The psychological reactions after witnessing a killing in public in a Danish high school.

0.11 MB·English

by  Elklit, Ask

(cid:2)BASIC RESEARCH ARTICLE
The psychological reactions after
witnessing a killing in public in a
Danish high school
Ask Elklit1* and Sessel Kurdahl2
1National Centre forPsychotraumatology, UniversityofSouthern Denmark, Odense, Denmark;
2Ringkøbing-Skjern Municipality, Ringkøbing, Denmark
Background: School killings attract immense media and public attention but psychological studies
surroundingthese events arerare.
Objective: To examine the prevalence of posttraumatic stress disorder (PTSD) and possible risk factors of
PTSDin320Danishhighschoolstudents(meanage18years)7monthsafterwitnessingayoungmankilling
his formergirlfriend infrontof a largeaudience.
Method:ThestudentsansweredtheHarvardTraumaQuestionnaire(HTQ),theCrisisSupportScale(CSS),
andthe TraumaSymptom Checklist (TSC).
Results: Prevalence of PTSD 7 months after the incident was 9.5%. Furthermore, 25% had PTSD at a
subclinicallevel.Intimacywiththedeceasedgirl;feelingfear,helplessness,orhorrorduringthekilling;lack
of expressive ability; feeling let down by others; negative affectivity; and dissociation predicted 78% of the
varianceof the HTQtotal scores.
Conclusion:ItispossibletoidentifystudentswhoaremostlikelytosufferfromPTSD.Thisknowledgecould
beusedto intervene earlyon toreduceadversities.
Keywords: witnessingschoolkilling;PTSD;socialsupport;riskfactors
For the abstract or full text in other languages, please see Supplementary files under Article Tools
online
Received:4 October 2012; Revised:18 December 2012; Accepted:18 December 2012; Published:9 January 2013
Although they are less frequent than other types Also,beingawitnesstoakillinghadasignificantimpact
of trauma, school shootings and killings result onmostpeople,butthedegreeoftraumawasinfluenced
in massive public attention because it is easy to byindividualfactors.
imagine yourself or your children as targets in daily Most importantly, the effects of witnessing a killing
environments. Nevertheless, rather few studies havebeen at school and possible risk factors for development of
performed on this phenomenon and its psychological PTSDhavetoourknowledgeneverbeenstudiedwithina
consequences for the witnesses. Some studies have been highschoolpopulation.Althoughthereissomeresearch
carried out on witnessed killings in the workplace or on the effect of disasters on adolescents (e.g., Dogan,
inpublic,mainlywithadults(Creamer,1989;Elklit,1994; 2011; Ghazali, Elklit, Yaman, & Ahmad, 2012) and we
Johnson,North, & Smith, 2002; North, Smith, McCool, knowquitealotaboutthenegativeeffectsofcommunity
& Shea, 1989; North, Smith, & Spitznagel, 1994, 1997; violence (i.e., witnessing of killing, robbing, threats with
North,Spitznagel,&Smith,2001)butalsowithchildren weapons, etc.) on adolescents’ development, studies of
(Pynoos, Frederick, Nader, & Arroyo, 1987; Schwarz & communityviolencearenormallyconductedinhigh-risk
Kowalski, 1991). All of the studiespointed to symptoms adolescent populations, who have experienced multiple
of posttraumatic stress disorder (PTSD) as a possible traumatic events, andwhere the possible risk factors for
consequence of witnessing killings or murders in public. the development of psychopathology are confounded,
EuropeanJournalofPsychotraumatology2013. #2013AskElklitandSesselKurdahl.ThisisanOpenAccessarticledistributedunderthetermsofthe 1
CreativeCommonsAttribution-Noncommercial3.0UnportedLicense(http://creativecommons.org/licenses/by-nc/3.0/),permittingallnon-commercialuse,
distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.
Citation:EuropeanJournalofPsychotraumatology2013,4:19826-http://dx.doi.org/10.3402/ejpt.v4i0.19826
(pagenumbernotforcitationpurpose)
AskElklitandSesselKurdahl
for example, living in high-risk neighborhoods, having ing actual or threatened death, physical injury, or other
low socioeconomic status, living with parents involved threats to the physical integrity of the self or others;
in criminal activity, and possible substance abuse (Foy, criterion A2 requires that the initial response to the
Ritchie, & Conway, 2012; Stein, Jaycox, Kataoka, event involves fear, helplessness, or horror (American
Rhodes,&Vestal,2003).Almostnothingisknownabout Psychiatric Association, 1994). Criterion A2 is known
the effects of witnessing of a single incident of high to be a very good predictor of subsequent PTSD (e.g.,
schoolkillingintypical(non-high-risk)adolescentpopu- Brewin, Andrews, & Rose, 2000; Parslow et al., 2006).
lations. The goal of the current study was to identify In general, it appears that social support from family
demographic, peri-traumatic, and psychosocial factors and friends has a positive influence on the ability to
that influencedthe degreetowhichtheexperienceofthe cope with a trauma (Olff, 2012). Out of 14 separate risk
high school killing affected psychological adjustment factorsforPTSDinvestigatedinameta-analysis(Brewin
(expressedbyseverityof PTSDsymptoms)inadolescent et al., 2000), social support was the strongest predictor
students, 7 months after the incident. Well-established accounting for 40% of variance in PTSD severity. When
riskandprotectivefactorsfromthePTSDliteraturewere studies investigated both positive and negative support
chosen for investigation in the present study. elements, a negative social environment was a better
Gender plays an important role in the risk of trau- indicatorofPTSDsymptomatologythanlackofpositive
matization. Two meta-analyses (Brewin, Andrews, & support (Brewin & Holmes, 2003). The perception of
Valentine, 2000; Tolin & Foa, 2006) concluded that social support is also important in the prediction of
women are more likely than men to meet the criteria PTSD. In a study of 150 undergraduate students aged
for PTSD after a traumatic event, even though they 17(cid:2)22 years (mean age 19.3) who reported having
are less likely to experience potentially traumatic events experienced different types of trauma, Haden, Scarpa,
(seealsoOlff,Langeland,Draijer,&Gersons,2007).The Jones, and Ollendick (2007) investigated the moderating
twofold risk for PTSD among women could not be effectofperceivedsocialsupportonPTSDandperceived
attributed solely to a higher frequencyof specific, highly injury. The researchers concluded that the concept of
pathogenic potentially traumatic events such as sexual support, perceiving it or seeking it out, was one of
assault, but seemed to exist across trauma categories. In the most powerful variables essential for minimizing the
relation to witnessing public killings, North et al. (1994, severityof PTSD inyoung adults.
1997, 2001) reported a corresponding difference with Negative affectivity refers to a mood dimension and
20 and 36% of the men andwomen, respectively, having reflects a general tendency to react to and have a nega-
PTSD 1 month after the incident. The participants in tive perspective on the surrounding world and oneself
their study had all witnessed a gunman shooting in a (Krog & Duel, 2003; Watson & Clark, 1984). Negative
cafeteria, but early in the gunfire it became obvious that affectivity can either be seen as a state or as a person-
the shooter favored killing females. Therefore, it is un- ality factor similar to neuroticism (ibid.). Significant
clear whether this gender-related risk affected the differ- associations between PTSD and negative affectivity are
ence in the level of traumatization in male and female oftenfoundindifferenttraumapopulations(e.g.,Bennett,
witnesses in this particular study. Owen, Koutsakis, & Bisson, 2002; Feldner, Lewis, Leen-
The level of intimacy with the victim has been shown Feldner, Schnurr, & Zvolensky, 2006; Monson, Price,
to influence the witnesses’ PTSD severity. Dyregrov, Rodriguez,Ripley, &Warner,2004; Shapinsky, Rapport,
Frykholm, Lilled, Broberg and Holmberg (2003) investi- Henderson,&Axelrod,2005).
gated563adolescents’reactions(aged13(cid:2)19years;mean Dissociationmaybeunderstoodasresponsesoccurring
age15.4)7monthsafteradiscothequefireinSwedenthat in the course of a traumatic event, which cause a dis-
killed 63 young people. The levels of the participants’ ruption in the usually integrated functions of perception
traumatic reaction increased with increased intimacy, of the environment, memory, consciousness, or identity
thatis,knowingthedeceasedpersonally.Thesameresults (American Psychiatric Association, 1994; Bryant, 2007).
havebeenfoundinotherstudiesofchild,adolescent,and Dissociativeresponsesthatoccuratthetimeoftrauma
young adult witnesses of mass killings, natural disasters, (peritraumatic dissociation) have been described as the
and peers to adolescent suicide victims (Melhem et al., largest known risk factor for subsequent PTSD (Ozer,
2004;Parslow,Jorm,&Christensen, 2006;Pynoosetal., Best, Lipsey, & Weiss, 2003). However, Briere, Scott,
1987). This indicates that particular attention should be and Weathers (2005) showed that the less-investigated,
given to those who perceivedthemselves to be psycholo- trauma-specific dissociation, that occurs during or after
gically close to the deceased. an event and continues until the time of assessment
The A2 criterion in the PTSD diagnosis: The exposure (persistentdissociation),whentakenintoaccount,pushes
criterion in the PTSD diagnosis in DSM-IV has two aside the effect of peritraumatic dissociation on PTSD.
elements: Criterion A1 specifies that the person experi- Trauma-related persistent dissociation is also found to
enced, witnessed, or was confronted with events involv- be a substantial predictor of PTSD in other studies
2
Citation:EuropeanJournalofPsychotraumatology2013,4:19826-http://dx.doi.org/10.3402/ejpt.v4i0.19826
(pagenumbernotforcitationpurpose)
Psychologicalreactionsafterwitnessingakilling
(Ehlers & Clark, 2000; Halligan, Michael, Clark, & Denmark.Themajorityofthestudents(220(cid:3)69%)lived
Ehlers, 2003; Murray, Ehlers, & Mayou, 2002). with both of their parents, and 76 (24%) livedwith only
In spite of the fact that this is a retrospective study one parent. One fourth (82 persons) of the students had
(conducted7monthsaftertheincident),itsimportanceis experienced at least one potential traumatic event in
highlighted by the lackof other recorded studies of this their life and 45% (128 persons) had experienced, as a
specific trauma-type in adolescents. Given the relatively minimum, one important life eventduring the past year.
rare occurrence of public high school killings, and even
more rare scientific investigation of them, identifica- Procedure
tion of possible risk factors of PTSD in this particular The study took place 7 months after the incident. The
incident can be of value in guiding future screening questionnaireswerehandedouttothestudentsinsecond
programs to prevent chronification of symptoms in the and third grade at the high school and sent to the
adolescent victims. parents’ addresses of those students who graduated in
June (Danish high schools, ‘‘gymnasiums’’, have three
The incident grade levels).
The killing at the Hasseris High School in Aalborg,
Denmark,occurredonFriday,March3,2006,justwhen Instruments
theannualShrovetidepartywasgettingstarted.Afemale The questionnaire included questions concerning demo-
student ran through the front door into the open ward- graphyandperi-traumaticfactors(e.g.,spatialproximity
robe area closely followed by her former boyfriend with tothekilling,psychologicalclosenesstothevictim)along
whom she had broken up with 2 days earlier. He held a with the following instruments:
bloodstained knife in his hand and she was screaming. The Harvard Trauma Questionnaire*Part IV (HTQ;
After overpowering her in the middle of the student Mollica, Caspi-Yavin, Bollini, & Truong, 1992) consists
crowd,hestabbedherseveraltimesinthechest.Mostof of 16 items scored on a 4-point Likert scale. It measures
the students thought that what they saw was some kind theintensityofthethreecoresymptomgroups(intrusion,
oftheaterperformanceanddidnotinterfere.Theyslowly avoidance,andarousal)ofPTSD.TheDanishversionof
became aware of the severity, when the stabber fled and the HTQ has good psychometric properties, comparable
the police and an ambulance arrived. Almost 200 party to the original (Bach, 2003). The alphavalues for three
participants were detained inside the assembly hall and scales in the present study were 0.83 (intrusion), 0.70
a classroom for 3 hours until the police had found the (avoidance),and0.82(arousal);thetotalscalewithalpha
killer, whohad hunghimself inashednear hishome. In was0.93.Toqualifyforthefulldiagnosis,theappropriate
themeantime,itwasdifficultforthestudentstofindout numberof core symptoms should be]3 (‘‘often’’).
what had happened, and students and family members TheCrisisSupportScale(CSS)wasusedforratingthe
outside the high school had problems contacting their experience of perceived social support after a traumatic
friends and relatives inside the school because of an eventthroughsevenitems(Joseph,Andrews,Williams,&
overloaded mobile telephone network. The next day, Yule, 1992). The items include: (1) perceived availability
the headmaster and a crisis psychologist held an infor- ofsomeonetolisten;(2)contactwithpeopleinasimilar
mation meeting for all of the students, relatives, and situation; (3) the ability to express oneself; (4) received
teachers about the incident and three local psychologists sympathy and support; (5) practical support; (6) the
offered counseling for those in need during the follow- experienceofbeingletdown;and(7)generalsatisfaction
ing days. with social support. The items are rated on a 7-point
Likertscale,rangingfrom‘‘never’’to‘‘always.’’TheCSS
Method has been used in several disaster studies*it has good
internalconsistencyaswellasgooddiscriminatorypower.
Participants Elklit,Pedersen,andJind(2001)haveconfirmedthepsy-
All 415 students, who were attending the high school at chometricalreliabilityandvalidityoftheCSSinDanish.
the time of the incident, were asked to participate in the Alpha for the total CSS score in the present study was
study. A total of 320 (77%) returned the questionnaire. 0.73.
Therespondentswereaged16(cid:2)20years,withameanage The Trauma Symptom Checklist (TSC) (Briere &
of 17.99 years (SD(cid:3)1.05). Of these, 62.2% were female Runtz, 1989) measures the occurrence of psychological
(199 persons). The parents of the students were well symptomsassociatedwithtrauma.Inthepresentstudy,a
educated: 36% of the fathers and 26% of the mothers Danishversionwith26items(TSC-26)wasused(Krog&
hadauniversitydegree,andonly11%ofthefathersand Duel, 2003). The Danish TSC has good psychometric
9%ofthemothershadendedtheireducationafter9years qualities and is avalid instrument measuring the effects
in public school. Just five (2%) of the students did not of traumatization (ibid.). The TSC covers three dimen-
have Danish citizenship and 19 (6%) were born outside sions and is scored on a 4-point Likert scale. The alpha
3
Citation:EuropeanJournalofPsychotraumatology2013,4:19826-http://dx.doi.org/10.3402/ejpt.v4i0.19826
(pagenumbernotforcitationpurpose)
AskElklitandSesselKurdahl
values in the present study were a(cid:3)0.83 (negative PTSDsymptomatology
affectivity; 11 items), a(cid:3)0.83 (somatization; 10 items), Seven months after the killing, 28 persons (9.5%) met
and a(cid:3)0.70 (dissociation; 5 items); total TSC a(cid:3)0.91. the three symptom-cluster criteria of a PTSD diagnosis.
In addition, 74 persons (25%) met the criteria for two
PTSD symptom clusters and for a diagnosis of subclini-
Dataanalyses
cal PTSD.
Rates are reported as raw numbers and percentages.
Meansarereportedwithstandarddeviations.Thefollow-
Factorsassociatedwithposttraumatic
ing preliminary tests were conducted to test the associa-
symptomatology
tionbetweenthepresumedrisk-factorsandsymptomsof
PTSD: x2-analyses were performed to test associations
Correlations
between dichotomous variables. Correlations were esti-
Correlations were computed between HTQ total scores
mated with Pearson’s correlation coefficient. ANOVA
andallothervariablestoidentifyfactorsassociatedwith
analyses were used to investigate associations between
posttraumatic symptomatology as assessed by the HTQ.
dichotomousandscalevariablesandtoassesseffectsizes.
Significant correlations between HTQ total and demo-
Variables that had significant associations with PTSD
graphic and peri-traumatic variables included intimacy
symptoms were subsequently entered in a hierarchical
(r(cid:3)0.28, pB0.0005), CSS item 3: lack of expressive
regressionanalysis.
ability(r(cid:3)0.37,pB0.0005),CSSitem6:feelingletdown
(r(cid:3)0.49, pB0.0005), TSC: negative affectivity (r(cid:3)0.76,
Results pB0.0005), and dissociation (r(cid:3)0.69, pB0.0005). One-
way ANOVA analyses showed a positive relationship
Exposuretothekilling between the HTQ total score and female gender
Sixty-four persons(19%) were eyewitnesses to the crime; (F(1,278)(cid:3)40.88, pB0.0005; mean HTQ for males(cid:3)46.2
71 (22%) were in the assembly hall nextto the wardrobe (SD 9.6) and for females(cid:3)57.2 (SD 16.1; h2(cid:3)12.8) and
area;48(15%)wereelsewhereinthehighschool;18(6%) the A2 criterion (F(3,276)(cid:3)44.95, pB0.0005; HTQ for
were outside, but went inside; 38 (12%) were outside no A2(cid:3)41.6 (SD 10.5) and for A2(cid:3)55.9 (SD 14.5);
andnevergotin;85(27%)didnottakepartintheparty; h2(cid:3)14.8).
one person did not answer. The students outside could
see the dead body as the front door was made of glass. Regression analysis
During the killing, 189 (59%) reported feeling helpless Thefirststepconsistedofgender,andthesecondstepof
andpowerless,161(50%)felthorrorand8(3%)thought peri-traumatic factors (intimacy and A2), because these
they were going to die. These feelings from now on three factors were assumed to stay constant across the
referred to as ‘‘A2’’. courseofthe7monthssincetheexposuretotheincident.
The third step consisted of aspects of social support
(CSS item three and CSS item six), and the fourth step
Intimacywiththedeceased
of individual differences in reactions to trauma (TSC
Three percent (9 persons) knew the murdered female
negative affectivity and dissociation). The final model
student‘‘verywell’’and6%(18persons)knewher‘‘well’’.
included six variables, which in all explained 78% of the
Nine percent (28 persons) knew her ‘‘some’’, 21% (68
total HTQ score (Table 1). The contribution of gender
persons) ‘‘a little’’, and 61% (195 persons) did not know
was no longer significant in the last step. The variables
her at all. Two persons did not answer this question.
with the strongest individual contribution to the level of
PTSD symptoms were (in orderof appearance) negative
Posttraumasocialsupport affectivity, persistent dissociation, and lack of social
The CSS total score had an average of 40 (SD(cid:3)5.7) support (‘‘feeling let down’’).
indicating that the students in general experienced good
social support. During the following days, there were Discussion
mourning ceremonies and class talks with counselors. Sevenmonthsaftertheincident,35%ofthestudentswere
In addition, 35% (112 persons) spoke individually with traumatized to at least a subclinical degree. This degree
a psychologist at the high school, 19% (61 persons) of traumatization existed in spite of the fact that every
with their curriculum counselor, and 10% (32 persons) student in addition to the collective interventions was
with the principal. Eight percent (25 persons) were in offered individual sessions with a crisis psychologist
contact with a crisis psychologist outside the school closely after the incident (which 1/3 accepted). The
and 7% (21 persons) with their general practitioner. Six Cochrane Collaboration made a comprehensive review
percent (16 persons) indicated that they still felt in need (Rose, Bisson, Churchill, & Wessely, 2002) of a single
of psychological treatment 7 months after the killing. session individual debriefing and found no evidence of
4
Citation:EuropeanJournalofPsychotraumatology2013,4:19826-http://dx.doi.org/10.3402/ejpt.v4i0.19826
(pagenumbernotforcitationpurpose)
Psychologicalreactionsafterwitnessingakilling
Table1. Hierarchicalregressionanalyses forvariables stable individual factors (i.e., personality traits) and
predicting posttraumaticseverityin witnesses toapublic therefore possible risk factors that influence the indivi-
killing(n(cid:3)269) dual reactions to trauma. Difficulties in labeling and
communicating emotions are central ingredients of
Step Variable b F df Adj.R2 DR2 alexithymia.Negativeaffectivity/neuroticismisrelatedto
alexithymia(Wise&Mann,1994),andthelackofability
1 Gender 0.35*** 38.0 1,268 0.12*** toexpressemotionsaboutthetraumaticeventmightvery
2 Gender 0.21*** 51.2 3,266 0.36 0.24*** well be the facet of the overallneurotic reactionpattern.
Intimacy 0.17** Although no firm conclusions about the causality of
A2 0.45***
the factors can be made because of the cross-sectional
3 Gender 0.15** 62.1 5,264 0.53 0.17***
designofthisstudy,negativeaffectivityseemstoactasa
Intimacy 0.14**
predisposing factor to more severe traumatic reactions
A2 0.39***
in the present adolescents. This is the most likely inter-
Lackofexpressive (cid:4)0.26***
pretation in light of the abundant former research with
ability
adults that points to the same associations between
Feelingletdown 0.29***
traumaandstabletraitslikenegativeaffectivity.
4 Gender 0.06 140.2 7,262 0.78 0.25***
In light of the retrospective nature of this study, the
Intimacy 0.08*
predictive power of the peri-traumatic factors (A2 crite-
A2 0.18***
rion, and the intimacy with the victim) might have been
Lackofexpressive (cid:4)0.12***
overridden by the simultaneous measuring of factors,
ability
which at the time of the assessment (7 months after the
Feelingletdown 0.21***
trauma), are probably indicators of developing patholo-
TSCnegative 0.36***
gical reactions to trauma (e.g., dissociation, and lack of
affectivity
socialsupport).Thismeansthatthefeelingofhorrorand
TSCdissociation 0.33***
helplessness during the trauma, and intimate knowledge
of the victim, might have had stronger predictive power
*p(cid:2)0.01;**pB0.001;***pB0.0005.
than dissociation and feelings of being let down, had
itasausefultreatmentforthepreventionofPTSDafter they been measured more closely to the incident itself.
traumaticincidents. As pointed out in the introduction, the A2 criterion and
We found that a considerable amount (78%) of the theintimateknowledgeofthevictimarealsowell-known
variance in the level of traumatic response could be risk factors for the development of PTSD in other
predicted with six well-known risk factors from the traumatizedpopulations.Inaddition,theyarebotheasily
general PTSD literature. First, because of the cross- assessed. Therefore, we propose that they be used as the
sectional nature of this study and the delayed measure- firstlineofscreeninginthemoreimmediateaftermathof
ment of the traumatic effects, some of the significant similarincidents.
predictors could be interpreted as a result of the Furthermore, the possibility that witnesses in risk of
traumatization, rather than risk factors. That is, lack of developing PTSD to some degree might have a lack of
social support can be seen as a consequence of having expressive ability needs to be considered when offering
developedPTSDsymptoms,whichinturnhinderssocial crisis intervention as this might lower the probability of
support seeking (Kaniasty & Norris, 2008). Also, at the the witnesses seeking help. In Denmark, after a poten-
time of the measurement, the dissociation levels can be tially traumatic event such as rape, death of loved ones,
seen as a representation of pathogenic traumatic reac- physicalassault,accidents,andothers,crisisintervention
tions akin to PTSD. This is the most logical explanation is available for the general population (Elklit & Nielsen,
because the very definition of PTSD includes some 2008). Taking a lack of expressive ability into account,
dissociativesymptoms.Innaturalremissionfromtrauma, those at risk should then not only be offered help but
dissociationsymptomsfadeawaywithtime(Dalenberg& also be strongly advised to seek help in a proactive way.
Carlson, 2012). Persistent dissociation can therefore, in This is an especially important consideration in relation
the present population, like in many other traumatized to adolescent witnesses, whose further development
populations(seetheintroduction), mostlikelybeindica- might be jeopardized by persistent traumatic symptoms.
tive of a treatment demanding reaction to trauma ex- Ehlers and Clark (2003) support the need for strategies
posure. This is an important point, because it enables to identify people who are unlikely to recover on their
us to identify adolescents in need of more intensive own. Similarly, the Cochrane review (Rose et al., 2002)
treatment. concludes that the compulsory debriefing of victims of
Contrary to this, negative affectivity and lack of trauma should cease and the focus of early intervention
expressive ability can almost certainly be interpreted as should instead, like we suggest, be targeted at the risk
5
Citation:EuropeanJournalofPsychotraumatology2013,4:19826-http://dx.doi.org/10.3402/ejpt.v4i0.19826
(pagenumbernotforcitationpurpose)
AskElklitandSesselKurdahl
group. The development of screening instruments and Brewin, C. R., & Holmes, E. A. (2003). Psychological theories of
intervention programs is therefore of great importance. posttraumatic stress disorder, Clinical Psychology Review.
SpecialIssue:PostTraumaticStressDisorder,23(3),339(cid:2)376.
The generalizability of the results of this study is
Briere, J., & Runtz, M. (1989). The trauma symptom checklist
limited by: the cross-sectional design, lack of control
(TSC-33):Earlydataonanewscale. JournalofInterpersonal
group, and the time between incident and assessment, Violence,4(2),151(cid:2)163.
as well as the use of self-reporting. Further research Briere, J., Scott, C., & Weathers, F. (2005). Peritraumatic and
is therefore needed of killings in a school context to persistent dissociation in the presumed etiology of PTSD.
investigatethepredictorsofPTSDinadolescentwitnesses AmericanJournalofPsychiatry,162(12),2295(cid:2)2301.
Bryant,R.A.(2007).Doesdissociationfurtherourunderstanding
tokillingsandothertypesofpublictraumas.Evenso,the
ofPTSD?JournalofAnxietyDisorders,21(2),183(cid:2)191.
associations between aspects of trauma and subsequent
Creamer,M.(1989).Post-traumaticstressdisorder:Somediagnostic
traumatic reactions, which are known from the adult and clinical issues. Australian and New Zealand Journal of
populations, arebeing confirmedinadolescentwitnesses Psychiatry,23(4),517(cid:2)522.
of public killings in the present study. This provides us Dalenberg, C., & Carlson, E. B. (2012). Dissociation in posttrau-
matic stress disorder part II: How theoretical models fit the
with valuable information that can be acted upon, until
empirical evidence and recommendations for modifying the
wehavebetter studies to informus.
diagnostic criteria for PTSD. Psychological Trauma: Theory,
Research, Practice, and Policy, 14, 551(cid:2)569. doi:10.1037/
a0027900
Conclusion
Dogan, A. (2011). Adolescents’ posttraumatic stress reactions and
This studyconfirms that being awitness to a killing can behavior problems following Marmara earthquake. European
be avery traumatic experience and identifies six central Journal of Psychotraumatology, 2, 1(cid:2)9. doi:10.3402/ejpt.v2i0.
peri- and posttraumatic variables, which explain a sub- 5825
Dyregrov, A., Frykholm, A., Lilled, L., Broberg, A. G., &
stantial amount of the variation in PTSD. Screening for
Holmberg, I. (2003). The Go¨teborg discotheque fire, 1998.
the variables, that is, intimacy with the deceased, A2
ScandinavianJournalofPsychology,44(5),449(cid:2)457.
criterion, lack of expressive ability, feeling let down by Ehlers,A.,&Clark,D.M.(2000).Acognitivemodelofposttrau-
others, negative affectivity, and dissociation in the wit- matic stress disorder. Behaviour Research and Therapy, 38(4),
nesses after a similar incident may well be useful to 319(cid:2)345.
Ehlers,A.,&Clark,D.M.(2003).Earlypsychologicalinterventions
identify those who will develop PTSD and thereby
foradultsurvivorsoftrauma:Areview.BiologicalPsychiatry,
constitute a group on which further psychological treat-
53(9),817(cid:2)826.
mentshouldbeconcentrated.
Elklit, A. (1994). Skuddramaet pa˚ Aarhus Universitet: en analyse
af de psykologiske efterreaktioner og mestringsstrategier
[TheshootingdramaattheUniversityofAarhus:Ananalysis
ofthepsychologicalimpactandcopingstrategies].Psykologisk
Acknowledgement
SkriftserieAarhus,19(3),1(cid:2)76.
Elklit, A., & Nielsen, L. H. (2008). Public subsidized assistance
The authors wish to express their gratitude to the students who
for psychological counseling in The Danish Public National
participatedandtothegoodwilloftheHasserisHighSchoolstaff
HealthInsuranceService.BriefTreatmentandCrisisInterven-
towardfacilitatingcontactwiththestudents.
tion,8(4),352(cid:2)357.doi:10.1093/brief-treatment/mhn022
Elklit,A.,Pedersen,S.S.,&Jind,L.(2001).Thecrisissupportscale:
Psychometricqualitiesandfurthervalidation.Personalityand
References IndividualDifferences,31(8),1291(cid:2)1302.
Feldner,M.T.,Lewis,S.F.,Leen-Feldner,E.W.,Schnurr,P.P.,&
Zvolensky, M. J. (2006). Anxiety sensitivity as a moderator
AmericanPsychiatricAssociation.(1994).Diagnosticandstatistical
of the relation between trauma exposure frequency and
manualofmentaldisorder(4thed.).Washington,DC:Author.
posttraumatic stress symptomatology. Journal of Cognitive
Bach,M.E.(2003).EnempiriskbelysningoganalyseafEmotional
Psychotherapy,20(2),201(cid:2)213.
Numbing som eventual selvstændig faktor i PTSD [An
Foy, D., Ritchie, I., & Conway, A. (2012). Trauma exposure,
empirical analysis of emotional numbing as a potential
posttraumatic stress, and comorbidities in female adolescent
independent factor in PTSD]. Psykologisk Studieskriftserie,
6(1),1(cid:2)199. offenders: Findings and implications from recent studies.
Bennett, P., Owen, R. L., Koutsakis, S., & Bisson, J. (2002). EuropeanJournalofPsychotraumatology,3,1(cid:2)13.doi:10.3402/
Personality, social context and cognitive predictors of post- ejpt.v3i0.17247
traumatic stress disorder in myocardial infarction patients. Ghazali, S. R., Elklit, A., Yaman, K., & Ahmad, M. (2012).
Psychology&Health,17(4),489(cid:2)500. SymptomsofPTSDandits predictivefactorsamongadoles-
Brewin, C. R., Andrews, B., & Rose, S. (2000). Fear, helplessness, cents exposed to tsunami disaster in northern peninsular
and horror in posttraumatic stress disorder: Investigating Malaysia: Four years after natural disaster. Journal of Loss
DSM-IV criterion A2 in victims of violent crime. Journal of andTrauma,18,260(cid:2)274.doi:10.1080/15325024.2012.688703
TraumaticStress,13(3),499(cid:2)509. Haden,S. C., Scarpa,A.,Jones, R. T., &Ollendick,T. H.(2007).
Brewin,C.R.,Andrews,B.,&Valentine,J.D.(2000).Meta-analysis Posttraumatic stress disorder symptoms and injury: The
of risk factors for posttraumatic stress disorder in trauma- moderating role of perceived social support and coping for
exposedadults.JournalofConsultingandClinicalPsychology, young adults. Personality and Individual Differences, 42(7),
68(5),748(cid:2)766. 1187(cid:2)1198.
6
Citation:EuropeanJournalofPsychotraumatology2013,4:19826-http://dx.doi.org/10.3402/ejpt.v4i0.19826
(pagenumbernotforcitationpurpose)
Psychologicalreactionsafterwitnessingakilling
Halligan, S. L., Michael, T., Clark, D. M., & Ehlers, A. (2003). Olff,M.(2012).Bondingaftertrauma:Ontheroleofsocialsupport
Posttraumatic stress disorder following assault: The role of andtheoxytocinsystemintraumaticstress.EuropeanJournal
cognitiveprocessing,traumamemory,andappraisals.Journal OfPsychotraumatology,3,1(cid:2)11.doi:10.3402/ejpt.v3i0.18597
ofConsultingandClinicalPsychology,71(3),419(cid:2)431. Olff, M., Langeland, W., Draijer, N., & Gersons, B. P. R. (2007).
Johnson, S. D., North, C. S., & Smith, E. M. (2002). Psychiatric Genderdifferencesinposttraumaticstressdisorder.Psycholo-
disorders among victims of a courthouse shooting spree: A gicalBulletin,133(2),183(cid:2)204.
three-yearfollow-upstudy.CommunityMentalHealthJournal, Ozer, E. J., Best, S. R., Lipsey, T. L., & Weiss, D. S. (2003).
38(3),181(cid:2)194. Predictors of posttraumatic stress disorder and symptoms in
Joseph, S., Andrews, B., Williams, R., & Yule, W. (1992). Crisis adults:Ameta-analysis.PsychologicalBulletin,129(1),52(cid:2)73.
support and psychiatric symptomatology in adult survivors
Parslow,R.A.,Jorm,A.F.,&Christensen,H.(2006).Associations
of the Jupiter cruise ship disaster. British Journal of Clinical
of pre-trauma attributes and trauma exposurewith screening
Psychology,31(1),63(cid:2)73.
positive for PTSD: Analysis of a community-based study of
Kaniasty,K.,&Norris,F.H.(2008).Longitudinallinkagesbetween
2085youngadults.PsychologicalMedicine,36(3),387(cid:2)395.
perceived social support and posttraumatic stress symptoms:
Pynoos,R.S.,Frederick,C.,Nader,K.,&Arroyo,W.(1987).Life
Sequential roles of social causation and social selection.
threatandposttraumaticstressinschool-agechildren.Archives
JournalofTraumaticStress,21,274(cid:2)281.
ofGeneralPsychiatry,44(12),1057(cid:2)1063.
Krog,T., &Duel,M. (2003).Envalideringogrevidering[Traume
Rose,S.,Bisson,J.,Churchill,R.,Wessely,S.(2002).Psychological
symptomcheckliste(TSC)].PsykologiskStudieskriftserie,6(4),
debriefing for preventing post traumatic stress disorder
1(cid:2)164.
(PTSD).CochraneDatabaseSystematicReview,2,CD000560.
Melhem,N.M.,Day,N.,Shear,M.K.,Day,R.,Reynolds,C.F.,III,
Schwarz, E. D., & Kowalski, J. M. (1991). Malignant memories:
& Brent, D. (2004). Predictors of complicated grief among
adolescents exposed to a peer’s suicide. Journal of Loss & PTSDinchildrenandadultsafteraschoolshooting.Journalof
Trauma. Special Issue: Risk and Resiliency Following Trauma theAmericanAcademyofChild&AdolescentPsychiatry,30(6),
andTraumaticLoss,9(1),21(cid:2)34. 936(cid:2)944.
Mollica,R.F.,Caspi-Yavin,Y.,Bollini,P.,&Truong,T.(1992).The Shapinsky,A.C.,Rapport,L.J.,Henderson,M.J.,&Axelrod,B.N.
Harvard trauma questionnaire: Validating a cross-cultural (2005). Civilian PTSD scales: Relationships with trait char-
instrument for measuring torture, trauma, and posttraumatic acteristicsandeverydaydistress.Assessment,12(2),220(cid:2)230.
stressdisorderinIndochineserefugees.JournalofNervousand Stein, B. D., Jaycox, L. H., Kataoka, S., Rhodes, H. J., & Vestal,
MentalDisease,180(2),111(cid:2)116. K.D. (2003).Prevalence ofchildandadolescentexposureto
Monson, C. M., Price, J. L., Rodriguez, B. F., Ripley, M. P., & community violence. Clinical Child and Family Psychology
Warner, R. A. (2004). Emotional deficits in military-related Review, 6(4), 247(cid:2)264. doi:http://dx.doi.org/10.1023/B:CCFP.
PTSD: An investigation of content and process disturbances. 0000006292.61072.d2
JournalofTraumaticStress,17(3),275(cid:2)279. Tolin, D. F., & Foa, E. B. (2006). Sex differences in trauma and
Murray, J., Ehlers, A., & Mayou, R. A. (2002). Dissociation and posttraumaticstressdisorder:Aquantitativereviewof25years
post-traumaticstressdisorder:Twoprospectivestudiesofroad ofresearch.PsychologicalBulletin,132(6),959(cid:2)992.
trafficaccidentsurvivors.BritishJournalofPsychiatry,180(4), Watson, D., & Clark, L. A. (1984). Negative affectivity: The
363(cid:2)368. disposition to experience aversive emotional states. Psycholo-
North,C.S.,Smith,E.M.,McCool,R.E.,&Shea,J.M.(1989). gicalBulletin,96(3),465(cid:2)490.
Short-term psychopathology in eyewitnesses to mass murder.
Wise, T. N., & Mann, L. S. (1994). The relationship between
Hospital&CommunityPsychiatry,40(12),1293(cid:2)1295.
somatosensory amplification, alexithymia, and neuroticism.
North, C. S., Smith, E. M., & Spitznagel, E. L. (1994). Posttrau-
JournalofPsychosomaticResearch,38(6),515(cid:2)521.doi:http://
maticstressdisorderinsurvivorsofamassshooting.American
dx.doi.org/10.1016/0022-3999(94)90048-5
JournalofPsychiatry,151(1),82(cid:2)88.
North, C. S., Smith, E. M., & Spitznagel, E. L. (1997). One-year
follow-upofsurvivorsofamassshooting.AmericanJournalof *AskElklit
Psychiatry,154(12),1696(cid:2)1702. Campusvej55,5320OdenseM
North,C.S.,Spitznagel,E.L.,&Smith,E.M.(2001).Aprospective Denmark
study of coping after exposure to a mass murder episode. Email:[email protected]
AnnalsofClinicalPsychiatry,13(2),81(cid:2)87.
7
Citation:EuropeanJournalofPsychotraumatology2013,4:19826-http://dx.doi.org/10.3402/ejpt.v4i0.19826
(pagenumbernotforcitationpurpose)