Content uploaded by Jesse R Cougle
Author content
All content in this area was uploaded by Jesse R Cougle on Jan 21, 2016
Content may be subject to copyright.
1
Cougle, J. R., Kilpatrick, D. G., & Resnick, H. (2012). Defining traumatic events: Research
findings and controversies. In J. G. Beck, & D. Sloan (Eds.), The Oxford Handbook of Traumatic
Stress Disorders (pp. 11-27). Oxford, UK: Oxford University Press.
Defining Traumatic Events
Research Findings and Controversies
Jesse R. Cougle. Dean G. Kilpatrick, and Heidi Resnick
ABSTRACT
Since the posttraumatic stress disorder (PTSD) diagnosis was introduced in the DSM-III
(American Psychiatric Association [APA], 1980) controversy has existed over what constitutes a
Criterion A stressor, or a potentially traumatic event. Such definitions have considerable
research, clinical, and legal implications. In this chapter, we review the history of Criterion A,
conditional risk of PTSD associated with specific stressors, and situational and demographic
variables likely to increase risk of PTSD. Significant attention is given to the controversies
surrounding Criterion A, as well as limitations of extant research. We conclude with specific
recommendations for future inquiry with particular emphasis on the importance of
comprehensive assessment of Criterion A stressor exposure and PTSD assessment that captures
the effects of exposure to more than one Criterion A stressor.
Key Words: Traumatic events; posttraumatic stress disorder; risk factors; Diagnostic and
Statistical Manual; Criterion A; diagnosis
Introduction
Traumatic event exposure is an essential component of posttraumatic stress disorder (PTSD).
Controversy has existed over what defines a potentially traumatic event (PTE) ever since the
2
PTSD diagnosis was added to DSM-III (APA, 1980). In the DSM diagnosis of PTSD, Criterion
A, or the stressor criterion, plays a key gatekeeping role because it determines which stressor
events can be assessed for PTSD (Kilpatrick, Resnick, & Acierno, 2009; Weathers & Keane,
2007). Thus, stressor events that are included in a given definition of Criterion A are defined as
meeting the threshold for PTSD eligibility and become potentially traumatic events (PTEs);
events that are not included, by definition, do not meet this eligibility threshold. Therefore, how
Criterion A is defined has a potential impact on PTSD caseness and prevalence. This impact on
PTSD prevalence and caseness is potential because, as has been described elsewhere (Kilpatrick
et al., 2009), the proportion of those who actually develop PTSD after exposure to given types of
stressor events (i.e., Criterion A vs. others) is an empirical question that is determined by data
and not by a DSM definition. For an individual being assessed for clinical or forensic purposes,
the definition of Criterion A is important because it establishes eligibility to receive a PTSD
diagnosis in a clinical case and whether psychological harm in the form of PTSD was either
caused or aggravated by a given stressor event in a forensic or compensation case. In summary,
how Criterion A is defined is a high-stakes issue because the definition has the potential to
impact PTSD prevalence, PTSD caseness, access to treatment for PTSD, and whether PTSD can
be diagnosed in forensic or compensation cases.
As has been noted elsewhere (Kilpatrick et al., 2009), the question of how to define
Criterion A raises both philosophical and empirical issues. Philosophical and value judgments
are involved in deciding what the primary purpose of Criterion A should be. Should it include all
stressor events that are capable of producing PTSD or should it include only those events that
have a high probability of producing PTSD? Is it important to define Criterion A so that
individuals having experienced less than truly terrible events are eligible to be evaluated for the
3
PTSD diagnosis? Does expanding the types of stressor events included in Criterion A undermine
efforts to encourage resilience or does doing so make it possible for some individuals who need
it to get access to services? None of these questions can be answered by data alone because how
they are answered is based more on philosophy and values than on data.
In this chapter, we summarize some of the controversies and research findings regarding
this criterion. We also review evidence regarding conditional risk of PTSD associated with
different types of PTEs, situational determinants of risk associated with PTEs, and demographic
variables likely to increase risk. We conclude with recommendations for future research.
Criterion A: A Brief History
The ongoing controversy regarding Criterion A is reflected in the changing definitions of this
criterion in each successive edition of the DSM. Table 2.1 provides the current DSM-IV and
proposed DSM-5 definitions of Criterion A. In DSM-III, a Criterion A potentially traumatic event
(PTE) was defined as the “existence of a recognizable stressor that would evoke significant
symptoms of distress in almost everyone” (APA, 1980, p. 238). The accompanying text of this
criterion described the stressor as evoking “symptoms of distress in most people” (p. 236) and
described the experience as outside the range of normal experiences, specifically excluding
bereavement, chronic illness, business losses, or marital conflict. By describing the criterion in
this manner, PTSD was conceptualized as more a product of discrete events than of a
combination of events and individual vulnerabilities. This definition was criticized by many
because it is clear that some of the most severe events do not lead to longer term symptoms in
most people. Further, we now know that exposure to many PTEs, such as interpersonal violence,
accidents, and natural disasters, are quite common and cannot truly be considered outside the
4
range of normal experiences (Kessler, Sonnega, Bromet, Hughes, & Nelson, 1995; Resnick,
Kilpatrick, Dansky, Saunders, & Best, 1993).
[table 2.1 here]
In the DSM-III-R (APA, 1987), the Criterion A definition was restricted somewhat. The
description of traumatic events as being “outside the range of usual human experience” and
“markedly distressing to almost everyone” (p. 250) was retained, and a more detailed list of
qualifying events was provided: “a serious threat to one’s life or physical integrity; serious threat
or harm to one’s children, spouse, or other close relatives and friends; sudden destruction of
one’s home or community; or seeing another person who has recently been, or is being seriously
injured or killed as the result of an accident or physical violence” (p. 247–248). The DSM-III-R
also emphasized the subjective aspects of traumatic event experience as involving “intense fear,
terror, and helplessness” (p. 247). Further, the descriptive text introduced a new class of
traumatic events involving indirect exposure, including “learning about a serious threat or harm
to a close friend or relative” (p. 248). Some of the same criticisms were made of this revision; in
particular, the description of events “outside the range of usual human experience” was criticized
on similar grounds as its predecessor (Davidson & Foa, 1991).
With the DSM-IV (APA, 1994), Criterion A was split into two parts (see table 2.1 for
definitions and relevant accompanying text). The first part states that the individual
“experienced, witnessed, or was confronted with an event or events that involved actual or
threatened death or serious injury, or the threat to physical integrity of self or others” (APA,
2000, p. 467). The list of qualifying events was also expanded to include “being diagnosed with
a life-threatening illness”; “learning about the sudden unexpected death of a family member or a
close friend”; and “learning that one’s child has a life-threatening disease” (p. 464). Inclusion of
5
these new events has been criticized by some as leading to a “conceptual bracket creep” in the
types of events that would qualify as traumatic (McNally, 2003). The additional wording that
describes being “confronted” with a distressing event has also been criticized for its vagueness.
For example, theoretically it is possible that simply viewing an event on television (e.g., the 9/11
attacks) might be viewed by some as sufficient to constitute “confrontation.” Controversy over
the broadening definitions of PTEs will be discussed later.
The second feature of Criterion A (i.e., A2) in DSM-IV states that the individual must
respond to the event with “intense fear, helplessness, or horror.” The subjective evaluation of the
event as traumatic is fundamental to the A2 criterion. Weathers and Keane (2007) provided
useful terminology for distinguishing between these features of Criterion A: meeting the A1
criterion qualifies an event as being potentially traumatic, while potentially traumatic events that
meet the A2 criterion can be considered traumatic.
Criterion A2 has generated controversy as well. Some have argued that traumatic events
are often characterized by other emotions, including guilt, shame, or disgust (Adler, Wright,
Bliese, Eckford, & Hoge, 2008; Brewin, Andrews, & Rose, 2000; Roemer, Orsillo, Borkovec, &
Litz, 1998). In addition, it appears that A2 has little positive predictive value, though events that
do not meet A2, rarely result in PTSD (Breslau & Kessler, 2001; Schnurr, Spiro, Vielhauer,
Findler, & Hamblen, 2002), confirming that A2 has negative predictive power.
Recommendations for Criterion A1 changes in the PTSD diagnosis range from
eliminating it altogether and emphasizing greater specificity in terms of key re-experiencing
symptoms related to events that the individual perceived as threatening to physical or
psychological well-being (Brewin, Lanius, Novac, Schnyder, & Galea, 2009), to retaining
Criterion A and allowing for additional empirical study of which events and/or event
6
characteristics are most likely to increase risk of PTSD following exposure (Kilpatrick et al.,
2009). In either case, emphasis on functional impairment criteria was also recommended (Brewin
et al., 2009; Kilpatrick et al., 2009). Other recommendations are to retain but narrow the
criteria—for example, eliminating events that are indirectly experienced (McNally, 2009).
Preliminary drafts of the proposed revisions for DSM-5 (see table 2.1) indicate that
Criterion A2 will be dropped and further specificity will be added to the definition of qualifying
events (APA, 2010). Such events must involve direct exposure—either experiencing or
witnessing—or learning of a violent or accidental event occurring to a close relative or friend.
Repeated exposure to relevant details (e.g., seeing photographs of injuries or death as part of
one’s job) may also qualify as a Criterion A event. Exposure to events via media, including
television, movies, or pictures, is explicitly excluded from meeting this criterion.
Criterion A Events: Too Broad or Too Narrow?
One aspect of the debate over Criterion A is whether the definition is too broad or too narrow.
Critics who argue it is too broad have singled out categories of secondary exposure, including
learning about injury, or unexpected or violent death occurring to others, as particularly
problematic (McNally, 2003, 2009). This expansion, which occurred in DSM-III-R and was
made more explicit in DSM-IV, is thought by some to have diluted the construct of PTSD by
including low-level stressors that are not truly traumatic. This potential problem is thought to
lead to significant heterogeneity in PTSD samples. Further, some critics argue that unnecessarily
broad definitions of trauma may lead researchers and clinicians to pathologize normal reactions
to stressful events. Critics have pointed to research suggesting that PTSD can develop due to
inappropriate sexual remarks, killing livestock, and uncomplicated delivery as being particularly
egregious examples of “bracket creep” (McNally, 2009). However, much of the research
7
purporting to demonstrate that PTSD develops following exposure to minor events suffers from a
host of methodological problems, including failure to assess comprehensively for exposure to
other PTEs, lack of evaluation of PTSD relative to such events, and failure to use state-of-the-
science PTSD assessment instruments.
Evidence that the DSM-IV Criterion A1 definition has led to significant increases in cases
of PTSD is mixed. Data from the National Comorbidity Survey-Replication, which assessed
PTSD using DSM-IV criteria and a longer list of qualifying events, revealed lower prevalence of
lifetime PTSD (6.8%; Kessler et al. 2005) than the earlier survey that relied on DSM-III-R
criteria (7.8%; Kessler et al., 1995). In the DSM-IV field trial (Kilpatrick et al., 1998) various
definitions of Criterion A were evaluated in a sample that included patients seeking treatment
related to a PTE (N = 400), as well as adults in the general population (N = 128). Lifetime
histories of high- and low-magnitude events were evaluated, with high-magnitude events
meeting Criterion A according to the DSM-III-R. The prevalence of PTSD in this sample was not
affected by more or less restrictive definitions of Criterion A.
Breslau and Kessler (2001) also examined changes in Criterion A prevalence as well as
changes in PTSD prevalence owing to the expanded definition of Criterion A in a sample of
2181 adults. They found the lifetime prevalence of Criterion A1 exposure to be quite high
(89.6%), and they noted a 59.2% increase in the total number of events experienced that was
accounted for by the expanded DSM-IV definition of A1. Further, the conditional probability of
PTSD from DSM-III and DSM-III-R Criterion A events was greater than that for DSM-IV
criterion A1 events (11.1% vs. 7.3%). Overall, they noted that 37.8% of all PTSD cases were due
to stressors added to the DSM-IV, with most of these PTSD cases being due to learning of the
unexpected death of a close friend or relative. This investigation had certain limitations,
8
however. First, as they note, “there is no precise way to quantify the impact of the broader range
of stressors in DSM-IV versus DSM-III or DSM-III-R because the explicated added stressors in
the DSM-IV overlap, in part, with stressors in earlier definitions” (p. 701). They were unable to
isolate PTSD cases due to new DSM-IV stressors, since the five new stressors they considered,
including violent death of a family member, could have also qualified as Criterion A events
under DSM-III and DSM-III-R (Weathers & Keane, 2007). Further, given that they only
examined PTSD in relation to most bothersome and a randomly selected event, they were unable
to truly ascertain whether overall lifetime PTSD prevalence would be increased by the addition
of the new DSM-IV Criterion A1 events.
Very recently, investigators examined whether PTSD prevalence would be affected by
using a nonrestrictive definition of Criterion A and determining PTSD diagnosis based on
symptom, duration, and impairment criteria (Kilpatrick et al., 2009). This question was
addressed in large household probability samples of U.S. adolescents and Florida adults. Overall,
they found that very few cases of PTSD occurred in the absence of Criterion A1 events, thus
providing little evidence that bracket creep affects PTSD prevalence.
There is some evidence to suggest that non-Criterion A1 stressors may lead to PTSD or at
least significant PTSD symptoms (e.g., Gold, Marx, Soler-Baillo, & Sloan, 2005; Long et al,
2008). Such data have been taken by some to mean that Criterion A1 should be broadened to
include other stressors. However, much of the supporting evidence is limited by use of self-
report symptom measures and a failure to include assessment of or adequately control for
lifetime PTE exposure and/or related prior PTSD. Many relevant studies also failed to include
full diagnostic criteria including functional impairment, and assessed PTSD only in reference to
subjectively defined “worst” events rather than in reference to index Criterion A and non-
9
Criterion A events occurring across individuals (e.g., first, most recent, randomly selected).
Additional exploration of whether some non-Criterion A events are actually more likely to occur
in some cases due to impaired functioning might be useful. In addition, research to evaluate to
what extent prior PTSD or Criterion A event exposure increases risk of symptom report in
reference to non-Criterion A events would be helpful.
Must Traumatic Events Be Directly Experienced?
An additional controversy over Criterion A relates to whether traumatic events must be directly
experienced. Beginning in DSM-III-R, events involving indirect exposure were listed as
potentially meeting Criterion A. Among the categories of specific events was “learning about a
serious threat or harm to a close friend or relative” (APA, 1987, p. 248). The list of types of
indirect exposure was expanded in DSM-IV, though each of these types was included, at least
implicitly, in the DSM-III-R. Epidemiological studies indicate that indirect exposure in the form
of unexpected death to a loved one constitutes the most common Criterion A1 event, as well as
the most common worst stressful event to which PTSD symptoms are anchored and evaluated
(Breslau et al., 1998; Roberts, Gilman, Breslau, Breslau, & Koenen, 2011).
The inclusion of indirect exposure in Criterion A1 is controversial because some argue
that this led to substantial increases in the percentage of the population with lifetime event
exposure in some studies (e.g., Breslau & Kessler, 2001). These exposure events have also been
associated with lower risk of PTSD. For example, Breslau and Kessler (2001) found that events
involving learning about harm or unexpected death to others were associated with lower
conditional probability of PTSD than events involving direct exposure (7.3% vs. 11.1%).
It should be noted, however, that death of a family member due to criminal or vehicular
homicide has been found to be associated with risk of PTSD and levels of distress similar to
10
prevalence among those with direct assault/event exposure (Amick-McMullan, Kilpatrick, &
Resnick, 1991; Freedy, Resnick, Kilpatrick, Dansky, & Tidwell, 1994; Zinzow, Rheingold,
Hawkins, Saunders, & Kilpatrick, 2009). Thus, failure to separate out such types of events from
other “confronted” events limits conclusions that may be drawn related to overall prevalence of
secondary exposure to these types of PTE. Critics who are wary of the broadening definitions of
trauma are especially concerned about the inclusion of indirect exposure in Criterion A1. Many
view the assessment of indirect exposure (learning about harm befalling a family member or
close friend) to be sufficiently vague so as to allow for the assignment of a PTSD diagnosis due
to various forms of information, including learning about the 9/11 attacks via television
(McNally, 2009). However, since such events very likely do not involve harm occurring to a
close friend or relative, according to DSM-IV criteria, a PTSD diagnosis would not be assigned.
Likewise, the fact that these events are defined as capable of producing PTSD does not mean that
there is a high probability that they will do so.
Preliminary drafts of the proposed revisions in DSM-5 have retained indirect exposure as
a qualifying event but with some specifications. According to these proposed revisions, PTEs
may be learned to have “occurred to a close relative or close friend” and “in such cases the actual
or threatened death must have been violent or accidental” (APA, 2010, para. 1). The “violent or
accidental” qualifier should exclude those events involving the death of a loved one due to a
severe medical condition (e.g., heart attack, life-threatening illness). It remains to be seen
whether these revisions will lead to significant reductions in lifetime Criterion A or PTSD
prevalence.
Assessment of PTSD in Relation to Which Criterion A1 Event?
11
PTSD symptoms are generally assessed in relation to specific Criterion A1 stressors; however, it
is quite common for individuals to experience multiple PTEs (Kessler et al., 1995). The issue of
which Criterion A1 event should be the focus of the PTSD assessment is complex. Clinical and
epidemiological assessments of PTSD typically involve the presentation of a list of Criterion A1
events on which individuals are asked to provide history (e.g., Kessler et al., 1995). Among those
events respondents endorse, they are asked to focus on the worst or most bothersome event and
then PTSD symptoms are assessed in relation to this event.
The “worst event” assessment method is a practical solution to the difficulty imposed by
assessing PTSD for people who have experienced multiple PTEs. Evidence suggests that few
individuals who fail to meet for PTSD diagnosis in relation to their worst event meet it due to
other traumatic events. For example, Breslau, Davis, Peterson, and Schultz (1997) evaluated
PTSD in response to both worst event and up to two additional events and found that the events
designated as worst accounted for 84% of total lifetime PTSD cases.
This method has certain limitations. Worst events are by definition perceived to be more
extreme than the types of same events to which individuals are usually exposed. Generally
speaking, a car accident that is considered a worst event will likely be more severe and
distressing than a typical car accident, and it will likely carry with it greater risk of PTSD.
Indeed, Breslau, Peterson, Poisson, Schultz, and Lucia (2004) found that the risk of PTSD
associated with car accidents was 6.5% when it was a worst event, although it was only 2.3% as
a randomly selected event. Thus, this method is prone to overstating the conditional risk of
PTSD following exposure to different traumatic events.
The “worst event” method also does not take into account the impact of multiple or
ongoing PTEs. Traumatic events experienced in the context of combat or domestic violence, in
12
particular, can be quite varied. PTSD in response to the worst lifetime event may remit prior to
the onset of a subsequent occurrence of PTSD due to a less distressing related event. Even
though respondents may report few past-month PTSD symptoms in response to a “worst”
incident, they may still meet diagnostic criteria for current PTSD due to a less bothersome event.
Thus, anchoring PTSD symptoms to a worst event that occurred long ago could give a false
impression of remission.
Respondents may choose worst events for reasons unrelated to the PTSD symptoms they
produced. For example, some may be more likely to choose the unexpected death of a loved one
as the worst of many events since this event represents the loss of someone close to them and
may have special significance that other Criterion A1 events do not possess. Indeed, researchers
have found that the unexpected death of a loved one was overrepresented among worst events
identified as such based on its actual prevalence in the sample (Breslau et al., 2004). Some worst
events may also be easier to disclose than an event that is actually most distressing to the
respondent (e.g., sexual assault).
Another factor that has not been adequately addressed in the context of history of
multiple types or incidents of exposure to PTEs is more complex, or aggregate, patterns of
PTSD symptoms (Kilpatrick et al., 2009). Specifically, intrusion and avoidance symptoms may
occur in reference to multiple traumatic events. Thus, an individual who has experienced several
traumatic events during his or her life (e.g., sexual assault as a child, physical assault as an adult,
injurious accident as an adult), may have intrusive and/or avoidance symptoms that incorporate
any or all of these events. This issue may also be relevant in the context of what might be
traditionally categorized as single types of PTEs that comprise multiple event experiences. For
example, an individual in military service might experience military sexual trauma, exposure to
13
improvised explosive device attacks, and being under enemy fire or witnessing death or injury of
a friend or others. PTSD diagnostic criteria might be considered aggregate in such instances if
the person has intrusion and/or avoidance symptoms incorporating several different traumatic
events. The reliance on the worst-event method of assessment may miss such cases in which
multiple PTE exposure is present. Potential methods for the assessment of PTSD in reference to
multiple events will be discussed in the concluding section.
Challenges to Evaluating Conditional Risk of PTSD Due to Different
Traumatic Events
Two basic methodologies have been used to determine conditional risk of PTSD associated with
different Criterion A events. One method involves structured interviews of large, representative
samples that may or may not have been exposed to a number of different Criterion A events
(e.g., Kessler et al., 1995). In these studies, lifetime PTE history is generally assessed and
lifetime and current PTSD is evaluated in relation to the most bothersome PTE. The second
method involves the recruitment of samples that were exposed to specific PTEs, usually very
recently, and the evaluation of PTSD in response to these events. Both methods include different
challenges, which we will discuss below.
Nationally representative surveys, such as the National Comorbidity Survey (Kessler et
al., 1995), are often used to determine prevalence rates of PTE exposure, as well as lifetime and
current PTSD diagnoses. These surveys often rely or report on worst-event methods for
determining PTSD diagnoses. This method, as already discussed, is problematic in that it tends
to overstate the conditional risk of PTSD associated with different events (Breslau et al., 2004).
It is also likely to neglect to evaluate the risk associated with stressors of lower magnitude. To
address this limitation, certain studies have attempted to also evaluate PTSD anchored to a
14
randomly selected event. This method is thought to provide a more accurate assessment of the
true risk of PTSD associated with various events.
Epidemiological studies very rarely capture PTSD in response to recent events. Many
cases of lifetime and current PTSD are determined with reference to events that occurred years
prior to the assessment. Naturally, certain limitations will arise from relying on such
retrospective assessments. For example, prospective research has demonstrated that half of
individuals meeting diagnostic criteria for depression at an earlier point in time did not recall
their depressive episodes before age 21 (Wells & Horwood, 2004). Those with remitted PTSD
should also be less likely to recall experiencing symptoms constituting the PTSD diagnosis,
especially if the symptoms remitted many years ago.
A recent analysis by Moffit and colleagues (2010) speaks directly to the problems of
ascertaining lifetime psychiatric diagnoses through retrospective cross-sectional methods. These
investigators compared prevalence rates of diagnoses in the longitudinal Dunedin New Zealand
birth cohort study with rates obtained in two U.S. National Comorbidity Surveys and found that
lifetime prevalence rates were approximately doubled when assessed prospectively relative to
retrospectively. Further, past-year prevalence rates were comparable across surveys, which
suggests that these marked differences are not due to differences in culture or assessment
instruments. However, this study did not report on PTSD as a separate category, thus it is
difficult to ascertain how PTSD specific prevalence would be affected.
An additional method for determining conditional risk of PTSD involves assessing
individuals who are exposed to specific traumatic events. Such investigations typically involve
recently exposed samples and different recruitment methods (e.g., emergency rooms) than the
kind used in epidemiological studies (e.g., household probability sampling). Since these studies
15
tend to rely on recently exposed samples, they are less vulnerable to the limitations of recall
listed above. In addition, since they assess PTSD anchored to specific events, they are not subject
to the problems inherent in worst-event assessment methods.
These investigations also have certain limitations, however. First, the assessment point
for current PTSD varies widely between studies. For example, recent studies have examined
early risk factors for current PTSD at six months posttrauma (Ehring, Ehlers, & Glucksman,
2008; Kleim, Ehlers, & Glucksman, 2007). However, diagnoses occurring at this point are
considered chronic PTSD according to DSM-IV (APA, 1994). Such study designs do not allow
investigators to determine conditional risk of ever having PTSD, which only requires one month
duration of symptoms. Similar studies have examined the conditional risk of current PTSD using
one-year postevent assessments (Irish et al., 2009). In short, these studies assess varying degrees
of PTSD chronicity rather than risk. This complicates efforts to determine conditional risk.
Focused studies examining PTSD due to specific PTE are also limited by the
representativeness of individuals typically assessed. For example, the assessment of PTSD due to
rape is complicated by the fact that most women do not report such events to authorities and do
not present to hospitals following such events (Resnick et al., 1993). Researchers must often rely
on advertisements or referrals through support agencies (e.g., Halligan, Michael, Clark, &
Ehlers, 2003). In addition, recruitment of assault or accident survivors through hospital
emergency rooms is likely to involve more severe types of Criterion A1 events, which may lead
to overestimation of the risk of PTSD associated with these events.
Conditional Risk of PTSD Associated with Specific Criterion A1 Events
The risk of PTSD associated with Criterion A1 events varies considerably by event type. Below
we review the conditional risk of PTSD associated with different events. Estimates will be
16
reviewed from both large-scale, representative epidemiological studies and recently exposed
samples.
Accidents
Studies of accident-related PTSD often recruit participants from patients admitted to the
emergency room. Varying prevalence of PTSD following traumatic injury have been found,
possibly due to the heterogeneous nature of the injuries for which these samples are admitted.
For example, Zatzick and colleagues (2007) found relatively high rates of PTSD (23%) at 12
months postinjury in their large sample, though O’Donnell et al. (2008) found that only 8% met
for PTSD in their sample at the 12-month assessment point. Epidemiological studies have found
conditional probabilities of PTSD that were quite low for accidents that were worst or only
events (6.3% men, 8.8% women; Kessler et al., 1995). Breslau et al. (2004) also found
conditional risk of PTSD following car accidents that were low for both worst (6.3%) and
randomly selected events (2.3%), though PTSD following “other serious accidents” was more
common (worst: 21.6%; randomly selected: 16.8%).
Prospective studies of accident-related PTSD tend to restrict recruitment to individuals
who are admitted to the emergency room who also exhibit significant injury. Many studies
exclude patients who are admitted with only mild cuts or abrasions (Ehring et al., 2008) or who
are admitted to the hospital for less than 24 hours (O’Donnell et al., 2008). These criteria may
lead to the inflation of PTSD rates. This range restriction also likely contributes to the
inconsistent findings regarding the association between injury severity and PTSD, which will be
discussed later.
Interpersonal Violence
17
Physical and sexual assault are consistently associated with some of the highest probabilities of
PTSD. Epidemiological studies have found high conditional probabilities of PTSD due to rape
for both men (65%) and women (45.9%; Kessler et al., 1995). The conditional risk of PTSD was
also high when rape was a randomly selected event (49%; Breslau et al., 2004). Prospective
studies of female rape victims have also found high rates of PTSD, with one study revealing that
65% met criteria at one month posttrauma, while 47% met at three months (Rothbaum, Foa,
Riggs, Murdock, & Walsh, 1992).
Lower conditional risk of PTSD has been found for physical assault in some studies. For
example, Breslau et al. (1997) found the conditional risk of PTSD due to being “badly beaten
up” to be 31.9% when using randomly selected event criteria. Despite the fact that these events
are more common in men than women (11.1% vs. 6.9%), the risk of PTSD due to physical
assault is much lower in men (1.8% vs. 21.3% for women; Kessler et al., 1995). Event-focused
prospective studies typically combine physical and sexual assault victims, though one
prospective study of physical assault victims found risk of PTSD to be at 22% at six months
postassault (Elklit & Brink, 2004). Riggs, Rothbaum, and Foa (1995) examined the longitudinal
course of symptoms following physical assault and found that while 71% of women and 50% of
men met symptom (but not duration) criteria for PTSD at the initial assessment (M = 18.68 days
postassault), no men and only 21% of women met criteria for PTSD three months later. As noted
by Pimlott-Kubiak and Cortina (2003), women may also be exposed to different types of
interpersonal aggression and/or different frequencies or differential risk of injuries in association
with physical assault. They found that, controlling for patterns of interpersonal aggression
experienced, gender differences were not observed in terms of physical or mental health
problems within the large National Violence Against Women sample. Such factors related to
18
differential characteristics or patterns of exposure should be studied further when considering
gender or other potential additional risk factors for PTSD.
Conditional risk of PTSD associated with childhood sexual and physical abuse has also
been studied. For example, using a worst-event method, Kessler et al. (1995) found the risk of
lifetime PTSD due to childhood sexual abuse to be 12.2% for men and 26.5% for women, and
the risk of PTSD due to childhood physical abuse to be 22.3% for men and 48.5% for women.
High prevalence of lifetime PTSD was found among a representative sample of women who had
experienced childhood physical assault compared to those who had not experienced childhood
physical assault (53.8% vs. 11.2%; Duncan, Saunders, Kilpatrick, Hanson, & Resnick, 1996).
Studies of children referred for psychiatric evaluation found higher rates of PTSD among those
who were sexually abused (42.3%) compared to those who had suffered nonsexual abuse (8.7%;
McLeer, Callaghan, Henry, & Wallen, 1994). Using data from a household probability sample of
adolescents ages 12 to 17, Kilpatrick, Saunders, and Smith (2003) found higher prevalence of
lifetime PTSD in association with a history of sexual assault compared to no sexual assault
among boys (28.2% vs. 5.4%) and girls (29.8% vs. 7.1%). Lifetime PTSD prevalence was also
higher among those who reported a history of physical assault or physically abusive punishment
compared to those without such history among boys (15.2% vs. 3.1%) and among girls (27.4%
vs. 6%). Other data indicate that interpersonal violence events that first occur in childhood may
increase later risk of PTSD in adulthood following subsequent stressor exposure (e.g., Binder et
al., 2008; Smith, Saunders, & Kilpatrick, 2008).
Combat
Epidemiological studies indicate that combat is the most commonly nominated worst event
among men with DSM-III-R PTSD (Kessler et al., 1995). Though combat-related exposure is
19
rare among men (6.4% lifetime prevalence), the conditional risk of PTSD associated with worst-
event combat is quite high (38.8%; Kessler et al., 1995). Studies focusing on combat veterans
have found varying rates of combat-related PTSD, though this has been a source of some
controversy. For example, the National Vietnam Veterans Readjustment Study (NVVRS) found
that 30.9% of a sample of 1200 veterans had developed PTSD during their lifetimes (Kulka et
al., 1988). A reanalysis of a subsample of the NVVRS using extremely rigorous criteria revealed
that 18.7% had developed war-related PTSD during their lifetimes (Dohrenwend, Turner, Turse
et al., 2006). Recently, approximately 12–20% of Iraq combat veterans were found to have
PTSD postdeployment, depending on the criteria used; these rates were lower than the rates
found following deployment to Afghanistan (6.2–11.5%; Hoge et al., 2004). Both of these
studies found strong dose relationships between duration or frequency of combat exposure and
likelihood of developing PTSD.
Natural Disaster
Studies of PTSD following natural disaster are less common than those for interpersonal
violence or accidents. However, they are useful in that the occurrence and timing of the Criterion
A1 events are objectively verifiable. They also typically involve large populations who have
been exposed and who can be followed over time. The assumption is that such exposures are
uniformly distributed across samples, though obviously there will be varying degrees of
exposure and varying degrees of risk depending on exposure severity.
Epidemiological studies indicate that natural disasters constitute “worst events” for a
small minority of individuals reporting lifetime PTSD (5.2% for men, 3.5% for women; Kessler
et al., 1995). In addition, the conditional risk of PTSD associated with natural disasters is quite
low (3.8% for randomly selected event, 4.1% for worst event; Breslau et al., 2004). Focused
20
studies of specific trauma-exposed populations, such as Florida residents affected by hurricanes,
have also found low population-based prevalence rates of disaster-related PTSD (3.6%; Acierno
et al., 2007). However, high rates of hurricane-related PTSD (22.5%) were recently reported in a
sample of Hurricane Katrina survivors (Galea, Tracy, Norris, & Coffey, 2008). These higher
rates are likely due to the high death toll, severe property damage, and relocation that were
unique to Katrina.
Witnessed Violence
Exposure to witnessed violence represents one of the most common types of Criterion A1 events.
Data from the National Comorbidity Survey found that 35.6% of men and 14.5% of women
witnessed someone being badly injured or killed (Kessler et al., 1995); witnessed violence was
also the second most common worst event reported by men with lifetime PTSD (combat being
the most common). Conditional risk of PTSD associated with witnessed violence varies, likely
due to the variability in severity of exposure. For example, Breslau et al. (2004) found the risk to
be 7.3% and 11.4% for randomly selected and worst-event traumas, respectively. However,
North, Smith, and Spitznagel (1994) found that 20% of men and 36% of women who had been
exposed to a mass shooting spree met diagnostic criteria for PTSD one month later. Seeing
someone accidently injured or killed would likely lead to lower risk of PTSD than witnessed
violence involving intentional acts of harm in which the exposed individual may have actually
been in serious danger. These characteristics are not typically distinguished in epidemiological
studies. However, Kilpatrick et al. (2003) found that witnessed violence increased PTSD risk
after controlling for the effects of direct exposure to violence.
Unexpected Death of a Loved One
21
Though unexpected death of a loved one was arguably considered a Criterion A event in DSM-
III-R, it was given greater emphasis in DSM-IV and was specifically assessed in epidemiological
studies using DSM-IV criteria (e.g., Kessler et al., 2005). As discussed previously, when
Criterion A was broadened to include this event, it led to substantial increases in PTSD cases in
one study (Breslau & Kessler, 2001). Unexpected death of a loved one represents the most
common specific Criterion A1 in recent epidemiological studies (Breslau et al., 1998; Roberts et
al., 2011) and is the event most likely to be identified as “worst” (Breslau et al., 2004). The
conditional risk of PTSD associated with such events is in the moderate range for Criterion A1
events, with 14.3% meeting PTSD criteria in the Detroit Area Survey (Breslau et al., 1998). One
epidemiological study assessed homicide-related PTSD in family members and friends of
victims of criminal homicide and alcohol-related vehicular homicide and found that 23.3% of
immediate family survivors experienced PTSD at some point in their lives, while 4.8% met
criteria for PTSD in the past six months (Amick-McMullan et al., 1991).
Focused longitudinal studies of individuals who had recently experienced the unexpected
death of a loved one are complicated by the difficulties inherent in identifying and enrolling
those who may still be experiencing significant grief reactions. A few noteworthy studies have
been conducted, however. For example, one investigation examined rates of PTSD among
bereaved parents of 12- to 28-year-olds who had suffered violent death (accident, suicide, or
homicide) and found that 5% of fathers and 34% of mothers met criteria for PTSD four months
after their child’s death, while 14% of fathers and 21% of mothers met criteria two years
following the event (Murphy et al., 1999). The rates of PTSD at four months were significantly
associated with cause of death, with the highest rates among mothers and fathers who had lost
their child to homicide (mothers: 60%, fathers: 40%). An additional study of widows and
22
widowers found at two months postloss that 10% whose spouse had died of chronic illness met
diagnostic criteria for PTSD, 9% whose spouse died unexpectedly had PTSD, and 36% of those
whose spouses died from accidents or suicides met criteria for PTSD (Zisook, Chentsova-Dutton,
& Shuchter, 1998).
Overall, there appears to be greater conditional risk of PTSD associated with violent
unexpected deaths than unexpected deaths, broadly considered. These findings have implications
for the new Criterion A definition put forward in preliminary revisions of the DSM-5. The new
Criterion, as mentioned earlier, specifies that the unexpected deaths must be violent or accidental
in nature; thus, medically related unexpected deaths would not qualify. This will likely have the
effect of reducing the prevalence of exposure to unexpected death of a loved one as well as
increasing the conditional risk associated with this event.
Situational Determinants of PTSD Risk
Various features of Criterion A events are associated with increased risk of PTSD. These will be
discussed next.
Direct Threat versus Witnessed Threat
Among the situational determinants of risk includes the perception that either the individual or
someone else is at risk during the traumatic event. As previously discussed, evidence suggests
that events involving direct threat to the individual (e.g., physical assault) are likely to lead to
greater PTSD symptoms than those in which the individual witnesses threat, serious injury, or
death to someone else (i.e., witnessed violence). However, Alden, Regambal, & Laposa (2008)
found that emergency department healthcare workers who experienced direct threat events
reported comparable PTSD symptoms to those who experienced witnessed threat events. The
direct-threat group reported greater arousal symptoms and work-related consequences, including
23
an unwillingness to take overtime shifts, negative feelings toward the hospital, and job
dissatisfaction. Interestingly, the witnessed-threat group appraised their PTSD symptoms more
negatively than the direct-threat group, which provides evidence of the role of trauma type in
influencing symptom presentation.
Severity of Potentially Tramatic Event
Meta-analytic findings suggest traumatic event severity, including measures of combat exposure
among combat veterans and factors such as injury within crime victim samples, is a potent risk
factor for PTSD (Brewin, Andrews, & Valentine, 2000). Interestingly, the effect sizes for this
risk factor vary according to whether the study design was prospective or retrospective, with
retrospective studies yielding greater associations between severity and PTSD risk. Such findings
suggest that PTSD diagnosis may negatively bias retrospective recall of traumatic events.
However, it is also possible that more severe traumatic events are those that are both most likely
to be remembered and to result in PTSD.
There is marked variability in the assessment of PTE severity across studies and PTE
types. Civilian PTE severity is typically assessed using more subjective criteria than that used for
combat or motor vehicle accidents (Brewin, Andrews, & Valentine, 2000). Studies of motor
vehicle accident survivors often use objective measures of injury severity. This perhaps explains
why a strong relationship between combat severity and PTSD exists (Brewin, Andrews, &
Valentine, 2000), though the relationship between PTE severity and PTSD is less consistently
demonstrated for motor vehicle accidents (Taylor & Koch, 1995). This is also evidence for the
importance of subjective appraisal of PTEs in determining risk.
Analyses of the role of injury severity in risk of PTSD are often complicated by range
restriction. Many individuals who suffer motor vehicle accidents are not injured severely enough
24
to present to emergency rooms and be recruited for risk-factor studies, whereas those who are
most severely injured (e.g., in need of life-saving surgery) are often unable to participate in such
studies. One recent prospective study examined a large sample of individuals hospitalized for
traumatic injuries and found that mild traumatic brain injury, objectively identified, increased
risk for PTSD at 3 months and 12 months postinjury (Bryant et al., 2010).
Pain and Ongoing Stressors
Level of pain reported shortly after traumatic injury has been found to increase risk of PTSD
(Norman, Stein, Dimsdale, & Hoyt, 2008). More prolonged, chronic pain following traumatic
events is also related to PTSD chronicity (Bryant, Marosszeky, Crooks, Baguley, & Gurka,
1999). This has led some to hypothesize a mutual maintenance model in which PTSD and
chronic pain are maintained by common or mutually reinforcing factors (e.g., behavioral
avoidance, arousal; Sharp & Harvey, 2001).
There is also evidence that stressors resulting from Criterion A1 events increase risk of
PTSD. For example, Blanchard and colleagues (1996) found prospective evidence that litigation
related to motor vehicle accidents increases risk of PTSD, although it was also true that more
serious cases were more likely to result in lawsuits. Additionally, researchers found that
hurricane-related financial loss and postdisaster stressors were associated with greater risk of
PTSD among survivors of Hurricane Katrina (Galea et al., 2008). The mechanisms by which
postdisaster stressors increase risk of PTSD have not been studied extensively, but it is possible
that increased risk may be a consequence of increased anxiety or hyperarousal produced by such
stressors (i.e., symptom overlap). Postevent stressors might also serve as reminders and lead to
more negative perceptions of the event.
Perceived Responsibility for the Trauma
25
Attributions of responsibility for different traumatic events have been studied in relation to risk
for PTSD. For example, Delahanty et al. (1997) found that those who blamed others for their
motor vehicle accidents experienced greater PTSD symptoms at 6 months and 12 months
postaccident than a control group that experienced minor accidents; individuals who blamed
themselves did not differ from the control group in PTSD symptoms at these later assessments.
These findings were subsequently replicated by Hickling, Blanchard, Buckley, and Taylor
(1999), who found that those who blamed themselves for their motor vehicle accidents recovered
more quickly than those who blamed others. Researchers also found that self-blame for traumatic
events leading to severe burns was associated with lower risk of acute stress disorder than other-
blame in a sample of hospitalized burn victims (Lambert, Difede, & Contrada, 2004). Some
research has found self-blame to predict increased risk of PTSD (Filipas & Ullman, 2006) or
more severe psychological distress (Weaver & Clum, 1995) among sexual assault or abuse
survivors, which implicates the importance of considering event type, as well as rationality of
attributions when assessing their potential role.
Demographic Risk Factors for PTSD Following Traumatic Event
Exposure
Certain demographic variables have also emerged as risk factors for PTSD following traumatic
event exposure, though the associations between some of these variables (e.g., age, ethnicity) and
increased risk are inconsistent. This literature will be discussed next.
Gender
There is now an abundance of evidence indicating that women are at greater risk of PTSD
compared to men. In the National Comorbidity Survey, differences in risk of DSM-III-R PTSD
were noted among those exposed to trauma, with 20.4% of women meeting PTSD diagnostic
26
criteria compared to 8.4% of men (Kessler et al., 1995).Tolin and Foa’s (2006) meta-analysis
demonstrated the consistency of this relationship across 25 years of research. Greater conditional
risk of PTSD among women is found even when controlling for trauma type. For example, one
prospective study found that women had rates of PTSD that were over twice as high as men
following a motor vehicle accident (Fullerton et al., 2001). These findings complement
epidemiological studies that have found comparable gender differences in the prevalence rates of
other anxiety disorders (Kessler et al., 1994). However, as noted earlier, differential
characteristics of assault (e.g., physical assault) or different patterns of exposure to a range of
events may also account for some observed gender differences in PTSD prevalence (Pimlott-
Kubiak & Cortina, 2003) and ongoing research should consider and attempt to control for such
factors when evaluating gender differences in PTSD.
Age
There is some evidence to suggest that conditional risk of PTSD declines with older age. For
example, Brewin, Andrews, and Valentine’s (2000) meta-analysis indicated that younger age at
the time of exposure was associated with greater risk of PTSD. Interestingly, they found that
younger age only predicted increased risk of PTSD in military samples, though it was not
associated with increased risk in civilian samples. Younger age was also more strongly related to
increased PTSD risk for men than for women, though this was likely a product of the military
study findings, which mostly examined all-male samples.
Ethnicity
There is some evidence implicating a relationship between ethnic minority status and increased
risk for PTSD. Meta-analytic findings have demonstrated such a relationship, though it was
stronger in military samples than in civilian samples and was present in male but not female
27
samples (Brewin, Andrews, & Valentine, 2000). Race may act as a proxy for other demographic
risk factors, including socioeconomic status or exposure severity. Consistent with this
explanation, Breslau et al. (1998) found a much higher conditional risk of PTSD among
nonwhites compared to whites (14.0% vs. 7.3%), though this difference was made nonsignificant
when controlling for other demographic variables. An analysis of the National Vietnam Veterans
Readjustment Survey revealed that, compared to whites, elevated rates of PTSD among blacks
were accounted for by greater exposure, and higher rates of PTSD among Hispanics were
accounted for by a combination of greater exposure, younger age, lesser education, and lower
scores on the Armed Forces Qualification Test (Dohrenwend et al., 2008).
Recently published data from the National Epidemiological Survey on Alcohol and
Related Conditions (N = 34,653) found the highest prevalence of PTSD among blacks (8.7%)
and the lowest among Asians (4.0%), with Hispanics (7.0%) and whites (7.4%) having
prevalence rates in the middle (Roberts et al., 2011). Analyses of conditional risk found that
blacks were at greater risk of PTSD and Asians were at lower risk of PTSD than whites, after
adjusting for gender, age at interview, and trauma characteristics.
Income and Education
Lower socioeconomic status has been linked to increased conditional risk for PTSD (Brewin,
Andrews, & Valentine, 2000). This relationship has been demonstrated prospectively in event-
focused studies. For example, Irish et al. (2008) found that lower income was modestly
predictive of greater PTSD symptoms at six weeks and one year following a motor vehicle
accident. Lower education is also associated with increased risk for PTSD. A Vietnam twin
registry study revealed that less than high school education at the time of entry into the military
was uniquely predictive of combat-related PTSD, even after controlling for other risk factors
28
(Koenen et al., 2002). Lower income may be associated with fewer resources for coping with
the trauma, and both lower income and lower educational attainment may be proxies for low
intelligence, an established risk factor for PTSD following trauma (Brewin, Andrews, &
Valentine, 2000).
Conclusions
We have hoped to discuss many issues of relevance to the topic of what constitutes a traumatic
event. These issues have important implications for the assessment of PTSD. Many of the
controversies and limitations we mentioned might be resolved as further studies are conducted.
Next we list some future areas of research that would be beneficial to address.
Future Directions
1. How Should We Address the Issue of Cumulative Exposure?
The question of how to address cumulative exposure to multiple stressor events—PTEs,
traumatic events, and lesser stressors—over a lifetime requires additional attention both
conceptually and with respect to measurement. All major epidemiological studies (e.g., Breslau
& Kessler, 2001; Kessler et al., 1995; Kilpatrick, Acierno, Resnick, Saunders, & Best, 2000;
Kilpatrick et al., 2009; Resnick et al., 1993) have found that many individuals have been exposed
to multiple PTEs and traumatic events, but few studies include comprehensive assessment of
PTSD in reference to all potential qualifying PTEs as well as other stressors. The PTSD
assessment strategy most often used in cases of exposure to multiple events is either to assess
PTSD with regard to the “worst” event and/or to a randomly selected event. Neither of these
approaches permits adequate assessment of potential cumulative exposure to PTEs or of whether
cumulative exposure to PTEs may account for PTSD symptoms that may appear to result from
exposure to a recent event that is not a PTE. There has been little to no research investigating
29
whether an “aggregate PTSD” due to multiple events differs in important ways from PTSD
linked to a specific event. How prevalent is the aggregate type? Is the consideration of this type
clinically or empirically useful? Such questions deserve further inquiry.
Criterion A was born in controversy, and it is unlikely that the proposed changes in its
definition in the DSM-5 will resolve this controversy, particularly the arguments about the
definition that involve philosophical issues and value judgments. However, some of the
questions about Criterion A do lend themselves to empirical investigation. We believe that the
most important questions about Criterion A cannot be answered without a new strategy for
assessing PTSD that incorporates a way to assess the potential impact of having experienced
more than one PTE when that occurs. Here we review key issues related to optimal or improved
strategy to assess PTSD in reference to multiple PTE exposure history, as well as a newly
developed PTSD assessment measure designed to implement that approach. We then summarize
points regarding additional areas of research that are needed.
Need for a PTSD assessment strategy that addresses exposure to multiple PTEs
When the PTSD diagnosis was first established in 1980, most traumatic stress researchers and
clinicians were focused on a single type of PTE (e.g., accidents, child abuse, combat, disasters,
or violence against women). Exposure to all such events was considered to be rare, so there was
little consideration of the possibility that a substantial number of individuals may have been
exposed to more than one PTE throughout their lives. Therefore, if PTE exposure is assumed to
be rare and exposure to multiple PTE types and/or multiple exposure to PTEs of the same type
are thought to be exceedingly rare, then a PTSD assessment approach that focuses on changes in
symptomatology and functioning after exposure to a specific PTE makes a great deal of sense
because most people having experienced a PTE will have experienced only one event. However,
30
as noted throughout the chapter, exposure to multiple events is common, making the assessment
task more complicated. A range of qualifying PTEs and other stressor events may occur across
the life span, including interpersonal violence events that first occur in childhood and which may
moderate risk of PTSD to subsequent events (e.g., Binder et al., 2008; Smith et al., 2008).
Getting a more comprehensive picture of whether PTSD develops following specific qualifying
or other stressor events and what the trajectory or course of reported symptoms is among those
with multiple-event exposure would be helpful in addressing key questions about Criterion A.
One approach to accomplishing this would be to use an assessment measure that includes
comprehensive evaluation of event exposure history and chronology, and that allows for
assessment of PTSD symptom onset or worsening in reference to any referent events. The
proposed DSM-5 criteria for PTSD explicitly state that it is necessary to demonstrate that PTSD
symptoms are associated with traumatic events, defined as having begun or worsened after the
traumatic event(s) (APA, 2010). This requirement means that it is important for PTSD diagnosis
to establish whether each PTSD symptom occurred or got worse after a specific PTE occurred. In
cases involving more than one PTE, assessment would appear to require assessing whether
PTSD symptoms are associated with (i.e., the symptom began after or was worsened by) each
PTE experienced by the individual. There are two potential ways to approach this diagnostic
requirement.
The first way is to use a traditional PTSD assessment instrument such as the Clinician-
Administered PTSD Schedule (CAPS; Blake et al., 1995) or PTSD Checklist (Weathers, Litz,
Herman, Huska, & Keane, 1993) and apply it to every PTE reported by an individual. Given that
the proposed PTSD diagnosis for DSM-5 now has 20 PTSD symptoms, this assessment strategy
would be lengthy when applied to multiple PTEs and might prove to be burdensome. Particularly
31
in cases with PTE exposure during childhood and subsequent PTE exposure as an adult, those
assessed would have to make difficult, and perhaps impossible, judgments about when a
symptom first occurred and whether it worsened after each PTE.
A second approach toward PTSD symptom assessment involves comprehensive
assessment of lifetime PTE exposure in conjunction with assessment of PTSD symptoms in
reference to any PTE reported. The procedure used in the National Women’s Study (NWS)
PTSD Module (Kilpatrick, Acierno, Resnick, Saunders, & Best, 1997; Resnick et al., 1993), is
one such approach that has been shown to be a reliable and valid measure that yields comparable
results to those obtained with structured diagnostic interviews such as the SCID (Kilpatrick et al.,
1998).
A proposed method to assess PTSD in reference to history of single or multiple
PTE exposure
We recently modified the NWS PTSD Module and constructed a new self-report PTSD measure,
the National Stressful Events Survey (NSES) PTSD Module (Kilpatrick, Resnick, Baber, Guille,
& Gros, 2010), that is designed to measure PTSD using the proposed DSM-5 Criteria. The NSES
measure was constructed for use in a large Web survey, but it can also be used in a structured
interview format. Specifically, the assessment strategy used has four parts. First, a
comprehensive assessment of lifetime exposure to PTEs is conducted. Second, additional
descriptive information is obtained about a subset of PTEs experienced—for example, the first,
most recent, and worst PTE experienced. Third, individuals with any lifetime PTE exposure are
asked whether they have ever had each PTSD symptom at any time during their lives. Fourth, for
each lifetime symptom endorsed, a series of follow-up questions are asked. For PTSD symptoms
that include direct reference to a PTE (e.g., recurrent distressing dreams that are related to the
32
content of the event; inability to remember an important aspect of the traumatic event), a follow-
up question asks the respondent to identify which of any lifetime PTE or PTEs was/were
involved in the symptom content. For other PTSD symptoms not directly referencing PTEs (e.g.,
hypervigilance, problems with concentration), a follow-up question asks if the symptom either
happened or got worse after a PTE, and if so, to identify each lifetime PTE that is applicable.
Other follow-ups include obtaining information about symptom recency and about how
disturbing the symptom has been during the past month. This approach allows for evaluation of
meeting criteria in reference to any lifetime PTE reported. For example, number of intrusion,
persistent avoidance, negative alterations in cognitions and mood, alterations in arousal and
reactivity symptoms in reference to specific events can be computed and those events associated
with meeting necessary criteria for diagnosis may be determined. Additional questions are
included to assess functional impairment and duration criteria.
Questions related to PTSD in association with complex history of PTE exposure may be
possible to address using this approach. In some cases, individuals with a history of exposure to
multiple PTEs may meet full symptom criteria in reference to several events. It may also be
possible that an individual may report symptoms in reference to multiple events but not fully
meet criteria in reference to any single PTE. The latter pattern might provoke additional needed
discussion regarding the issue of “aggregate PTSD” if it was clear, for example, that all
necessary criterion symptoms occurred within the same time frame (e.g., past month), but
constituted a mix of PTEs as reference events.
2. What Additional Research Is Needed about Bereavement and PTSD?
The proposed DSM-5 definition of Criterion A includes learning about events involving the
death of a relative or close friend if the death was violent or accidental. This definition is more
33
restrictive than in DSM-IV because it excludes deaths due to natural causes. In addition,
proposals have been put forth advocating the inclusion of a prolonged grief (PG) or complicated
grief (CG) diagnosis within the DSM-5. There is overlap between some of the criteria and
symptoms of the proposed PG and CG diagnoses with those of the proposed PTSD diagnosis, but
two types of additional research are needed to clarify PTSD as a response to learning about the
deaths of loved ones and how PTSD relates to PG and/or CG. First, research is needed that tests
the assumption in the DSM-5 proposed definition that development of PTSD is limited to those
who learn about accidental or violent deaths versus learning about deaths occurring due to
natural causes. Without information comparing PTSD associated with learning about different
types of deaths, it will be difficult to determine whether restricting the types of deaths to
accidents or violent deaths is supported by data. Second, research is needed about the degree to
which there is overlap between PTSD and the proposed PG or CG diagnoses following
bereavement associated with different types of deaths.
3. How Meaningful Are Recollections of Distant Traumatic Events and Their
Associated Sequelae?
Recent epidemiological evidence suggests that the prevalence of some psychiatric diagnoses
doubles if prospective rather than retrospective designs are used (Moffit et al., 2010), although
this study did not specifically address PTSD. Much of what we know about the consequences of
various traumatic events relies on retrospective studies, and findings from such studies may
underestimate the prevalence of exposure to PTEs. Expansive, longitudinal studies that are able
to capture the occurrence of a variety of recent traumatic events can address the problems
associated with recall failure to some extent and should provide more accurate estimates of
lifetime PTSD prevalence. However, given that exposure to some types of PTEs begins early in
34
life and given the high cost of longitudinal research with probability samples, there is still an
important role for well-designed retrospective studies.
References
Acierno, R., Ruggiero, K. J., Galea, S., Resnick, H. S., Koenen, K., Roitzsch, J., de Arellano, M.,
et al. (2007). Psychological sequelae resulting from the 2004 Florida hurricanes:
Implications for postdisaster intervention. American Journal of Public Health, 97(Suppl.
1), S103–108.
Adler, A. B., Wright, K. M., Bliese, P. D., Eckford, R., & Hoge, C. W. (2008). A2 Diagnostic
criterion for combat-related posttraumatic stress disorder, Journal of Traumatic Stress,
21, 301–308.
Alden, L. E., Regambal, M. J., & Laposa, J. M. (2008). The effects of direct versus witnessed
threat on emergency department healthcare workers: Implications for PTSD Criterion A.
Journal of Anxiety Disorders, 22, 1337–1346.
American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders
(3rd ed.). Washington, DC: Author.
American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders
(3rd ed., rev.). Washington, DC: Author.
35
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders
(4th ed.). Washington, DC: Author.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders
(4th ed., text rev.). Washington, DC: Author.
American Psychiatric Association. (2010). Proposed Revision – APA DSM-5. Retrieved from:
http://www.dsm5.org/ProposedRevisions/Pages/ proposedrevision.aspx?rid=165.
Amick-McMullan, A., Kilpatrick, D. G., & Resnick, H. S. (1991). Homicide as a risk factor for
PTSD among surviving family members. Behavior Modification, 15, 545–559.
Binder, E. B., Bradley, R. G., Liu, W., Epstein, M. P., Deveau, T. C., Mercer, K. B., et al.
(2008). Association of FKBP5 polymorphisms and childhood abuse with risk of
posttraumatic stress disorder symptoms in adults. Journal of the American Medical
Association, 299, 1291–1305.
Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., Gusman, F. D., Charney, D. S., &
Keane, T. M. (1995). The development of a clinician-administered PTSD scale. Journal
of Traumatic Stress, 8, 75–90.
Blanchard, E. B., Hickling, E. J., Taylor, E., Loos, W. R., Forneris, C. A., & Jaccard, J. (1996).
Who develops PTSD from motor vehicle accidents? Behaviour Research and Therapy,
34, 1–10.
Breslau, N., Davis, G. C., Peterson, E. L., and Schultz, L. (1997). Psychiatric sequelae of
posttraumatic stress disorder in women. Archives of General Psychiatry, 54, 81–87.
Breslau, N., & Kessler, R. C. (2001). The stressor criterion in DSM–IV posttraumatic stress
disorder: An empirical investigation. Biological Psychiatry, 50, 699–704.
36
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.
Breslau, N., Peterson, E. L., Poisson, L. M., Schultz, L. R., & Lucia, V. C. (2004). Estimating
post-traumatic stress disorder in the community: Lifetime perspective and the impact of
typical traumatic events. Psychological Medicine, 34, 889–898.
Brewin, C. R., Andrews, B., & Rose, S. (2000). Fear, helplessness, and horror in posttraumatic
stress disorder: Investigating DSM-IV criterion A2 in victims of violent crime. Journal of
Traumatic Stress, 13(3), 499–509.
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.
Brewin, C.R., Lanius, R.A., Novac, A., Schnyder, U., & Galea, S. (2009). Reformulating PTSD
for DSM-V: Life After Criterion A. Journal of Traumatic Stress, 22, 366–373.
Bryant, R. A., Marosszeky, J. E., Crooks, J., Baguley, I. J., & Gurka, J. A. (1999). Interaction of
posttraumatic stress disorder and chronic pain following traumatic brain injury. Journal
of Head Trauma Rehabilitation, 14, 588–594.
Bryant, R. A., O’Donnell, M. L., Creamer, M., McFarlane, A. C., Clark, C. R., & Silove, D.
(2010). The psychiatric sequelae of traumatic injury. American Journal of Psychiatry,
167, 312–320.
Davidson, J. R. T., & Foa, E. B. (1991). Diagnostic issues in PTSD: Considerations for the DSM
IV. Journal of Abnormal Psychology, 100, 346–355.
37
Delahanty, D. L., Herberman, H. B., Craig, K. J., Hayward, M. C., Fullerton, C. S., Ursano, R. J.,
et al. (1997). Acute and chronic distress and posttraumatic stress disorder as a function of
responsibility for serious motor vehicle accidents. Journal of Consulting and Clinical
Psychology, 65, 560–567.
Dohrenwend, B. P., Turner, J. B., Turse, N. A., Adams, B. G., Koenen, K. C., & Marshall, R.
(2006). The psychological risks of Vietnam for U.S. veterans: A revisit with new data
and methods. Science, 313, 979-982.
Duncan, R. D., Saunders, B. E., Kilpatrick, D. G., Hanson, R. F., & Resnick, H. S. (1996).
Childhood physical assault as a risk factor for PTSD, depression, and substance abuse:
Findings from a national survey. American Journal of Orthopsychiatry, 66, 437–48.
Ehring, T., Ehlers, A., & Glucksman, E. (2008). Do cognitive models help in predicting the
severity of posttraumatic stress disorder, phobia, and depression after motor vehicle
accidents? A prospective longitudinal study. Journal of Consulting and Clinical
Psychology, 76, 219−230.
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.
Filipas, H. H., & Ullman, S. E. (2006). Child sexual abuse, coping responses, self-blame,
posttraumatic stress disorder, and adult sexual revictimization. Journal of Interpersonal
Violence, 21, 652–672.
Freedy, J. R., Resnick, H. S., Kilpatrick, D. G., Dansky, B. S., & Tidwell, R. P. (1994). The
psychological adjustment of recent crime victims in the criminal justice system. Journal
of Interpersonal Violence, 9, 450–468.
38
Fullerton, C. S., Ursano, R. J., Epstein, R. S., Crowley, B., Vance, K., Kao, T. C., Dougall, A., &
Baum, A. (2001). Gender differences in posttraumatic stress disorder after motor vehicle
accidents. American Journal of Psychiatry, 158(9), 1486–1491.
Galea, S., Tracy, M., Norris, F., & Coffey, S. (2008). Financial and social circumstances, the
incidence and course of PTSD in Mississippi during the first two years after Hurricane
Katrina. Journal of Traumatic Stress, 21, 357–368.
Gold, S. D., Marx, B. P., Soler-Baillo, J. M., & Sloan, D. M. (2005). Is life stress more traumatic
than traumatic stress? Journal of Anxiety Disorders, 19, 687–698.
Halligan, S. L., Michael, T., Clark, D. M., & Ehlers, A. (2003). Posttraumatic stress disorder
following assault: The role of cognitive processing, trauma memory, and appraisals.
Journal of Consulting and Clinical Psychology, 71, 419–431.
Hickling, E. J., Blanchard, E. B., Buckley, T. C., & Taylor, A. E. (1999). Effects of attribution of
responsibility for motor vehicle accidents on severity of PTSD symptoms, ways of
coping, and recovery over six months. Journal of Traumatic Stress, 12, 345 – 353.
Hoge, C. W., Castro, C. A., Messer, S. C., McGurk, D., Cotting, D. I., & Koffman, R. L. (2004).
Combat duty in Iraq and Afghanistan, mental health problems and barriers to care. New
England Journal of Medicine, 351, 13–22.
Irish, L., Ostrowski, S. A., Fallon, W. F., Spoonster, E., Van Dulmen, M., Sledjeski, E. M., &
Delahanty, D. L. (2008). Trauma history characteristics and subsequent PTSD symptoms
in motor vehicle accident victims. Journal of Traumatic Stress, 21, 77–384.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005).
Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National
Comorbidity Survey Replication. Archives of General Psychiatry, 62, 593–602.
39
Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., Wittchen,
H.-U. & Kendler, K. S. (1994). Lifetime and 12-month prevalence of DSM-III-R
psychiatric disorders in the United States: Results from the National Comorbidity Survey.
Archives of General Psychiatry, 51, 8–19.
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.
Kilpatrick, D. G., Acierno, R., Resnick, H. S., Saunders, B. E., & Best, C. L. (1997). A 2-year
longitudinal analysis of the relationship between violent assault and substance use in
women. Journal of Consulting and Clinical Psychology, 65, 834–847.
Kilpatrick, D. G., Acierno, R. E., Resnick, H. S., Saunders, B. E., & Best, C. L. (2000). Risk
factors for adolescent substance abuse and dependence: Data from a national sample.
Journal of Consulting and Clinical Psychology, 68, 19–30.
Kilpatrick, D. G., Resnick, H. S., & Acierno, R. (2009). Should PTSD Criterion A be retained?
Journal of Traumatic Stress, 22, 374–383.
Kilpatrick, D. G., Resnick, H. S., Baber, B., Guille, C., & Gros, K. (2010). National Stressful
Events Survey PTSD Module. Unpublished instrument.
Kilpatrick, D. G., Resnick, H. S., Freedy, J. R., Pelcovitz, D., Resick, P., Roth, S., et al. (1998).
Posttraumatic stress disorder field trial: Evaluation of the PTSD construct-Criteria A
through E. In T. Widiger, A. Frances, H. Pincus, R. Ross, M. First, W. Davis, et al.
(Eds.), DSM-IV Sourcebook, Vol. 4 (pp. 803–844). Washington, DC: American
Psychiatric Press.
40
Kilpatrick, D. G., Saunders, B. E., & Smith, D. W. (2003). Youth Violence: Prevalence and
Implications (Research in Brief NCJ 194972). Washington, DC: National Institute of
Justice.
Kleim, B., Ehlers, A., & Glucksman, E. (2007). Early predictors of chronic post-traumatic stress
disorder in assault survivors. Psychological Medicine, 37, 1457–1468.
Koenen, K. C., Harney, R., Lyons, M. J., Wolfe, J., Simpson, J. C., Goldberg, J., et al. (2002). A
twin registry study of familial and individual risk factors for trauma exposure and
posttraumatic stress disorder. Journal of Nervous and Mental Disease, 190, 209–218.
Kulka, R. A., Schlenger, W. E., Fairbank, I. A., Hough, R. L., Jordan, B. K., Marmar, C. R. and
Weiss, D. S. (1988). Contractual Report of Findings from the National Vietnam Veterans
Readjustment Study. Research Triangle Park, NC: Research Triangle Institute.
Lambert, J., Difede, J., & Contrada, R. (2004). The relationship of attribution of responsibility to
acute stress disorder among hospitalized burn patients. Journal of Nervous and Mental
Disease, 192, 304–312.
Long, M. E., Elhai, J. D., Schweinle, A., Gray, M. J., Grubaugh, A. L., & Frueh, B. C. (2008).
Differences in posttraumatic stress disorder diagnostic rates and symptom severity
between Criterion A1 and non-Criterion A1 stressors. Journal of Anxiety Disorders, 22,
1255–1263.
McLeer, S., Callaghan, M., Henry, D., & Wallen, J. (1994). Psychiatric disorders in sexually
abused children. Journal of the American Academy of Child and Adolescent Psychiatry,
33, 313–319.
McNally, R. J. (2003). Progress and controversy in the study of posttraumatic stress disorder.
Annual Review of Psychology, 54, 229–252.
41
McNally, R. J. (2009). Can we fix PTSD in DSM-V? Depression and Anxiety, 26, 597–600.
Moffit, T. E., Caspi, A., Taylor, A., Kokaua, J., Milne, B. J., Polanczyk, G., & Poulton, R.
(2010). How common are common mental disorders? Evidence that lifetime prevalence
rates are doubled by prospective versus retrospective ascertainment. Psychological
Medicine, 40, 899–909.
Murphy, S. A., Braun,T., Tillery, L., Cain, K. C., Johnson, L. C., & Beaton, R. D. (1999). PTSD
among Bereaved parents following the violent deaths of their 12- to 28-year-old children:
A longitudinal prospective analysis. Journal of Traumatic Stress, 12, 273–291.
Norman, S. B., Stein, M. B., Dimsdale, J. E., & Hoyt, D. B. (2008). Pain in the aftermath of
trauma is a risk factor for a post-traumatic stress disorder. Psychological Medicine, 38,
533–542.
North, C. S., Smith, E. M., & Spitznagel, E. L. (1994). Post-traumatic stress disorder in survivors
of a mass shooting. American Journal of Psychiatry, 15, 82–88.
O’Donnell, M. L., Creamer, M. C., Parslow, R., Elliott, P., Holmes, A. C. N., Ellen, S., et al.
(2008). A predictive screening index for posttraumatic stress disorder and depression
following traumatic injury. Journal of Consulting and Clinical Psychology, 76(6), 923–
932.
Pimlott-Kubiak, S., & Cortina, L. M. (2003). Gender, victimization, and outcomes:
Reconceptualizing risk. Journal of Consulting and Clinical Psychology, 71, 528–539.
Resnick, H. S., Kilpatrick, D., Dansky, B., Saunders, B., & Best, C. (1993). Prevalence of
civilian trauma and posttraumatic stress disorder in a representative national sample of
women. Journal of Consulting and Clinical Psychology, 61, 984–991.
42
Riggs, D., Rothbaum, B. O., & Foa, E. (1995). A prospective examination of symptoms of
posttraumatic stress disorder in vicitms of nonsexual assault. Journal of Interpersonal
Violence, 10, 201–214.
Roberts, A. L., Gilman, S. E., Breslau, J., Breslau, N., & Koenen, K. C. (2011). Race/ethnic
differences in exposure to traumatic events, development of post-traumatic stress
disorder, and treatment-seeking for post-traumatic stress disorder in the United States.
Psychological Medicine, 41, 71-83.
Roemer, L., Orsillo, S. M., Borkovec, T. D., & Litz, B. T. (1998). Emotional response at the time
of a potentially traumatizing event and PTSD symptomatology: A preliminary
retrospective analysis of the DSM-IV criterion A-2. Journal of Behavior Therapy and
Experimental Psychiatry, 29, 123–130.
Rothbaum, B. O., Foa, E. B., Riggs, D. S., Murdock, T., & Walsh, W. (1992). A prospective
evaluation of post-traumatic stress disorder in rape victims. Journal of Traumatic Stress,
5, 455–475.
Schnurr, P. P., Spiro, A., Vielhauer, M. J., Findler, M. N., & Hamblen, J. L. (2002). Trauma in
the lives of older men: Findings from the normative aging study. Journal of Clinical
Geropsychology, 8, 175–187.
Sharp, T. J., & Harvey, A. G. (2001). Chronic pain and Posttraumatic Stress Disorder: Mutual
maintenance? Clinical Psychology Review, 21, 857–877.
Smith, D. W., Saunders, B. E., & Kilpatrick, D. G. (2008). Prevalence and correlates of dating
violence in a national sample of adolescents. Journal of the American Academy of Child
and Adolescent Psychiatry, 47, 755–762.
43
Taylor, S., & Koch, W. J. (1995). Anxiety disorders due to motor vehicle accidents: Nature and
treatment. Clinical Psychology Review, 15, 721–738.
Tolin, D. F., & Foa, E. B. (2006). Sex differences in trauma and posttraumatic stress disorder: A
quantitative review of 25 years of research. Psychological Bulletin, 132, 959–992.
Weaver, T. L., & Clum, G. A. (1995). Psychological distress associated with interpersonal
violence: A meta-analysis. Clinical Psychology Review, 15, 115–140.
Weathers, F. W., & Keane, T. M. (2007). The Criterion A problem revisited: Controversies and
challenges in defining and measuring psychological trauma. Journal of Traumatic Stress,
20, 107–121.
Weathers, F., Litz, B., Herman, D., Huska, J., & Keane, T. (1993, October). The PTSD Checklist
(PCL): Reliability, validity, and diagnostic utility. Paper presented at the Annual
Convention of the International Society for Traumatic Stress Studies, San Antonio, TX.
Wells, J. E., & Horwood, L. J. (2004). How accurate is recall of key symptoms of depression? A
comparison of recall and longitudinal reports. Psychological Medicine, 34, 1001–1011.
Zatzick, D. F., Rivara, F. P., Nathans, A. B., Jurkovich, G. J., Wang, J., Fan, M., Russo, J.,
Salkever, D. S., & MacKenzie, E. J. (2007). A nationwide US study of post-traumatic
stress after hospitalization for physical injury. Psychological Medicine, 37, 1469–1480.
Zinzow, H. M., Rheingold, A. A., Hawkins, A. O., Saunders, B. E., & Kilpatrick, D. G. (2009).
Losing a loved one to homicide: Prevalence and mental health correlates in a national
sample of young adults. Journal of Traumatic Stress, 22, 20–27.
Zisook, S., Chentsova-Dutton, Y., & Shuchter, S. R. (1998). PTSD following bereavement.
Annals of Clinical Psychiatry, 10, 157–163.
44
Table 2.1. Criterion A definitions and accompanying text from DSM-IV and proposed revisions for DSM-5
DSM-IV
DSM-5 proposed revisions
The person has been exposed to a traumatic event in which both of the following
were present:
(1) the person experienced, witnessed, or was confronted with an event or
events that involved actual or threatened death or serious injury, or a threat to the
physical integrity of self or others.
(2) the person’s response involved intense fear, helplessness, or horror. Note:
In children, this may be expressed instead by disorganized or agitated behavior.
The person was exposed to one or more of the following event(s): death or threatened death
actual or threatened serious injury, or actual or threatened sexual violation, in one or more o
the following ways*:
1. Experiencing the event(s) him/herself
2. Witnessing, in person, the event(s) as they occurred to others
3.
Learning that the event(s) occurred to a close relative or close friend; in such cases
the actual or threatened death must have been violent or accidental
4. Experiencing repeated or extreme exposure to aversive details o
f the event(s) (e.g.,
first responders collecting body parts; police officers repeatedly exposed to details of child
abuse); this does not apply to exposure through electronic media, television, movies, or
pictures, unless this exposure is work related.
Accompanying text:
• . . . exposure to an extreme traumatic stressor involving direct personal
experience of an event that involves actual or threatened death or serious injury, or
other threat to one’s physical integrity; or witnessing an event that involves death,
injury, or a threat to the physical integrity of another person; or learning about
unexpected or violent death, serious harm, or threat of death or injury experienced
by a family member or other close associate (Criterion A1).
• The person’s response to the event must involve intense fear, helplessness,
or horror (or in children, the response must involve disorganized or agitated
behavior) (Criterion A2).
• Traumatic events that are experienced directly include, but are not limited
to, military combat, violent personal assault (sexual assault, physical attack,
robbery, mugging), being kidnapped, being taken hostage, terrorist attack, torture,
Accompanying text:
Not yet available
45
incarceration as a prisoner of war or in a concentration camp, natural or manmade
disasters, severe automobile accidents, or being diagnosed with a life-threatening
illness. For children, sexually traumatic events may include developmentally
inappropriate sexual experiences without threatened or actual violence or injury.
Witnessed events include, but are not limited to, observing the serious injury or
unnatural death of another person due to violent assault, accident, war, or disaster
or unexpectedly witnessing a dead body or body parts. Events experienced by
others that are learned about include, but are not limited to, violent personal assault,
serious accident, or serious injury experienced by a family member or a close
friend; learning about the sudden, unexpected death of a family member or a close
friend; or learning that one’s child has a life-threatening disease.
* For children, inclusion of loss of parent or other attachment figure is being considered.
DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (American Psychiatric Association, 1994).
DSM-5 = Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (proposed revisions, retrieved from:
http://www.dsm5.org/ProposedRevisions/Pages/proposedrevision.aspx?rid=165).