Posttraumatic stress disorder
Posttraumatic stress disorder (PTSD) become familiar into the world
psychiatric nomenclature in 1978 (World Health Organization, 1978) with
the issuance of the ICD-9, documenting the cross-cultural recognition
of the typical symptomatic response to exposure to traumatic life events
[1, 2] The characteristic core of the disorder includes the
the distressing oscillation between intrusion and avoidance. Intrusion is
characterized by nightmares, unbidden visual images of the trauma or its
aftermath while awake, intrusive thoughts about aspects of the traumatic
event, sequelae, or self-conceptions. Avoidance is typified by deliberate efforts to not think about the event, not talk about the event, and avoiding reminders of the event. Also characteristic are more active attempts to
push memories and recollections of the event or its aftermath out of mind
by increasing the use of alcohol or drugs, overworking, or other strategies
designed to divert attention or to so exhaust someone that he or she is
temporarily untouched by the intrusive phenomenology. In addition to
the frank avoidance, Horowitz also described emotional numbing as a not
uncommon sequel to exposure to a traumatic life event[2, 3]
Despite the usefulness of the original IES, complete assessment of the
response to traumatic events required tracking of response in the domain
of hyperarousal symptoms. Beginning with data from a longitudinal
study of the response of emergency services personnel to traumatic
events, including the Loma Prieta earthquake ( Weiss, Marmar,
Metzler, & Ronfeldt, 1995)[4], a set of seven additional items, with six to tap
the domain of hyperarousal, and one to parallel the DSM-III-R and now
DSM-IV diagnostic criteria for PTSD were developed, piloted, and then
used. These additional seven items were interspersed with the existing
seven intrusion and eight avoidance items of the original IES using a table
of random numbers to establish placement. The IES-R comprises these 22
items, and was originally presented in the first edition of this reference
work (Weiss & Marmar, 1997). [5]
An important consideration in the construction of the revised IES
was to maintain comparability with the original version of the measure as
much as was possible. Consequently, the one-week time frame to which
the instructions refer in measuring symptomatic response was retained,
as was the original scoring scheme of frequency – 0, 1, 3, and 5 for the
responses of “Not at all,” “Rarely,” “Sometimes,” and “Often.” The only
modification to the original items that were made was to change the item
“I had trouble falling asleep or staying asleep” from its double-barreled
status into two separate items. The first is simply “I had trouble staying
asleep” and because of a somewhat higher correlation between it and the
remaining intrusion items it was assigned to represent the original item
in the Intrusion subscale. The second item, “I had trouble falling asleep”
was assigned to the new Hyperarousal subscale because of its somewhat
higher correlation with the other hyperarousal items, it is somewhat lower
correlation with the intrusion items, and its more apparent link with
hyperarousal than with intrusion. The six new items comprising the
Hyperarousal subscale target the following domains: anger and irritability,
jumpiness and exaggerated startle response, trouble concentrating, psychophysiological
arousal upon exposure to reminders, and hypervigilance. As
mentioned earlier, the one new intrusion item taps the dissociative-like
reexperiencing captured in true flashback-like experiences. The reader is
referred to Weiss (2004) [6] for a summary of the internal consistency of the
three subscales, all of which were strong, the pattern of item-total correlations,
test-retest stability, which was also satisfactory, and commonality of
the interitem correlations.
Based on the experience with those data, and considerations of
the insufficiency of frequency as a completely summarizing marker for
self-report, the over-weighting of responses of “Sometimes” and “Often”
in the scoring scheme, the IES-R moulted into a measure with the following
characteristics: (1) the directions were modified so that the respondent
is not asked about the frequency of symptoms in the past 7 days but is
instead asked to report the degree of distress of the symptom in the past
7 days; (2) the response format was modified to a 0–4 response format with
equal intervals − 0 = Not at all, 1 = A little bit, 2 = Moderately, 3 = Quite a
bit, 4 = Extremely – rather than the unequal intervals of the original scale;
and (3) the subscale scoring was changed from the sum of the responses
to the mean of the responses, allowing the user to immediately identify
the degree of symptomatology merely by examining the subscale scores,
since they are presented in the same metric as the item responses,
something the original scale did not. According to Derogatis, (1994), [7] these changes brought the IES-R in parallel format to the SCL-90-R () allowing for direct comparison
of endorsement of symptom levels across these two instruments.
For a variety of reasons, many measures that have
contributed to the growing cross-cultural literature in traumatic stress
and PTSD were initially developed in English. The IES-R being no exception.
Consequently, for use with samples whose native language is not
English, a translation of the measure is required. Given this requirement,
it is useful to review some of the issues involved in that process.
Mallinckrodt and Wang (2004) [8] borrowed from Hambleton’s work on
the difference between literal translation and what is described as the adaptation
of items from one language to another. For example, the English
phrase “go on automatic pilot” if translated literally into German, will not
give the sense of engaging in behaviours without active deliberation that is
only recognized after the fact. Thus, if only a literal translation were
adopted, the reliability and validity would be compromised. Mallinckrodt
and Wang (p. 369) present Hulin’s view that “[a] pair of items from the
original scale and its adapted version are said to be equivalent when two
individuals with the same amount or level of the construct being measured
have equal probabilities of making the same response to the different
language versions of the same item.”
Flaherty et al. (1988) [9] suggested that there were five levels of equivalence
that an adapted measure should possess to show that it has
cross-cultural validity. The first, content equivalence, involves establishing
that the content domain of items is relevant and appropriate for both cultures.
The second, they describe as semantic equivalence, establishing that
each item of the new measure communicates the meaning of its parallel
item on the original scale. The third is more methodological: technical
equivalence addresses the question of whether the data collection method
(e.g., self-report) produces comparable results in each culture. The fourth,
criterion equivalence, involves evidence of parallel comparisons to within cultural
norms. The fifth and final equivalence is conceptual. This addresses
whether the intended construct or phenomenon has the same meaning in each
culture. The claim of ordered equivalence posits that subsequent levels
of equivalence cannot be achieved in the absence of equivalence in all
prior levels.
The efficiency and directness of the IES-R have led scholars in a variety of
different countries and cultures to produce versions in non-English languages.
A review of the literature revealed that the work accomplished for
the international versions approached the recommendations of
Mallinckrodt and Wang (2004)[8] to varying degrees. It is, of course, an
empirical question as to whether the detailed and extensive approach
suggested by these authors would produce a more reliable or valid version
than a more manageable approach. Published data are in the literature
for formal translations as well as ad hoc translations in the context of
an investigation of another question. Of the former, the following versions
can be found (listed alphabetically): Chinese (Wu & Chan, 2003),[10] French
(Brunet, St-Hilaire, Jehel, & King, 2003)[11], German (Maercker & Schuetzwohl,
1998),[12] Japanese (Asukai et al., 2002),[13] and Spanish (Baguena et al., 2001)[14].
A Bosnian version of the IES-R is described in a study that compares
refugees to nonrefugee (Hunt & Gakenyi, 2005).[15] and there is no validated Arabic version which follows Mallinckrodt and Wang protocol[8] and as there is need to assess the impact of different events on our society which compatible without language and culture to raise the bar of scale accuracy when used in our community