74
47
M
ot
h
er
tr
ea
te
d
vi
ol
en
tl
y
10
10
34
31
41
59
38
—
10
86
62
M
em
be
r
im
pr
is
on
ed
27
1
29
29
40
62
42
37
—
86
64
m
ed
ia
n
29
.5
25
.4
40
.5
48
.5
38
32
8.
5
80
54
.5
ra
n
ge
(2
2–
54
)
(1
9–
52
)
(3
4–
47
)
(3
9–
62
)
(3
1–
50
)
(2
3–
39
)
(6
–1
0)
(6
5–
93
)
(4
0–
74
)
*N
um
be
r
ex
po
se
d
to
fi
rs
t
ca
te
go
ry
.
Fo
r
ex
am
pl
e,
am
on
g
pe
rs
on
s
w
h
o
w
er
e
ps
yc
h
ol
og
ic
al
ly
ab
us
ed
,
52
%
w
er
e
al
so
ph
ys
ic
al
ly
ab
us
ed
.
M
or
e
pe
rs
on
s
w
er
e
a
se
co
n
d
ca
te
go
ry
th
an
w
ou
ld
be
ex
pe
ct
ed
by
ch
an
ce
(P
,
.0
01
;
ch
i-s
qu
ar
e)
.
250 American Journal of Preventive Medicine, Volume 14, Number 4
Assessment of the Influence of Exclusions
In the sensitivity analysis where missing information for
a category of childhood exposure was considered as no
exposure, the direction and strength of the associations
between the number of childhood exposures and the
risk factors and disease conditions were nearly identical
(data not shown). Thus, the results we present appear
to be unaffected by our decision to exclude persons for
whom information on any category of childhood expo-
sure was incomplete.
Discussion
We found a strong dose response relationship between
the breadth of exposure to abuse or household dysfunc-
tion during childhood and multiple risk factors for
several of the leading causes of death in adults. Disease
conditions including ischemic heart disease, cancer,
chronic lung disease, skeletal fractures, and liver dis-
ease, as well as poor self-rated health also showed a
graded relationship to the breadth of childhood expo-
sures. The findings suggest that the impact of these
adverse childhood experiences on adult health status is
strong and cumulative.
The clear majority of patients in our study who were
exposed to one category of childhood abuse or house-
hold dysfunction were also exposed to at least one
other. Therefore, researchers trying to understand the
long-term health implications of childhood abuse may
benefit from considering a wide range of related ad-
verse childhood exposures. Certain adult health out-
comes may be more strongly related to unique combi-
nations or the intensity of adverse childhood exposures
than to the total breadth of exposure that we used for
our analysis. However, the analysis we present illustrates
the need for an overview of the net effects of a group of
complex interactions on a wide range of health risk
behaviors and diseases.
Several potential limitations need to be considered
when interpreting the results of this study. The data
about adverse childhood experiences are based on
self-report, retrospective, and can only demonstrate
associations between childhood exposures and health
risk behaviors, health status, and diseases in adulthood.
Second, some persons with health risk behaviors or
diseases may have been either more, or less, likely to
report adverse childhood experiences. Each of these
issues potentially limits inferences about causality. Fur-
thermore, disease conditions could be either over- or
under-reported by patients when they complete the
medical questionnaire. In addition, there may be me-
diators of the relationship between childhood experi-
ences and adult health status other than the risk factors
we examined. For example, adverse childhood experi-
ences may affect attitudes and behaviors toward health
and health care, sensitivity to internal sensations, or
physiologic functioning in brain centers and neuro-
transmitter systems. A more complete understanding
of these issues is likely to lead to more effective ways
to address the long-term health problems associated
with childhood abuse and household dysfunction.
However, our estimates of the prevalence of child-
Table 3. Prevalence of categories of adverse childhood exposures by demographic characteristics
Characteristic
Sample size
(N)
Number of categories (%)a
0 1 2 3 4
Age group (years)
19–34 807 35.4 25.4 17.2 11.0 10.9
35–49 2,063 39.3 25.1 15.6 9.1 10.9
50–64 2,577 46.5 25.2 13.9 7.9 6.6
$65 2,610 60.0 24.5 8.9 4.2 2.4
Genderb
Women 4,197 45.4 24.0 13.4 8.7 8.5
Men 3,859 53.7 25.8 11.6 5.0 3.9
Raceb
White 6,432 49.7 25.3 12.4 6.7 6.0
Black 385 38.8 25.7 16.3 12.3 7.0
Hispanic 431 42.9 24.9 13.7 7.4 11.2
Asian 508 66.0 19.0 9.9 3.4 1.7
Other 300 41.0 23.5 13.9 9.5 12.1
Educationb
No HS diploma 480 56.5 21.5 8.4 6.5 7.2
HS graduate 1,536 51.6 24.5 11.3 7.4 5.2
Any college 2,541 44.1 25.5 14.8 7.8 7.8
College graduate 3,499 51.4 25.1 12.1 6.1 5.3
All participants 8,056 49.5 24.9 12.5 6.9 6.2
aThe number of categories of exposure was simply the sum of each of the seven individual categories that were assessed (see Table 1).
bPrevalence estimates adjusted for age.
Am J Prev Med 1998;14(4) 251
hood exposures are similar to estimates from nationally
representative surveys, indicating that the experiences
of our study participants are comparable to the larger
population of U.S. adults. In our study, 23.5% of
participants reported having grown up with an alcohol
abuser; the 1988 National Health Interview Survey
estimated that 18.1% of adults had lived with an alcohol
abuser during childhood.41 Contact sexual abuse was
reported by 22% of respondents (28% of women and
16% of men) in our study. A national telephone survey
of adults in 1990 using similar criteria for sexual abuse
estimated that 27% of women and 16% of men had
been sexually abused.12
There are several reasons to believe that our esti-
mates of the long-term relationship between adverse
childhood experiences and adult health are conserva-
tive. Longitudinal follow-up of adults whose childhood
abuse was well documented has shown that their retro-
spective reports of childhood abuse are likely to under-
estimate actual occurrence.47,48 Underestimates of
childhood exposures would result in downwardly bi-
ased estimates of the relationships between childhood
exposures and adult health risk behaviors and dis-
eases. Another potential source of underestimation
of the strength of these relationships is the lower
number of childhood exposures reported by older
persons in our study. This may be an artifact caused
by premature mortality in persons with multiple
adverse childhood exposures; the clustering of mul-
tiple risk factors among persons with multiple child-
hood exposures is consistent with this hypothesis.
Thus, the true relationships between adverse child-
hood exposures and adult health risk behaviors,
health status, and diseases may be even stronger than
those we report.
An essential question posed by our observations is,
“Exactly how are adverse childhood experiences linked
to health risk behaviors and adult diseases?” The link-
Table 4. Number of categories of adverse childhood exposure and the adjusted odds of risk factors including current
smoking, severe obesity, physical inactivity, depressed mood, and suicide attempt
Health problem
Number
of
categories
Sample
size
(N)a
Prevalence
(%)b
Adjusted
odds
ratioc
95%
confidence
interval
Current smokerd 0 3,836 6.8 1.0 Referent
1 2,005 7.9 1.1 ( 0.9–1.4)
2 1,046 10.3 1.5 ( 1.1–1.8)
3 587 13.9 2.0 ( 1.5–2.6)
4 or more 544 16.5 2.2 ( 1.7–2.9)
Total 8,018 8.6 — —
Severe obesityd
(BMI $ 35)
0 3,850 5.4 1.0 Referent
1 2,004 7.0 1.1 ( 0.9–1.4)
2 1,041 9.5 1.4 ( 1.1–1.9)
3 590 10.3 1.4 ( 1.0–1.9)
4 or more 543 12.0 1.6 ( 1.2–2.1)
Total 8,028 7.1 — —
No leisure-time
physical activity
0 3,634 18.4 1.0 Referent
1 1,917 22.8 1.2 ( 1.1–1.4)
2 1,006 22.0 1.2 ( 1.0–1.4)
3 559 26.6 1.4 ( 1.1–1.7)
4 or more 523 26.6 1.3 ( 1.1–1.6)
Total 7,639 21.0 — —
Two or more weeks of
depressed mood in
the past year
0 3,799 14.2 1.0 Referent
1 1,984 21.4 1.5 ( 1.3–1.7)
2 1,036 31.5 2.4 ( 2.0–2.8)
3 584 36.2 2.6 ( 2.1–3.2)
4 or more 542 50.7 4.6 ( 3.8–5.6)
Total 7,945 22.0 — —
Ever attempted suicide 0 3,852 1.2 1.0 Referent
1 1,997 2.4 1.8 ( 1.2–2.6)
2 1,048 4.3 3.0 ( 2.0–4.6)
3 587 9.5 6.6 ( 4.5–9.8)
4 or more 544 18.3 12.2 (8.5–17.5)
Total 8,028 3.5 — —
aSample sizes will vary due to incomplete or missing information about health problems.
bPrevalence estimates are adjusted for age.
cOdds ratios adjusted for age, gender, race, and educational attainment.
dIndicates information recorded in the patient’s chart before the study questionnaire was mailed.
252 American Journal of Preventive Medicine, Volume 14, Number 4
ing mechanisms appear to center on behaviors such as
smoking, alcohol or drug abuse, overeating, or sexual
behaviors that may be consciously or unconsciously
used because they have immediate pharmacological or
psychological benefit as coping devices in the face of
the stress of abuse, domestic violence, or other forms of
family and household dysfunction. High levels of expo-
sure to adverse childhood experiences would expect-
edly produce anxiety, anger, and depression in chil-
dren. To the degree that behaviors such as smoking,
alcohol, or drug use are found to be effective as coping
devices, they would tend to be used chronically. For
Table 5. Number of categories of adverse childhood exposure and the prevalence and risk (adjusted odds ratio) of health
risk factors including alcohol or drug abuse, high lifetime number of sexual partners, or history of sexually
transmitted disease
Health problem
Number
of
categories
Sample
size
(N)a
Prevalence
(%)b
Adjusted
odds
ratioc
95%
confidence
interval
Considers self an
alcoholic
0 3,841 2.9 1.0 Referent
1 1,993 5.7 2.0 (1.6–2.7)
2 1,042 10.3 4.0 (3.0–5.3)
3 586 11.3 4.9 (3.5–6.8)
4 or more 540 16.1 7.4 (5.4–10.2)
Total 8,002 5.9 — —
Ever used illicit drugs 0 3,856 6.4 1.0 Referent
1 1,998 11.4 1.7 (1.4–2.0)
2 1,045 19.2 2.9 (2.4–3.6)
3 589 21.5 3.6 (2.8–4.6)
4 or more 541 28.4 4.7 (3.7–6.0)
Total 8,029 11.6 — —
Ever injected drugs 0 3,855 0.3 1.0 Referent
1 1,996 0.5 1.3 (0.6–3.1)
2 1,044 1.4 3.8 (1.8–8.2)
3 587 2.3 7.1 (3.3–15.5)
4 or more 540 3.4 10.3 (4.9–21.4)
Total 8,022 0.8 — —
Had 50 or more
intercourse partners
0 3,400 3.0 1.0 Referent
1 1,812 5.1 1.7 (1.3–2.3)
2 926 6.1 2.3 (1.6–3.2)
3 526 6.3 3.1 (2.0–4.7)
4 or more 474 6.8 3.2 (2.1–5.1)
Total 7,138 4.4 — —
Ever had a sexually
transmitted diseased
0 3,848 5.6 1.0 Referent
1 2,001 8.6 1.4 (1.1–1.7)
2 1,044 10.4 1.5 (1.2–1.9)
3 588 13.1 1.9 (1.4–2.5)
4 or more 542 16.7 2.5 (1.9–3.2)
Total 8023 8.2 — —
aSample sizes will vary due to incomplete or missing information about health problems.
bPrevalence estimates are adjusted for age.
cOdds ratios adjusted for age, gender, race, and educational attainment.
dIndicates information recorded in the patient’s chart before the study questionnaire was mailed.
Table 6. Relationship between number of categories of childhood exposure and number of risk factors for the leading
causes of deatha
Number of categories
Sample
size
% with number of risk factors
0 1 2 3 4
0 3,861 56 29 10 4 1
1 2,009 42 33 16 6 2
2 1,051 31 33 20 10 4
3 590 24 33 20 13 7
$4 545 14 26 28 17 7
Total 8,056 44 31 15 7 3
aRisk factors include: smoking, severe obesity, physical inactivity, depressed mood, suicide attempt, alcoholism, any drug use, injected drug use,
$50 lifetime sexual partners, and history of a sexually transmitted disease.
Am J Prev Med 1998;14(4) 253
example, nicotine is recognized as having beneficial
psychoactive effects in terms of regulating affect49 and
persons who are depressed are more likely to
smoke.50,51 Thus, persons exposed to adverse child-
hood experiences may benefit from using drugs such as
nicotine to regulate their mood.49,52
Consideration of the positive neuroregulatory effects
of health-risk behaviors such as smoking may provide
biobehavioral explanations53 for the link between ad-
verse childhood experiences and health risk behaviors
and diseases in adults. In fact, we found that exposure
to higher numbers of categories of adverse childhood
experiences increased the likelihood of smoking by the
age of 14, chronic smoking as adults, and the presence
of smoking-related diseases. Thus, smoking, which is
medically and socially viewed as a “problem” may, from
the perspective of the user, represent an effective
immediate solution that leads to chronic use. Decades
later, when this “solution” manifests as emphysema,
cardiovascular disease, or malignancy, time and the
tendency to ignore psychological issues in the manage-
ment of organic disease make improbable any full
understanding of the original causes of adult disease
(Figure 2). Thus, incomplete understanding of the
possible benefits of health risk behaviors leads them to
be viewed as irrational and having solely negative
consequences.
Because adverse childhood experiences are common
and they have strong long-term associations with adult
health risk behaviors, health status, and diseases, in-
creased attention to primary, secondary, and tertiary
prevention strategies is needed. These strategies in-
clude prevention of the occurrence of adverse child-
hood experiences, preventing the adoption of health
risk behaviors as responses to adverse experiences
during childhood and adolescence, and, finally, help-
ing change the health risk behaviors and ameliorating
the disease burden among adults whose health prob-
lems may represent a long-term consequence of ad-
verse childhood experiences.
Table 7. Number of categories of adverse childhood exposure and the prevalence and risk (adjusted odds ratio) of heart
attack, cancer, stroke, COPD, and diabetes
Disease conditiond
Number
of
categories
Sample
size
(N)a
Prevalence
(%)b
Adjusted
odds
ratioc
95%
confidence
interval
Ischemic heart disease 0 3,859 3.7 1.0 Referent
1 2,009 3.5 0.9 (0.7–1.3)
2 1,050 3.4 0.9 (0.6–1.4)
3 590 4.6 1.4 (0.8–2.4)
4 or more 545 5.6 2.2 (1.3–3.7)
Total 8,022 3.8 — —
Any cancer 0 3,842 1.9 1.0 Referent
1 1,995 1.9 1.2 (1.0–1.5)
2 1,043 1.9 1.2 (1.0–1.5)
3 588 1.9 1.0 (0.7–1.5)
4 or more 543 1.9 1.9 (1.3–2.7)
Total 8,011 1.9 — —
Stroke 0 3,832 2.6 1.0 Referent
1 1,993 2.4 0.9 (0.7–1.3)
2 1,042 2.0 0.7 (0.4–1.3)
3 588 2.9 1.3 (0.7–2.4)
4 or more 543 4.1 2.4 (1.3–4.3)
Total 7,998 2.6 — —
Chronic bronchitis or
emphysema
0 3,758 2.8 1.0 Referent
1 1,939 4.4 1.6 (1.2–2.1)
2 1,009 4.4 1.6 (1.1–2.3)
3 565 5.7 2.2 (1.4–3.3)
4 or more 512 8.7 3.9 (2.6–5.8)
Total 7,783 4.0 — —
Diabetes 0 3,850 4.3 1.0 Referent
1 2,002 4.1 1.0 (0.7–1.3)
2 1,046 3.9 0.9 (0.6–1.3)
3 587 5.0 1.2 (0.8–1.9)
4 or more 542 5.8 1.6 (1.0–2.5)
Total 8,027 4.3 — —
aSample sizes will vary due to incomplete or missing information about health problems.
bPrevalence estimates are adjusted for age.
cOdds ratios adjusted for age, gender, race, and educational attainment.
dIndicates information recorded in the patient’s chart before the study questionnaire was mailed.
254 American Journal of Preventive Medicine, Volume 14, Number 4
Primary prevention of adverse childhood experi-
ences has proven difficult54,55 and will ultimately re-
quire societal changes that improve the quality of family
and household environments during childhood. Recent
research on the long-term benefit of early home visitation
on reducing the prevalence of adverse childhood experi-
ences is promising.56 In fact, preliminary data from the
ACE Study provided the impetus for the Kaiser Health
Plan to provide funding to participate at 4 locations
(including San Diego County, California) in the Com-
monwealth Fund’s “Healthy Steps” program. This pro-
gram extends the traditional practice of pediatrics by
adding one or more specialists in the developmental and
psychosocial dimensions of both childhood and parent-
hood. Through a series of office visits, home visits, and a
telephone advice line for parents, these specialists develop
close relationships between children and their families
from birth to 3 years of age. This approach is consistent
with the recommendation of the U.S. Advisory Board on
Child Abuse and Neglect that a universal home visitation
program for new parents be developed57,58 and provides
an example of a family-based primary prevention effort
that is being explored in a managed care setting. If these
types of approaches can be replicated and implemented
on a large scale, the long-term benefits may include,
somewhat unexpectedly, substantial improvements in
overall adult health.
Secondary prevention of the effects of adverse child-
hood experiences will first require increased recogni-
tion of their occurrence and second, an effective un-
derstanding of the behavioral coping devices that
commonly are adopted to reduce the emotional impact
of these experiences. The improbability of giving up an
immediate “solution” in return for a nebulous long-
term health benefit has thwarted many well-intended
preventive efforts. Although articles in the general
medical literature are alerting the medical community
to the fact that childhood abuse is common,59 adoles-
cent health care is often inadequate in terms of psycho-
social assessment and anticipatory guidance.60 Clearly,
comprehensive strategies are needed to identify and
intervene with children and families who are at risk for
these adverse experiences and their related out-
comes.61 Such strategies should include increased com-
munication between and among those involved in
family practice, internal medicine, nursing, social work,
pediatrics, emergency medicine, and preventive medi-
cine and public health. Improved understanding is also
needed of the effects of childhood exposure to domes-
tic violence.19,62 Additionally, increased physician train-
ing63 is needed to recognize and coordinate the man-
agement of all persons affected by child abuse,
domestic violence, and other forms of family adversity
such as alcohol abuse or mental illness.
In the meantime, tertiary care of adults whose health
problems are related to experiences such as childhood
abuse5 will continue to be a difficult challenge. The
relationship between childhood experiences and adult
health status is likely to be overlooked in medical
practice because the time delay between exposure
Table 8. Number of categories of adverse childhood exposure and the prevalence and risk (adjusted odds ratio) of skeletal
fracture, hepatitis or jaundice, and poor self-rated health
Disease condition
Number
of
categories
Sample
size
(N)a
Prevalence
(%)b
Adjusted
odds
ratioc
95%
confidence
interval
Ever had a skeletal
fracture
0 3,843 3.6 1.0 Referent
1 1,998 4.0 1.1 (1.0–1.2)
2 1,048 4.5 1.4 (1.2–1.6)
3 587 4.0 1.2 (1.0–1.4)
4 or more 544 4.8 1.6 (1.3–2.0)
Total 8,020 3.9 — —
Ever had hepatitis or
jaundice
0 3,846 5.3 1.0 Referent
1 2,006 5.5 1.1 (0.9–1.4)
2 1,045 7.7 1.8 (1.4–2.3)
3 590 10.2 1.6 (1.2–2.3)
4 or more 543 10.7 2.4 (1.8–3.3)
Total 8,030 6.5 — —
Fair or poor self-rated
health
0 3,762 16.3 1.0 Referent
1 1,957 17.8 1.2 (1.0–1.4)
2 1,029 19.9 1.4 (1.2–1.7)
3 584 20.3 1.4 (1.1–1.7)
4 or more 527 28.7 2.2 (1.8–2.7)
Total 7,859 18.2 — —
aSample sizes will vary due to incomplete or missing information about health problems.
bPrevalence estimates are adjusted for age and gender.
cOdds ratios adjusted for age, gender, race, and educational attainment.
dIndicates information recorded in the patient’s chart before the study questionnaire was mailed.
Am J Prev Med 1998;14(4) 255
during childhood and recognition of health problems
in adult medical practice is lengthy. Moreover, these
childhood exposures include emotionally sensitive top-
ics such as family alcoholism29,30 and sexual abuse.64
Many physicians may fear that discussions of sexual
violence and other sensitive issues are too personal
even for the doctor-patient relationship.65 For example,
the American Medical Association recommends screen-
ing of women for exposure to violence at every en-
trance to the health system;66 however, such screening
appears to be rare.67 By contrast, women who are asked
about exposure to sexual violence say they consider
such questions to be welcome and germane to routine
medical care,68 which suggests that physicians’ fears
about patient reactions are largely unfounded.
Clearly, further research and training are needed to
help medical and public health practitioners under-
stand how social, emotional, and medical problems are
linked throughout the lifespan (Figure 2). Such re-
search and training would provide physicians with the
confidence and skills to inquire and respond to patients
who acknowledge these types of childhood exposures.
Increased awareness of the frequency and long-term
consequences of adverse childhood experiences may
also lead to improvements in health promotion and
disease prevention programs. The magnitude of the
difficulty of introducing the requisite changes into
medical and public health research, education, and
practice can be offset only by the magnitude of the
implications that these changes have for improving the
health of the nation.
We thank Naomi Howard for her dedication to the ACE Study.
This research is supported by the Centers for Disease Control
and Prevention via cooperative agreement TS-44-10/12 with the
Association of Teachers of Preventive Medicine.
References
1. Springs F, Friedrich WN. Health risk behaviors and
medical sequelae of childhood sexual abuse. Mayo Clin
Proc 1992;67:527–32.
2. Felitti VJ. Long-term medical consequences of incest,
rape, and molestation. South Med J 1991;84:328–31.
3. Felitti VJ. Childhood sexual abuse, depression and family
dysfunction in adult obese patients: a case control study.
South Med J 1993;86:732–6.
4. Gould DA, Stevens NG, Ward NG, Carlin AS, Sowell HE,
Gustafson B. Self-reported childhood abuse in an adult
population in a primary care setting. Arch Fam Med
1994;3:252–6.
5. McCauley J, Kern DE, Kolodner K, Schroeder AF, et al.
Clinical characteristics of women with a history of child-
hood abuse. JAMA 1997;277:1362–8.
Figure 2. Potential influences throughout the lifespan of adverse childhood experiences.
256 American Journal of Preventive Medicine, Volume 14, Number 4
6. Mortality patterns: United States, 1993. Morb Mortal
Wkly Rep 1996;45:161–4.
7. McGinnis JM, Foege WH. Actual causes of death in the
United States. JAMA 1993;270:2207–12.
8. Landis J. Experiences of 500 children with adult sexual
deviation. Psychiatr Q 1956;30(Suppl):91–109.
9. Straus MA, Gelles RJ. Societal change and change in
family violence from 1975 to 1985 as revealed by two
national surveys. J Marriage Family 1986;48:465–79.
10. Wyatt GE, Peters SD. Methodological considerations in
research on the prevalence of child sexual abuse. Child
Abuse Negl 1986;10:241–51.
11. Berger AM, Knutson JF, Mehm JG, Perkins KA. The
self-report of punitive childhood experiences of young
adults and adolescents. Child Abuse Negl 1988;12:251–
62.
12. Finkelhor D, Hotaling G, Lewis IA, Smith C. Sexual abuse
in a national survey of adult men and women: prevalence,
characteristics, and risk factors. Child Abuse Negl 1990;
14:19–28.
13. Egelend B, Sroufe LA, Erickson M. The developmental
consequence of different patterns of maltreatment. Child
Abuse Negl 1983;7:459–69.
14. Finkelhor D, Browne A. The traumatic impact of child
sexual abuse Am J Orthopsychiatry. 1985;55:530–41.
15. Beitchman JH, Zucker KJ, Hood JE, DaCosta GA, Akman
D, Cassavia E. A review of the long-term effects of sexual
abuse. Child Abuse Negl 1992;16:101–18.
16. Hibbard RA, Ingersoll GM, Orr DP. Behavioral risk,
emotional risk, and child abuse among adolescents in a
nonclinical setting. Pediatrics 1990;86:896–901.
17. Nagy S, Adcock AG, Nagy MC. A comparison of risky
health behaviors of sexually active, sexually abused, and
abstaining adolescents. Pediatrics 1994;93:570–5.
18. Cunningham RM, Stiffman AR, Dore P. The association
of physical and sexual abuse with HIV risk behaviors in
adolescence and young adulthood: implications for pub-
lic health. Child Abuse Negl 1994;18:233–45.
19. Council on Scientific Affairs. Adolescents as victims of
family violence. JAMA 1993;270:1850–6.
20. Nelson DE, Higginson GK, Grant-Worley JA. Physical
abuse among high school students. Prevalence and cor-
relation with other health behaviors. Arch Pediatr Ado-
lesc Med 1995;149:1254–8.
21. Mullen PE, Roman-Clarkson SE, Walton VA, Herbison
GP. Impact of sexual and physical abuse on women’s
mental health. Lancet 1988;1:841–5.
22. Drossman DA, Leserman J, Nachman G, Li Z, et al. Sexual
and physical abuse in women with functional or organic
gastrointestinal disorders. Ann Intern Med 1990;113:
828–33.
23. Harrop-Griffiths J, Katon W, Walker E, Holm L, Russo J,
Hickok L. The association between chronic pelvic pain,
psychiatric diagnoses, and childhood sexual abuse. Ob-
stet Gynecol 1988;71:589–94.
24. Briere J, Runtz M. Multivariate correlates of childhood
psychological and physical maltreatment among univer-
sity women. Child Abuse Negl 1988;12:331–41.
25. Briere J, Runtz M. Differential adult symptomatology
associated with three types of child abuse histories. Child
Abuse Negl 1990;14:357–64.
26. Claussen AH, Crittenden PM. Physical and psychological
maltreatment: relations among types of maltreatment.
Child Abuse Negl 1991;15:5–18.
27. Moeller TP, Bachman GA, Moeller JR. The combined
effects of physical, sexual, and emotional abuse during
childhood: long-term health consequences for women.
Child Abuse Negl 1993;17:623–40.
28. Bryant SL, Range LM. Suicidality in college women who
were sexually and physically punished by parents. Vio-
lence Vict 1995;10:195–201.
29. Zeitlen H. Children with alcohol misusing parents. Br
Med Bull 1994;50:139–51.
30. Dore MM, Doris JM, Wright P. Identifying substance
abuse in maltreating families: a child welfare challenge.
Child Abuse Negl 1995;19:531–43.
31. Ethier LS, Lacharite C, Couture G. Childhood adversity,
parental stress, and depression of negligent mothers.
Child Abuse Negl 1995;19:619–32.
32. Spaccarelli S, Coatsworth JD, Bowden BS. Exposure to
family violence among incarcerated boys; its association
with violent offending and potential mediating variables.
Violence Vict 1995;10:163–82.
33. McCloskey LA, Figueredo AJ, Koss MP. The effects of
systemic family violence on children’s mental health.
Child Dev 1995;66:1239–61.
34. Brent DA, Perper JA, Moritz G, Schweers J, Balach L,
Roth C. Familial risk factors for adolescent suicide: a
case-control study. Acta Psychiatr Scand 1994;89:52–8.
35. Shaw DS, Vondra JI, Hommerding KD, Keenan K, Dunn
M. Chronic family adversity and early child behavior
problems: a longitudinal study of low income families.
J Child Psychol Psychiatry 1994;35:1109–22.
36. U.S. Department of Health and Human Services. Physical
activity and health: A report of the Surgeon General.
Atlanta, Georgia. U.S. Department of Health and Human
Services, Centers for Disease Control and Prevention,
National Center for Chronic Disease Prevention and
Health Promotion; 1996.
37. Rivara FP, Mueller BA, Somes G, Mendoza CT, Rushforth
NB, Kellerman AL. Alcohol and illicit drug abuse and the
risk of violent death in the home. JAMA 1997;278:569–
75.
38. Dillman DA. Mail and telephone surveys: the total design
method. New York: John Wiley & Sons; 1978.
39. Straus M, Gelles RJ. Physical violence in American fami-
lies: risk factors and adaptations to violence in 8,145
families. New Brunswick: Transaction Press; 1990.
40. Wyatt GE. The sexual abuse of Afro-American and White-
American women in childhood. Child Abuse Negl 1985;
9:507–19.
41. National Center for Health Statistics. Exposure to alco-
holism in the family: United States, 1988. Advance Data,
No. 205. U.S. Department of Health and Human Services,
Washington, DC; September 30, 1991.
42. Siegel PZ, Frazier EL, Mariolis P, et al. Behaviorial risk
factor surveillance, 1991; Monitoring progress toward the
Nation’s Year 2000 Health Objectives. Morb Mortal Wkly
Rep 1992;42(SS-4).1–15.
43. Crespo CJ, Keteyian SJ, Heath GW, Sempos CT. Leisure-
time physical activity among US adults: Results from the
Am J Prev Med 1998;14(4) 257
Third National Health and Nutrition Examination Sur-
vey. Arch Intern Med 1996;156:93–8.
44. Robins LN, Helzer JE, Groughan J, Ratliff K. National
Institute of Mental Health Diagnostic Interview Schedule:
its history, characteristics, and validity. Arch Gen Psychi-
atry 1981;38:381–9.
45. Idler E, Angel RJ. Self-rated health and mortality in the
NHANES I Epidemiologic Follow-up Study. Am J Pub
Health 1990;80:446–52.
46. SAS Procedures Guide. SAS Institute Inc. Version 6, 3rd
edition, Cary, NC: SAS Institute; 1990.
47. Femina DD, Yeager CA, Lewis DO. Child abuse: adoles-
cent records vs. adult recall. Child Abuse Negl 1990;14:
227–31.
48. Williams LM. Recovered memories of abuse in women
with documented child sexual victimization histories.
J Traumatic Stress 1995;8:649–73.
49. Carmody TP. Affect regulation, nicotine addiction, and
smoking cessation. J Psychoactive Drugs 1989;21:331–42.
50. Anda RF, Williamson DF, Escobedo LG, Mast EE, Giovino
GA, Remington PL. Depression and the dynamics of
smoking. A national perspective. JAMA 1990;264:1541–5.
51. Glassman AH, Helzer JE, Covey LS, Cottler LB, Stetner F,
Tipp JE, Johnson J. Smoking, smoking cessation, and
major depression. JAMA 1990;264:1546–9.
52. Hughes JR. Clonidine, depression, and smoking cessa-
tion. JAMA 1988;259:2901–2.
53. Pomerlau OF, Pomerlau CS. Neuroregulators and the
reinforcement of smoking: towards a biobehavioral expla-
nation. Neurosci Biobehav Rev 1984;8:503–13.
54. Hardy JB, Street R. Family support and parenting educa-
tion in the home: an effective extension of clinic-based
preventive health care services for poor children. J Pedi-
atr 1989;115:927–31.
55. Olds DL, Henderson CR, Chamberlin R, Tatelbaum R.
Preventing child abuse and neglect: a randomized trial of
nurse home visitation. Pediatrics 1986;78:65–78.
56. Olds DL, Eckenrode J, Henderson CR, Kitzman H, et al.
Long-term effects of home visitation on maternal life
course and child abuse and neglect: Fifteen-year fol-
low-up of a randomized trial. JAMA 1997;278:637–43.
57. U.S. Advisory Board on Child Abuse and Neglect. Child
abuse and neglect: critical first steps in response to a
national emergency. Washington, DC: U.S. Government
Printing Office; August 1990; publication no. 017-092-
00104-5.
58. U.S. Advisory Board on Child Abuse and Neglect. Creat-
ing caring communities: blueprint for an effective federal
policy on child abuse and neglect. Washington, DC: U.S.
Government Printing Office; September 1991.
59. MacMillan HL, Fleming JE, Trocme N, Boyle MH, et al.
Prevalence of child physical and sexual abuse in the
community. Results from the Ontario Health Supple-
ment. JAMA 1997;278:131–5.
60. Rixey S. Family violence and the adolescent. Maryland
Med J 1994;43:351–3.
61. Chamberlin RW. Preventing low birth weight, child
abuse, and school failure: the need for comprehensive,
community-wide approaches. Pediatr Rev 1992;13:64–71.
62. Kashani JH, Daniel AE, Dandoy AC, Holcomb WR. Family
violence: impact on children. J Am Acad Child Adolesc
Psychiatry 1992;31:181–9.
63. Dubowitz H. Child abuse programs and pediatric resi-
dency training. Pediatrics 1988;82:477–80.
64. Tabachnick J, Henry F, Denny L. Perceptions of child
sexual abuse as a public health problem. Vermont, Sep-
tember 1995. Morb Mortal Wkly Rep 1997;46:801–3.
65. Sugg NK, Inui T. Primary care physicians’ response to
domestic violence. Opening Pandora’s box. JAMA 1992;
267:3157–60.
66. Council on Scientific Affairs. American Medical Associa-
tion Diagnostic and Treatment Guidelines on Domestic
Violence. Arch Fam Med 1992;1:38–47.
67. Hamberger LK, Saunders DG, Hovey M. Prevalence of
domestic violence in community practice and rate of
physician inquiry. Fam Med 1992;24:283–7.
68. Friedman LS, Samet JH, Roberts MS, Hans P. Inquiry
about victimization experiences. A survey of patient pref-
erences and physician practices. Arch Int Med 1992;152:
1186–90.
258 American Journal of Preventive Medicine, Volume 14, Number 4
Posttraumatic Stress in Youth
Experiencing Illnesses and Injuries:
An Exploratory Meta-Analysi
s
Shoshana Y
.
Kahana, Norah C. Feeny,
Eric A. Youngstrom, and Dennis Drotar
tumors, and non-Hodgkin’s lymphoma (Woodruff
et al., 2004). More than 11,000 children are diagnosed
with new cancers each year in the United States,
and there are currently an estimated 250,000 pedi-
atric cancer survivors. Furthermore, from 1988 to
present, between 7% and 9.5% of all organ trans-
plants conducted annually in the United States wer
e
performed with pediatric populations (Transpla
n
t
DataSource, 2001).
Despite the high prevalence rates of both injuri
es
and illnesses among youth, little is understood about
the psychological sequelae that are related to both
conditions. Recently, researchers have posited that
posttraumatic stress disorder (PTSD) may be a use-
ful construct to conceptualize some of the psycho-
logical distress and loss of functioning that is
present in a significant subset of individuals following
illnesses or injuries. Consistent with the Diagnostic
and Statistical Manual, Fourth Edition (DSM-IV;
American Psychiatric Association, 2001), a serious
or life-threatening illness or injury usually contai
ns
several traumatic elements, including a “real or
The incidence of injuries and illnesses (e.g.,
cancer and organ transplantation) among
youth is high. Injuries have been the largest
cause of morbidity and mortality among children in
the United States for many years (Guyer et al.,
1999). Close to 20 million children suffer uninten-
tional injuries annually, and motor vehicle acciden
ts
(MVAs) and fires are among the leading causes of
unintentional injuries in youth through the ages of 19
(Public Health Policy Advisory Board, 1999). With
respect to medical illnesses, recent epidemiological
data suggest that the incidence of certain childhood
cancers has steadily increased since 1990, including
acute lymphoblastic leukemia, central nervous system
To date there is no quantitative review of predictors of
posttraumatic stress disorder (PTSD) symptoms in youth
experiencing illnesses or injuries. This article presents
a
meta-analysis of variables associated with the develop-
ment of PTSD among those youth. Twenty-six studies
were included: 18 involving children experiencing
injuries and 8 with pediatric illnesses. Among injured
youth, socioeconomic status and social impairment were
small to moderate correlates of PTSD, whereas depres-
sive and anxious symptoms, dissociation, acute stress dis-
order, and the appraisal of trauma severity and life threat
displayed large effect sizes with PTSD severity. Among ill
youth, social support and the appraisal of illness severity
and life threat emerged as small to moderate predictors
of posttraumatic symptoms. The current findings are
exploratory in nature, as a primary limitation of the cur-
rent study includes the limited number of independe
nt
studies that have evaluated these predictors. Current
findings further our understanding of PTSD through
exploration of possible indicators of at-risk youth who
have experienced illness and injury.
Keywords: PTSD; youth; injuries; medical illness;
predictors
From Case Western Reserve University, Cleveland, Ohio (SYK,
NCF, EAY, DD); Rainbow Babies & Children’s Hospital,
Division of Behavioral Pediatrics & Psychology, Cleveland,
Ohio (DD).
Address correspondence to: Shoshana Y. Kahana, Case Western
Reserve University, Department of Psychology, 11220 Bellflower
Rd., Cleveland, OH 44106; e-mail: shoshana.kahana@case.edu.
Traumatology
Volume 12 Number 2
June 2006 148-161
© 2006 Sage Publications
10.1177/1534765606294562
http://tmt.sagepub.com
hosted at
http://online.sagepub.com
148
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e
A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
t
io
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r
t
he
p
er
so
na
l
u
se
o
f t
he
in
di
vi
du
al
u
se
r
a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
threatened death or a threat to the physical integrity
of self and others.” In addition to the injury or illness
itself, life-threatening conditions and their treat-
ments typically involve a series of potential traumas,
such as the life threat inherent in some of the “high-
tech” procedures involved in treatment, the repeated
intrusions of the procedures, and the experienced
pain (Stuber, Nader, Houskamp, & Pynoos, 1996).
Moreover, both injuries and illnesses frequently
evoke intense feelings of fear, uncertainty, and help-
lessness about future outcomes and functioning.
Although many individuals exposed to trauma do
not develop PTSD, a substantial minority do. Indeed,
researchers have reported significant rates of PTS
D
as well as posttraumatic stress symptoms (PTSSs) in
children experiencing various illness-related trau-
mas, such as cancer (Kazak et al., 2004) and liver
transplants (Shemesh et al., 2000), or injuries, such
as MVAs (Bryant, Mayou, Wiggs, Ehlers, & Stores,
2004), traumatic brain injury (TBI; Levi, Drotar,
Yeates, & Taylor, 1999), and burns (Saxe, Stoddard,
& Sheridan, 1998). Pelcovitz et al. (1998) reported
that 35% of adolescents with cancer were diagnosed
with PTSD, whereas 34.5% of youth involved in
MVAs met criteria for PTSD (Stallard, Velleman, &
Baldwin, 1998).
It is important to note that the clinical signifi-
cance of subthreshold or partial PTSD (i.e., exhibit-
ing some but not all PTSD symptoms) also has
begun to be addressed in the literature. For example,
research has noted that among youth who had expe-
rienced acute physical injury, many suffered with
subthreshold levels of PTSD, which was associated
with internalizing symptoms, such as depression and
anxiety (Aaron, Zaglul, & Emery, 1999). In addition,
Marshall et al. (2001) reported that the presence of
subthreshold PTSSs significantly increased the ris
k
for suicidal ideation, comorbidity, and overall greater
impairment. Thus, the clinical relevance of sub-
threshold PTSD is an important construct to con-
sider among individuals exposed to various traumas.
Because of the relatively high incidence of PTSD
among youth experiencing injuries and illnesses, it is
crucial to identify children at high risk for the devel-
opment of PTSD after these traumas. If left
untreated, PTSD may affect many domains, including
social functioning (Berman, Kurtines, Silverman, &
Serafini, 1996; Stallard, Velleman, Langsford, &
Baldwin, 2001), physical recovery from illnesses or
injuries (Stallard et al., 2001), and academic function-
ing (Reinherz, Giaconia, Lefkowitz, Pakiz, & Frost,
1993). The identification of potential predictors of
PTSD among youth experiencing illness- or injury-
related traumas may also facilitate efforts to effec-
tively address emergent psychiatric symptoms.
Currently, we know a good deal about predictors
for the development of PTSD in adults, but we know
substantially less about predictors for youth exposed to
illness- or injury-related traumas. Some early research
(La Greca, Silverman, Vernberg, & Prinstein, 1996;
Vernberg, Silverman, La Greca, & Prinstein, 1996
)
has been influential in introducing a classification
model for the predictors of PTSD in traumatized
youth, which includes the child’s characteristics
(i.e., gender, age), characteristics of the stressor
(i.e., amount of time exposed to stressor, life threat
involved), characteristics of the environment (i.e.,
access to social support, family psychiatric history),
and cognitive processing (i.e., coping) of the trau-
matic event. Although some qualitative work has
described various predictors of PTSD among youth
with illnesses and injuries (e.g., Stuber, Shemesh, &
Saxe, 2003), no research to date has derived quanti-
tative estimates for these predictors. Similarly,
meaningful descriptive data across various studies,
such as the aggregation of the number and types of
PTSSs, has not been reported for youth experienc-
ing illnesses and injuries.
Finally, although illnesses and injuries clearly share
some significant commonalities (e.g., can involve life
threat, affect family functioning) and have often been
treated as the same construct, the literature has not
addressed differences that may exist between these
two traumas. In fact, it is likely that individuals who
experience illness as compared to injury are exposed to
different traumatic events. Furthermore, it is the expo-
sure to these events that defines the traumatic experi-
ence of illnesses and injuries. For example, among
youth exposed to more acute traumas, typically involv-
ing injuries, the aftermath may include additional
traumatic stressors, such as aspects of the emergency
medical setting and/or the deaths of other individua
ls
involved in the traumatic event. Conversely, for youth
with illnesses, there are the uncertainties of second
malignancies (cancer) or organ rejection (transplantees)
as well as the painfulness of certain treatments (e.g.,
chemotherapy). In short, combining injuries and ill-
nesses into the same construct may obscure some of
the subtle yet important nuances that are inherent
to each.
The aim of this study is twofold. The authors
synthesize descriptive information along with PTSD
Meta-Analysis of PTSD in Youth / Kahana et al. 149
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
prevalence rates and symptoms among youth who
were exposed to illnesses or injuries. In addition, the
authors conduct a meta-analysis to identify consis-
tent variables associated with the development of
PTSD among youth experiencing illnesses or injuries.
To address what we perceive to be the gap in the lit-
erature, the phenomenology of PTSD among youth
experiencing illnesses and injuries will be investi-
gated separately and, when possible, will be directly
compared to each other.
Method
Literature Search
The authors conducted comprehensive literature
searches using various medical and psychological
bibliographic databases, such as PsycINFO and
MEDLINE, to find articles that reported the preva-
lence rates as well as data on the predictors for PTSD
(as defined by DSM-III, DSM-III-R, and DSM-IV) in
youth between the ages of 6 and 19. Search terms
such as trauma, posttraumatic stress disorder, post-
traumatic stress symptoms, PTSD, PTSS, stress, child,
adolescent, youth, medical, traffic accidents, motor
vehicle accidents, injury, cancer, burns, transplant,
predictor(s), prevalence, and epidemiology were uti-
lized in the search for articles. In addition, the refer-
ence sections of all of these identified articles were
examined to glean additional articles. Finally, several
researchers noted for their work in pediatric trau-
matic stress were contacted to access data that were
included in manuscripts in press.
Criteria for Inclusion and Exclusion
The studies included for the current review used
quantitative methods to examine the prevalence
rates of PTSD, as well as the potential predictors of
PTSD, in youth experiencing illnesses or injuries. In
accordance with the current classifications of
numerous government, medical, and public policy
organizations (e.g., National Center for
Injury
Prevention and Control), ill youth were defined as
those experiencing relatively acute medical illnesses,
such as cancer or liver transplants, whereas injury-
related traumas were defined as those involving
MVAs, burns, or brain injury. Injuries that were the
result of violent physical interpersonal traumas,
such as assaults and muggings, were not included in
this study. A predictor was defined as any variable
that contributed to variability in PTSD diagnostic
status or symptom severity. The authors opted for
the use of terms such as predictors and correlates, as
compared to risk factors, owing to the developing
and emergent state of the current medical trauma
literature.
Several additional inclusionary and exclusionary
criteria were applied to the current study. Only
English-language articles published in peer-reviewed
journals (from 1980 onward) were included, as the
diagnosis of PTSD as applied to children only
emerged in the 1980s. The use of only published
studies has received some recent support from
research suggesting that a publication bias does not
necessarily exist among certain medical and social
science journals (Olson et al., 2002). Finally, this
review included studies that assessed for a DSM
diagnosis of PTSD or PTSD symptom severity.
Articles were excluded on the following grounds:
(a) they did not utilize a diagnostic instrument
or measure that could assess all relevant PTSSs,
(b) they were based on case studies, (c) they were
based on a sample that was primarily composed of
individuals older than 19 years of age, and (d) they
did not include sufficient statistical data to compute
meaningful analyses. As such, 26 empirical studies
were included in this review—18 of which described
injured youth, whereas 8 involved youth with various
illnesses (see Table 1 for a description of studies).
Methodological Considerations
The studies included in this meta-analysis utilized a
wide range of research designs and effect size esti-
mates. As such, the current authors employed various
strategies to ensure methodological accuracy. For
example, if the same sample was used in several stud-
ies, it was only included once in the meta-analysis,
whereas multiple dependent variables based on the
same sample were aggregated across studies. This
was done to avoid a significant distortion of the
standard error estimates that typically results when
treating nonindependent studies as independent
(Gleser & Olkin, 1994). In addition, when a study
reported multiple effect size estimates for the same
general construct, they were averaged to calculate
an overall effect size.
To compute prevalence rates of PTSD, the authors
decided that scores falling in the moderate and
severe ranges of continuous assessment measures
would qualify as roughly meeting PTSD diagnostic
150 Traumatology / Vol. 12, No. 2, June 2006
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
Ta
bl
e
1.
D
es
cr
ip
ti
on
o
f
S
tu
di
es
I
nc
lu
de
d
in
t
he
M
et
a-
A
na
ly
si
s
S
am
pl
e
S
iz
e
of
Ty
pe
o
f
Tr
au
m
at
iz
ed
C
om
pa
ri
so
n
T
im
e
C
ur
re
nt
o
r
D
ia
gn
os
ti
c
A
ut
ho
r
S
am
pl
e
Yo
ut
h
G
ro
up
s
F
ra
m
e
L
if
et
im
e
In
fo
rm
an
ts
In
st
ru
m
en
ts
B
ro
w
n,
M
ad
an
–
S
w
ai
n,
&
L
am
be
rt
(
20
03
)
B
ut
le
r,
R
iz
zi
,
&
H
an
dw
er
ge
r
(1
99
6)
K
az
ak
e
t
al
.
(1
99
7)
a
K
az
ak
e
t
al
.
(2
00
4)
b
Pe
lc
ov
it
z
et
a
l.
(1
99
8)
S
he
m
es
h
et
a
l.
(2
00
0)
S
tu
be
r,
N
ad
er
,
H
ou
sk
am
p,
&
P
yn
oo
s
(1
99
6)
c
W
al
ke
r,
H
ar
ri
s,
B
ak
er
,
K
el
ly
,
&
H
ou
gh
to
n
(1
99
9)
A
ar
on
,
Za
gl
ul
,
&
E
m
er
y
(1
99
9)
B
ry
an
t,
M
ay
ou
,
W
ig
gs
, E
hl
er
s,
&
S
to
re
s
(2
00
4)
d
A
do
le
sc
en
t
su
rv
iv
or
s
of
ch
i
ld
ho
od
c
an
ce
r
Pe
di
at
ri
c
ca
nc
er
su
rv
iv
or
s
C
hi
ld
ho
od
l
eu
ke
m
ia
su
rv
iv
or
s
A
do
le
sc
en
t
su
rv
iv
or
s
of
ch
ild
ho
od
c
an
ce
r
C
an
ce
r
L
iv
er
t
ra
ns
pl
an
ts
Pe
di
at
ri
c
ca
nc
er
(
bo
ne
m
ar
ro
w
t
ra
ns
pl
an
ts
;
B
M
T
)
L
iv
er
t
ra
ns
pl
an
ts
P
hy
si
ca
l
in
ju
ri
es
M
VA
52 72 13
0
15
0
23 19 30 18 40 81
C
an
ce
r
su
rv
iv
or
s
vs
.
he
al
th
y
co
nt
ro
ls
P
T
S
D
v
s.
n
on
-P
T
S
D
C
an
ce
r
su
rv
iv
or
s
vs
.
he
al
th
y
co
nt
ro
ls
S
ur
vi
vo
rs
r
ec
ei
vi
ng
ps
yc
ho
so
ci
al
t
re
at
m
en
t
vs
.
su
rv
iv
or
w
ai
t-
lis
t
co
nt
ro
ls
P
hy
si
ca
l
in
ju
ry
;
hi
gh
-r
is
k
vs
.
lo
w
-r
is
k
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
L
iv
er
t
ra
ns
pl
an
te
es
v
s.
y
ou
th
w
it
h
ei
th
er
c
hr
on
ic
a
st
hm
a
or
t
ho
se
u
nd
er
go
in
g
E
N
T
op
er
at
io
ns
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
R R R R
a
nd
P
R R P R P P
C
ur
re
nt
a
nd
L
if
et
im
e
U
nc
le
ar
C
ur
re
nt
a
nd
L
if
et
im
e
C
ur
re
nt
a
nd
L
if
et
im
e
C
ur
re
nt
a
nd
L
if
et
im
e
C
ur
re
nt
?
U
nc
le
ar
L
if
et
im
e
an
d
C
ur
re
nt
C
ur
re
nt
C
ur
re
nt
C
hi
ld
a
nd
P
ar
en
t
Pa
re
nt
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
C
hi
ld
a
nd
P
ar
en
t
(o
nl
y
ch
ild
f
or
P
T
S
D
)
C
hi
ld
C
hi
ld
a
nd
P
ar
en
t
(o
nl
y
ch
ild
f
or
P
T
S
D
)
P
T
S
D
-R
I
P
S
S
IE
S
/P
T
S
D
-R
I/
T
S
C
C
P
T
S
D
-R
I/
IE
S
-R
/
S
C
ID
S
C
ID
/D
IS
U
C
L
A
-P
T
S
D
-R
I
C
P
T
S
D
-R
I
C
P
T
S
-R
I
P
T
S
D
-R
I
P
T
S
D
-R
I/
(I
E
S
)
151
(c
on
ti
nu
ed
)
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
Ta
bl
e
1
(c
on
ti
nu
ed
)
S
am
pl
e
S
iz
e
of
Ty
pe
o
f
Tr
au
m
at
iz
ed
C
om
pa
ri
so
n
T
im
e
C
ur
re
nt
o
r
D
ia
gn
os
ti
c
A
ut
ho
r
S
am
pl
e
Yo
ut
h
G
ro
up
s
F
ra
m
e
L
if
et
im
e
In
fo
rm
an
ts
In
st
ru
m
en
ts
D
av
is
s
et
a
l.
(2
00
0)
de
V
ri
es
e
t
al
.
(1
99
9)
D
i
G
al
lo
,
B
ar
to
n,
&
P
ar
ry
-J
on
es
(1
99
7)
e
D
yb
,
H
ol
en
,
B
ra
en
ne
,
In
dr
ed
av
ik
,
&
A
ar
se
th
(
20
03
)
G
er
ri
ng
e
t
al
.
(2
00
2)
K
as
sa
m
-A
da
m
s
&
W
in
st
on
(
20
04
)f
K
ep
pe
l-
B
en
so
n,
O
lle
nd
ic
k,
&
B
en
so
n
(2
00
2)
L
ev
i,
D
ro
ta
r,
Ye
at
es
,
&
T
ay
lo
r
(1
99
9)
M
ax
e
t
al
.
(1
99
8)
M
at
he
r,
Ta
te
,
&
H
an
na
n
(2
00
3)
M
cD
er
m
ot
t
&
C
vi
ta
no
vi
ch
(2
00
0)
A
cc
id
en
ta
l
in
ju
ri
es
M
VA
M
VA
M
VA
C
lo
se
d
he
ad
i
nj
ur
y
M
VA
M
VA
S
ev
er
e
an
d
m
od
er
at
e
T
B
I
T
B
I
M
VA
/(
T
B
I)
M
VA
48 10
2
57 16 95 17
7
50 ~
71
44 43 26
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
T
B
I
vs
.
or
th
op
ed
ic
i
nj
ur
y
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
;
T
B
I
vs
.
no
n-
T
B
I
P
T
S
D
v
s.
n
on
-P
T
S
D
P P P P P P R P P P P
C
ur
re
nt
C
ur
re
nt
C
ur
re
nt
?
C
ur
re
nt
C
ur
re
nt
C
ur
re
nt
C
ur
re
nt
a
nd
L
if
et
im
e
C
ur
re
nt
C
ur
re
nt
C
ur
re
nt
?
C
ur
re
nt
C
hi
ld
a
nd
P
ar
en
t
Pa
re
nt
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
a
nd
P
ar
en
t
U
nc
le
ar
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
a
nd
P
ar
en
t
C
A
P
S
-C
A
/
P
C
L
-C
/P
R
P
C
L
P
T
S
D
-R
I/
R
IE
S
C
P
T
S
-R
I
D
IC
A
C
A
P
S
-C
A
D
IC
A
-R
-C
;
D
IC
A
-R
-P
C
P
T
S
D
-R
I/
P
T
S
S
K
-S
A
D
S
C
P
T
S
-R
I
P
T
S
D
-R
I
152
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
M
ir
za
,
B
ha
dr
in
at
h,
G
oo
dy
er
,
&
G
ilm
ou
r
(1
99
8)
S
ax
e,
S
to
dd
ar
d,
&
S
he
ri
da
n
(1
99
8)
g
S
ta
lla
rd
,
Ve
lle
m
an
,
&
B
al
dw
in
(
19
98
)h
S
to
dd
ar
d,
N
or
m
an
,
M
ur
ph
y,
&
B
ea
rd
sl
ee
(
19
89
)i
Zi
nk
&
M
cC
ai
n
(2
00
3)
M
VA
B
ur
ns
M
VA
B
ur
ns
M
VA
11
9
72 11
9
30 14
3
P
T
S
D
v
s.
n
on
-P
T
S
D
P
T
S
D
v
s.
n
on
-P
T
S
D
S
po
rt
s
in
ju
ri
es
;
M
V
A
P
T
S
D
v
s.
M
VA
n
on
-P
T
S
D
P
T
S
D
v
s.
n
or
m
al
c
on
tr
ol
s
P
T
S
D
v
s.
n
on
-P
T
S
D
P P P R P
U
nc
le
ar
C
ur
re
nt
C
ur
re
nt
C
ur
re
nt
a
nd
L
if
et
im
e
C
ur
re
nt
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
C
hi
ld
C
hi
ld
a
nd
P
ar
en
t
C
hi
ld
a
nd
P
ar
en
t
P
T
S
D
-R
I/
K
-S
A
D
S
C
P
T
S
D
-R
I
C
A
P
S
-C
A
D
IC
A
/S
C
ID
D
IC
A
-R
N
O
T
E
: P
T
S
D
=
p
os
tt
ra
um
at
ic
s
tr
es
s
di
so
rd
er
; P
T
S
D
-R
I
=
P
os
tt
ra
um
at
ic
S
tr
es
s
D
is
or
de
r-
R
ea
ct
io
n
In
de
x;
P
S
S
=
P
T
S
O
S
ym
pt
om
S
ca
le
;
I
E
S
=
I
m
pa
ct
o
f
E
ve
nt
S
ca
le
; T
S
C
=
T
ra
um
a
S
ym
pt
om
c
he
ck
lis
t;
C
P
T
S
D
-R
I
=
C
hi
ld
P
os
tt
ra
um
at
ic
S
tr
es
s
D
is
or
de
r-
R
ea
ct
io
n
In
de
x;
I
E
S
-R
=
I
m
pa
ct
o
f
E
ve
nt
s
S
ca
le
-R
ev
is
ed
;
S
C
ID
=
S
tr
uc
tu
re
d
C
lin
ic
al
In
te
rv
ie
w
f
or
D
S
M
-I
V;
D
IS
=
D
ia
gn
os
ti
c
In
te
rv
ie
w
S
ch
ed
ul
e;
U
C
L
A
-P
T
S
D
-R
I
=
U
C
L
A
P
os
tt
ra
um
at
ic
S
tr
es
s
D
is
or
de
r-
R
ea
ct
io
n
In
de
x;
C
P
T
S
-R
I
=
C
hi
ld
P
os
tt
ra
um
at
ic
S
tr
es
s
R
ea
ct
io
n
In
de
x;
C
A
P
S
-C
A
=
C
lin
ic
ia
n-
A
dm
in
is
te
re
d
P
T
S
D
S
ca
le
(
C
hi
ld
a
nd
A
do
le
sc
en
t
ve
rs
io
n)
;
P
C
L
=
P
os
tt
ra
um
at
ic
S
tr
es
s
D
is
or
de
r
C
he
ck
lis
t;
P
C
L
-C
=
Po
st
tr
au
m
at
ic
S
tr
es
s
D
is
or
de
r
C
he
ck
lis
t-
C
iv
ili
an
V
er
si
on
;
R
IE
S
=
R
ev
is
ed
I
m
pa
ct
o
f
E
ve
nt
S
ca
le
;
D
IC
A
=
D
ia
gn
os
ti
c
In
te
rv
ie
w
f
or
C
hi
ld
re
n
an
d
A
do
le
sc
en
ts
;
D
IC
A
-R
-C
=
D
ia
gn
os
ti
c
In
te
rv
ie
w
f
or
C
hi
ld
re
n
an
d
A
do
le
sc
en
ts
-R
ev
is
ed
-C
hi
ld
V
er
si
on
; D
IC
A
-R
-P
=
D
ia
gn
os
ti
c
In
te
rv
ie
w
f
or
C
hi
ld
re
n
an
d
A
do
le
sc
en
ts
-R
ev
is
ed
-P
ar
en
t
Ve
rs
io
n;
K
-S
A
D
S
=
S
ch
ed
ul
e
fo
r
A
ff
ec
ti
ve
D
is
or
de
rs
a
nd
S
ch
iz
op
hr
en
ia
f
or
S
ch
oo
l-
A
ge
C
hi
ld
re
n.
a.
B
as
ed
o
n
da
ta
f
ro
m
G
er
ri
ng
e
t
al
.
(2
00
2)
;
K
ep
pe
l-
B
en
so
n,
O
lle
nd
ic
k,
a
nd
B
en
so
n
(2
00
2)
.
b.
B
as
ed
o
n
da
ta
f
ro
m
G
er
ri
ng
e
t
al
.
(2
00
2)
;
K
ep
pe
l-
B
en
so
n,
O
lle
nd
ic
k,
a
nd
B
en
so
n
(2
00
2)
;
M
at
he
r,
Ta
te
,
an
d
H
an
na
n
(2
00
3)
;
S
ta
lla
rd
,
Ve
lle
m
an
,
an
d
B
al
dw
in
(
19
98
).
c.
B
as
ed
o
n
da
ta
f
ro
m
G
er
ri
ng
e
t
al
.
(2
00
2)
;
M
at
he
r,
Ta
te
,
an
d
H
an
na
n
(2
00
3)
;
S
ta
lla
rd
,
Ve
lle
m
an
,
an
d
B
al
dw
in
(
19
98
).
d.
B
as
ed
o
n
da
ta
f
ro
m
D
av
is
s
et
a
l.
(2
00
0)
;
K
as
sa
m
-A
da
m
s
an
d
W
in
st
on
(
20
04
).
e.
B
as
ed
o
n
da
ta
f
ro
m
B
ry
an
t,
M
ay
ou
,
W
ig
gs
,
E
hl
er
s,
a
nd
S
to
re
s
(2
00
4)
;
S
ax
e,
S
to
dd
ar
d,
a
nd
S
he
ri
da
n
(1
99
8)
.
f.
B
as
ed
o
n
da
ta
f
ro
m
A
ar
on
,
Za
gl
ul
,
an
d
E
m
er
y
(1
99
9)
;
M
at
he
r,
Ta
te
,
an
d
H
an
na
n
(2
00
3)
;
M
cD
er
m
ot
t
an
d
C
vi
ta
no
vi
ch
(
20
00
).
g.
B
as
ed
o
n
da
ta
f
ro
m
G
er
ri
ng
e
t
al
.
(2
00
2)
;
S
ta
lla
rd
,
Ve
lle
m
an
,
an
d
B
al
dw
in
(
19
98
).
h.
B
as
ed
o
n
da
ta
f
ro
m
B
ry
an
t,
M
ay
ou
,
W
ig
gs
,
E
hl
er
s,
a
nd
S
to
re
s
(2
00
4)
;
M
cD
er
m
ot
t
an
d
C
vi
ta
no
vi
ch
(
20
00
);
S
ta
lla
rd
,
Ve
lle
m
an
,
an
d
B
al
dw
in
(
19
98
).
i.
B
as
ed
o
n
da
ta
f
ro
m
D
av
is
s
et
a
l.
(2
00
0)
;
G
er
ri
ng
e
t
al
.
(2
00
2)
;
K
ep
pe
l-
B
en
so
n,
O
lle
nd
ic
k,
a
nd
B
en
so
n
(2
00
2)
;
S
ax
e,
S
to
dd
ar
d,
a
nd
S
he
ri
da
n
(1
99
8)
;
S
ta
lla
rd
,
Ve
lle
m
an
,
an
d
B
al
dw
in
(
19
98
).
153
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
154 Traumatology / Vol. 12, No. 2, June 2006
criteria. Many research and clinical settings tend to
interpret scores in these ranges as indicative of clin-
ically meaningful levels of PTSS or PTSD (e.g., Levi
et al., 1999). If multiple time point assessments
were conducted, the estimates closest to the trauma
(within at least 1 month to allow for the emergence
of PTSD symptoms) were included. Any follow-up
data related to the trauma and rates of PTSD were
included when it was available.
Subthreshold PTSD has been defined differently
within the literature, such as the endorsement of at
least one, two, or three PTSSs (Marshall et al.,
2001) or of at least two of the three symptom clus-
ters (Aaron et al., 1999). The authors employed the
criteria of endorsing at least two PTSSs to meet cri-
teria for subthreshold PTSD, as this was basically
consistent with the methodologies noted above. Also
important to note is that there was often a great deal
of variability in the comparison groups used in the
studies reviewed in this article (e.g., normal con-
trols, comparing those developing PTSD vs. those
who do not after the experience of a trauma; see
Table 1). As such, the various PTSD predictors and
correlates described within this article are to be
interpreted cautiously. Because very few studies
examined similar variables that related to either of
the traumas, a predictor needed to be examined in at
least two independent studies for inclusion in the
current analysis (per Rosenthal, 1991).
Data Analytic Plan
For each study that met inclusion criteria, the
authors coded descriptive information including the
mean age of youth at the time of PTSD assessment,
percentage of the sample that was female, mean
PTSD, and partial PTSD prevalence rates along with
other study characteristics, such as type of trauma
and informant of PTSS. Next, predictors for the
development of PTSD among youth exposed to ill-
nesses and injuries were evaluated. Studies that uti-
lized between-group designs typically reported t, F,
and chi-square statistics, whereas studies using cor-
relational designs reported either Pearson r or phi
statistics. Per Rosenthal (1991, 1994) and Hedges
and Olkin (1985), all of the statistics were converted
to Cohen’s d to yield a single common measure of
effect size. The use of effect size statistics for both
group differences and correlational data is consistent
with current standard statistical practices (Lipsey &
Wilson, 2001). Conventional social sciences research
generally interprets Cohen’s d effect size values as .2
for small, .5 for medium, and .8 for large effects,
with higher d values indicating a stronger relation-
ship with PTSD.
Q statistics were examined to test for homogene-
ity among the effect sizes associated with any given
predictor. In general, significant Q statistics indicate
that the variability among effect sizes is greater than
what would result from subject-level sampling error
alone. As such, for those predictors with significant
Q statistics, the authors chose to report effect sizes
individually (rather than as part of an aggregate). This
approach is consistent with statistical standards (e.g.,
Durlak, 1995). Moreover, because of the exploratory
nature of the current manuscript, this approach
seemed preferable than simply utilizing a random
effects model, which would have obscured the poten-
tial significant differences between certain effect
sizes. For those predictors in which the Q statistics
were not significant, the effect sizes were averaged
(aggregated) and weighted by the sample sizes asso-
ciated with the predictors.
In accordance with the rules described above,
several predictors and correlates were culled from the
literature. For both groups, the associations between
mean age of youth, gender rates, and PTSD preva-
lence rates were computed as was the lag between
the trauma and the appraisal of symptoms (see Table 2).
For youth experiencing injuries, 8 unique predictors
were identified: (a) socioeconomic status (SES),
(b) Internalizing scale on the Child Behavior Checklist
(CBCL; Achenbach, 1991), (c) comorbid depressive
symptoms, (d) comorbid anxious symptoms, (e) comorbid
dissociation, (f) acute stress disorder (ASD), (g) social
impairment, and (h) appraisal of trauma severity
(and/or life appraisal). For youth with illnesses, the
years since active treatment, appraisal of trauma
severity (and/or life appraisal), and social support
emerged as important PTSD predictors (see Table 3).
Results
Injury
As Table 2 demonstrates, 11 of the 18 studies (61%)
involved MVAs, whereas the remainder included
participants experiencing burns (n = 2), TBIs or
closed-head injuries (n = 3), and a variety of physical
injuries (n = 2). The mean age of participants was
11.3 years (SD = 1.41 years), and 41% of the studies’
participants were female (weighted by sample size).
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
Meta-Analysis of PTSD in Youth / Kahana et al. 155
The average sample size was approximately 70.39
(SD = 46.38) participants. Caucasians composed the
large majority of the study samples at 70%, whereas
African American and Hispanic populations constituted
24% and 5%, respectively, of the samples. Finally,
the average lag time between the occurrence of the
trauma and the assessment of PTSS/PTSD was 4.97
(SD = 3.77) months.
Table 2. Demographics and Prevalence Rates of PTSD Among Pediatric Injury and Illness Populations
Injury Population (n = 18 studies) Weighted Mean Illness Population (n = 8 studies) Weighted Mean
Age 11.30 (1.41) Mean age 13.39 (2.67)
Percentage female 41.01 Percentage female 50.29
PTSD prevalence rates 19.82 (9.94) PTSD prevalence rates 12.04 (7.87)
Partial PTSD rates 25.03 (8.06) Partial PTSD rates 37.58 (11.98)
Sample size 70.39 (46.38) Sample size 61.75 (51.96)
PTSD correlates Pearson’s r PTSD correlates Pearson’s r
Mean age .04 Mean age –.48
Gender (female) .09 Gender .04
Gender (and partial PTSD) .50 Gender (and partial PTSD) .88
Trauma and assessment lag –.26 Years since active treatment –.63*
NOTE: PTSD = posttraumatic stress disorder.
*p < .05.
Table 3. Predictors of PTSD Among Pediatric Injury and Illness Populations
Cohen’s d Cohen’s d
SESa –0.45 Appraisalj .39
CBCL Internalizing Scaleb 1.03 Social Supportk .14
Depressive Symptomsc 1.06, 1.09, 1.57
Anxious Symptomsd 0.90, 0.90, 1.64, 1.96
Dissociatione 3.53, 1.19
Acute Stress Disorderf 0.94, 1.35
Appraisal of Life Threatg 0.82
Trauma Severityh –0.39, 0.19, 0.44, 0.85, 1.32
Social Impairmenti 0.41
NOTE: Cohen’s d effect size values are d = .2 for small, .5 for medium, and .8 for large effects. The effect sizes for Depressive
Symptoms, Anxious Symptoms, Dissociation, Acute Stress Disorder, and Trauma Severity are presented separately (instead of as part
of a weighted mean), as Q statistics indicated a significant amount of heterogeneity between them. PTSD = posttraumatic stress dis-
order; SES = socioeconomic status; CBCL = Child Behavior Checklist.
a. Based on data from Gerring et al. (2002); Keppel-Benson, Ollendick, and Benson (2002).
b. Based on data from Aaron, Zaglul, and Emery (1999); Mather, Tate, and Hannan (2003); McDermott and Cvitanovich (2000).
c. Based on data from Gerring et al. (2002); Mather, Tate, and Hannan (2003); Stallard, Velleman, and Baldwin (1998).
d. Based on data from Gerring et al. (2002); Keppel-Benson, Ollendick, and Benson (2002); Mather, Tate, and Hannan (2003);
Stallard, Velleman, and Baldwin (1998).
e. Based on data from Bryant, Mayou, Wiggs, Ehlers, and Stores (2004); Saxe, Stoddard, and Sheridan (1998).
f. Based on data from Daviss et al. (2000); Kassam-Adams and Winston (2004).
g. Based on data from Bryant, Mayou, Wiggs, Ehlers, and Stores (2004); McDermott and Cvitanovich (2000); Stallard, Velleman,
and Baldwin (1998).
h. Based on data from Daviss et al. (2000); Gerring et al. (2002); Keppel-Benson, Ollendick, and Benson (2002); Saxe, Stoddard,
and Sheridan (1998); Stallard, Velleman, and Baldwin (1998).
i. Based on data from Gerring et al. (2002); Stallard, Velleman, and Baldwin (1998).
j. Based on data from Brown, Madan-Swain, and Lambert (2003); Shemesh et al. (2000); Stuber et al. (1996).
k. Based on data from Brown, Madan-Swain, and Lambert (2003); Pelcovitz et al. (1998).
1. One study (Brown, Madan-Swain, & Lambert, 2003) contained a few childhood cancer survivors who were 23 years old. Despite
this, the study was included because it was mostly composed of adolescents.
2. Several studies were excluded because the injury versus illness categories were not differentiated (Balluffi et al., 2004; Landolt,
Vollrath, Ribi, Gnehm, & Stennhauser, 2003).
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
156 Traumatology / Vol. 12, No. 2, June 2006
Average PTSD prevalence rates ranged from 0%
to 37.5%, with a mean of 19.82% (SD = 9.94)
weighted by sample size. Of the 18 studies involving
injuries, more than half (n = 11) provided enough
information to calculate that 25% of the samples
met criteria for partial PTSD. With respect to spe-
cific clusters, 50.47%, 32.53%, and 17.90% of the
participants endorsed sufficient symptoms to meet
criteria for the re-experiencing, arousal, and avoid-
ance clusters, respectively. A paired sample t test
revealed that all these rates were significantly different
from each other (i.e., arousal vs. avoidance, t(4) = 3.66,
p < .05; reexperiencing vs. avoidance, t(4) = 10.13,
p < .005; reexperiencing vs. arousal, t(4) = 3.13, p <
.05). It is conceivable that the greater number of
youth meeting criteria for the re-experiencing clus-
ter is due to the fact that only 1 re-experiencing
symptom needs to be endorsed to meet diagnostic
criteria. Indeed, of the three studies that reported
percentages of specific symptoms, the re-experiencing
symptoms of intrusive recollections and psychological
exposure to trauma-related cues were among the most
highly endorsed symptoms, with weighted percents
of 61.46% and 51.32%, respectively.
Among youth experiencing injuries, PTSD
prevalence rates were not strongly associated with
mean age of youth, r(15) = .04, n.s., or female gen-
der, r(15) = .09, n.s. (see Table 2). However, females
exhibited relatively high rates of subthreshold PTSD
at r(10) = .50, n.s. There was a statistical trend for
PTSD prevalence rates to be inversely related with
the lag between the occurrence of the trauma and
the assessment of symptoms at r(16) = –.26, p < .10.
This suggests that injured youth tended to exhibit
lower rates of PTSD as more time lapsed between
the experience of the trauma and the assessment
of PTSD.
Among youth experiencing injuries, SES was a
significant, though small to moderate, predictor of
PTSS (Q = 3.35 [1], n.s.; mean weighted d = –.45
[95% confidence interval (CI) = –0.30 to –0.60]).
This suggests that youth from a lower SES were
somewhat more likely to develop PTSSs after expe-
riencing an injury. PTSS and PTSD were also often
related to the experience of other internalizing dis-
orders. Specifically, injured children with PTSSs
tended to endorse significantly higher clinical eleva-
tions on the CBCL Internalizing scale (Q = 3.55 [2],
n.s.; mean weighted d = 1.03 [95% CI = 0.85 to
1.21]) as well as a greater number of anxiety symp-
toms (Q =58.85 [3], p < .0001; ds = 0.90, 0.90,
1.64, and 1.96, respectively) as compared to youth
who sustained injuries but who did not endorse
PTSSs. Similarly, the endorsement of significant
depressive symptoms was also strongly related to
increased PTSS rates (Q =18.26 [2], p < .0005; ds =
1.06, 1.09, and 1.57), with large to very large effect
sizes indicating considerable comorbidity between
PTSD and depressive symptoms.
With two effect size estimates in the large to very
large range, dissociation served as a very robust pre-
dictor of PTSD diagnosis (Q = 65.34 [1], p < .0001;
ds = 3.53 and 1.19, respectively) among youth sus-
taining injuries. This result is logical, as dissociation
can often be a core symptom of PTSD. ASD was also
a strong predictor of PTSSs among injured youth,
with two reported effect sizes in the large range (Q =
6.34 [1], p < .05; ds = 0.94 and 1.35, respectively).
Finally, injured youth with PTSD demonstrated
small to moderate impairments in social functioning
(Q = .08 [1], n.s.; mean weighted d = .41 [95% CI =
0.39 to 0.43]).
A particularly interesting finding is that the
individual’s subjective appraisal of the life threat
involved in the injury was a robust predictor of
PTSD (Q = 4.78 [2], n.s.; mean weighted d = .82
[95% CI = 0.67 to 0.97]). This indicates that among
injured youth, the perception of threat to one’s life
or physical integrity was usually strongly related to
the development of PTSSs. Somewhat concordant
with this finding was the relationship between
PTSSs and the vaguely defined “trauma severity”
construct, which was rated either by the clinician,
parent, or youth. Various effect sizes were reported
for this association (Q = 85.58 [4], p < .0001; ds =
–0.39, 0.19, 0.44, 0.85, and 1.32), with some in the
small to moderate range and others in the large to
very large range. It is likely that the varied perspec-
tives of different raters contributed, at least in part,
to the variability in the effect sizes. For example, a
clinician might focus on the more objective severity
(e.g., acuity, intensity, length of exposure to trauma)
associated with an injury, whereas the parents and
child will likely appraise the trauma severity more
subjectively (e.g., involvement of any pain). In addi-
tion, because severity was defined somewhat differ-
ently between studies, it was difficult to identify
meaningful trends regarding specific informants
attached to the varied effect sizes.
Illness
Six of the eight studies involved pediatric or adolescent
cancer survivors, whereas the other two focused on
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
pediatric liver transplantees. The mean age of partic-
ipants was 13.39 years (SD = 2.67 years), and about
half (50.29%) of the participants were female. The
average sample size for youth experiencing illnesses
was 61.75 (SD = 51.96) participants. More than
three quarters (75%) of the study participants were
Caucasian, and the remaining 25% were divided
among African Americans, Hispanics, Asians, and
Other. Finally, the mean number of years since active
treatment had ended and the assessment of PTSD/
PTSS was 3.65 (SD = 2.04) years.
Among ill youth, the average PTSD prevalence
rates ranged from 0% to 32%, with a mean of 12.04%
(SD = 7.87) weighted by sample size. Insufficient
data precluded providing information about the
symptom clusters among youth experiencing ill-
nesses. Among these youth, female gender was not
associated with PTSD prevalence rates, r(6) = .04,
n.s. However, the mean age of the youth appeared to
be inversely related with PTSD prevalence, r(7) =
–.48, n.s., suggesting that the younger the youth, the
more likely they were to display PTSSs. Subthreshold
PTSD rates for ill youth were fairly high at 37.58%
(SD = 11.98), with females tending to exhibit rela-
tively high rates at r(3) = .88, n.s.
Of the six studies that described youth who had
been survivors of (as compared to youth still under-
going) illnesses or procedures, there was a robust
and statistically significant association between
PTSD prevalence rates and the time since youth had
finished their active treatments and/or medical pro-
cedures, r(6) = –.63, p < .05. That is, youth exhib-
ited lower rates of PTSD when more time had lapsed
between the illness/procedure and the assessment of
their trauma symptoms. This suggests that among
youth experiencing illnesses, PTSD is likely to occur
shortly following traumatic exposure (with delayed
onset less common), or alternatively, that trauma
symptoms diminish over time. Because most studies
that involve ill youth do not pinpoint the time when
PTSSs have emerged, and evaluate current as com-
pared to past symptoms, the data more likely indi-
cate that symptoms decrease with time, and, thus
measure chronicity.
The association between perceived severity of
trauma (and/or life threat) and PTSSs was small to
moderate (Q = 5.47 [2], n.s.; mean weighted d = .39;
95% CI = 0.16 to 0.62). Three studies contributed
to this finding; one study involved a clinician’s
assessment and the other two were made by the
youth. Finally, the social support construct (broadly
defined as support from one’s family and friends)
was investigated within two pediatric illness studies.
Among ill youth, social support exhibited a relatively
small relationship with the development of PTSS
(Q = .08 [1], n.s.; mean weighted d = .14; 95% CI =
0.11 to 0.17). This is surprising, as past literature
has noted the importance of social support as a
buffer against the development of PTSD. It is possi-
ble that this finding is an artifact of the protracted
assessment lag that occurred for these youth.
Differences Between the
Groups on Prevalence Rates
and Time of Assessment
Several important differences between the injury
and illness groups emerged in this review. First,
a chi-square test of proportions indicated that the
prevalence rates of PTSD differed significantly
between youth experiencing injury versus illness
(χ2 = 14.45 [1], p < .0005). Injured youth exhibited
a mean of 19.82% (SD = 9.94), whereas ill youth
displayed a mean of 12.04% (SD = 7.87). Another
important finding is that the mean effect size differ-
ences between the groups on the perceived life
threat and/or assessment of trauma severity variable
were approaching statistical significance (difference
between independent ds; z = 1.90, p = .06), with
children in the injury group endorsing higher rates
of perceived trauma severity and/or life threat. This
disparity might be attributed to significant differ-
ences in the mean lag of PTSD assessment for
youth in both groups (t = 8.55 [1,215], p < .0001).
Specifically, youth experiencing illnesses were often
followed up years after the diagnosis and treatments
for their illness, whereas youth exposed to injuries
were often assessed months after their experienced
trauma.
The availability of Time 2 PTSD assessment data
in a subset of studies involving injured youth made it
possible to formally examine this hypothesis. Nine of
the 18 studies (50%) measured PTSS and PTSD
prevalence rates at multiple time points, allowing for
a comparison of PTSD prevalence rates at different
time points among the same trauma group. Within
these 9 studies, Time 1 assessments occurred an
average of 3.40 (SD = 3.38) months after the trauma
occurrence, whereas Time 2 assessments occurred
an average of 6.48 (SD = 2.64) months after the
trauma. The mean PTSD prevalence rates of 24.61%
(SD = 9.15) at Time 1 were significantly higher than
Meta-Analysis of PTSD in Youth / Kahana et al. 157
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
the PTSD prevalence rates of 16.25% (SD = 9.77) at
Time 2 (χ2 = 19.61 [1], p < .0005). This finding sup-
ports the notion that delayed onset of PTSSs is not
common and likely occurs only in a minority of
injured youth. In addition, it appears that among
injured youth, PTSSs subside with time, a pattern
consonant with that manifested by ill youth.
Discussion
The goal of the current meta-analysis was to review
and synthesize studies addressing PTSD in youth
exposed to medical illness or injury. Several interest-
ing findings emerged from this investigation. Among
the most important is that the injury- and illness-
related trauma literatures have not examined many
similar or overlapping predictors of PTSD. The
sparseness of replication data with respect to poten-
tial predictors makes it difficult to investigate mean-
ingful comparisons across studies, either between
the phenomenology of PTSD or PTSD predictors.
This review highlights that a minority of youth
experiencing illnesses or injuries typically developed
PTSD or PTSSs. The prevalence rates of PTSD
appeared higher among youth experiencing injuries
as compared to illness. However, it is crucial to high-
light that this result might be the product of
methodological artifact. Specifically, although some
of the cancer and transplant studies followed chil-
dren for a considerable number of years, follow-up
periods for injured youth were shorter and more
proximal to the traumatic event, thus resulting in a
possible inflation of differences in PTSD rates
between the groups. In fact, support for this notion
comes from Time 2 assessment data among injured
youth, where PTSD rates were significantly lower at
follow-up as compared to the initial assessment.
Age at the time of PTSD assessment appears to
be more closely related to the development of
PTSD/PTSSs among youth with illnesses as com-
pared to those with injuries. Specifically, the older
the ill youth, the less likely they were to exhibit sig-
nificant PTSSs. This might be consistent with
research suggesting that older youth possess more
cognitive capacities to implement coping strategies
after a trauma than do younger youth (Joseph,
Brewin, Yule, & Williams, 1993). Alternatively, the
relationship between mean age and PTSD rates
within the illness group might also be the result of
the time gap between the termination of active med-
ical treatment and the evaluation of symptoms. That
is, the lag between PTSD assessment and trauma
occurrence may be masking the actual association
between age and PTSD.
Although females were more likely than males to
display PTSSs (and subthreshold PTSD), the current
meta-analysis indicated there did not appear to be
significant gender differences in rates of diagnostic
PTSD among youth experiencing injuries and ill-
nesses. Although the adult literature consistently
indicates that women are at higher risk for develop-
ing PTSD following various traumas (Breslau et al.,
1998; Kessler, Sonnega, Bromet, Hughes, & Nelson,
1995), these data do not typically include medically
ill samples and thus might not be the most appropriate
for comparison with the current findings. Another
explanation might be that the degree of trauma severity,
whether subjective or objective, might have differed
between males and females, thus leading to differ-
ences in symptom rates. Although there are reported
differences in exposure to various traumas among
males and females, with males typically exposed to
greater rates of community and physical violence
(e.g., Jaycox et al., 2002) and females to greater rates
of sexual abuse (e.g., Putnam, 2003), there is no lit-
erature comparing the severity of experienced traumas
between the genders.
Internalizing symptoms, particularly those asso-
ciated with depression and anxiety, appear to be
highly comorbid with PTSD in injured youth. This is
helpful clinical information, as it suggests that
PTSSs will typically not occur in isolation from
other psychiatric symptoms. This finding is consis-
tent with research that reports high levels of comor-
bidity among adults with PTSD (e.g., Kessler et al.,
1995). Finally, although temporal order of onset is
difficult to establish, it is possible that preexisting
internalizing disorders, such as anxiety, serve as a
diathesis that places people at higher risk for devel-
oping PTSD after a trauma (Bradley, 2000).
The appraisal of life threat and trauma severity
is a construct that requires further study. There are
many components to consider, including but not
limited to, medical late effects, future necessary
treatments, and other individuals directly affected
by the trauma. The relationship between the appraisal
of trauma severity and PTSD was moderate to large
in both the illness and injury groups. However, given
the ambiguity of the construct (i.e., is severity sim-
ply a proxy for life threat?), it is difficult to make
conclusive interpretations about these data. Future
studies would be well served to integrate both objective
158 Traumatology / Vol. 12, No. 2, June 2006
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
Meta-Analysis of PTSD in Youth / Kahana et al. 159
and subjective (e.g., including collateral family reports)
trauma severity ratings to make comparisons between
trauma groups.
Research conducted to date has important
implications for identifying those youth who are at
risk for developing PTSD in response to illnesses or
injuries. First, it is important for clinicians and
researchers to assess uniform predictors and corre-
lates, including appraisal of life threat, academic
functioning, family functioning, social support, cog-
nitive strategies, dissociation, and perceived levels of
stress. It is only through replication by independent
research groups that predictors will gain credibility.
Further potential support for the role of these iden-
tified factors can be useful to practitioners in assess-
ing and preventing psychiatric symptoms and/or
improving functioning. Perhaps with more conclu-
sive research of predictors and moderators of PTSD
with youth experiencing illnesses and injuries, clini-
cians and programs serving these youth (such as
schools and medical clinics) will have a better
understanding of services and types of treatment
that need to be provided.
Second, studies should include more ethnically
diverse youth, as the composition of many of the
studied samples was predominantly Caucasian. Third,
to facilitate comparisons across studies, researchers
should clearly describe the rationale and methods of
selecting populations, as understanding potential
sample limitations is vital for drawing generalizations
from study results. Fourth, to understand the famil-
ial context of PTSD, family members affected by
the trauma should be assessed for PTSD and other
psychopathology. Finally, the recent literature has
addressed the importance of posttraumatic growth in
the context of an experienced trauma (Milam, Ritt-
Olson, & Unger, 2004; Salter & Stallard, 2004). As
such, assessing the resilience and strengths of youth
undergoing a trauma would allow for a more com-
prehensive understanding of the youth’s psychologi-
cal functioning after the trauma.
This study has several limitations. Various poten-
tial predictors could not be analyzed because they
were not included in a sufficient number of inde-
pendent studies. Unfortunately, much of the pedi-
atric trauma research has been hindered by a lack of
measurement of uniform variables. Specifically, it
appears that of the 11 variables for which effect
sizes were calculated, 5 of them only included infor-
mation from only two independent studies. In addi-
tion, the literature regarding illness (with only eight
studies meeting criteria and two predictors) may be
less advanced than the literature pertaining to
injuries (with 18 studies and eight identified predic-
tors). Moreover, several studies reported statistics
(e.g., multiple regressions) about PTSD predictors
that were not included in this review because the
values depended on the constellation of covariates
used in the original study. Another significant limi-
tation is that the comparison groups (e.g., normal
controls, all traumatized) often differed between the
injury and illness samples, rendering comparisons
across the trauma groups quite difficult. Finally, null
findings not reported in the literature might limit
the generalizability of the findings from this review.
In summary, there is much more research
that needs to be conducted before conclusive state-
ments can be made about the predictors for the
development of PTSD in trauma-exposed youth.
Despite the limitations discussed above, data sug-
gest that among injured youth, SES and social
impairment were small to moderate correlates of
PTSD, whereas depressive and anxious symptoms,
dissociation, ASD, and the appraisal of trauma
severity and life threat served as robust predictors
of PTSD. Among ill youth, social support and the
appraisal of illness severity and life threat emerged
as small to moderate predictors of PTSSs. Rather
than constituting a rigid set of predictors, these fac-
tors should be conceptualized as a rough heuristic
that clinicians can employ when appraising trauma-
exposed youth. Accordingly, clinicians should closely
monitor at-risk individuals who exhibit many of the
predictors identified in this article in an attempt to
prophylactically address psychiatric symptoms before
they become chronic.
Acknowledgments
The authors would like to thank Sheridan Stull, BA,
Judith Geraci, BA, Jennifer Goodpaster, BA, and
Kristen Walter, BA, for their aid in the collection of
articles included in the current manuscript. In addi-
tion, the authors wish to express appreciation to the
Lance Armstrong Center for Survivors of Pediatric
Illness for their continued support and funding.
References
*Studies preceded by an asterisk were included in the meta-
analysis.
*Aaron, J., Zaglul, H., & Emery, R. E. (1999). Posttraumatic
stress in children following acute physical injury. Journal
of Pediatric Psychology, 24, 335-343.
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
Achenbach, T. M. (1991). Manual for the Child Behavior
Checklist/4-18 and 1991 profile. Burlington: University of
Vermont, Department of Psychiatry.
American Psychiatric Association. (2001). Diagnostic and
statistical manual of mental disorders (4th ed.). Washington,
DC: Author.
Balluffi, A., Kassam-Adams, N., Kazak, A., Tucker, M.,
Dominguez, T., & Helfaer, M. (2004). Traumatic stress in
parents of children admitted to the pediatric intensive
care unit. Pediatric Care Medicine, 5, 547-553.
Berman, S. L., Kurtines, W. M., Silverman, W. K., & Serafini,
L. T. (1996). The impact of exposure to crime and violence
on urban youth. American Journal of Orthopsychiatry, 66,
329-336.
Bradley, S. J. (2000). Affect regulation and the development
of psychopathology. New York: Guilford.
Breslau, N., Kessler, R. C., Chilcoat, H. D., Schultz, L. R., Davis,
G. C., & Andreski, P. (1998). Trauma and posttraumatic stress
disorder in the community: The 1996 Detroit Area Survey of
Trauma. Archives of General Psychiatry, 55, 626-632.
*Brown, R. T., Madan-Swain, A., & Lambert, R. (2003).
Posttraumatic stress symptoms in adolescent survivors of
childhood cancer and their mothers. Journal of Traumatic
Stress, 16, 309-318.
*Bryant, B., Mayou, R., Wiggs, L., Ehlers, A., & Stores, G.
(2004). Psychological consequences of road traffic acci-
dents for children and their mothers. Psychological
Medicine, 34, 335-346.
*Butler, R. W., Rizzi, L. P., & Handwerger, B. A. (1996).
Brief report: the assessment of posttraumatic stress disor-
der in pediatric cancer patients and survivors. Journal of
Pediatric Psychology, 21, 499-504.
*Daviss, W. B., Mooney, D., Racusin, R., Ford, J. D.,
Fleischer, A., & McHugo, G. J. (2000). Predicting post-
traumatic stress after hospitalization for pediatric injury.
Journal of the American Academy of Child and Adolescent
Psychiatry, 39, 576-583.
*de Vries, A. P., Kassam-Adams, N., Cnaan, A., Sherman-
Slate, E., Gallagher, P. R., & Winston, F. K. (1999).
Looking beyond the physical injury: posttraumatic stress
disorder in children and parents after pediatric traffic
injury. Pediatrics, 104, 1293-1299.
*Di Gallo, A., Barton, J., & Parry-Jones, W. L. (1997). Road
traffic accidents: Early psychological consequences in
children and adolescents. British Journal of Psychiatry,
170, 358-362.
Durlak, J. A. (1995). Understanding meta-analysis. In L. A.
Grimm & P. R. Yarnold (Eds.), Reading and understanding
multivariate statistics (pp. 319-352). Washington, DC:
American Psychological Association.
*Dyb, G., Holen, A., Braenne, K., Indredavik, M. S., &
Aarseth, J. (2003). Parent-child discrepancy in reporting
children’s post-traumatic stress reactions after a traffic
accident. Nordic Journal of Psychiatry, 57, 339-344.
*Gerring, J. P., Slomine, B., Vasa, R. A., Grados, M., Chen,
A., Rising, W., et al. (2002). Clinical predictors of
posttraumatic stress disorder after closed head injury in
children. Journal of the American Academy of Child and
Adolescent Psychiatry, 41, 157-165.
Gleser, L. J., & Olkin, I. (1994). Stochastically dependent
effect sizes. In H. Cooper & L. V. Hedges (Eds.), The
handbook of research synthesis (pp. 339-355). New York:
Russell Sage Foundation.
Guyer, B., Hoyert, D. L., Martin, J. A., Ventura, S. J.,
MacDorman, M. F., & Strobino, D. M. (1999). Annual sum-
mary of vital statistics-1998. Pediatrics, 102, 1333-1349.
Hedges, L. V., & Olkin, I. (1985). Statistical methods for
meta-analysis. San Diego, CA: Academic Press.
Jaycox, L. H., Stein, B. D., Kataoka, S. H., Wong, M., Fink,
A., Escudero, P., et al. (2002). Violence exposure, post-
traumatic stress disorder, and depressive symptoms
among recent immigrant schoolchildren. Journal of the
American Academy of Child and Adolescent Psychiatry, 41,
1104-1110.
Joseph, S. A., Brewin, C. R., Yule, W., & Williams, R. (1993).
Causal attributions and post-traumatic stress in adolescents.
Journal of Child Psychology and Psychiatry, 34, 247-253.
*Kassam-Adams, N., & Winston, F. K. (2004). Predicting
child PTSD: The relationship between acute stress disorder
and PTSD in injured children. Journal of the American
Academy of Child and Adolescent Psychiatry, 43, 403-411.
*Kazak, A. E., Alderfer, M. A., Streisand, R., Simms, S.,
Rourke, M. T., Barakat, L. P., et al. (2004). Treatment of
posttraumatic stress symptoms in adolescent survivors of
childhood cancer and their families: A randomized clinical
trial. Journal of Family Psychology, 18, 493-504.
*Kazak, A. E., Barakat, L. P., Meeske, K., Christakis, D.,
Meadows, A. T., Casey, R., et al. (1997). Posttraumatic
stress, family functioning, and social support in survivors of
childhood leukemia and their mothers and fathers. Journal
of Consulting and Clinical Psychology, 65, 120-129.
*Keppel-Benson, J. M., Ollendick, T. H., & Benson, M. J.
(2002). Post-traumatic stress in children following motor
vehicle accidents. Journal of Child Psychology and
Psychiatry, 43, 203-212.
Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., &
Nelson, C. B. (1995). Posttraumatic stress disorder in the
National Comorbidity Survey. Archives of General
Psychiatry, 52, 1048-1060.
La Greca, A., Silverman, W. K., Vernberg, E. M., & Prinstein,
M. J. (1996). Symptoms of posttraumatic stress in children
after Hurricane Andrew: A prospective study. Journal of
Consulting and Clinical Psychology, 64, 712-723.
Landolt, M. A., Vollrath, M., Ribi, K., Gnehm, H. E., &
Sennhauser, F. H. (2003). Incidence and associations of
parental and child posttraumatic stress symptoms in pedi-
atric patients. Journal of Child Psychology and Psychiatry,
44, 1199-1207.
*Levi, R. B., Drotar, D., Yeates, K. O., & Taylor, H. G.
(1999). Posttraumatic stress symptoms in children fol-
lowing orthopedic or traumatic brain injury. Journal of
Clinical Child Psychology, 28, 232-243.
160 Traumatology / Vol. 12, No. 2, June 2006
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
Meta-Analysis of PTSD in Youth / Kahana et al. 161
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis.
Thousand Oaks, CA: Sage.
Marshall, R. D., Olfson, M., Hellman, F., Blanco, C.,
Guardino, M., & Struening, E. L. (2001). Comorbidity,
impairment, and suicidality in subthreshold PTSD.
American Journal of Psychiatry, 158, 1467-1473.
*Mather, F. J., Tate, R. L., & Hannan, T. J. (2003). Post-
traumatic stress disorder in children following road traffic
accidents: A comparison of those with and without mild
traumatic brain injury. Brain Injury, 17, 1077-1087.
*Max, J. E., Castillo, C. S., Robin, D. A., Lindgren, S. D.,
Smith, W. L. Jr., Sato, Y., et al. (1998). Posttraumatic stress
symptomatology after childhood traumatic brain injury.
Journal of Nervous and Mental Disease, 186, 589-596.
*McDermott, B. M., & Cvitanovich, A. (2000). Posttraumatic
stress disorder and emotional problems in children fol-
lowing motor vehicle accidents: An extended case series.
Australian and New Zealand Journal of Psychiatry, 34,
446-452.
Milam, J. E., Ritt-Olson, A., & Unger, J. B. (2004).
Posttraumatic growth among adolescents. Journal of
Adolescent Research, 19, 192-204.
*Mirza, K. A. H., Bhadrinath, B. R., Goodyer, I. M., &
Gilmour, C. (1998). Post-traumatic stress disorder in chil-
dren and adolescents following road traffic accidents.
British Journal of Psychiatry, 170, 358-362.
Olson, C. M., Rennie, D., Cook, D., Dickersin, K., Flanagin,
A., Hogan, J. W., et al. (2002). Publication bias in edito-
rial decision making. Journal of the American Medical
Association, 287, 2825-2828.
*Pelcovitz, D., Libov, B. G., Mandel, F., Kaplan, S.,
Weinblatt, M., & Septimus, A. (1998). Posttraumatic
stress disorder and family functioning in adolescent can-
cer. Journal of Traumatic Stress, 11, 205-221.
Public Health Policy Advisory Board. (1999). Health and the
American child: Risks, trends, and priorities for the 21st
century. Washington, DC: Author.
Putnam, F. W. (2003). Ten-year research update review:
child sexual abuse. Journal of the American Academy of
Child and Adolescent Psychiatry, 42, 269-278.
Reinherz, H. Z., Giaconia, R. M., Lefkowitz, E. S., Pakiz, B.,
& Frost, A. K. (1993). Prevalence of psychiatric disorders
in a community population of older adolescents. Journal
of the American Academy of Child and Adolescent
Psychiatry, 32, 369-377.
Rosenthal, R. (1991). Meta-analytic procedures for social
research (Rev. ed.). Newbury Park, CA: Sage.
Rosenthal, R. (1994). Parametric measures of effect size. In
H. Cooper & L. V. Hedges (Eds.), The handbook of research
synthesis (pp. 231-244). New York: Russell Sage Foundation.
Salter, E., & Stallard, P. (2004). Posttraumatic growth in
child survivors of a road traffic accident. Journal of
Traumatic Stress, 17, 335-340.
*Saxe, G. N., Stoddard, F. J., & Sheridan, R. (1998). PTSD
in children with burns: A longitudinal study. Burn Care &
Rehabilitation, Part 2, S206.
*Shemesh, E., Lurie, S., Stuber, M. L., Emre, S., Patel, Y.,
Vohra, P., et al. (2000). A pilot study of posttraumatic
stress and nonadherence in pediatric liver transplant
recipients. Pediatrics, 105, E29.
*Stallard, P., Velleman, R., & Baldwin, S. (1998). Prospec-
tive study of post-traumatic stress disorder in children
involved in road traffic accidents. British Medical Journal,
317, 1619-1623.
Stallard, P., Velleman, R., Langsford, J., & Baldwin, S.
(2001). Coping and psychological distress in children
involved in road traffic accidents. British Journal of
Clinical Psychology, 40, 197-208.
*Stoddard, F. J., Norman, D. K., Murphy, J. M., & Beardslee,
W. R. (1989). Psychiatric outcome of burned children
and adolescents. Journal of the American Academy of
Child and Adolescent Psychiatry, 28, 589-595.
Stuber, M. L., Christakis, D. A., Houskamp, B., & Kazak, A.
E. (1996). Posttrauma symptoms in childhood leukemia
survivors and their parents. Psychosomatics, 37, 254-261.
*Stuber, M. L., Nader, K. O., Houskamp, B. M., & Pynoos,
R. S. (1996). Appraisal of life threat and acute trauma
responses in pediatric bone marrow transplant patients.
Journal of Traumatic Stress, 9, 673-686.
Stuber, M. L., Shemesh, E., & Saxe, G. N. (2003).
Posttraumatic stress responses in children with life-
threatening illnesses. Child and Adolescent Psychiatric
Clinics of North America, 12, 195-209.
Transplant DataSource. Richmond, VA: United Network for
Organ Sharing; 2001. Retrieved from www.unos.org.
Vernberg, E. M., Silverman, W. K., La Greca, A. M., &
Prinstein, M. J. (1996). Prediction of posttraumatic stress
symptoms in children after hurricane Andrew. Journal of
Abnormal Psychology, 105, 237-248.
*Walker, A. M., Harris, G., Baker, A., Kelly, D., & Houghton,
J. (1999). Post-traumatic stress responses following liver
transplantation in older children. Journal of Child
Psychology and Psychiatry, 40, 363-374.
Woodruff, T. J., Axelrad, D. A., Kyle, A. D., Nweke, O., Miller,
G. G., & Hurley, B. J. (2004). Trends in environmentally
related childhood illnesses. Pediatrics, 113, 1133-1140.
*Zink, K. A., & McCain, G. C. (2003). Post-traumatic stress
disorder in children and adolescents with motor vehicle-
related injuries. Journal for Specialists in Pediatric Nursing,
8, 99-106.
Th
is
d
oc
um
en
t i
s c
op
yr
ig
ht
ed
b
y
th
e A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
tio
n
or
o
ne
o
f i
ts
a
lli
ed
p
ub
lis
he
rs
.
Th
is
a
rti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r t
he
p
er
so
na
l u
se
o
f t
he
in
di
vi
du
al
u
se
r a
nd
is
n
ot
to
b
e
di
ss
em
in
at
ed
b
ro
ad
ly
.
Early predictors of chronic post-traumatic stress
disorder in assault survivors
BIRGIT KLEIM 1, ANKE EHLERS 1* AND EDWARD GLUCKSMAN 2
1 Institute of Psychiatry, Department of Psychology, King’s College London, London, UK ; 2 Accident and
Emergency Department, King’s College Hospital London, London, UK
ABSTRACT
Background. Some studies suggest that early psychological treatment is effective in preventing
chronic post-traumatic stress disorder (PTSD), but it is as yet unclear how best to identify trauma
survivors who need such intervention. This prospective longitudinal study investigated the prog-
nostic validity of acute stress disorder (ASD), of variables derived from a meta-analysis of risk
factors for PTSD, and of candidate cognitive and biological variables in predicting chronic PTSD
following assault.
Method. Assault survivors who had been treated for their injuries at a metropolitan Accident and
Emergency (A&E) Department were assessed with structured clinical interviews to establish diag-
noses of ASD at 2 weeks (n=222) and PTSD at 6 months (n=205) after the assault. Candidate
predictors were assessed at 2 weeks.
Results. Most predictors significantly predicted PTSD status at follow-up. Multivariate logistic
regressions showed that a set of four theory-derived cognitive variables predicted PTSD best
(Nagelkerke R2=0.50), followed by the variables from the meta-analysis (Nagelkerke R2=0.37)
and ASD (Nagelkerke R2=0.25). When all predictors were considered simultaneously, mental
defeat, rumination and prior problems with anxiety or depression were chosen as the best combi-
nation of predictors (Nagelkerke R2=0.47).
Conclusion. Questionnaires measuring mental defeat, rumination and pre-trauma psychological
problems may help to identify assault survivors at risk of chronic PTSD.
INTRODUCTION
After experiencing a violent traumatic event,
such as an assault or terrorist attack, most
people show some symptoms of distress, but
only a minority develop persistent symptoms of
sufficient severity to warrant a diagnosis of post-
traumatic stress disorder (PTSD) (Galea et al.
2002; Rubin et al. 2005; Shalev & Freedman,
2005). Most people recover from trauma with-
out formal intervention (Kessler et al. 1995),
and there is evidence that brief early inter-
ventions given to all trauma survivors do not
prevent chronic PTSD (for reviews, see
NCCMH, 2005; Rose et al. 2002). However,
early trauma-focused cognitive-behavioural
treatment of trauma survivors with high levels
of PTSD symptoms is effective (for recent re-
views, see NCCMH, 2005; Bisson & Cohen,
2006), and this treatment may be even more
effective when delivered within the first weeks
post-trauma than at later time points (Shalev
et al. 2006). This has led to recommendations
that early trauma-focused cognitive-behavioural
treatment should be offered to trauma survivors
with severe symptoms, and that screening should
be considered for populations at risk of chronic
PTSD (e.g. NCCMH, 2005).
This raises the question of how best to
identify trauma survivors who are at risk of
* Address for correspondence: Professor Anke Ehlers,
Department of Psychology (PO77), Institute of Psychiatry, De
Crespigny Park, London SE5 8AF, UK.
(Email : anke.ehlers@iop.kcl.ac.uk)
Psychological Medicine, 2007, 37,
1457
–1467. f 2007 Cambridge University Press
doi:10.1017/S0033291707001006 First published online 22 June 2007 Printed in the United Kingdom
1457
developing chronic PTSD and need inter-
vention. One of the motivations for introducing
the diagnosis of acute stress disorder (ASD) into
DSM-IV (APA, 1994) was to help to identify,
within the first weeks post-trauma, those in-
dividuals who are likely to go on to develop
PTSD (Marshall et al. 1999). A range of em-
pirical studies found indeed that the majority of
trauma survivors with ASD go on to develop
PTSD (e.g. Harvey & Bryant, 1998; Brewin et al.
1999). However, several authors have recently
questioned the predictive validity of the ASD
diagnosis for subsequent PTSD, and have par-
ticularly challenged its sensitivity (McNally et al.
2003; Creamer et al. 2004). Studies with differ-
ent trauma populations found that on average
only 46% of those who developed PTSD had
suffered from ASD in the initial month after
trauma exposure (McNally et al. 2003) ; pro-
portions ranged from 10% (severely injured ac-
cident survivors; Schnyder et al. 2001) to 72%
(survivors of mild traumatic brain injury;
Harvey & Bryant, 2000). The low sensitivity
poses a problem for clinical purposes as it may
lead to a failure to identify individuals in need of
treatment.
A review by Bryant (2003) concluded that
psychological or biological reactions in the
acute trauma phase may provide more accurate
means of predicting PTSD than symptom-based
measures such as the ASD diagnosis, but this
hypothesis remains to be tested. The present
paper explored the question of whether risk
factors identified in previous research may help
to improve the early identification of trauma
survivors who are likely to develop PTSD.
The first set of predictors under consideration
was based on a meta-analysis of risk factors for
PTSD (Ozer et al. 2003). The meta-analysis
identified seven variables as the best empirically
established predictors of PTSD; most of these
had also emerged as risk factors in an earlier
meta-analysis (Brewin et al. 2000): peritrau-
matic dissociation (weighted effect size : r=
0.35), perceived life threat during trauma, peri-
traumatic emotional responses (both r’s=0.26),
history of trauma, history of psychological prob-
lems prior to the trauma, family history of psy-
chopathology (all r’s=0.17), and post-trauma
social support (r=x0.28). However, Ozer et al.
(2003) used the term ‘predictor’ in a statisti-
cal sense and included many cross-sectional
studies, some of which were conducted many
years the trauma. It remains to be tested
whether the risk factors identified in their meta-
analysis consistently predict chronic PTSD
when evaluated in a prospective longitudinal
design.
The second set of candidate predictors was
derived from a cognitive model of PTSD (Ehlers
& Clark, 2000). This model emphasizes factors
that maintain PTSD in order to explain why
some people develop persistent symptoms
whereas others recover on their own. A series of
prospective longitudinal studies showed that the
cognitive factors specified in the model predict
chronic PTSD at 6 months, 1 year and 3 years
(Ehlers et al. 1998b, 2003; Dunmore et al. 2001;
Mayou et al. 2002; Murray et al. 2002; Halligan
et al. 2003; Michael et al. 2005). The presumed
maintaining factors also predicted PTSD
severity at follow-up over and above what
could be predicted from initial PTSD symptom
severities. These include the candidate pre-
dictors chosen for the present study (with cor-
relations with PTSD severity at 6 months to 1
year) : mental defeat (Dunmore et al. 2001;
r=0.64), nowness of trauma memories (Michael
et al. 2005; r=0.58), negative appraisals of the
self, including negative interpretation of PTSD
symptoms (Ehlers et al. 1998b ; Dunmore et al.
2001; Halligan et al. 2003; r’s=0.43–0.62), and
rumination about the trauma (Ehlers et al.
1998b ; r=0.48).
Finally, the study included resting heart
rate as an easy-to-measure potential biological
predictor that could be used in screening pro-
grammes. A recent review found that eight out
of 10 studies showed that elevated heart rate in
the acute trauma phase (measured at varying
time points between admission to the emergency
department and 1 month afterwards) is associ-
ated with later PTSD. Elevated heart rates may
be a marker of a generalized fear response in
those who are at risk of PTSD, and may reflect
the strength of the noradrenergic response to the
trauma (Bryant, 2006). However, there was
great variability in heart rate levels and the
strength of association with subsequent PTSD
between studies (Bryant, 2006). Bryant et al.
(2000) found that a diagnosis of ASD or a rest-
ing heart rate of >90 beats/min (measured
within 20 days post-trauma) predicted PTSD
with high sensitivity (88%) and specificity
1458 B. Kleim et al.
(85%). Heart rate accounted for 4% of the
variance of PTSD severity at follow-up.
The purpose of the present prospective
longitudinal study was to compare the predic-
tive power of the ASD diagnosis, the set of seven
best predictors from Ozer et al.’s (2003) meta-
analysis, the set of four theory-derived cognitive
variables, and basal heart rate in the prediction
of PTSD at 6 months. The 13 predictors were
assessed at 2 weeks after injury in an assault. We
investigated which constellation of variables
would be most predictive of PTSD at 6 months.
METHOD
Sample characteristics
Participants had been injured in an assault
and were recruited from consecutive attendees
of King’s College Hospital’s Accident and
Emergency (A&E) Department, London, UK.
A total of 1063 assault survivors who attended
the A&E department during the recruitment
period (July 2003 to December 2004) were con-
tacted, of whom 389 did not fulfil inclusion cri-
teria and 197 declined to take part. Another 255
were initially interested, but failed to schedule or
attend the research session within the designated
time period. The remaining 222 assault sur-
vivors participated in a diagnostic session at 2
weeks, and 205 of these (92%) were interviewed
again at 6 months. Exclusion criteria were cur-
rent psychosis, assault occurring in the context
of ongoing domestic violence, and inability to
remember the assault (e.g. due to head injury).
Assaults were mainly physical (99%), and 1%
were sexual assaults. Demographic character-
istics are listed in Table 1. The study sample was
compared on demographic and assault charac-
teristics with a random sample of identical size,
drawn from 2785 assault survivors admitted to
the same A&E department during a period of 1
year. There were no significant differences in
age, sex or seriousness of injuries as measured
by triage category. However, the study sample
comprised more Caucasian participants com-
pared to the random A&E sample, x2(df=1)=
17.88, p<0.001, 58.1% v. 36.9%#.
Education correlated negatively with PTSD
at 6 months (r=x0.25, p<0.001) and age cor-
related positively (r=0.15, p<0.05). None of
the other demographic variables distinguished
between participants with and without PTSD at
6 months (all p’s>0.07). With respect to assault
characteristics, the PTSD group reported less
alcohol or drug consumption during the assault
[x2(1, n=204)=11.71, p=0.001], but the groups
did not differ significantly in other variables
(e.g. injury severity, assault duration, and
proximity to victims’ home).
Table 1. Demographic, clinical and assault
characteristics for participants who completed
assessments at both 2 weeks and 6 months
(n=205)
Variable
Sex, n (%)
Male 140 (68.0)
Female 65 (32.0)
Age, mean (S.D.) 35.0 (11.5)
Ethnicity, n (%)
Caucasian 119 (58.0)
Black 67 (32.7)
Other 19 (9.3)
Socio-economic statusa, n (%)
Very low income (£5000 or less) 59 (28.8)
Low income (£5000–£15000) 56 (27.3)
Moderate income (£15000–£30000) 43 (21.0)
High income (over £30000) 35 (17.1)
Refused information 12 (5.9)
Marital status, n (%)
Single 136 (66.3)
Married 39 (19.0)
Divorced/separated 23 (11.2)
Widowed 2 (1.0)
No information 5 (2.5)
Education, mean (S.D.) 14.1 (4.7)
Acute stress disorder at 2 weeks, n (%)
Fulfilled diagnostic criteria 34 (16.6)
Post-traumatic stress disorder at
6 months, n (%)
Fulfilled diagnostic criteria 49 (23.9)
Days since assault, mean (S.D.) 17.5 (7.65)
Number of assailants, n (%)
1 107 (52)
2 or more 97 (48)
Weapon involved, n (%)
Yes 106 (52)
No 98 (48)
Survivor under influence of
alcohol/drugs, n (%)
Yes 87 (43)
No 117 (57)
S.D., Standard deviation.
Assault characteristics were missing for one participant.
a Combined household income.
# The figure of 36.9% presents an underestimate of the number of
Caucasian participants. It is based on A&E department codes that
require the staff to code ethnicity in 80 categories based on country of
origin. In practice, the category of ‘other’ was often used for non-
British participants regardless of their ethnic background.
Early predictors of chronic PTSD 1459
Measures
Diagnostic interviews
ASD and PTSD diagnoses were established with
standard structured clinical interviews: the
Acute Stress Disorder Scale (Bryant & Harvey,
2000) and the Structured Clinical Interview
for DSM-IV (SCID; First et al. 1996). All in-
terviews were conducted by B.K. under the
supervision of A.E. Inter-rater reliability was
excellent (k=0.97 for ASD, k=0.82 for PTSD;
based on n=56 interviews, two trained raters).
Predictors from Ozer et al.’s (2003)
meta-analysis
Psychological problems prior to trauma. Par-
ticipants reported whether or not they had had
problems with anxiety or depression in the past.
This self-report measure correlated significantly
with the SCID assessment of a history of major
depression (r=0.49, p<0.001).
Family history of psychological problems. Par-
ticipants reported whether any of their close
family members had ever had a problem with
anxiety or depression.
Perceived social support following trauma. A
modified subscale from the Crisis Support Scale
(CSS; Joseph, 1999) assessed social support
soon after the assault (a=0.75). Participants
rated seven different aspects of social support
(e.g. ‘Whenever you wanted to talk, how often
was there someone willing to listen?’) on a scale
from 1 (never) to 7 (always). An original item
enquiring about personal contact with other
trauma survivors was omitted because it did not
seem to be relevant for the majority of assault
survivors.
Perceived life threat. Participants rated how
much they feared for their life during the as-
sault, on a scale from 0 (not at all) to 4 (very
strongly). This rating has been shown to predict
PTSD (Dunmore et al. 2001; Halligan et al.
2003).
Peritraumatic emotional response. This was
measured on a seven-item scale (Halligan et al.
2003; a=0.89). Participants rated the extent to
which they had experienced the following
emotions during the assault : terrified, ashamed,
helpless, fearful, guilty, horrified, frightened,
each on a scale from 0 (not at all) to 4 (very
strongly).
Peritraumatic dissociation. Dissociation during
the assault was assessed with the State
Dissociation Questionnaire (Murray et al. 2002;
a=0.88), a nine-item scale assessing different
aspects of dissociation such as derealization,
depersonalization, detachment, altered time
sense, emotional numbing, and reduction of
awareness in surroundings. The scale has been
shown to have good reliability and predictive
validity in previous studies (e.g. Halligan et al.
2002).
Theory-derived cognitive predictors
Four more variables were chosen on the basis
of earlier studies to each represent one of the
four hypothesized cognitive domains specified
in Ehlers & Clark’s (2000) model of PTSD.
The model states that people with PTSD
experience a sense of current threat that arises
from two sources : (1) excessively negative ap-
praisals of the trauma and/or its consequences,
and (2) certain characteristics of the trauma
memory that lead to easy cue-driven retrieval
(i.e. unintentional triggering by matching stim-
uli) of aspects of the trauma memory. These
memory characteristics are thought to result
from (3) the quality of cognitive processing
during the event. Finally, a range of (4) cogni-
tive and behavioural strategies (e.g. rumination,
thought suppression, excessive precautions,
and ongoing dissociation) that individuals use
to control the threat and symptoms are thought
to maintain PTSD.
Negative appraisals of the self. Negative ap-
praisals of the self were measured with the
Negative Thoughts about the Self subscale
of the Posttraumatic Cognitions Inventory
(PTCI; 21 items; a=0.93; Foa et al. 1999).
The scale measures generalized negative
appraisals of the trauma and its aftermath
(e.g. ‘My reactions since the event mean that I
am going crazy’, ‘I have permanently changed
for the worse ’, ‘I am a weak person’) and has
been shown to have good reliability, convergent
validity and to discriminate well between trau-
matized people with and without PTSD (Foa
et al. 1999).
1460 B. Kleim et al.
‘Nowness ’ of trauma memories. Trauma mem-
ories are thought to be disjointed from other
autobiographical memories, leading to a deficit
in the awareness that they reflect an experience
of the self in the past (Ehlers et al. 2004).
Perceived ‘nowness’ of traumatic memories
was measured with the corresponding item from
the Assault Memory Questionnaire (Halligan
et al. 2003) : ‘When I remember the assault,
it is like happening again, here and now’.
Self-reported ‘nowness’ correlates well with in-
terview assessment (r=0.84; Hackmann et al.
2004), and the retest reliability is r=0.68
(Speckens et al. 2006).
Mental defeat. Mental defeat refers to pro-
cessing the trauma as a complete loss of per-
sonal autonomy and has been shown to predict
PTSD in survivors of assault and torture
(Dunmore et al. 1999, 2001; Ehlers et al. 2000).
The Mental Defeat Scale (Dunmore et al. 1999,
2001; a=0.90) is an 11-item self-report ques-
tionnaire. Participants rated the extent to
which statements such as ‘I no longer felt like
a human being’ or ‘In my mind, I gave up’ ap-
plied to them at some time during the assault,
each on a scale from 0 (not at all) to 4 (very
strongly).
Rumination. Rumination about the trauma
and its consequences was measured with the
eight-item subscale of the Response to
Intrusions Questionnaire (Clohessy & Ehlers,
1999; Murray et al. 2002; a=0.84). Participants
rated how often they did things such as ‘I think
about why the event happened to me’ or ‘I go
over and over the assault ’, when memories
of the assault popped into their mind, each on
a scale from 0 (never) to 3 (always). The scale
has been shown to have good reliability and
predictive validity (Murray et al. 2002; Ehring
et al. 2006).
Resting heart rate
Heart inter-beat intervals were registered with
an S810 Polar watch (Polar Electro, Vantaa,
Finland) for a period of 3 min, while partici-
pants remained seated. Participants were in-
structed to sit quietly, not to speak and to keep
their eyes open. Heart rate data were available
for 177 participants (80%). The Polar device
has been shown to reliably assess inter-beat
intervals during rest as well as activity
(e.g. Laukkanen & Virtanen, 1998). None of
the participants were taking beta-blockers or
other medication affecting the cardiovascular
system.
Procedure
The study was approved by the local ethics
committees, and participants gave written
consent. Participants completed a research
session comprising the completion of ques-
tionnaires assessing the psychological predictor
variables, heart rate assessment and the
ASD Scale (Bryant & Harvey, 2000) interview.
At 6 months, the SCID (First et al. 1996) was
conducted over the telephone by the same
interviewer. Participants received £50 as reim-
bursement for their time.
Statistical analysis
SPSS version 13.0 (SPSS Inc., Chicago, IL,
USA) was used for most analyses ; bootstrap
analyses were implemented by Stata Statistics
and Data Analysis Package version 9.0
(Stata Corporation, College Station, TX, USA).
Predictors were z-standardized to allow direct
comparison of odds ratios (ORs). Association
between candidate predictors and PTSD
were first investigated with univariate logistic
regression analyses. Hierarchical logistic re-
gression analysis tested which of the predictors
explained PTSD at 6 months over and
above what could be predicted from the ASD
diagnosis, by forcing ASD into the equation
in the first step and the predictor in the
second step. In addition, two multivariate
logistic regression analyses tested the predictive
power of the variable sets from Ozer et al.’s
(2003) meta-analysis and from the cognitive
model (including only significant univariate
predictors). Finally, a logistic regression analy-
sis of all significant univariate predictors
identified the combination of predictors that
best predicted PTSD at 6 months; this model
determined which variables to add to the
model based on the likelihood ratio test.
Reliability of univariate results was established
by calculating confidence intervals (CIs) for
ORs of all predictors with bootstrap resampling
(Efron & Tibshirani, 1986), which involved
drawing 1000 random subsamples of the total
sample.
Early predictors of chronic PTSD 1461
RESULTS
ASD and PTSD prevalence
At 2 weeks, 17% of the participants (n=37/222)
met diagnostic criteria forASD, and at 6months,
24% (n=49/205) met diagnostic criteria for
PTSD.
Early predictors of chronic PTSD
Table 2 shows the associations between the
predictors assessed at 2 weeks and PTSD at
6 months. Most univariate associations were
significant. Participants with ASD were at in-
creased risk for PTSD at 6 months compared
to those without ASD (non-transformed OR
11.7, w=0.46). Table 2 shows that although the
majority of participants with ASD went on to
develop chronic PTSD (68%), ASD had low
sensitivity (47%) in predicting PTSD.
Most of the predictors derived from Ozer
et al.’s (2003) meta-analysis were significantly
associated with a greater PTSD risk (prior
Table 2. Associations of predictors at 2 weeks with PTSD diagnosis at 6 months
Predictor
Diagnosis at 6 months
Univariate logistic regression
Multiple logistic regression Sign predictor
after controlling
for ASDPTSD No PTSD ORa 95% CI
Boot-
strapped
CI ORa 95% CI
Acute stress disorder R2=0.25 (95% CI 0.15–0.35), x2=36.89***
ASD diagnosis
Yes 23 11 2.50*** 1.84–3.41 1.13–2.96 2.50*** 1.84–3.41 N.A.
No 26 145
Risk factors from the
Ozer et al. (2003)
meta-analysis
R2=0.37 (95% CI 0.24–0.47), x2=48.22***
Prior trauma
Yes 46 142 1.47 0.87–2.47 0.28–1.44 Not includedb 0.225
No 1 46
Prior psychological
problems
Yes 31 45 2.10*** 1.50–2.95 0.84–3.18 2.06** 1.35–3.16 0.001
No 16 107
Family history of
psychological problems
Yes 14 37 1.15 0.83–1.58 0.75–2.34 Not includedb 0.623
No 32 115
Perceived life threat,
mean (S.D.)
2.25 (1.59) 1.05 (1.46) 2.12*** 1.52–2.96 0.78–2.27 1.22 0.75–1.98 0.006
Peritraumatic emotional
response, mean (S.D.)
2.61 (0.96) 1.66 (1.04) 2.77*** 1.85–4.14 0.83–3.06 2.43** 1.32–4.50 0.001
Perceived social support,
mean (S.D.)
3.35 (1.34) 3.79 (1.17) 0.70* 0.50–1.00 0.29–1.40 0.64* 0.42–3.16 0.112
Peritraumatic dissociation,
mean (S.D.)
2.24 (0.94) 1.31 (1.09) 2.35*** 1.65–3.52 1.04–3.04 1.42 0.86–2.35 0.011
Theory-derived cognitive
measures
R2=0.50 (95% CI 0.38–0.59), x2=68.91***
Mental defeat, mean (S.D.) 2.15 (1.09) 0.89 (0.80) 3.76*** 2.51–5.65 1.29–3.51 1.74 0.97–3.11 0.000
Nowness of trauma
memories, mean (S.D.)
2.21 (1.53) 0.73 (1.10) 2.89*** 1.99–4.19 1.18–4.81 1.67* 1.00–2.79 0.000
Negative appraisals of the
self, mean (S.D.)
3.47 (1.20) 2.19 (0.96) 3.18*** 2.09–4.82 1.36–4.75 1.99* 1.13–3.50 0.000
Rumination, mean (S.D.) 1.79 (0.60) 1.00 (0.58) 4.19*** 2.52–7.01 1.16–3.26 1.92* 1.04–3.56 0.000
Biological measure R2=0.04 (95% CI 0.00–0.11), x2=4.90*
Resting heart rate,
mean (S.D.)
71.80 (12.35) 67.10 (10.80) 1.50* 1.04–2.15 0.77–2.99 1.50* 1.04–2.15 0.049
PTSD, Post-traumatic stress disorder ; ASD, acute stress disorder ; S.D., standard deviation; N.A., not applicable ; CI, confidence interval.
a Predictors are z-standardized to make odds ratios comparable ; b as non-significant in univariate analysis.
Questionnaire ranges : mental defeat 0–4, nowness of trauma memories 0–4, negative appraisals about the self 1–7, rumination 0–4,
perceived life threat 0–4, perceived support 1–7, peritraumatic dissociation 0–4.
*** p<0.001, ** p<0.01, * p<0.05.
1462 B. Kleim et al.
psychological problems, less post-trauma social
support, greater perceived threat to life, peri-
traumatic emotional responses and dissociation
during the assault). Associations with prior
trauma (non-transformed OR 4.54, 95% CI
0.58–35.45, w=0.11) and family history of
psychological problems (non-transformed OR
1.36, 95% CI 0.66–2.82, w=0.06) did not
reach significance. All theory-derived cognitive
variables (mental defeat, nowness of memories,
negative appraisals of the self, and rumination
about the trauma) and resting heart rate pre-
dicted PTSD. Bootstrap analyses for 1000 sub-
samples showed that ORs and CIs replicated
well. Table 2 also shows which of the variables
predicted PTSD at 6 months over and above
what could be predicted from an ASD diagnosis
at 2 weeks. These were the four theory-derived
cognitive variables, prior psychological prob-
lems, perceived life threat, peritraumatic
emotional response and dissociation, and rest-
ing heart rate.
Next, we compared the predictive power
of the sets of variables under investigation
(Table 2). The theory-derived cognitive vari-
ables explained 50% of the variance (sensitivity
0.60, specificity 0.95, 88% correct classifi-
cations). CIs indicated that the strength of the
prediction was significantly larger than the 25%
explained by an ASD diagnosis (sensitivity 0.47,
specificity 0.93, 82% correct classifications),
and the 4% explained by resting heart rate at 2
weeks (sensitivity 0.00, specificity 1.00). The es-
tablished predictors identified in Ozer et al.’s
(2003) meta-analysis explained 37% of the
variance (sensitivity 0.44, specificity 0.95, 85%
correct classifications), which was not signifi-
cantly different from the predictive power of the
ASD diagnosis.
When all significant predictors were con-
sidered in an overall multiple logistic regression
analysis, mental defeat (OR 2.07, p=0.014,
95% CI 1.16–3.70), rumination about the trau-
ma (OR 2.99, p=0.002, 95% CI 1.50–5.96) and
prior psychological problems (OR 1.95,
p=0.014, 95% CI 1.14–3.31) were chosen as the
best combination of predictors by the forward
likelihood ratio test (x2=47.37, p=0.000).
Together, the three variables explained 47% of
the variance (sensitivity 0.57, specificity 0.96)
and classified 88% of the cases correctly into
PTSD and non-PTSD status. The Hosmer and
Lemeshow goodness-of-fit test showed that the
model fit the data well (x2=9.86, p=0.275). Not
selected were ASD (after last step p=0.184),
perceived threat to life (p=0.452), peritraumatic
dissociation (p=0.941), peritraumatic emotion-
al response (p=0.695), nowness of trauma
memories (p=0.138), negative appraisals about
the self (p=0.589) and resting heart rate
(p=0.828).
DISCUSSION
This prospective longitudinal study found that
17% of 222 injured assault survivors met diag-
nostic criteria for ASD at 2 weeks after the
event, and at 6 months, 24% had PTSD. These
prevalences are similar to those reported in
other studies of assault survivors (e.g. Harvey
& Bryant, 1998; Brewin et al. 1999). While the
majority of participants with ASD went on to
develop chronic PTSD (68%), ASD had low
sensitivity (47%) in predicting PTSD, again in
accord with previous studies (Brewin et al. 1999;
Creamer et al. 2004; Elklit & Brink, 2004; see
McNally et al. 2003 for a review). One of the
reasons for the low sensitivity of ASD may be
that not all assault survivors at risk for PTSD
may have had the three dissociative symptoms
required for a diagnosis. Overall, the usefulness
of ASD as a criterion for identifying people at
risk of PTSD was limited.
This raises the question of whether there are
other ways of identifying people at risk for
chronic PTSD (e.g. Shalev et al. 1997; Brewin
et al. 2002). The present study tested three
groups of candidate predictors of PTSD ident-
ified in previous research, namely the seven best
established predictors of PTSD according to the
meta-analysis of Ozer et al. (2003), four pre-
dictors derived from a cognitive model of PTSD
(Ehlers & Clark, 2000), and resting heart rate as
a candidate biological predictor. In line with
previous findings, nearly all of these variables
predicted PTSD at 6months (Ehlers et al. 1998b ;
Brewin et al. 2000; Bryant et al. 2000; Dunmore
et al. 2001; Murray et al. 2002; Halligan et al.
2003; Ozer et al. 2003). The results extend pre-
vious findings in several ways. (1) They demon-
strated that the predictors not only are
associated with concurrent PTSD but also pre-
dict chronic PTSD prospectively ; this strength-
ens Ozer et al.’s (2003) findings, which were
Early predictors of chronic PTSD 1463
based largely on cross-sectional studies. (2) The
variables were predictive even when assessed
2 weeks after the trauma, which extends the
results of previous prospective longitudinal
studies where the initial assessment took place
later (e.g. Dunmore et al. 2001). (3) The study
allowed a direct comparison of the strength
of the prediction of different sets of variables
suggested as relevant in the literature. (4)
Recruitment from an A&E department ensured
a sample that was representative in terms of the
severity of assaults and most other demographic
characteristics.
Overall, the results replicated well those of the
Ozer et al. (2003) meta-analysis. As in the meta-
analysis, peritraumatic variables (peritraumatic
dissociation and emotions, perceived life threat)
and social support tended to be stronger pre-
dictors than pre-trauma characteristics (see
also Brewin et al. 2000). The finding that the
relationships between prior trauma and a family
history of psychological problems and PTSD
did not reach significance in this study should
not be overinterpreted. Given the modest PTSD
rate in the sample, our study had low power for
detecting very small effect sizes. Restrictions in
range may have played a role as the majority of
the participants reported prior trauma.
However, it is also possible that the type of
trauma or timing of assessment plays a role in
the strength of the relationship of these vari-
ables with PTSD, as other studies of single-
event trauma have also failed to establish a
relationship between family history and PTSD
(Blanchard & Hickling, 2004) and as Ozer et al.
(2003) identified three previous studies of recent
trauma survivors (f4 months), all of which
found no relationship between trauma history
and PTSD (r’s=0.00–0.03).
The set of predictors from the Ozer et al.
(2003) meta-analysis did not significantly pre-
dict chronic PTSD better than the ASD diag-
nosis. A possible explanation may be that these
variables mainly predict the onset of symptoms,
but are less useful for the prediction of long-
term outcome as maintaining factors are not
included (e.g. Ehlers & Steil, 1995). By contrast,
the theory-derived cognitive factors each pre-
dicted PTSD at 6 months over and above what
can be predicted from an ASD diagnosis at
2 weeks, and together explained more variance
than ASD. This finding may reflect the fact
that PTSD symptoms show substantial natural
recovery in the first months after trauma
(e.g. Rothbaum et al. 1992; Kessler et al. 1995),
and factors that impede recovery and maintain
PTSD symptoms may therefore be of particular
interest in predicting chronic PTSD (e.g. Ehlers
& Steil, 1995; Ehlers & Clark, 2000).
The logistic regression analysis that con-
sidered all candidate predictors simultaneously
identified a combination of two of such theory-
derived maintaining variables (mental defeat,
rumination) and a vulnerability variable (psy-
chological problems prior to the assault) as
the best predictor set. This suggests that the
combination of vulnerability and maintaining
factors may provide the best prediction of
chronic PTSD. The combination of these three
predictors predicted diagnostic status at 6
months correctly in 88% of the cases. Together,
the three variables explained almost half of the
variance in PTSD diagnosis at 6 months, com-
pared to 25% of explained variance by the ASD
diagnosis.
In line with previous research, resting heart
rate at 2 weeks predicted PTSD at 6 months, but
only explained a small proportion of the vari-
ance. Bryant (2006) proposed multiple path-
ways to PTSD development, some of which are
characterized by elevated and others by lowered
heart rate levels (alongside dissociative re-
sponses). Such differences in response pattern
would be consistent with the present results, but
limit the usefulness of basal heart rate as a way
of early screening for risk of chronic PTSD.
Heart rate may be more predictive if it is as-
sessed sooner after the trauma, although the
currently available data show great variation for
the predictive power of such early assessments
(Bryant, 2006). It is also possible that the parti-
cipants’ heart rate may have been influenced by
memories triggered by the questionnaires that
were part of the research session, although such
responses would have been expected to increase
rather than decrease the relationship between
heart rate and PTSD.
The study had strengths and limitations.
Among the strength are the prospective longi-
tudinal design, the relatively large number of
participants, the excellent follow-up rate, the
use of reliable structured interviews to establish
diagnoses, the replication of the results in
bootstrapping analyses, and the early initial
1464 B. Kleim et al.
assessment. Nevertheless, further research will
need to establish how well the present results
generalize to other populations with different
PTSD rates and other forms of trauma before
the results are applied to clinical practice. The
sample showed a moderate PTSD rate of 24%
at follow-up, and had low power to detect very
small effect sizes. Thirteen potential predictors
were considered, which raises the possibility of
false positive associations. However, all of the
predictors had been shown to be associated with
PTSD risk in previous studies, and the present
sample replicated earlier results well and showed
good replication in bootstrapping analyses.
Although our analyses indicated that the sample
was fairly representative of assault survivors
attending the A&E department, it comprised
relatively fewer people of ethnic minorities
compared to the population of assault survivors
treated at this hospital (nevertheless, the 42%
exceeded the 24% ethnic minorities living in
London according to the 2001 Census). The
concept of mental defeat, one of the strongest
predictors in this study, was developed with
survivors of interpersonal violence such as
assault and torture (e.g. Dunmore et al. 1999,
2001; Ehlers et al. 1998a, 2000), and other pre-
dictors may be more relevant for survivors of
other traumas such as accidents or natural
disaster. A further possible limitation was that
PTSD diagnoses at 6 months were based on
a diagnostic interview conducted over the tele-
phone, rather than a face-to-face diagnostic
interview. It may be argued that telephone in-
terviews make participants somewhat less forth-
coming in providing information about their
symptoms than face-to-face interviews. How-
ever, the interviewer had already established a
relationship with the participants at the first as-
sessment, which makes it unlikely that partici-
pants withheld information from her later.
Finally, personal and family history of problems
with anxiety or depression was assessed with
simple self-report questions, and more sophisti-
cated assessments may have given different re-
sults.
In conclusion, this study found that PTSD
at 6 months after injury in an assault can be
predicted from early assessment at 2 weeks.
A set of four theory-derived cognitive variables
predicted PTSD better than the ASD diagnosis.
Logistic regression analysis suggested that
the combination of mental defeat, rumination
and a history of problems with anxiety or
depression predicted PTSD best. These concepts
can be measured with only 20 self-report
questions. However, it remains to be tested how
the three selected scales perform on their
own, as they were selected from a larger group
of 11 significant predictors in this study. If
replicated, our results suggest that assault
survivors are at heightened PTSD risk if they
report mental defeat at the time of the event,
if they ruminate about the trauma, and if they
report prior problems with depression or anxi-
ety. Early assessment of these predictors
appears straightforward and may be useful
for screening programmes in general practice
or communities affected by trauma. They
may help to identify those trauma survivors
who are most likely to develop PTSD and
who would benefit from early intervention.
Putting the experience of mental defeat in
perspective and reducing rumination appear
to be promising targets for such interventions.
ACKNOWLEDGEMENTS
This study was funded by a grant from the
Psychiatry Research Trust and a Wellcome
Trust Principal Research Fellowship to Anke
Ehlers. We thank Thomas Ehring, Silke Frank,
Inga Boellinghaus, Emma Briddon, Anke
Weidmann, Ines Sengstock, Johanna Hissbach,
Jennifer Baumeister, Stephanie Spengler and the
staff of King’s College Accident and Emergency
Department for their help.
DECLARATION OF INTEREST
None.
REFERENCES
APA (1994). Diagnostic and Statistical Manual of Mental Disorders
(4th edn). American Psychiatric Association: Washington, DC.
Bisson, J. I. & Cohen, J. A. (2006). Disseminating early interventions
following trauma. Journal of Traumatic Stress 19, 583–595.
Blanchard, E. B. & Hickling, E. J. (2004). After the Crash:
Psychological Assessment and Treatment of Survivors of Motor
Vehicle Accidents (2nd edn). American Psychiatric Association:
Washington, DC.
Brewin, C. R., Andrews, B., Rose, S. & Kirk, M. (1999). Acute stress
disorder and posttraumatic stress disorder in victims of violent
crime. American Journal of Psychiatry 156, 360–366.
Brewin, C. R., Andrews, B. & Valentine, J. D. (2000). Meta-analysis
of risk factors for posttraumatic stress disorder in trauma-
exposed adults. Journal of Consulting and Clinical Psychology 68,
748–766.
Early predictors of chronic PTSD 1465
Brewin, C. R., Rose, S., Andrews, B., Green, J., Tata, P., McEvedy,
C., Turner, S. & Foa, E. B. (2002). Brief screening instrument for
post-traumatic stress disorder. British Journal of Psychiatry 181,
158–162.
Bryant, R. A. (2003). Early predictors of posttraumatic stress dis-
order. Biological Psychiatry 53, 789–795.
Bryant, R. A. (2006). Longitudinal psychophysiological studies of
heart rate: mediating effects and implications for treatment.
Annals of the New York Academy of Sciences 1071, 19–26.
Bryant, R. A. & Harvey, A. G. (2000). Acute Stress Disorder: A
Handbook of Theory, Assessment, and Treatment. American
Psychological Association: Washington, DC.
Bryant, R. A., Harvey, A. G., Guthrie, R. M. &Moulds, M. L. (2000).
A prospective study of psychophysiological arousal, acute stress
disorder, and posttraumatic stress disorder. Journal of Abnormal
Psychology 109, 341–344.
Clohessy, S. & Ehlers, A. (1999). PTSD symptoms, response to
intrusive memories and coping in ambulance service workers.
British Journal of Clinical Psychology 38, 251–265.
Creamer, M., O’Donnell, M. L. & Pattison, P. (2004). The relation-
ship between acute stress disorder and posttraumatic stress dis-
order in severely injured trauma survivors. Behaviour Research and
Therapy 43, 315–328.
Dunmore, E., Clark, D. M. & Ehlers, A. (1999). Cognitive factors
involved in the onset and maintenance of posttraumatic stress
disorder (PTSD) after physical or sexual assault. Behaviour
Research and Therapy 37, 809–829.
Dunmore, E., Clark, D. M. & Ehlers, A. (2001). A prospective inves-
tigation of the role of cognitive factors in persistent posttraumatic
stress disorder (PTSD) after physical or sexual assault. Behaviour
Research and Therapy 39, 1063–1084.
Efron, B. & Tibshirani, R. (1986). Bootstrap methods for standard
errors, confidence intervals and other measures of statistical
accuracy. Statistical Science 1, 54–77.
Ehlers, A. & Clark, D. M. (2000). A cognitive model of posttraumatic
stress disorder. Behaviour Research and Therapy 38, 319–345.
Ehlers, A., Clark, D. M., Dunmore, E., Jaycox, L., Meadows, E.
& Foa, E. B. (1998a). Predicting response to exposure treatment in
PTSD: the role of mental defeat and alienation. Journal of
Traumatic Stress 11, 457–471.
Ehlers, A., Hackmann, A. & Michael, T. (2004). Intrusive re-
experiencing in post-traumatic stress disorder : phenomenology,
theory, and therapy. Memory 12, 403–415.
Ehlers, A., Maercker, A. & Boos, A. (2000). Posttraumatic stress
disorder following political imprisonment: the role of mental de-
feat, alienation, and perceived permanent change. Journal of
Abnormal Psychology 109, 45–55.
Ehlers, A., Mayou, R. A. & Bryant, B. (1998b). Psychological
predictors of chronic posttraumatic stress disorder after
motor vehicle accidents. Journal of Abnormal Psychology 107,
508–519.
Ehlers, A., Mayou, R. A. & Bryant, B. (2003). Cognitive predictors of
posttraumatic stress disorder in children: results of a prospective
longitudinal study. Behaviour Research and Therapy 41, 1–10.
Ehlers, A. & Steil, R. (1995). Maintenance of intrusive memories in
posttraumatic stress disorder: a cognitive approach. Behavioural
and Cognitive Psychotherapy 23, 217–249.
Ehring, T., Ehlers, A. & Glucksman, E. (2006). Contribution of cog-
nitive factors to the prediction of post-traumatic stress disorder,
phobia and depression after motor vehicle accidents. Behaviour
Research and Therapy 44, 1699–1716.
Elklit, A. & Brink, O. (2004). Acute stress disorder as a predictor of
post-traumatic stress disorder in physical assault victims. Journal
of Interpersonal Violence 19, 709–726.
First, M. B., Spitzer, R. L., Gibbon, M. & Williams, J. B. W. (1996).
Structured Clinical Interview for DSM-IV Axis I Disorders.
American Psychiatric Press : Washington, DC.
Foa, E. B., Ehlers, A., Clark, D. M., Tolin, D. F. & Orsillo, S. M.
(1999). The Posttraumatic Cognitions Inventory (PTCI) :
development and validation. Psychological Assessment 11, 303–
314.
Galea, S., Ahern, J., Resnick, H., Kilpatrick, D., Bucuvalas, M., Gold,
J. & Vlahov, D. (2002). Psychological sequelae of the September 11
terrorist attacks in New York City. New England Journal of
Medicine 346, 982–987.
Hackmann, A., Ehlers, A., Speckens, A. & Clark, D. M. (2004).
Characteristics and content of intrusive memories in PTSD and
their changes with treatment. Journal of Traumatic Stress 17,
231–240.
Halligan, S. L., Clark, D. M. & Ehlers, A. (2002). Cognitive pro-
cessing, memory, and the development of PTSD symptoms: two
experimental analogue studies. Journal of Behavior Therapy and
Experimental Psychiatry 33, 73–89.
Halligan, S. L., Michael, T., Clark, D. M. & Ehlers, A. (2003).
Posttraumatic stress disorder following assault : the role of cogni-
tive processing, trauma memory, and appraisals. Journal of
Consulting and Clinical Psychology 71, 419–431.
Harvey, A. G. & Bryant, R. A. (1998). The relationship between acute
stress disorder and posttraumatic stress disorder: a prospective
evaluation of motor vehicle accident survivors. Journal of
Consulting and Clinical Psychology 66, 507–512.
Harvey, A. G. & Bryant, R. A. (2000). A two-year pro-
spective evaluation of the relationship between acute stress
disorder and posttraumatic stress disorder following mild
traumatic brain injury. American Journal of Psychiatry 157, 626–
628.
Joseph, S. (1999). Social support and mental health following trau-
ma. In Post-traumatic Stress Disorder: Concepts and Therapy
(ed. W. Yule), pp. 71–91. Wiley : Chichester.
Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M. & Nelson, C. B.
(1995). Posttraumatic stress disorder in the National Comorbidity
Survey. Archives of General Psychiatry 52, 1048–1060.
Laukkanen, R. M. T. & Virtanen, P. (1998). Heart rate monitors :
state of the art. Journal of Sports Sciences 16, 3–7.
Marshall, R. D., Spitzer, R. & Liebowitz, M. R. (1999). Review and
critique of the new DSM-IV diagnosis of acute stress disorder.
American Journal of Psychiatry 156, 1677–1685.
Mayou, R. A., Ehlers, A. & Bryant, B. (2002). Posttraumatic stress
disorder after motor vehicle accidents : 3-year follow-up of a
prospective longitudinal study. Behaviour Research and Therapy
40, 665–675.
McNally, R. J., Bryant, R. A. & Ehlers, A. (2003). Does early
psychological intervention promote recovery from posttraumatic
stress? Psychological Science in the Public Interest 4, 45–79.
Michael, T., Ehlers, A., Halligan, S. & Clark, D. M. (2005).
Unwanted memories of assault : what intrusion charac-
teristics predict PTSD? Behaviour Research and Therapy 43, 613–
628.
Murray, J., Ehlers, A. & Mayou, R. A. (2002). Dissociation and
post-traumatic stress disorder : two prospective studies of road
traffic accident survivors. British Journal of Psychiatry 180, 363–
368.
NCCMH (2005). Post-traumatic stress disorder: the management of
PTSD in adults and children in primary and secondary care.
National Collaborating Centre for Mental Health. National
Institute for Clinical Excellence (NICE): London, UK.
Ozer, E. J., Best, S. R., Lipsey, T. L. & Weiss, D. S. (2003).
Predictors of posttraumatic stress disorder and symptoms in
adults : a meta-analysis. Psychological Bulletin 129, 52–73.
Rose, S., Bisson, J., Churchill, R. & Wesseley, S. (2002). Psycho-
logical debriefing for preventing posttraumatic stress disorder. The
Cochrane Database of Systematic Reviews. Issue 2. Article No.:
CD00560.
Rothbaum, B. O., Foa, E. B., Riggs, D. S., Murdock, T. &
Walsh, W. (1992). A prospective examination of posttraumatic
stress disorder in rape victims. Journal of Traumatic Stress 5,
455–475.
Rubin, G. J., Brewin, C. R., Greenberg, N., Simpson, J. &
Wessely, S. (2005). Psychological and behavioural reactions to
the bombings in London on 7 July 2005: cross sectional survey of a
representative sample of Londoners. British Medical Journal 331,
606.
1466 B. Kleim et al.
Schnyder, U., Moergli, H., Klaghofer, R. & Buddeberg, C. (2001).
Incidence and prediction of posttraumatic stress disorder symp-
toms in severely injured accident victims. American Journal of
Psychiatry 158, 594–599.
Shalev, A. Y. & Freedman, S. (2005). PTSD following terrorist at-
tacks: a prospective evaluation. American Journal of Psychiatry
170, 558–564.
Shalev, A. Y., Freedman, S., Adessky, R. & Watson, P. (2006). Who
needs care, who wants care, who is helped by early intervention:
5600 trauma survivors’ results. Abstract no. 159877. In
Symposium at the 22nd Annual Meeting of the International
Society for Traumatic Stress Studies, Hollywood, CA, USA.
Shalev, A. Y., Freedman, S., Peri, T., Brandes, D. & Sahar, T. (1997).
Predicting PTSD in trauma survivors: prospective evaluation of
self-report and clinician-administered instruments. British Journal
of Psychiatry 170, 558–564.
Speckens, A. E., Ehlers, A., Hackmann, A. & Clark, D. M. (2006).
Changes in intrusive memories associated with imaginal reliving in
posttraumatic stress disorder. Journal of Anxiety Disorders 20,
328–341.
Early predictors of chronic PTSD 1467
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Am J Psychiatry 166:3, March 2009 293
Reviews and Overviews
ajp.psychiatryonline.org
Systematic Review and Meta-Analysis of Multiple-Session
Early Interventions Following Traumatic Events
Neil P. Roberts, D.Clin.Psy.
Neil J. Kitchiner, M.Sc.
Justin Kenardy, Ph.D.
Jonathan I. Bisson, D.M.,
F.R.C.Psych.
Objective: The authors sought to deter-
mine the efficacy of multiple-session psy-
chological interventions to prevent and
treat traumatic stress symptoms beginning
within 3 months of a traumatic event.
Method: Nine computerized databases
were searched, and manual searches
were conducted of reference lists of se-
lected articles as well as two journals. In
addition, key researchers in the field were
contacted to determine whether they
were aware of other relevant studies. The
reviewers identified randomized con-
trolled trials of multiple-session psycho-
logical treatments aimed at preventing or
reducing traumatic stress symptoms in in-
dividuals within 3 months of exposure to
a traumatic event. Details of the studies
were independently extracted by two re-
viewers, and outcome data were entered
into the Review Manager software pack-
age. Quality assessment was also con-
ducted by two researchers independently.
Results: Twenty-five studies examining a
range of interventions were identified.
For treatment of individuals exposed to a
trauma irrespective of their symptoms,
there was no significant difference be-
tween any intervention and usual care.
For treatment of traumatic stress symp-
toms irrespective of diagnosis, trauma-fo-
cused cognitive-behavioral therapy (CBT)
was more effective than waiting list or
supportive counseling conditions. The dif-
ference was greatest for treatment of
acute stress disorder and acute posttrau-
matic stress disorder.
Conclusions: Trauma-focused CBT within
3 months of a traumatic event appears to
be effective for individuals with traumatic
stress symptoms, especially those who
meet the threshold for a clinical diagnosis.
(Am J Psychiatry 2009; 166:293–301)
There is now a large body of literature to show that a va-
riety of traumatic experiences can cause significant psy-
chological difficulties for large numbers of people (see ref-
erences 1–6, for example). Many individuals show great
resilience in the face of such experiences and will manifest
short-lived or subclinical stress reactions that diminish
over time (7), and most people recover without medical or
psychological assistance (8). Nevertheless, a range of psy-
chological difficulties may develop following trauma in
some of those who have been exposed, including acute
stress disorder and posttraumatic stress disorder (PTSD).
The rate of acute stress disorder has been reported to be
13% in motor vehicle accident survivors (9) and 19% in
victims of violent crime (10). Reported rates of acute PTSD
(defined as PTSD symptoms for less than 3 months) have
varied across different trauma populations, from 23% in
motor vehicle accident victims (6) to 47% in rape victims
(11). Epidemiological research suggests that one-third of
individuals who develop acute PTSD remain symptomatic
for 6 years or longer (12). The impact on social, interper-
sonal, and occupational functioning can be marked and
enduring for those who develop chronic PTSD (13).
Over the past three decades, clinicians have been in-
creasingly involved in attempts to develop interventions
that might mitigate the effects of trauma and prevent the
onset of chronic PTSD. Although psychological debriefing
(also known as critical incident stress debriefing) was a
widely used intervention for some years, it came under in-
creasing scrutiny in the 1990s. Systematic reviews (14, 15)
failed to find evidence for the efficacy of single-session in-
dividual debriefing, and many experts in the field now
caution against its use (16, 17). Increasingly the field has
turned its attention to other models of intervention (13,
18–21). A common theme has been the suggestion that ef-
forts should be focused on identifying those most at risk of
developing ongoing problems in the aftermath of trau-
matic incidents and directing resources and interventions
mainly to them.
Although there is considerable evidence for the efficacy
of multiple-session trauma-focused psychological inter-
ventions to treat chronic PTSD (22–24), the early use of
such interventions has not received the same level of scru-
tiny. A number of randomized controlled trials have been
conducted with early interventions, but the issues of how
effective they are, whom to offer intervention to, the tim-
ing of intervention, and the mode of intervention remain
contentious. To help clarify the potential utility of these
interventions, we performed a systematic review and
meta-analysis of randomized controlled trials of psycho-
294 Am J Psychiatry 166:3, March 2009
EARLY INTERVENTIONS FOLLOWING TRAUMATIC EVENTS
ajp.psychiatryonline.org
logical interventions aimed at preventing or treating PTSD
within 3 months of a traumatic event.
Method
Data Sources
The electronic databases MEDLINE, ClinPSYC, PsychLit, EM-
BASE, PILOTS, LILACS, PSYNEBS, SocioFile, CINAHL, and the Co-
chrane Depression, Anxiety, and Neurosis Group Trials Register
were searched until July 2007 using the Cochrane optimal ran-
domized controlled trial search strategy combined with the fol-
lowing keywords: PTSD, trauma, acute stress disorder, acute
posttraumatic stress disorder, acute post-traumatic stress disor-
der, early intervention, early psychological intervention, preven-
tative, prevention, PTSD prevention, crisis, crisis intervention,
psychological first aid, cognitive, behaviour, behavior, behav-
ioural, behavioral, cognitive-behavioural, cognitive-behavioral,
exposure, eye movement desensitization and reprocessing, psy-
chological, psychotherapy, psychodynamic, stress inoculation,
relaxation, anxiety management, psychoeducation, collaborative
care, collaborative intervention, recovery, facilitating recovery,
critical incident stress debriefing, debriefing, critical incident
stress management, counseling, counselling, supportive counsel-
ling, and supportive counseling. Manual searches were under-
taken of the Journal of Traumatic Stress and the Journal of Con-
sulting and Clinical Psychology. Reference lists of studies
identified in the search, related review articles, and management
guidelines were scrutinized. Internet searches of known web sites
and discussion forums were conducted, and key researchers in
TABLE 1. Summary of Meta-Analysis of Results for Interventions for PTSD Symptoms Within 3 Months After a Traumatic Eventa
Comparison Follow-Up
Number
of Trials
Total
N
Interventions beginning within 1 month for all exposed to
the traumatic event
Any intervention vs. usual care (PTSD symptoms self-report) Posttreatment (37, 40–43) 5 431
3–6 months after trauma (41–43) 3 234
Any intervention vs. usual care (PTSD diagnosis) Posttreatment (38–43) 6 470
3–6 months after trauma (38, 39, 41, 43) 4 303
Interventions beginning within 3 months for individuals
with traumatic stress symptoms
Structured writing vs. minimal intervention (PTSD symptoms
self-report)
Posttreatment (53; Bugg et al., unpublished data)
3 months after trauma (Bugg et al., unpublished data)
2
1
146
104
Structured writing vs. minimal intervention (PTSD diagnosis) Posttreatment (53) 1 42
Trauma-focused CBT vs. waiting list (PTSD symptoms
clinician rated)
Posttreatment (18, 47, 50–52; Öst et al., unpublished data)
3–5 months after trauma (51, 52)
6
2
423
138
9–11 months after trauma (50, 51) 2 73
12–18 months after trauma (18) 1 152
Trauma-focused CBT vs. waiting list (PTSD diagnosis) Posttreatment (18, 47, 50–53; Öst et al., unpublished data) 7 515
3–5 month follow-up (51, 52) 2 141
9–11 months after trauma (50, 51) 2 54
12–18 months after trauma (18) 1 116
Trauma-focused CBT vs. supportive counseling (PTSD
symptoms self-report)
Posttreatment (30, 31, 46, 51)
3–6 months follow-up (30, 31, 46, 51)
4
4
195
176
2–4 years after trauma (Clinician Administered PTSD Scale)
(33, 34)
2 94
Trauma-focused CBT vs. supportive counseling (PTSD
diagnosis)
Posttreatment (30, 31, 35, 46, 51)
3–6 months after trauma (30, 31, 35, 46, 51)
5
5
251
200
3–4 years after trauma (33, 34) 2 94
Interventions for individuals with acute stress disorder or
acute PTSD within 3 monthsd
Trauma-focused CBT vs. waiting list (PTSD symptoms
clinician rated)
Posttreatment (47, 50, 52; Öst et al., unpublished data)
3–5 months after trauma (52)
4
1
210
95
9–11 months after trauma (50) 1 12
Trauma-focused CBT vs. waiting list (PTSD diagnosis) Posttreatment (47, 50, 52, 53; Öst et al., unpublished data) 5 254
3–5 months after trauma (52) 1 95
9–11 months after trauma (50) 1 12
Trauma-focused CBT vs. supportive counseling (PTSD
symptoms clinician rated)
Posttreatment (30, 31, 46)
3–6 months follow-up (30, 31, 46)
3
3
138
134
2–4 years after trauma (Clinician Administered PTSD Scale)
(33, 34)
2 94
Trauma-focused CBT vs. supportive counseling (PTSD
diagnosis)
Posttreatment (30, 31, 35, 46)
3–6 months after trauma (30, 31, 35, 46)
4
4
191
158
3–4 years after trauma (33, 34) 2 94
a PTSD=posttraumatic stress disorder; CBT=cognitive-behavioral therapy.
b Relative risk of diagnosis of PTSD; a value of 1 indicates that the intervention was the same as the control condition; a value <1 indicates that
the intervention was better than the control condition; and a value >1 indicates that the control condition was better than the intervention.
c The standardized mean difference in continuous PTSD symptom score was used when different outcome measures were used in the studies
included; the weighted mean difference in continuous PTSD symptom score was used when the same outcome measure was used in the
studies included. A score of 0 indicates that there was no difference between the intervention and the control condition, a score <0 indicates
that the intervention was better, and a score >0 indicates that the control condition was better.
d These analyses are subanalyses of the analyses of interventions beginning within 3 months for individuals with traumatic stress symptoms.
*p<0.05.
Am J Psychiatry 166:3, March 2009 295
ROBERTS, KITCHINER, KENARDY, ET AL.
ajp.psychiatryonline.org
the field were contacted to determine whether they were aware of
other relevant studies.
Study Selection
All abstracts were read independently by two of the reviewers
to determine whether they potentially met the inclusion criteria.
If either reviewer thought the study potentially met the criteria,
the full manuscript was obtained and read independently by
three of the reviewers. To be included, a study had to be a ran-
domized controlled trial that considered one or more defined
psychological interventions or treatments (excluding single-ses-
sion interventions) aimed at preventing or reducing traumatic
stress symptoms following events that appeared to fulfill DSM-IV
criterion A1 for PTSD or acute stress disorder in comparison with
a placebo control, other control (e.g., usual care or waiting list
control), or alternative psychological treatment condition. All
studies had to have been completed and analyzed by September
2007. Presence or absence of symptoms, sample size, language,
and publication status were not used to determine whether a
study should be included. The review considered studies involv-
ing adults only.
Data Extraction
A data extraction sheet was designed to record data, which
were then entered into the Review Manager software package,
version 4.2 (25). Information extracted included demographic de-
tails of participants, details of the traumatic event, the random-
ization process, the interventions used, dropout rates, and out-
come data. Quality was assessed by rating studies according to
randomization, allocation concealment, blinding of assessors,
and intention to treat. Each study was also rated using quality as-
sessment criteria derived from the literature on key methodolog-
ical issues pertinent to psychological intervention research on
traumatic stress (26–29). Data were extracted and quality as-
sessed by two reviewers independently. Any disagreements were
discussed with a third reviewer and a consensus achieved.
Data Synthesis
Given the differences between studies with respect to partici-
pants’ symptom severity and the interval between exposure to a
traumatic event and commencement of the intervention, we sep-
arated the trials into three groups based on work previously con-
ducted in this area (16): studies that offered an intervention to
any individual exposed to a traumatic event irrespective of their
symptoms with the aim of preventing PTSD; studies providing in-
terventions begun within 3 months with the aim of preventing
PTSD or ongoing distress in individuals with traumatic stress
symptoms; and studies providing interventions begun within 3
months with the aim of preventing PTSD or ongoing distress in
individuals with acute stress disorder or acute PTSD.
In order to combine information from several studies, all inter-
ventions offered to any individual exposed to a traumatic event
with the aim of preventing PTSD were considered together. The
efficacy of trauma-focused cognitive-behavioral therapy (CBT)
was considered in individuals with traumatic stress symptoms.
Trauma-focused CBT was defined as any intervention that fo-
cused on the trauma using exposure to trauma memories and
trauma reminders with or without cognitive therapy and other
cognitive-behavioral techniques. The exposure-based therapy
with anxiety management (30) and exposure-based therapy with
hypnosis (31) arms in two studies were combined with the expo-
sure therapy arm to generate a single mean and standard devia-
tion. The combined results were then compared with the waiting
list arm to avoid double counting.
The data were analyzed for summary effects using Review Man-
ager. Continuous outcomes were analyzed using weighted mean
differences when all trials measured outcome on the same scale.
When some trials measured outcomes on different scales, stan-
dardized mean differences were used, based on the assumption
that all scales measure the same underlying symptom or condition.
Relative risk was calculated for categorical outcome measures, and
95% confidence intervals were computed for all outcomes.
Available case analysis and intent-to-treat analysis with impu-
tation using the last-observation-carried-forward method were
performed when enough information was available. In cases
where the information presented in the paper was inadequate to
perform these analyses, further information was requested from
the lead author.
Heterogeneity between studies was assessed by observing the
I2 test of heterogeneity, which measures the percentage of varia-
tion that is not due to chance (32). An I2 of less than 30% was
taken to indicate mild heterogeneity, and a fixed-effects model
Relative
Riskb 95% CI
Standardized or
Weighted Mean
Difference in
PTSD Scorec 95% CI
0.05 –0.26 to 0.36
0.26 0.00 to 0.52
0.82 0.56 to 1.20
0.59 0.27 to 1.25
–0.28
1.10
–1.22 to 0.65
–3.80 to 6.00
0.61 0.25 to 1.47
–0.54
–0.22
–0.93 to –0.16*
–0.56 to 0.11
–0.85 –2.49 to 0.79
–6.01 –12.44 to 0.42
0.72 0.5 to 1.05
0.64 0.42 to 0.99*
0.42 0.03 to 5.23
0.74 0.36 to 1.51
–0.95
–0.62
–1.66 to –0.23*
–0.94 to –0.31*
–0.85 –1.29 to –0.40*
0.56 0.29 to 1.06
0.37 0.20 to 0.67*
0.28 0.13 to 0.58*
–0.85
–0.88
–1.35 to –0.34*
–6.52 to 4.76
–33.67 –52.77 to –14.57*
0.54 0.31 to 0.95*
0.85 0.50 to 1.44
0.09 0.01 to 1.35
–20.66
–0.77
–28.81 to –12.51*
–1.14 to –0.40*
–0.85 –1.29 to –0.40*
0.44
0.26
0.22 to 0.86*
0.16 to 0.45*
0.28 0.13 to 0.58*
296 Am J Psychiatry 166:3, March 2009
EARLY INTERVENTIONS FOLLOWING TRAUMATIC EVENTS
ajp.psychiatryonline.org
was used. When the I2 was 30% or greater, a random-effects
model was used. A visual inspection of the forest plots was used
as a test of the robustness of these findings.
Results
Two hundred fifty titles and abstracts were identified
through the search process, and 49 papers were reviewed
in detail by three of the authors independently to establish
whether they met the inclusion criteria. Twenty-five stud-
ies were found to meet the inclusion criteria, and another
two papers (33, 34) reported long-term follow-up on three
studies (30, 31, 35). Twenty-four studies were reported in
English and one (36) in French. A flow diagram of the sys-
tematic review and a table summarizing the characteris-
tics of the studies are presented in the data supplement
that accompanies the online edition of this article.
Synthesis of Results
The outcomes for individual studies are indicated in Ta-
ble S1 in the online data supplement. The postinterven-
tion and follow-up results of the meta-analyses for com-
parisons that included more than one study are listed in
Table 1, and examples of forest plots are presented in Fig-
ures 1–3. The outcomes reported are rates of PTSD and se-
verity of PTSD (clinician rated unless unavailable, in
which case self-report data are used).
Studies offering intervention to individuals involved
in a traumatic event irrespective of symptoms.
Eight studies (36–42, 43) evaluated brief psychosocial in-
terventions aimed at preventing PTSD in individuals ex-
posed to a specific traumatic event. All started within 1
month of the trauma. Meta-analysis of the studies with
sufficient data available showed no significant differences
between those who received an intervention and those
who did not (see Table 1; see also Figure S1 in the online
data supplement). The only statistically significant differ-
ences observed for specific interventions were in favor of
the waiting list control group over adapted critical inci-
dent stress debriefing for self-reported PTSD symptoms
immediately after the trauma (41) and for preventive
FIGURE 1. Forest Plot Comparing Any Multiple-Session Early Intervention and Treatment as Usual for Prevention of PTSD
in Individuals Exposed to a Traumatic Event, With PTSD Diagnosis as Outcome Measurea
a Total events: treatment, 36; control, 43. Test for heterogeneity: χ2=7.97, df=5, p=0.16, I2=37.2%. Test for overall effect: z=1.01, p=0.31.
FIGURE 2. Forest Plot Comparing Trauma-Focused Cognitive-Behavioral Therapy and Waiting List in Individuals Exposed to
a Traumatic Event Who Have Symptoms of PTSD or Acute Stress Disorder, With PTSD Diagnosis as Outcome Measurea
a Total events: treatment, 119; control, 142. Test for heterogeneity: χ2=21.01, df=6, p=0.002, I2=71.4%. Test for overall effect: z=1.71, p=0.09.
Study Treatment Control
Favors Treatment Favors Control Weight
(%)
Relative Risk (Fixed)
(95% CI)
Ryding et al. 1998 (39) 2/50 1/49 2.31 1.96 (0.18, 20.92)
Gidron et al. 2001 (44) 1/8 4/9 8.61 0.28 (0.04, 2.02)
Zatzick et al. 2001 (43) 2/14 6/15 13.25 0.36 (0.09, 1.48)
Ryding et al. 2004 (40) 10/82 15/65 38.28 0.53 (0.25, 1.10)
Gamble et al. 2005 (38) 18/50 16/53 35.53 1.19 (0.69, 2.07)
Marchand et al. 2006 (41) 3/33 1/42 2.01 3.82 (0.42, 35.04)
Total 237 233 100.00 0.82 (0.56, 1.20)
0.1 1.0 10.05.02.00.50.2
Relative Risk (Fixed)
(95% CI)
Study Treatment Control
Favors Treatment Favors Control Weight
(%)
Relative Risk (Fixed)
(95% CI)
Bryant et al. 2008 (47) 10/30 23/30 15.82 0.43 (0.25, 0.75)
Öst (unpublished) 3/23 13/20 7.71 0.20 (0.07, 0.60)
Ehlers et al. 2003 (50) 1/6 5/6 3.55 0.20 (0.03, 1.24)
Bisson et al. 2004 (48) 35/76 29/76 19.18 1.21 (0.83, 1.76)
Foa et al. 2006 (51) 21/31 19/30 19.39 1.07 (0.74, 1.54)
Sijbrandij et al. 2007 (52) 40/79 43/64 21.05 0.75 (0.57, 0.99)
van Emmerik et al. 2008 (53) 9/21 10/23 13.30 0.99 (0.50, 1.94)
Total 266 249 100.00 0.72 (0.50, 1.05)
0.1 1.0 10.05.02.00.50.2
Relative Risk (Fixed)
(95% CI)
Am J Psychiatry 166:3, March 2009 297
ROBERTS, KITCHINER, KENARDY, ET AL.
ajp.psychiatryonline.org
counseling over monitoring or usual care for clinician-as-
sessed PTSD symptoms 3 months after trauma (38). Two
studies (44, 45) evaluated a two-session memory structur-
ing intervention in individuals who had been involved in a
motor vehicle accident and had a heart rate >95 bpm in
the emergency department. No statistically significant dif-
ferences were found between the interventions and the
control conditions.
Studies offering intervention to individuals with
traumatic stress symptoms within 3 months after a
traumatic event. Fifteen studies (30, 31, 35, 46–55; A.
Bugg et al., unpublished 2007 data; L. Öst et al., unpub-
lished 2004 data) evaluated interventions for individuals
who had traumatic stress symptoms within 3 months after
a traumatic event. No statistically significant differences
were observed between structured writing and minimal
intervention. Statistically significant differences were ob-
served in favor of trauma-focused CBT over the waiting list
condition and supportive counseling at posttreatment as-
sessment. Follow-up data were incomplete, but statisti-
cally significant differences were present at several time
points, particularly over supportive counseling.
Studies offering intervention to individuals with a di-
agnosis of acute stress disorder or acute PTSD within
3 months after a traumatic event. Eleven studies (30,
31, 35, 46, 47 , 49, 50, 52, 53; Bugg et al., unpublished 2007
data; Öst et al., unpublished 2004 data) offered interven-
tions to individuals with a diagnosis of acute stress disorder
or acute PTSD. The analyses of these studies are subanaly-
ses of the analyses of symptomatic individuals described
above, with four studies excluded. Statistically significant
differences were observed in favor of trauma-focused CBT
over the waiting list control condition and supportive coun-
seling. The only evidence to support any other form of
treatment was for cognitive restructuring, which was signif-
icantly better than the waiting list condition but less effec-
tive than trauma-focused CBT in one study (47).
Methodological Quality of Included Studies
Only 12 studies adequately described a method of allo-
cation judged to make no bias possible (31, 37, 38, 43, 46–
48, 50, 52, 53, 55; Bugg et al., unpublished 2007 data). Re-
porting of adequate concealment procedures was present
in only seven studies (31, 37, 43, 47, 48, 50; Bugg et al., un-
published 2007 data). Adequate blinding of the assessor of
outcome measures was present in 16 studies (30, 31, 35, 38,
41, 43, 44–48, 50, 51, 53, 55); Bugg et al., unpublished 2007
data). Loss to follow-up was fully reported with reasons by
group in 16 studies (31, 35, 38, 41, 43, 46–54; Bugg et al., un-
published 2007 data; Öst et al., unpublished 2004 data).
The overall quality of the studies in relation to the other
methodological and reporting factors considered was
variable. Fewer than 10 studies fully reported whether
training was offered to assessors and how performance,
supervision, or reliability checks of assessors were per-
formed (four studies), whether treatment fidelity was
independently checked and adequate (five studies),
whether power calculation was reported (four studies),
whether follow-up extended beyond 6 months (nine stud-
ies), and whether there were any side effects (one study).
In order to determine the impact of quality on outcome,
the three studies included in the analysis of trauma-fo-
cused CBT versus waiting list for individuals with trau-
matic stress symptoms (47, 48 , 50) that achieved the high-
est quality ratings for the four Cochrane quality criteria
were included in a sensitivity analysis. This resulted in a
reduction in the magnitude of intervention effect, with su-
periority over waiting list control only approaching statis-
tical significance (three studies, N=224, standardized
mean difference –0.75, 95% CI=–1.53 to 0.02).
Discussion
Main Findings
There was no evidence that a multiple-session interven-
tion aimed at everyone, irrespective of their symptoms,
FIGURE 3. Forest Plot Comparing Trauma-Focused Cognitive-Behavioral Therapy and Waiting List in Individuals Exposed to
a Traumatic Event Who Have Symptoms of PTSD or Acute Stress Disorder, Excluding Those Not Meeting Full Diagnostic Cri-
teria, With PTSD Diagnosis as Outcome Measurea
a Total events: treatment, 47; control, 76. Test for heterogeneity: χ2=12.11, df=4, p=0.02, I2=67.0%. Test for overall effect: z=2.15, p=0.03.
Study Treatment Control
Favors Treatment Favors Control Weight
(%)
Relative Risk (Fixed)
(95% CI)
Bryant et al. 2008 (47) 10/30 23/30 25.83 0.43 (0.25, 0.75)
Öst (unpublished) 3/23 13/20 14.74 0.20 (0.07, 0.60)
Ehlers et al. 2003 (50) 1/6 5/6 7.43 0.20 (0.03, 1.24)
Sijbrandij et al. 2007 (52) 24/50 24/45 29.26 0.86 (0.59, 1.28)
van Emmerik et al. 2008 (53) 9/21 10/23 22.75 0.99 (0.50, 1.94)
Total 130 124 100.00 0.54 (0.31, 0.95)
0.1 1.0 10.05.02.00.50.2
Relative Risk (Fixed)
(95% CI)
298 Am J Psychiatry 166:3, March 2009
EARLY INTERVENTIONS FOLLOWING TRAUMATIC EVENTS
ajp.psychiatryonline.org
following a traumatic event was effective. Trauma-focused
CBT was significantly better than waiting list or usual care
at reducing traumatic stress symptoms in individuals who
were symptomatic at entry into the study, but the magni-
tude of effect varied. The magnitude was largest for indi-
viduals who were diagnosed with acute stress disorder or
acute PTSD. Evidence of the benefits of trauma-focused
CBT for symptomatic individuals who did not meet full di-
agnostic criteria for these conditions was weak.
Heterogeneity
There was evidence of both clinical and statistical het-
erogeneity in the included studies.
Although all the trials attempted to reduce traumatic
stress symptoms, the nature of the interventions was quite
diverse. This was partially dealt with by separating inter-
ventions into predetermined groups, although we decided
to combine the interventions provided to individuals irre-
spective of their symptoms in order to maximize the infor-
mation available. There were more studies evaluating
trauma-focused CBT than other interventions, but the spe-
cific interventions in the trauma-focused CBT group were
not identical. All were trauma focused, but some interven-
tions primarily used a prolonged exposure paradigm (e.g.,
reference 48), whereas others (e.g., reference 50) primarily
used cognitive techniques with more limited exposure. It
was not possible to conduct a formal analysis comparing
studies with varying amounts of exposure because of lim-
ited detail regarding the specific amount of exposure deliv-
ered. However, there did not appear to be a direct relation-
ship between more exposure and improved outcome or
vice versa among interventions included in the trauma-fo-
cused CBT group. This seems at odds with the Institute of
Medicine’s recent conclusion that exposure therapy is the
only type of psychological treatment with sufficient evi-
dence of efficacy in the treatment of PTSD (56), although
exposure therapy was superior to cognitive restructuring
with no exposure in the single study included in this review
that directly compared these two interventions (47).
In addition, the total number of hours of intervention
provided varied from around 4 hours to around 16 hours.
There were also differences in the clinical populations, es-
pecially with regard to the severity of symptoms at entry to
the studies. On the basis of previous work on this topic
(16), we grouped trials in a clinically meaningful manner
according to the intervention and the clinical populations
included, but this approach is not empirically based,
which should be considered when interpreting the results.
We concluded that the trials grouped together were essen-
tially trying to measure the same thing and that it was
worthwhile summarizing their combined results, but the
variation means that caution should be applied when in-
terpreting the results (32).
Methodological Quality
The overall quality of the studies was varied. Using the
Cochrane quality criteria, 16 (64%) studies fully reported
loss to follow-up with reasons, 15 (60%) described using
appropriately blinded assessors to measure outcome, 12
(48%) described appropriate randomization with no bias
possible, and only seven (28%) reported adequate alloca-
tion concealment. The small sample sizes of most of the
studies are also an important limitation. However, the in-
tervention and control groups appeared well matched at
baseline in most studies, reducing the risk of the re-
ported unadjusted mean outcomes being influenced by
baseline differences.
Several studies, including those that provided more pos-
itive results, had strong methodological characteristics. A
meta-analysis of the highest-quality studies resulted in a
larger effect size (0.75) for the efficacy of trauma-focused
CBT versus waiting list control than the meta-analysis of
all studies irrespective of quality (0.54), but the former
meta-analysis just failed to reach statistical significance.
This is probably a power issue and, contrary to previous
research (57), does not suggest that poorer-quality studies
falsely elevated the apparent efficacy of the intervention.
The choice of control condition is particularly impor-
tant in early intervention research, where a reduction in
symptoms over the duration of the trial would be expected
given the natural course of traumatic stress reactions (58).
The development of a psychological treatment placebo is
very difficult, if not impossible, as is blinding of partici-
pants and therapists. Some of the waiting list and usual
care groups may have received some form of intervention
by virtue of contact through symptom monitoring, but
this was not properly evaluated, and it is not possible to
determine what impact this might have had on outcomes.
Only one study (47) reported adverse effects, and it is
unclear whether any occurred in other studies. The drop-
out rates were mostly no higher in the intervention groups
than in the control groups across the studies reviewed,
which suggests that the interventions did not cause major
adverse effects. However, the absence of tolerability as-
sessment is a key shortcoming in the trials identified and
one that has previously been noted in psychological treat-
ment studies of chronic PTSD (22).
Implications for Practice
The results suggest that no psychological intervention
can be recommended for routine use following traumatic
events. This is consistent with the results of single-session
interventions, although, in contrast to them (14), with the
possible exception of adapted critical incident stress de-
briefing, no evidence was found of any harm occurring as
a result of an intervention. Trauma-focused CBT was the
only early intervention with convincing evidence of effi-
cacy in reducing and preventing traumatic stress symp-
toms, but only for symptomatic individuals and particu-
larly for those who met the diagnostic criteria for acute
Am J Psychiatry 166:3, March 2009 299
ROBERTS, KITCHINER, KENARDY, ET AL.
ajp.psychiatryonline.org
stress disorder or acute PTSD. The less convincing evi-
dence in favor of trauma-focused CBT for all symptomatic
individuals raises some interesting clinical implications.
Positive outcomes in the meta-analysis for all symptom-
atic individuals appear to have been bolstered by the out-
comes from studies focusing specifically on individuals
meeting all diagnostic criteria for acute stress disorder and
acute PTSD. This suggests that the presence of a specific
diagnosis may be the most important predictor of who will
benefit from trauma-focused CBT. However, when plan-
ning how best to detect such individuals, it is important to
heed the research suggesting that merely screening for
acute stress disorder is problematic as it misses many in-
dividuals who go on to develop PTSD (59).
While the majority of symptomatic individuals are likely
to gain some benefit from trauma-focused CBT 1 to 3
months after a traumatic event, the magnitude of this ben-
efit may not be very large. Whether the magnitude of im-
provement is likely to be significant enough to justify the
routine provision of trauma-focused CBT to all symptom-
atic individuals is open to debate. The evidence suggests
that trauma-focused CBT should be offered to all who suf-
fer from acute stress disorder or acute PTSD and that lim-
iting it to this group can be justified, particularly when re-
sources are limited. The results for non-trauma-focused
CBT interventions were disappointing, but it remains pos-
sible that elements from them are effective, particularly if
used with more symptomatic individuals. For example,
behavioral reactivation (54) has clearly not yet been evalu-
ated with an adequately powered trial and would benefit
from further evaluation.
The results of this review support calls that have been
made for a stepped or stratified care system whereby those
with the most severe symptoms are offered more complex
interventions (18). The fact that trauma-focused CBT ap-
pears to be an effective treatment suggests that more work
should be done to determine whether it could be delivered
as part of a screening program after major traumatic events.
Implications for Research
Further well-designed randomized controlled trials of
trauma-focused CBT starting within the first 3 months af-
ter traumatic events with longer follow-up periods are
needed. Given the modest overall effects of trauma-fo-
cused CBT, the development and trialing of other psycho-
logical treatments are important. The finding that expo-
sure therapy was superior to cognitive restructuring in one
study requires replication, and the comparison of treat-
ments with more or less exposure should be pursued in
the future.
Most of the studies included in this review attempted to
evaluate individual psychological therapy. Given the im-
portant role of social support as a predictor of outcome
(60, 61), it would be of interest to examine interventions
aimed at couples and families to improve familial re-
sponse. It would also be of interest to evaluate forms of
community intervention and interventions aimed at im-
proving coping skills and enhancing positive and helpful
behaviors (J.I. Ruzek, unpublished 2007 paper). Future re-
search should also consider adverse events and tolerabil-
ity of treatment, carefully control for additional interven-
tion, and explore the optimal time to intervention, how
long treatment should last, and whether other techniques
can be incorporated into existing treatments to improve
their efficacy.
Received April 25, 2008; revisions received July 18 and Oct. 1, 2008;
accepted Oct. 27, 2008 (doi: 10.1176/appi.ajp.2008.08040590). From
Cardiff and Vale NHS Trust, Wales; University of Queensland, Austra-
lia; and Cardiff University, Wales. Address correspondence and reprint
requests to Dr. Bisson, Cardiff University, Monmouth House, Univer-
sity Hospital of Wales, Heath Park, Cardiff, CF14 4XW, Wales, UK;
bissonji@cf.ac.uk (e-mail).
All authors report no competing interests.
The authors thank the authors of studies in the review for providing
unpublished data, Helen Davies for administrative support, and Hugh
McGuire for help with searches and with translation. Jonathan Bisson
had full access to all of the data in the study and takes responsibility
for the integrity of the data and the accuracy of the data analysis.
References
1. Goenjian A: A mental health relief programme in Armenia af-
ter the 1988 earthquake: implementation and clinical obser-
vations. Br J Psychiatry 1993; 163:230–239
2. Green BL, Grace MC, Lindy JD, Gleser GC, Leonard AC, Crum-
mier TL: Buffalo Creek survivors in the second decade: com-
parison with unexposed and nonlitigant groups. J Appl Soc Psy-
chol 1990; 20:1033–1050
3. Kulka RA, Schlenger WE, Fairbank JA, Jordan BK, Hough RL,
Marmar CR, Weiss DS: Trauma and the Vietnam War Genera-
tion: Report of Findings From the Vietnam Veterans Study.
New York, Bruner/Mazel, 1990
4. Kilpatrick DG, Saunders BE, Veronen LJ, Best CL, Von JM: Crimi-
nal victimization: lifetime prevalence reporting to police, and
psychological impact. Crime Delinq 1987; 33:479–489
5. North CS, Smith EM, Spitznagel EL: Postraumatic stress disorder
in survivors of a mass shooting. Am J Psychiatry 1994; 151:82–
88
6. Ehlers A, Mayou RA, Bryant B: Psychological predictors of
chronic posttraumatic stress disorder after motor vehicle acci-
dents. J Abnorm Psychol 1998; 107:508–519
7. Bonanno GA: Loss, trauma, and human resilience: have we un-
derestimated the human capacity to thrive after extremely
aversive events? Am Psychol 2004; 59:20–28
8. McNally RJ, Bryant RA, Ehlers A: Does early psychological inter-
vention promote recovery from posttraumatic stress disorder?
Psychological Science in the Public Interest 2003; 4:45–79
9. Harvey AG, Bryant RA: The relationship between acute stress
disorder and posttraumatic stress disorder: a prospective eval-
uation of motor vehicle accident survivors. J Consult Clin Psy-
chol 1998; 66:507–512
10. Brewin CR, Andrews B, Rose S, Kirk M: Acute stress disorder
and posttraumatic stress disorder in victims of violent crime.
Am J Psychiatry 1999; 156:360–366
11. Rothbaum BO, Foa EB, Riggs DS, Murdock T, Walsh W: A pro-
spective examination of post-traumatic stress disorder in rape
victims. J Trauma Stress 1992; 5:455–475
12. Kessler R, Sonnega A, Bromet E, Hughes M, Nelson CB: Post-
traumatic stress disorder in the National Comorbidity Survey.
Arch Gen Psychiatry 1995; 52:1048–1060
300 Am J Psychiatry 166:3, March 2009
EARLY INTERVENTIONS FOLLOWING TRAUMATIC EVENTS
ajp.psychiatryonline.org
13. Litz BT, Gray MJ: Early intervention for trauma in adults: a
framework for first aid and secondary prevention, in Early In-
tervention for Trauma and Traumatic Loss. Edited by Litz BT.
New York, Guilford, 2004
14. Rose S, Bisson J, Churchill R, Wessely S: Psychological debriefing
for preventing post traumatic stress disorder (PTSD). Cochrane
Database Syst Rev 2005; issue 3
15. van Emmerik AA, Kamphuis JH, Hulsbosch AM, Emmelkamp
PM: Single session debriefing after psychological trauma: a
meta-analysis. Lancet 2002; 360:766–771
16. National Collaborating Centre for Mental Health: Post-Trau-
matic Stress Disorder: The Management of PTSD in Adults and
Children in Primary and Secondary Care. London, Gaskell,
2005
17. Australian Centre for Posttraumatic Mental Health: Australian
Guidelines for the Treatment of Adults With Acute Stress Disor-
der and Posttraumatic Stress Disorder. Melbourne, 2007
18. Bisson JI, Roberts N, Macho G: Service innovations: the Cardiff
Traumatic Stress Initiative: an evidence-based approach to
early psychological intervention following traumatic events.
Psychiatr Bull 2003; 27:145–147
19. Ehlers A, Clark DM: Early psychological interventions for adult
survivors of trauma: a review. Biol Psychiatry 2003; 53:817–
826
20. Litz B, Gray M, Bryant R, Adler A: Early intervention for trauma:
current status and future directions. Clin Psychol 2002; 9:112–
134
21. Gray MJ, Litz BT: Behavioral interventions for recent trauma.
Behav Modif 2005; 29:189–215
22. Bisson JI, Ehlers A, Matthews R, Pilling S, Richards D, Turner S:
Psychological treatments for chronic post-traumatic stress dis-
order: systematic review and meta-analysis. Br J Psychiatry
2007; 190:97–104
23. Bradley R, Greene J, Russ E, Dutra L, Westen D: A multidimen-
sional meta-analysis of psychotherapy for PTSD. Am J Psychia-
try 2005; 162:214–227
24. Bisson J, Andrew M: Psychological treatment of post-traumatic
stress disorder. Cochrane Database Syst Rev 2007; 3:CD003388
25. Cochrane Collaboration: Review Manager (RevMan) version 4.2
for Windows. Oxford, Cochrane Collaboration, 2003
26. Moncrieff J, Churchill R, Drummond C: Development of a qual-
ity assessment for trials of treatment for depression and neuro-
sis. Int J Methods Psychiatr Res 2001; 10:126–133
27. Foa EB, Meadows EA: Psychosocial treatments for posttrau-
matic stress disorder: a critical review. Annu Rev Psychol 1997;
48:449–480
28. Hertlein KM, Ricci RJ: A systematic research synthesis of EMDR
studies: implementation of the platinum standard. Trauma Vi-
olence Abuse 2004; 5:285–300
29. Kenardy J, Carr V: Imbalance in the debriefing debate: what we
don’t know outweighs what we do. Bulletin of the Australian
Psychological Society 1996; 17:4–6
30. Bryant RA, Sackville T, Dang ST, Moulds M, Guthrie R: Treating
acute stress disorder: an evaluation of cognitive behavior ther-
apy and supportive counseling techniques. Am J Psychiatry
1999; 156:1780–1786
31. Bryant RA, Moulds ML, Guthrie RM, Nixon RDV: The additive
benefits of hypnosis and cognitive-behavioral therapy in treat-
ing acute stress disorder. J Consult Clin Psychol 2005; 73:334–
340
32. Fletcher J: What is heterogeneity and is it important? Br Med J
2007; 334:94–96
33. Bryant RA, Moulds ML, Nixon RVD: Cognitive behaviour ther-
apy of acute stress disorder: a four-year follow-up. Behav Res
Ther 2003; 41:489–494
34. Bryant RA, Moulds ML, Nixon RDV, Mastrodomenico J, Felming-
ham K, Hopwood S: Hypnotherapy and cognitive behaviour
therapy of acute stress disorder: a 3-year follow-up. Behav Res
Ther 2006; 44:1331–1335
35. Bryant RA, Harvey AG, Dang ST, Sackville T, Basten C: Treatment
of acute stress disorder: a comparison of cognitive-behavioral
therapy and supportive counseling. J Consult Clin Psychol
1998; 66:862–866
36. André C, Lelord F, Légeron P, Reignier A, Delattre A: [Effective-
ness of early intervention on 132 bus drivers victims of aggres-
sions: a controlled trial]. L’Encephale 1997; 23:65–71 (French)
37. Kazak AE, Simms S, Alderfer MA, Rourke MT, Crump T, McClure
K, Jones P, Rodriguez A, Boeving A, Hwang W, Reilly A: Feasibil-
ity and preliminary outcomes from a pilot study of a brief psy-
chological intervention for families of children newly diag-
nosed with cancer. J Pediatr Psychol 2005; 30:644–655
38. Gamble J, Creedy D, Moyle W, Webster J, McAllister M, Dickson
P: Effectiveness of a counseling intervention after a traumatic
childbirth: a randomized controlled trial. Birth 2005; 32:11–19
39. Ryding E, Wijma K, Wijma B: Postpartum counselling after an
emergency caesarean. Clin Psychol Psychother 1998; 5:231–
237
40. Ryding El, Wiren E, Johansson G, Ceder B, Dahlstrom A: Group
counseling for mothers after emergency cesarean section: a
randomized controlled trial of intervention. Birth 2004; 31:
247–253
41. Marchand A, Guay S, Boyer R, Iucci S, Martin A, St-Hilaire M: A
randomized controlled trial of an adapted form of individual
critical incident stress debriefing for victims of an armed rob-
bery. Brief Treat Crisis Interv 2006; 6:122–129
42. Brom D, Kleber RJ, Hofman MC: Victims of traffic accidents: in-
cidence and prevention of post-traumatic stress disorder. J Clin
Psychol 1993; 49:131–140
43. Zatzick D, Roy-Byrne P, Russo JE, Rivara FP, Koike A, Jurkovich
GJ, Katon W: Collaborative interventions for physically injured
trauma survivors: a pilot randomized effectiveness trial. Gen
Hosp Psychiatry 2001; 23:114–123
44. Gidron Y, Gal R, Freedman S, Twiser I, Lauden A, Snir Y, Ben-
jamin J: Translating research findings to PTSD prevention: re-
sults of a randomized-controlled pilot study. J Trauma Stress
2001; 14:773–780
45. Gidron Y, Gal R, Givati G, Lauden A, Snir Y, Binjamin J: Interac-
tive effects of memory structuring and gender in preventing
posttraumatic stress symptoms. J Nerv Ment Dis 2007; 195:1–4
46. Bryant RA, Moulds M, Guthrie R, Nixon RDV: Treating acute
stress disorder following mild traumatic brain injury (letter).
Am J Psychiatry 2003; 160:585–587
47. Bryant RA, Mastrodomenico J, Felmingham KL, Hopwood S,
Kenny L, Kandris E, Cahill C, Creamer M: Treatment of acute
stress disorder: a randomized controlled trial. Arch Gen Psychi-
atry 2008; 65:659–667
48. Bisson JI, Shepherd JP, Joy D, Probert R, Newcombe RG: Early
cognitive-behavioural therapy for post-traumatic stress symp-
toms after physical injury. Br J Psychiatry 2004; 184:63–69
49. Echeburua E, de Corral P, Sarasua B, Zubizarreta I: Treatment
of acute posttraumatic stress disorder in rape victims: an ex-
perimental study. J Anxiety Disord 1996; 10:185–199
50. Ehlers A, Clark D, Hackmann A, McManus F, Fennell M, Herbert
C, Mayou R: A randomized controlled trial of cognitive therapy,
a self-help booklet, and repeated assessments as early inter-
ventions for posttraumatic stress disorder. Arch Gen Psychiatry
2003; 60:1024–1032
51. Foa EB, Zoellner LA, Feeny NC: An evaluation of three brief pro-
grams for facilitating recovery after assault. J Trauma Stress
2006; 19:29–43
52. Sijbrandij M, Olff M, Reitsma JB, Carlier IVE, de Vries MH, Ger-
sons BPR: Treatment of acute posttraumatic stress disorder
with brief cognitive behavioral therapy: a randomized con-
trolled trial. Am J Psychiatry 2007; 164:82–90
Am J Psychiatry 166:3, March 2009 301
ROBERTS, KITCHINER, KENARDY, ET AL.
ajp.psychiatryonline.org
53. van Emmerik AA, Kamphuis JH, Emmelkamp PM: Treating
acute stress disorder and posttraumatic stress disorder with
cognitive behavioral therapy or structured writing therapy: a
randomized controlled trial. Psychother Psychosom 2008; 77:
93–100
54. Wagner AW, Zatzick DF, Ghesquiere A, Jurkovich GJ: Behavioral
activation as an early intervention for posttraumatic stress dis-
order and depression among physically injured trauma survi-
vors. Cogn Behav Pract 2007; 4:341–349
55. Zatzick D, Roy-Byrne P, Russo J, Rivara F, Droesch R, Wagner A,
Dunn C, Jurkovich G, Uehara E, Katon W: A randomized effec-
tiveness trial of stepped collaborative care for acutely injured
trauma survivors. Arch Gen Psychiatry 2004; 61:498–506
56. Committee on Treatment of Posttraumatic Stress Disorder, In-
stitute of Medicine of the National Academies: Treatment of
Posttraumatic Stress Disorder: An Assessment of the Evidence.
Washington, DC, National Academies Press, 2008
57. Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher M, Tug-
well P, Klassen TP: Does quality of reports of randomised trials
affect estimates of intervention efficacy in meta-analyses? Lan-
cet 1998; 352:609–613
58. Galea S, Vlahov D, Resnick H, Ahern J, Susser E, Gold J, Bucuv-
alas M, Kilpatrick D: Trends of probable post-traumatic stress
disorder in New York City after the September 11 terrorist at-
tacks. Am J Epidemiol 2003; 158:514–524
59. Bryant R: Acute stress disorder. Psychiatry 2006; 5:238–239
60. Brewin CR, Andrew B, Valentine JD: Meta-analysis of risk fac-
tors for posttraumatic stress disorder in trauma-exposed
adults. J Consult Clin Psychol 2000; 68:748–766
61. Ozer EJ, Best SR, Lipsey TL, Weiss DS: Predictors of posttrau-
matic stress disorder and symptoms in adults: a meta-analysis.
Psychol Bull 2003; 129:52–73
HELPFUL TUTORIAL ON WAY TO WRITE THIS ASSIGNMENT.
Read scenario a few times, and highlight points.
Part A –
Structure of Part A. Essay form – nothing tricky about this.
Could make some assumptions in this part.
Can be a bit flexible in what is happening in scenario.
Do not spend too much time repeating as only 1500 words. Do not rewrite the scenario – assume the markers know this.
Client is driving the treatment. Is a good way to write assignment that way.
You can’t prescribe medication.
This assessment is not about physical health.
Planting a seed for community services.
Paragraph format – an essay.
Outline of additional information.
B – 500 words.
In letter – pay attention to Mark’s past trauma.
Write letter. Summarise findings – and make some suggestions.
Put citations in letter – need for assessment (would not normally do in a letter).
Letter – reference intervention (treatment suggestions).
Looking at follow up care – not all provided by GP.
Mark will read what you have written to GP.
In letter to GP- could say have conducted risk assessment so he could be discharged as safe.
Outline things. He is safe. Recommendations to GP.
Look at what is in your local area and could refer some of this in GP letter. Then some online stuff is available too.
Delivering a high-quality product at a reasonable price is not enough anymore.
That’s why we have developed 5 beneficial guarantees that will make your experience with our service enjoyable, easy, and safe.
You have to be 100% sure of the quality of your product to give a money-back guarantee. This describes us perfectly. Make sure that this guarantee is totally transparent.
Read moreEach paper is composed from scratch, according to your instructions. It is then checked by our plagiarism-detection software. There is no gap where plagiarism could squeeze in.
Read moreThanks to our free revisions, there is no way for you to be unsatisfied. We will work on your paper until you are completely happy with the result.
Read moreYour email is safe, as we store it according to international data protection rules. Your bank details are secure, as we use only reliable payment systems.
Read moreBy sending us your money, you buy the service we provide. Check out our terms and conditions if you prefer business talks to be laid out in official language.
Read moreOur specialists are always online to help you! We are available 24/7 via live chat, WhatsApp, and phone to answer questions, correct mistakes, or just address your academic fears.
See our T&Cs