!" #$"
#% &'!&()&
#*% +!'',!&,!./0//../!,1('
! "#$"! % & % %
' ()
% *+, - . .,
* '/0, !, - . 1
*
2
).& '
3& 0
, , ", 45 , , 678
9"/!!4/:"4/#
- & % 3& 3& % %, - % % & % 3& , -.
. , 0 .
. & % :
8-0; 6)0< 7
"99: & &
,
=
& %
&
%% % . , - .
% & .
#9 :5 . 3& . & % &
. &
%% , - &
% /
.
. & ,
-
& 3 %% &
, - % 3 $ . 0> ?
@
% ))'6A &
A %
& & % / B
B % B B%
B, 6 .
% % %& %% , -
% & . &
, >
&
.
. . 3 ,
*
3& ,
!" # $# %& '()# # *!+',-.
# C '/0% * !
68 ":!"/: :
678 9"/!!4/:"4/#
$$$$
/45! ($DD,3,DEF$$$$
/45!+
Papers included
I.
Larsson, G., Michel, P-O., Lundin, T. (2000). Systematic assessment
of mental health following various types of post-trauma support.
Military Psychology, 12, 121-135.
II.
Michel, P-O., Lundin, T., Larsson, G. (2003). Stress Reactions among
Swedish Peacekeeping Soldiers Serving in Bosnia: A Longitudinal
study. Journal of Traumatic Stress, 6, 589-593.
III.
Michel, P-O., Lundin, T., Larsson, G. (2004). Personality Disorders in
a Swedish Peacekeeping Unit. Accepted for publication in Nordic
Journal of Psychiatry.
IV.
Michel, P-O., Lundin, T., Larsson, G. (2005). Suicide among
Former Swedish Peacekeeping Personnel. Manuscript.
Contents
Introduction.....................................................................................................9
Mental health among soldiers through history ...........................................9
Primary interventions – “Forward Psychiatry” ........................................10
The Vietnam War and post-traumatic stress disorder (PTSD) ................11
Stress and mental health in personnel serving in peacekeeping
operations ..............................................................................................12
Early group interventions and psychological debriefing..........................14
Personality disorders among peacekeepers ..............................................17
Suicide in military settings.......................................................................18
Ethics........................................................................................................20
Aim of this thesis......................................................................................20
Material and methods....................................................................................21
Participants ...............................................................................................21
Paper I-III ............................................................................................21
Paper IV...............................................................................................21
Methods....................................................................................................23
Pre-deployment assessment (paper I) ..................................................23
Post-deployment assessment (paper I).................................................24
One-year follow up (paper II and III) ..................................................25
Register study (paper IV).....................................................................26
Statistics ...................................................................................................26
Results...........................................................................................................28
General Mental Health (papers I and II) ..................................................28
Traumatic events (paper I) .......................................................................28
Potential predictors for poor mental health (papers I and II) ..................29
Effect of different kinds of support (paper I) ...........................................30
Personality disorders and their impact on general mental health
(paper III) ..............................................................................................31
Suicide among former peacekeepers (paper IV) ......................................32
Dropouts (papers I-III) .............................................................................32
General discussion ........................................................................................33
Conclusions...................................................................................................38
Acknowledgements.......................................................................................39
References.....................................................................................................41
Abbreviations
BICEPS
Brevity, immediacy, centrality, expectancy,
proximity and simplicity
CI
Confidence interval
CISD
Critical incident stress debriefing
DIP-Q
DSM-IV & ICD-10 Personality Questionnaire
DSM-III (-III R, -IV) Diagnostic and Statistical Manual of Mental
Disorders. Third Edition (third edition revised and
fourth edition
FFPI
Five Factor Personality Inventory
GHQ-28
General Health Questionnaire (28 questions)
HGD
Historical group debriefing
ICD-7 (-8,-9WHO International Classification of Diseases
10)
Revision 7 (revisions 8 through 10)
IFOR
Implementation Force (of the Dayton Agreement in
the Balkans)
NATO
North Atlantic Treaty Organisation
NVVRS
National Vietnam Veterans Readjustment Study
PfP
Partnership for Peace (Non-member states attached to
NATO in peacekeeping operations)
PIE
Proximity, immediacy and expectancy
PTSD
Posttraumatic stress disorder
SMR
Standard Mortality Ratio
SOC
Sense of coherence
SD
Standard deviation
UK
United Kingdom
UN
United Nations
UNIFIL
United Nations Interim Forces in Lebanon
US
United States
USA
United States of America
WHO
World Health Organisation
WW I
World War I
WW II
World War II
Introduction
Mental health among soldiers through history
Over millions of years, humans have evolved the ability, when threatened
with danger, to respond with a stress reaction to prepare them to fight or to
flee. Throughout history soldiers have been common victims of this
experience, and the history of psychotraumatology primarily relates to the
development of knowledge in the military field. Shepard (2000) and
Weisaeth (2002) present an elegant description of this history. A short
summary of their work is outlined below. For furthermore details, see these
two authors.
Nightmares among soldiers were described already by Homer in the Iliad
(Shay, 1994). Field Marshall von Goltz described reactions among Danish
soldiers in the Battle of Lund in 1656 as “Terror Panicus”. In 17th Century
Switzerland (Hofer, 1678) and during the American Civil War in the 19th
Century, soldiers’ symptoms were referred to as ”Nostalgia”. It can be easily
seen that the descriptions of soldiers’ reactions to battle are a function of the
views of contemporary society. In the latter part of 19th Century, soldiers’
reactions focused primarily on somatic experiences, and the diagnosis
consequently changed to “irritable heart syndrome” or “soldier’s heart” (da
Costa, 1871). Other civilian examples are the terms “railway spine” and
“railway brain”, describing the injuries following accidents involving the
recently introduced new mode of transport, the train (Erichsen, 1866, 1886).
This focus on somatic origin would continue for a considerable period of
time and can obviously be explained by a lack of knowledge concerning the
relationship between mind and body at that time. There were just a handful
of far-sighted doctors who pointed out that these reactions might have a
psychological origin and perhaps be the result of pure fright (Page, 1883).
Another important aspect is that knowledge concerning soldiers’ reactions to
their experiences in the battlefield seems to have been lost between the wars
(Ahrenfeldt, 1958). Psychiatrists found during the 1905 Russo-Japanese
War, for example, that mental illness in soldiers worsened if they were
evacuated from the field. They called this phenomenon “evacuation
neurosis”. This lesson had to be learned repeatedly over the years that
followed.
9
The unwillingness to accept soldiers’ reactions as psychosocial aspects
continued and was evident in some cases as late as the Second World War
(WW II). In the First World War (WW I), many soldiers with mental
disorders were regarded as malingerers and deserters and dealt with
accordingly. About three hundred British soldiers were executed for failure
to function in battle. An important factor in these kinds of large-scale wars is
the need to keep manpower levels up, and many of the measures taken were
designed to meet this concern. Later in WW I the term “shell shock“ was
introduced (Myers, 1915). This state was thought to be the consequence of
organic changes in the brain after shelling. Later on it helped many soldiers
to obtain a diagnosis without being stigmatised as cowards. The works of
Oppenheim (1889) introducing the term “traumatic neurosis”, Charcot
(1887), Janet (1889, 1894) and later on Freud (1920/1964a) describe the
psychological effects of trauma. However, introducing psychological aspects
at that time represented another pitfall for military psychiatrists. The
vulnerability within the individual soldier became the primary source of the
problem, rather than his traumatic experiences (Bonhoeffer 1926). This view
was exploited at an early stage in Germany (“Rentenneurose”) to avoid
having to compensate soldiers financially after a war. It was not until the
second half of WW II that views on soldiers’ reactions changed, although
some still had problems with this. The term “psychoneurosis” was soon
turned into “psycho” by soldiers and persisted for a long time. General
Patton slapping a soldier in the face and calling him a coward because of an
earlier breakdown was perhaps a key moment for the American forces. Later
on, labelling reactions as “combat exhaustion” or “combat fatigue” probably
helped to reduce the risk of further stigmatisation and instead to encourage
acceptance.
Primary interventions – “Forward Psychiatry”
Thomas W. Salmon, an American psychiatrist, was sent to Europe before the
United States joined WW I, to study the Allies’ practices in taking care of
their soldiers. He found that there were problems in returning soldiers who
had been evacuated after breaking down at the front. The primary gain for
these soldiers was that their lives were saved. They also achieved a
secondary gain in that they could declare themselves, and be described by at
least some others, as suffering from an illness. Such soldiers often defined
themselves as failures and in addition were separated from their groups and
the support that came with them. Salmon consequently introduced ways of
dealing with this problem. His principles (Salmon, 1917), which became the
foundation of “Forward Psychiatry”, comprise “Proximity”, “Immediacy”
and “Expectancy” (PIE). Soldiers should be taken care of close to the front
immediately after their breakdown and with a high expectancy that they
10
would regain their strength and return to the front. Later on another acronym
was also used to describe forward psychiatry in the allied nations: BICEPS
(brevity, immediacy, centrality, expectancy, proximity and simplicity). This
“cure of functioning” was elaborated by Brock (1918) and later resulted in
principles of treatment, the “7 Rs”: Recognition, Respite, Rest, Recall,
Reassurance, Rehabilitation and Return to duty. At the beginning of WW II,
evacuating soldiers from the front line again was the primary form of
treatment. This obviously resulted in a major loss of manpower (Stouffer et
al. 1949). As mentioned earlier, the lessons learned in WW I that early
interventions have positive effects was forgotten. Later in WW II, however,
when shortage of manpower became critical it was necessary to re-evaluate
the situation. The use of early intervention again achieved higher return rates
and became the leading strategy (Shepard, 2000). In the Korean War the
history of WW II repeated itself. Salmon’s principles were once again
introduced to help soldiers and to maintain the level of manpower (Glass,
1954).
The Vietnam War and post-traumatic
stress disorder (PTSD)
Like the previous wars, the Vietnam War took place overseas from an
American point of view, but in many aspects the similarity ends there. The
war was fought during an era influenced by the Cold War and great political
and social changes in the Western Hemisphere. The idea behind and
meaningfulness of the war was questioned by many people, including in the
soldiers’ home countries. Due to a poorly organised American selection
system, vulnerable individuals were over-represented among soldiers. Drug
abuse and disciplinary problems consequently were common. Overrepresentation of individuals from lower social groups and minority groups
were frequent negative manifestations. Since opposition to the war was also
growing in the United States, soldiers returning home were mostly not
treated as heroes. This was in sharp contrast to the warm welcome soldiers
from previous wars had experienced. In the aftermath, many soldiers had to
deal with their war trauma in a society in which their endeavours were
questioned. The United States sent about 3 million soldiers to Vietnam over
the years. Around 50 000 of them died in the Vietnamese jungle and it is
estimated that approximately 100 000 committed suicide back home, after
the war. It is said that in all 800 000 to 900 000 former American soldiers
have suffered from what has since been diagnosed as post-traumatic stress
disorder (PTSD). The introduction of the diagnosis of PTSD in DSM-III,
DSM-III R and later in DSM-IV (American Psychiatric Association, 1980,
1987, 1994) became an instrument to meet the need to label the suffering
that Vietnam veterans were facing.
11
Those suffering from PTSD have experienced a significant event, which
is of a threatening or traumatic nature. In addition they repeatedly relive the
experience through nightmares or sudden mental images of the event, known
as flashbacks. The individual tries to overcome the reliving of the event by
avoiding any circumstances that remind him of the experience. Finally,
lapses in memory of important aspects of the trauma, problems with
sleeping, aggressive outbursts, difficulty in concentrating or jumpiness in
response to noise can occur.
The establishment of PTSD also led to increased research, initially on
Vietnam veterans and subsequently on natural or non-natural disasters as
well as in criminal violence (McFarlane, 1988b; Shore et al. 1986, 1989;
Steinglass & Gerrity, 1990; Kilpatrick and Resnick 1993; Smith & North,
1993). The study of Vietnam veterans in particular contributed greatly to the
understanding of PTSD (Kulka et al. 1990, 1991; Davidson & Fairbank,
1992). The incidence of PTSD in combat veterans varies in different studies.
Helzer et al. (1987) found it to be 6.3% in general among American combat
veterans. Solomon and Benbenishty (1986) found chronic PTSD rates as
high as 56% in Israeli soldiers 2 years after combat exposure in the war in
Lebanon in 1982. The most extensive epidemiological study with the aim of
discovering the long-term psychiatric effects of combat was the National
Vietnam Veterans Readjustment Study (NVVRS). The prevalence of PTSD
in Vietnam veterans up to 19 years after the war was 15% (Kulka et al.
1990). Later, in the first Gulf War, approximately 9% exhibited PTSD
(Rosenheck et al. 1992).
Psychiatric disorders after combat are positively associated with the
degree of trauma the soldiers have experienced, for instance witnessing or
participating in atrocities or being wounded (Ursano, 1981; Kulka et al.
1990, 1991; Sutker et al. 1991). Greater exposure to combat is also related to
higher rates of PTSD, depression and alcohol abuse (Kulka et al. 1990).
Stress and mental health in personnel serving in
peacekeeping operations
During the past decade, following the end of the Cold War, there has been an
increasing need for peacekeeping and peace-enforcement operations
throughout the world. It also appears that there has been a shift from pure
peacekeeping operations led by the United Nations (UN) towards more
peace-enforcement operations, sometimes commanded by NATO. This
change could be explained by an increasing number of conflicts within
countries, for instance in the Balkans. Weisaeth (2003) describes this as a
change from “blue” helmets to “green” helmets and also discusses the
differences in the psychological challenges in these different missions. He
argues that personnel in UN peacekeeping missions are sometimes subject to
12
higher levels of stress than in peace-enforcement operations and that the
former are predominantly a mix of political, humanitarian and military
objects. Hence, soldiers in peacekeeping missions tend to report a need for
more training than those in peace-enforcement missions (Johansson, 2001).
Research on stress and mental health among peacekeepers was not very
common in the past. However, since the increased need for peaceenforcement operations and the increased involvement of more influential
nations, such as the US, the UK and others, and also organisations such as
NATO, there have been an increasing number of studies on personnel
serving in peacekeeping operations. The mental health status of personnel
following these missions often seems to vary as a function of the intensity of
the conflict in which they serve. There is a risk of peacekeepers being
involved in combat situations. In a Swedish peacekeeping mission in the
Congo in the early sixties, 0.5% of the soldiers were killed and 2% were
wounded (Kettner, 1972). After Swedish soldiers had been exposed to
combat during the peacekeeping mission in Congo in the early sixties, 3.5%
of them had become unfit for further duty (Kettner, 1972). Kettner also
found that in the group exposed to combat those below the age of 21 fared
worst. In high-intensity conflicts with many war-like situations, the main
focus was on PTSD. In a study on US soldiers serving in Somalia, eight per
cent met the diagnostic criteria for PTSD (Litz et al. 1997). In another study,
one third of 3,461 active-duty US military personnel who also served in
Somalia met criteria for psychiatric caseness (Orsillo et al. 1998). A high
level of PTSD (16%) was found in a study of British soldiers serving in the
early stages of the conflict in the former Yugoslavia (Baggaley et al. 1999).
A confounder in this study was that most of the soldiers had previously
served in Northern Ireland. In a follow-up study of former Norwegian
peacekeepers, 5 per cent were suffering from PTSD more than six years after
service (Mehlum & Weisaeth, 2002).
In low-intensity conflicts, PTSD may not be the main problem. Instead
peacekeepers find themselves exposed to cumulative stress, resulting from
boring missions and the ambiguity often present in peacekeeping operations
(Lundin & Otto, 1989; Carlström, Lundin & Otto, 1990; Lundin & Otto,
1992; Elklit, 1998; MacDonald et al.1998; Huffman et al. 1999). Among the
stressors that peacekeepers encounter, frustration is very common due to the
problems in reaching negotiated solutions in these types of conflicts
(Weisaeth, 2003). Peacekeeping personnel being harassed and humiliated by
the warring parties or the local population is not uncommon either. In
Somalia 76% of the US soldiers reported that they had had stones thrown at
them and 65% that they had been shot at (Litz et al. 1997). During the period
of the UN mission in the Balkans between 1993 and 1995, 17% of Swedish
personnel reported that they had been fired upon and 27% that they had been
exposed to artillery shelling by the warring parties (Johansson, 2001). Being
shot at evokes strong aggressive impulses, and peacekeepers, in contrast to
13
peace-enforcement personnel, do not always have the ability to express their
anger verbally or through activity. This need for “neutrality” might increase
the risk of developing psychiatric problems (Weisaeth, 2003). Troops trained
for active combat duty who are engaged in peacekeeping missions also tend
to suffer boredom and end up in role conflicts (Rikhye et al. 1974; Segal et
al. 1990). In a British study by Hotopf et al. (2003), the main risk factors for
stress symptoms in personnel serving in Bosnia are: lower rank, being
deployed early in the campaign, having more deployment-related exposure,
and serving in staff duties. There is no protective effect of previous
deployments to the Falklands or Northern Ireland, and time off following
deployment is not protective. On the other hand, in a study on peacekeeping
in Kosovo, Maguen et al. (2004) find a low degree of post-deployment
psychopathology. Self-engagement can be one protective measure. Britt &
Bliese (2003) show that soldiers who are committed to their job reported less
elevation in reports of psychological distress than soldiers who are
disengaged from their job.
Research on predictors for mental health is mostly done on individuals
with PTSD in whom personality traits such as neuroticism are correlated
with subsequent development of PTSD (Breslau et al. 1991; McFarlane
1992; Schnurr et al. 1993; Carlier et al. 1997; Fauerbach et al. 2000). This is
also shown in peacekeeping personnel (Bramsen et al. 2000). Mehlum and
Weisaeth (2002) found that exposure to stressful events, lack of
meaningfulness during a mission and stressful life events after service
contributed to the development of post-traumatic stress reactions in former
peacekeepers. Although some research has been done, we need to know
more about what specific factors before, during and after deployment
influence general mental health in Swedish peacekeepers. It is hoped that
increased knowledge in this field could thus contribute to improvements in
selection processes.
Early group interventions and psychological debriefing
The view that early interventions could have positive effects rests partly on
descriptions by Caplan (1964). He considers that it might be possible to
avoid a negative outcome after traumatic events by using preventive
techniques. Others also show such positive effects in dealing with sorrow
and the death of relatives (Raphael, 1977; Parkes, 1980) and after accidents
(Bordow & Porrit, 1975). Others instead find that such preventive measures
do not show any positive effects (Polak et al. 1975). After the introduction of
the diagnosis PTSD in 1980, a great deal of attention was drawn to attempts
to prevent the development of mental disorders following traumatic events
(Raphael et al. 1983). A model known as “Critical Incident Stress
Debriefing” (CISD) was introduced by Mitchell (1983), implying a
14
structured way of taking care of emergency personnel after they have been
exposed to traumatic events such as disasters and accidents. In the last
decade this type of intervention has become quite common in civilian
contexts, such as after major disasters and following minor traumatic events
faced for instance by firemen, hospital staff, police officers and social
workers (Mitchell & Bray, 1990; Mitchell & Everly, 1995). This stressmanagement technique is recommended as suitable for groups exposed to
traumatic events and capable of preventing subsequent stress disorders
(Raphael, 1986; Armstrong et al. 1995). Most participants also tend to take a
positive view of these kinds of sessions, which is clearly confirmed in a
follow up of approximately 1 000 psychological debriefings held in various
professional groups in Sweden (Larsson & Österdahl, 1996).
Psychological debriefing as a tool following traumatic events also has a
long history in military settings. The chief historian of the U.S. Army during
WW II, S.L.A. Marshall, developed a method to obtain comprehensive
descriptions of combat events (Marshall, 1944, 1947; Spiller, 1980, 1988).
Marshall’s technique, known as “historical group debriefing” (HGD) was
primarily conceived as a method of gathering historical data and did not
include deliberate psychological interventions to reduce distress in
individuals or groups. However, HGD does include elements of cognitive
reconstruction, validation of individual experiences and supportive
acceptance and is found to have positive effects. Due to these experiences
and the fact that it is in line with Salmon’s principles, the CISD concept is
easily adopted in the military. This has been subsequently enhanced since
positive effects of psychological debriefing have been shown in various
military settings (Ford et al. 1993; Shalev, 1994; Lundin, 1995; Shalev et al.
1998). Psychological debriefing is thus becoming widely used in both
civilian and military settings.
It is natural and legitimate to scientifically investigate phenomena that are
in such wide use. Some critics argue that there is a lack of systematic
knowledge of how psychological debriefing functions and whether it makes
an impact on individuals who have experienced stressful or traumatic events.
Eight studies are included in a critical Cochrane review by Wessely et al.
(2000). In these studies individual participants were mainly recruited from
emergency rooms and had experienced different traumatic events, such as
burns or dog bites. These individuals were then randomised to either one
session of psychological debriefing, on average lasting about 44 minutes, or
alternatively a brief counselling session. Follow up was conducted 13
months later. The dropout rates in some of the studies included were high.
Some of the studies showed that there was no difference in the rate of PTSD
at follow up and in some studies an even higher rate of PTSD was found
among those who had been given a psychological debriefing session. In a
later review (Rose et al. 2002), the authors argue that psychological
debriefing can harm people and should be abandoned. These reviews have
15
been criticised for various reasons. The main criticism levelled at the
Cochrane reviews concerns the definition of psychological debriefing: when
psychological debriefing was introduced by Mitchell & Everly (1995) it was
outlined as a group intervention and a part of an intervention system
primarily designated for emergency personnel. Giving traumatised
individuals one session of psychological debriefing for about 44 minutes and
nothing more is regarded by some critics as unethical. Other problems with
the Cochrane review are the dropout rate and the fact that the individual’s
prior trauma history is not accounted for. On the other hand, other studies
also cast doubt on the efficacy of psychological debriefing (Bisson et al.
1997; Hobbs et al. 1996; Lee et al. 1996).
There are some problems in conducting research on psychological
debriefing. At a general level, concepts need to be clarified, potentially
favourable mechanisms need to be analysed and short- and long-term effects
need to be studied more closely (Dyregrov, 1989; Mitchell & Bray, 1990).
More specifically, common research problems in this area include lack of
prospective studies, small sample sizes, lack of control groups, varying types
of debriefing, varying times of assessment and varying methods of
assessment (Deahl et al. 1994; Mitchell & Bray, 1990). The lack of
prospective studies is particularly problematic because this prevents
evaluations of the importance of pre-trauma personality and existential
conditions for post-trauma reactions. Existing research in the general field of
stress and trauma suggests that personality characteristics affect how a
person copes with and reacts to traumatic events (Gal, 1995; Hewitt & Flett,
1996; Rosenbaum, 1988). Antonovsky’s (1987) concept, Sense of
Coherence, with its three components (comprehensibility, manageability and
meaningfulness), is another stable person-related characteristic. It includes
existential beliefs and is similarly seen to co-vary with health and well being
in stressful conditions (Antonovsky 1993; Larsson et al. 1994).
A notable exception to much of the aforementioned criticism is a study by
Deahl et al. (1994) of 62 British soldiers after the first Gulf War. About two
thirds of these soldiers received psychological debriefing; one third did not.
No difference regarding mental health was noted between these two groups 9
months after service. However, as pointed out by the authors themselves,
this is a small study that is also affected by other methodological problems
such as lack of detailed records of the debriefing sessions. Another problem
in this study appears to be the timing and context of the debriefing sessions.
A psychological debriefing was organised “as soon as possible, either in the
Gulf or on return to the UK” (p. 61). In another study by Deahl et al. (2000)
in which 106 British soldiers returning from UN peacekeeping duties in the
former Republic of Yugoslavia were studied, it is concluded that
psychological debriefing might well have been a benefit.
In spite of this ongoing discussion about psychological debriefing, it is
still found useful as one tool in the management of stress reactions in
16
military settings (National Institute of Mental Health, 2002; Veterans Health
Administration, Department of Veterans Affairs and Health Affairs,
Department of Defence, 2003). Since this practice is widely used, the
following is therefore recommended: psychological debriefing should be
used only in organised groups that have been briefed beforehand, for
instance the military; it should be seen as only part of a range of measures
following traumatic incidents; screening for high-risk individuals and those
with high symptom levels should precede the session; these identified
individuals should be excluded, and finally any participation in
psychological debriefing should be voluntary. In a later study on
psychological debriefing in military settings by Litz et al. (2004), again no
protective features against the risk of developing PTSD are found. However,
it is also confirmed that military personnel are in favour of psychological
debriefing and that it does no harm.
Leading researchers on psychological debriefing (e.g., Mitchell & Bray,
1990; and Mitchell & Everly) emphasise the timing and context of
psychological debriefing sessions. Soldiers or other professional groups who
have shared highly stressful experiences tend to talk to each other
spontaneously about it. More or less formally structured ventilation or
defusing sessions led by the ordinary group leader also occur frequently.
According to these researchers, psychological debriefings should be seen in
the context of such preceding forms of support. However, no data on peer
support or defusing sessions are provided in most of the studies mentioned
above. Furthermore, not many studies have been carried out on the effect of
the support at the lowest level of command, although positive findings in this
field have been discussed as experiences in Israel (e.g. see Gal, 1986;
Milgram, 1986; Solomon, 1993).
Personality disorders among peacekeepers
Personality disorders are not uncommon in civilian communities. Studies
have shown that the prevalence of personality disorders in the general
population varies in different studies between 9 and 13% (Zimmerman &
Coryell 1990; Maier et al. 1992; Ekselius et al. 1994; Torgersen et al. 2001;
Samuels et al. 2002). Different methods of assessment; interviews or
questionnaires are used in these studies. Paranoid personality disorders are
seen in less than 2 to 2.4% of a population (Torgersen et al. 2001; Samuels et
al. 2002; Ekselius 1994). Schizotypal personality disorders seem to vary in
populations between less than 1% and 5.6% (Zimmerman & Coryell 1990;
Torgersen et al. 2001; Samuels et al. 2002; Reich et al. 1989). Borderline
personality disorders vary in populations between 0.5% and 4.6% (Torgersen
et al. 2001; Samuels et al. 2002; Ekselius 1994). Reich et al. (1989) and
Zimmerman and Coryell (1990) found a prevalence of obsessive-compulsive
17
personality disorders of between 4% and 6.4%. A study that uses the same
questionnaire as the present study shows a prevalence of 11.1% personality
disorders in a Swedish community sample (Ekselius et al. 2001). In that
study paranoid personality disorders are found in 5.6%, schizotypal in 5.2%,
borderline in 5.4% and obsessive-compulsive in 7.7% of the population.
Few studies have been made of personality disorders in military settings.
However, one study by Gundersen and Hourani (2003) evaluated the
incidence of first hospitalisations for personality disorders in the U.S. Navy.
Rates per 100 000 person-years were registered in this study. Gender
differences were pronounced. The most common personality disorders
among men are (rates per 100 000 person-years): dependent 47.9, schizoid
32.4 and antisocial 30.7. Among women, the most common personality
disorders were histrionic 97.2, dependent 68.2 and schizoid 23.3 (rates per
100 000 person-years).
Individuals with personality disorders are at risk of ending up with other
problems. A high degree of personality disorders according to DSM-IV
(56%) was found in a population of 130 Swedish male prisoners (LongatoStadler et al. 2002). Other studies show a high frequency of borderline
personality disorders (68%) among patients with PTSD (Shea et al. 1999).
Despite this, outpatients with a diagnosis of co-morbid borderline
personality disorders and PTSD do not differ from outpatients with
borderline personality disorder without PTSD (Zlotnik et al. 2002). Often the
life situation is impaired in both groups. A co-morbid diagnosis of axis I
disorders and axis II personality disorders is shown to yield a poorer
outcome (Bank & Silk 2001), and the levels of quality of life also seem to be
reduced in individuals with personality disorders (Narud & Dahl 2002). We
also know that those with personality disorders have a greater lifetime
history of previous suicide attempts (Suominen et al. 2000; Pirkis et al.
1999) and that patients with borderline personality disorders do not
markedly differ from those suffering from a mood disorder in that sense
(Soloff et al. 2000).
Personality traits and personality disorders among peacekeepers and their
consequences are less well known. In view of what has been said above,
there is a need to increase our knowledge in this field.
Suicide in military settings
The suicide rate in Sweden has shown a steady and stable decline since the
1960s and 1970s. In 1999 the suicide rate was 20.9 cases (29.5 men and 12.7
women) among 100 000 inhabitants over the age of 15 (Wasserman, 2003).
This suicide rate puts Sweden in a middle-ranking position among the
European countries. Nevertheless, suicide is the most common cause of
death in the 15-44 age group in Sweden. During wartime, it is hypothesised
18
that the suicide rate is lower in the general population (Lester, 1991; Lester
& Yang, 1994)). A higher suicide rate is found among individuals in a
community that suffers from higher levels of stress or post-traumatic stress
disorder (PTSD) (Paykel et al. 1975; Davidson et al. 1991).
Suicide rates in military settings overall are lower than in the general
population (Kang & Bullman, 1996; Hourani et al. 1999) but depend on
what population and what situation is being studied. Kramer et al. (1994)
show a lower suicide rate among Vietnam veterans during war but an
increased risk after the war. Other studies of war veterans find an increased
incidence of accidents, suicide attempts and completed suicides often related
to PTSD (Hendin & Polliger Haas, 1991; Campion, 1996; Fontana &
Rosenheck, 1995; Bullman & Kang, 1994; Goodale, 1999). In studies of
conscripts in the Scandinavian countries, Finland (Schroderus et al. 1992),
Norway (Engelstad, 1968; Hytten & Weisaeth, 1989) and Sweden
(unpublished data), a lower suicide rate is found than in the general
population, in some studies half the rate. Such results are also found in
studies in Germany (Brickenstein, 1967), the United States (Rothberg et al.
1988) and Singapore (Lim & Ang 1992).
A few studies have looked at the suicide rates in peacekeepers. A Danish
study (Jessen et al. 2002) found 4 suicide cases in a contingent of 3 859
soldiers where the expected rate would have been 3 cases. However, two of
the suicides actually happened one month before deployment and the
number is also too small for any conclusions to be drawn. An increased risk
of suicide was found in Finnish peacekeeping personnel, in a group that for
various reasons was sent home before the end of the mission (Ponteva et al.
2000). In a case-control study (Wong et al. 2001) comparing 66 suicides
among Canadian peacekeepers between 1990 and 1995, with two control
groups, no increased risk of suicide in peacekeepers was found except
among a subgroup of air-force personnel. An interesting result was shown in
a Norwegian study of 22 275 peacekeepers (Thoresen et al. 2003). A
moderately increased risk of suicide was found among the former
peacekeepers compared to the general population. This risk was increased
among peacekeepers that were not married. The results of the latter study are
hard to explain since it is hypothesised that the suicide rate would be lower
among Norwegian peacekeepers, primarily due to selection processes.
Much has been said about mental health of former peacekeeping
personnel in Swedish media. Until now no studies have been conducted to
answer questions regarding suicide rates among former Swedish
peacekeepers. The aim of this work was to plug that gap.
19
Ethics
Ethical approval was granted both for the study on the battalion in Bosnia
and for the register study.
Aim of this thesis
Most of the peacekeeping operations Swedish personnel have taken part in
are low-intensity conflicts, and there is a lack of prospective studies focusing
more on general mental health than on PTSD in such settings. Factors
before, during and after deployment that can influence poor mental health in
Swedish peacekeepers need to be explored. Are peer support and defusing
sessions led by platoon leaders of any value when dealing with traumatic
experiences during service? We do not know much about the rate of
personality disorders and their impact on general mental health in
peacekeepers. Furthermore, the ultimate result of poor general mental health,
suicide among former peacekeepers, needs to be addressed. The general aim
of this thesis was therefore to shed some light on these aspects.
The specific aims were to:
Study the general mental health situation and the effect of traumatic
experiences among peacekeeping personnel by using a longitudinal
approach.
Identify factors that could be viewed as predictors for reported poor general
mental health.
Evaluate the influence of different forms of support on post-deployment
general mental health.
Investigate the rate of personality disorders in peacekeeping personnel and
their impact on general mental health.
Investigate the suicide rate among former Swedish peacekeeping personnel.
20
Material and methods
Participants
Paper I-III
Between 1993 and 1999, Sweden deployed 13 mechanised battalions in
Bosnia. Part of this work (papers I-III) was based on data collected from one
of them (the Sixth Battalion), which was deployed from March to October
1996 within the framework of a NATO-PfP-IFOR mission. Serving in the
battalion were 724 individuals. Except for the commander, all the personnel
were volunteers. Eleven per cent were regular army officers, and the
majority of the remainder were civilian former conscripts. Most of the
women in the battalion were non-conscript civilians who had attended a
three-week military combatant course. Of those taking part in this study, 180
(35%) had served in prior peacekeeping missions, some in more than one.
Unfortunately we did not have access to records of prior traumatic
experiences in that group.
All personnel were asked to complete a questionnaire before deployment,
immediately after returning, six months after returning, and after one year.
Of the 724 individuals in the unit, 510 (70.4%) took part at the beginning of
the study (paper I). Some cases were missing initially, for administrative
reasons, but were included later so that 514 (71%) (paper II) and 516
(71.3%) (paper III) took part in the later part of the study. At the 6-month
follow up 395 (77%) of the participants (55% of the entire unit) responded,
and at the 1-year follow up 365 (71%) of the participants (50% of entire
unit) responded. Complete responses on the GHQ-28 on all four assessment
occasions were obtained from 316 individuals (61%).
The mean age of the respondents at the pre-deployment assessment
was 27.6 years (SD = 6.4), with a range from 20 to 53 years. Most of
the participants (96%) were men.
Paper IV
The cohort studied in paper IV was made up of all Swedish individuals, both
men and women, who had served in international UN or NATO
peacekeeping missions from 1 May 1960 to 31 December 1999 and
consisted of 39 825 individuals. The vast majority were men. The
21
distribution of gender and age in the cohort at the end of the study was
normal over the five-year age groups except for a dip in the group of 50-54year-old men. The total numbers of deaths in different age groups were also
normally distributed when the study ended.
Since the purpose was to study suicides that specifically occurred after
serving in peacekeeping missions, we excluded 57 individuals who had died
mainly due to accidents during service abroad, so that the final number of
participants was 39 768. In Sweden every person has a unique personal
identity number. The personal identity numbers of the investigated cohort
were supplied by the Swedish international command, which was the unit
responsible for Swedish peacekeeping missions.
Swedish personnel signing up for participation in peacekeeping missions
are all volunteers and are often 20 – 25 years old. Most of the personnel are
former male conscripts selected in three steps before it is possible for them
to participate. The first step in the selection process takes place before
conscription at the age of 18. These young men undergo a thorough medical
evaluation in which their physical fitness is tested and their intellectual
capacity is screened. A psychological evaluation by a psychologist is also
performed in order to 1) disclose any personality traits that make the
individual unfit for military duty and 2) screen for leadership potential. The
second selection step is a result of the conscription service. Conscripts
interested in applying for participation in a peacekeeping mission have to be
good soldiers and be in the upper third of the marking scale when they end
their conscription service. A third and final selection takes place after
application and acceptance for peacekeeping service. During this period,
when the unit is formed and trained for the actual mission, the commanders
evaluate the individuals, with those who are found not to be fit enough, in a
broader sense, not being able to join.
Sweden contributed personnel to international peacekeeping missions
during the studied period in the following countries/regions: Gaza 19561966; Congo 1960-1964; Cyprus 1964-1993; Sinai 1973-1979; Lebanon
1980-1993; Saudi Arabia 1991; Kuwait 1991; Somalia 1993; the former
Yugoslavia (Bosnia, FYR Macedonia and Kosovo) 1993-2000. Of these
missions, those that could be characterised as being high-intensity conflicts,
or as including such elements, would be periods in Gaza, Congo, Lebanon,
Saudi Arabia, Kuwait and the early periods of the missions in the former
Yugoslavia. The rest could be regarded as low-intensity conflicts, although
still entailing a risk of service personnel being exposed to potentially
traumatic events.
After serving one or more 6-months periods in peacekeeping missions,
most individuals returned to their civilian life and employment. Military
personnel, comprising about 5 - 10% in a unit, went back to their regular
work in the Armed Forces back home.
22
Methods
Pre-deployment assessment (paper I)
The pre-deployment questionnaire included socio-demographic data and the
following three questions about pre-deployment stressors (Yes/No response
format): “Have you had long-term problems in your marriage/cohabitating
relationship or family?” (henceforth labelled “prior family problems”),
“Have you ever thought of taking your own life?” (henceforth labelled
“suicide thoughts”), and “Have you ever tried to kill yourself?” (henceforth
labelled “suicide attempt”). These were the same variables that were used in
the UNIFIL study by Weisaeth et al. (1993).
The following instruments were also used on the pre-deployment
assessment: Personality was mapped using the Five-Factor Personality
Inventory (FFPI; Hendriks 1997), which is based on what is known as the
Big-Five model of personality (see e.g., Costa & McCrae, 1985). The FFPI
consists of 100 items: 20 are designed to measure each of the following five
factors: Extraversion (e.g., “Takes time out to chat”), Agreeableness (e.g.,
“Goes out of his/her way for others”), Conscientiousness (e.g., “Loves order
and regularity”), Emotional Stability (e.g., “Thinks that all will be well”),
and Intellect/Autonomy (e.g., “Decides things on his/her own”). Responses
on all items could range from 1 (not at all applicable) to 5 (entirely
applicable). Scores could range from –40 (lowest degree) to 40 (highest
degree) on each of these scales. In this sample, Cronbach´s alpha ranged
between .72 and .82 on the 10 subscales constituting these scales.
Sense of coherence (SOC) was measured using Antonovsky´s (1987) 13item short version of the SOC questionnaire. Examples of items are “Do you
have the feeling that you don’t really care about what goes around you?”
(designed to measure meaningfulness) and “How often do you have feelings
that you are not sure that you can keep under control?” (designed to measure
manageability). All items had 7-point response scales with the anchors
defined. A scale score was computed by adding the scores of all individual
items. Scores on the SOC scale could range from 13 (lowest sense of
coherence) to 91 (highest sense of coherence), and a Cronbach´s alpha of .80
was obtained.
The General Health Questionnaire (GHQ-28; Goldberg & Hillier, 1979)
was used to assess general mental health on each assessment occasion
(papers I-III). The GHQ-28 consists of 28 items. Each item has 4 response
choices ranging from 0 (most favourable) to 3 points (least favourable).
Cronbach’s alpha ranged from .81 to .92 on the four assessments.
Following the suggestions of the test constructors (Goldberg & Hillier,
1979), an alternative scoring method was also used, where replies were
coded 0-0-1-1. Scores could range from 0 to 28. When this method is used, a
total score of 5 across all 28 items indicates that there is a 50% probability
that the respondent is a “psychiatric case” (McDowell & Newell, 1987, p.
23
40) or has a score “equivalent to the average patient referred to a
psychiatrist” (Bowling, 1995, p. 77).
Post-deployment assessment (paper I)
Directly upon returning home, all personnel took part in a three-day
homecoming programme, the aim of which was to sum up the mission,
screen for those needing individual support and prepare for return to
families. During this period we conducted a post-deployment assessment in
which the participants were asked the following question, eliciting six
behavioural responses regarding the occurrence of traumatic events during
their service in Bosnia: “Have you during your service been involved in a
particular event according to the list below which you experienced as very
stressful (circle Yes or No)?” The list contained the following six types of
events: “Any kind of firing very close,” “Threats with weapons pointed at
you,” “Taken prisoner or hostage,” “Seen wounded, maimed, or dead
people,” “Been involved in a serious accident (What…),” and “Other
situation (What…).” A composite service trauma exposure score was
computed by counting all individuals who had scored “Yes,” on at least one
of the above six items.
Respondents indicating yes on any of the traumatic event alternatives
were asked to respond yes or no to three follow-up items related to received
support. The question read: Did you receive any support or help of the
following types: (1) Peer support = A friend helped you by sitting down and
talking to you in connection with the event; (2) A ventilation session = Your
platoon leader (or similar leader) gathered your group the same day that the
event occurred and you went through what had happened and were given
opportunities to express your feelings; and (3) A debriefing session = A
group session 1-3 days after the event which was led by an external
counsellor with the aim of providing an opportunity to work through the
event regarding facts, thoughts and emotions. For each of the three support
types, a follow-up question regarding perceived quality of support was asked
if a participant had responded “Yes”. This question read: “How would you
describe the peer support/ventilation/debriefing session?” A 5-point response
scale, ranging from 1 (very poor) to 5 (very good) was used for these
questions. Drawing on the conceptual framework developed by Mitchell and
Bray (1990), a ventilation session is henceforth labelled defusing.
The post-deployment assessment also included one additional instrument
designed to measure reactions to traumatic events. The Impact of Event
Scale (IES), developed by Horowitz, Wilner and Alvarez (1979), was used.
The IES consists of 15 statements related to reactions to a particular event.
Seven items were designed to measure Intrusiveness (e.g., “I have dreamt
about it”), and eight were designed to measure Avoidance (e.g., “ I have
tried to avoid talking about what happened“). The response format was a
24
four-choice scale with numerical values of 0 (not at all), 1 (seldom), 3
(sometimes), and 5 (often), A scale score was computed by adding all 15
responses (Cronbach´s alpha = .87). Thus the scale score could range from 0
to 75. In one of the analyses, scores of 31 or higher were used to define
critical cases. According to Lundin (1995), scores between 31 and 40
indicated “a traumatic stress reaction with a certain likelihood of posttraumatic stress disorder” (p. 27, our translation).
One-year follow up (paper II and III)
Follow-up assessments were conducted by sending participants
questionnaires by mail six months and one year after they had returned from
duty. No reminders were sent. At the one-year follow up, three different
questionnaires were used. The GHQ-28 described above was used again to
measure general mental health.
Potential post-deployment stressors were registered by using a
questionnaire (Yes/No response format), made up of the following nine
questions: “Since arriving home from the mission in Bosnia I have”: “Had
relationship problems,” “Had financial problems,” “Experienced a breaking
up with girlfriend/boyfriend,” “Experienced a divorce/broken cohabitating
relationship,” “Had serious illness of my own,” “Experienced illness of a
close relative,” ”Experienced the death of a close relative,” “Experienced a
serious traffic accident,” “Had other problems.” A composite postdeployment index, “Stressful life events post-deployment,” was computed
by counting all individuals who had scored “Yes” on at least one of the
above mentioned nine items.
Personality disorders were measured using the DSM-IV & ICD-10
Personality Questionnaire (DIP-Q) constructed by Ottoson et al. (1995;
1998). DIP-Q is a 140-item true/false self-report questionnaire designed to
measure all ten DSM-IV and all ICD-10 personality disorders. The ICD-10
schizotypal disorder is also included. The diagnostic criteria for each
personality disorder are covered by 135 items. Twenty-one items are
reversed so that “false” indicates that the corresponding criterion is fulfilled.
In DIP-Q an impairment/distress scale (ID scale) that consists of 5 items and
a self-report version of the Global Assessment of Functioning Scale (GAF)
are used to define the general criteria for personality disorders. The cut-off
score for fulfilling the general criteria on the ID scale was set to two or
higher and a score of 70 or less on the GAF scale. To meet a categorical
diagnosis of personality disorders thus required that the number of criteria
for the specific personality disorder reached the level specified by DSM-IV
and ICD-10 and also that the general criteria were fulfilled. Although ICD10 personality disorders were also registered, the main focus in this study is
on DSM-IV personality disorders.
25
Register study (paper IV)
During the studied period Sweden has not had any professional army to
compare our cohort with, and we compared it instead with the total
population in the cause of death registry. One problem in our study was that
we did not have access to the real start date of the observation period for
each person. The start of the observation period was therefore set at the age
of 20 or the year of 1960. The observation period ended at the time of death,
at the age of 60 or at the end of 2001. We had no access to information on
the proportion of emigrants in our cohort, but it is believed not to have any
major influence on the results.
The classifications of causes of death are those of the WHO International
Classification of Diseases (revisions 7 to 10; World Health Organisation
1957, 1967; 1977; 1992). The causes of death were classified into five
categories: 1) Suicide (ICD-7 codes 970-979; ICD-8 codes E950-E959; ICD9 codes E950-E959; ICD-10 codes X60-X84); 2) Death by undetermined
intent (“unclear cases”) (ICD-7 codes none; ICD-8 codes E980-E989; ICD9 codes E980-E989; ICD-10 codes Y10-Y34); 3) Accidents (ICD-7 codes
800-936, 960-964; ICD-8 codes 800-946; ICD-9 codes 800-929; ICD-10
codes V01-X59); 4) Murder (ICD-7 codes 980-985; ICD-8 codes 960-969;
ICD-9 codes 960-969; ICD-10 codes X85-Y09); and 5) Other causes (ICD-7
codes none; ICD-8 codes 000-799; ICD-9 codes 001-799; ICD-10 codes AR).
Statistics
The studied population in papers I-III was made up of all personnel taking
part in Swedish peacekeeping units serving in low-intensity conflicts. The
target population should also include those who participate in peacekeeping
missions in the near future during which period the same selection
procedures would still be used. Although the studied battalion was randomly
chosen, there was no possibility of the individuals in the whole population
being selected, for practical reasons. Hence, the sample could not be seen as
a simple random sample but could instead be labelled as a convenience
sample. However, since the selection procedures have remained the same in
the Swedish Armed Forces for many years it would still seem reasonable to
carefully make predictions and estimations about the population.
T-tests were used to compare means and Chi square homogeneity tests
were used to compare groups for one variable. Comparisons of independent
groups were performed using Mann-Whitney tests and also one-way analysis
of variances with Scheffé tests as post hoc comparisons. Spearman
correlations were used to assess the bivariate relationships between the
number of traumatic events encountered during service and the pre- and
26
post-service measurements (papers I and II). Spearman correlations were
also computed between the quality ratings of received support following
traumatic events and the post-deployment measurements. A logistic
regression analysis was performed where the dichotomised overall GHQ-28
scores on the post-deployment assessment was the dependent variable (paper
I). The SOC, FFPI and overall pre-service GHQ-28 variables were entered
first. Numbers of traumatic events were entered in the second step. Type of
support received following a traumatic event, defined as categorical variable,
was entered in a third step. Logistic regression analyses were also performed
in order to find contributions to the odds of reporting impaired health at the
one-year follow up (papers II and III). Thus, dichotomised GHQ-28 scores
were again used as the dependent variable. In paper II the following
predictor variables were entered simultaneously: prior family problems,
suicide thoughts, suicide attempts, traumatic experiences during service and
problems after returning. Personality disorder (DSM-IV), prior family
problems, traumatic experiences and problems after returning were entered
in the logistic regression analysis in paper III. In the latter study the kappa
test was used to measure the concordance between DSM-IV and ICD-10
parts of the DIP-Q.
To compare our cohort in paper IV to the general population, we assumed
a Poisson distribution and calculated a standardised mortality rate (SMR),
which is the ratio between the observed and the expected number of cases.
The observed cases are the number of deaths that actually occurred in the
cohort during the studied period. For each calendar year, sex and age group,
the expected number of deaths was computed by multiplying the personyears at risk with the respective cause-specific rates of death. A 95%
confidence interval (CI) was then calculated. SMR was first calculated for
the different causes of death in the cohort over the studied period. Since the
studied period was very long (1960-1999), we divided this phase and
calculated SMR specifically for suicide in five-year groups.
Overall the significance level was set at p < .05.
27
Results
General Mental Health (papers I and II)
General mental health was measured before deployment, directly after
returning home, after 6 months and at the 1-year follow up. Before
deployment, the mean score, which can range from 0-28, was 1.28 (SD =
2.48) among participants responding on all measurement occasions. We
found no difference between the means on the GHQ-28 across the four
assessment occasions. Using a dichotomised scoring method, described in
the methods section, scores of 5 or higher were regarded as representing
poor mental health. Among this group the proportion of personnel meeting
these criteria before deployment was 7.6% (n = 39). These proportions did
not differ over time on the different measurement occasions.
Traumatic events (paper I)
181 soldiers (35% of the whole sample) reported occurrence of traumatic
events. The most frequently reported event was seeing wounded, dead or
maimed people (129). The most common kinds of events in the “other
situations” category (73), were mine accidents and witnessing fights between
local parties. The category of “serious accidents” (61) included episodes in
which fellow soldiers were killed or severely wounded when driving onto
mines or being involved in traffic accidents. 23 soldiers reported “any kind
of firing very close”, 9 individuals reported “threats with weapons pointed at
them” and 3 reported that they had been “taken hostage”. Within this group
of 181 soldiers, 89 had experienced one event, 72 had experienced two, 16
had experienced three, 3 had experienced four and 1 individual had
experienced five events.
We compared those (n = 181) who had reported traumatic experiences
with those who had not (n = 329). We found that the soldiers without
traumatic experiences scored higher on one of the subscales of the FFPI –
Agreeableness (19.61 vs. 18.08) before service. There was also a negative
correlation between the number of reported traumatic events and
Agreeableness. Directly after returning home those without traumatic
experiences reported more favourable mean scores on the total GHQ-28
28
(12.54 vs. 13.76), on the GHQ-28 subscale “Somatic Symptoms” and on the
“Anxiety and Insomnia” subscale.
The mean score on the IES-15 at the post-deployment assessment was 6.6
and did not change at the 6-month or 1-year assessment. Only 7 individuals
(1.4%) scored 31 or more on the IES-15 (“a certain likelihood of posttraumatic stress disorder”). Unfortunately we did not have any predeployment record of PTSD or IES-15 scores. We cannot exclude the
possibility that some of these 7 individuals had previous traumatic
experiences. Higher IES-15 scores were correlated to a higher number of
traumatic events.
Potential predictors for poor mental health
(papers I and II)
We used different ways to try to find any potential predictors for poor mental
health in our cohort. One way in which we did this was to compare the group
that reported poor mental health directly after deployment (GHQ-28 > 5)
with those who had no indication of poor mental health directly after
deployment (GHQ-28 < 5) (paper I). We found that soldiers who reported
poor mental health directly after service in Bosnia differed from the other
soldiers on several pre-deployment measurements. They had lower mean
scores on the SOC questionnaire (65.13 vs. 70.04) and on the Emotional
Stability scale of the FFPI (19.69 vs. 22.23). In addition, they had less
favourable mean scores on the GHQ-28 (16.34 vs. 11.61).
Another way to look for potential predictors was to dichotomise GHQ-28
scores on each measurement occasion and cross-tabulate them with the three
pre-deployment stress indices, (“prior family problems” “suicide thoughts”
and “suicide attempt”), the deployment trauma index (traumatic events that
the individual could have been involved in), and the nine post-deployment
stress indices (see methods section for details) (paper II). We found that
“prior family problems” and “suicide thoughts” before deployment coexisted
with poorer mental health after service. Exposure to traumatic events during
deployment coexisted with poorer mental health only at the post-deployment
and at the 6-month follow up. Among the post-deployment stressors,
“relation problems”, “financial problems” and “severe illness of close
relative” coexisted with poorer mental health at the 1-year follow up.
For further investigations of potential predictors, logistic regression
analyses were used to find out what contributed to poor mental health. In one
of these analyses a selection of the pre-deployment variables was entered in
the first step, the number of traumatic events encountered during service was
entered in the second step and the type of support received was entered in
the third step (paper I). We found that the only variable that was associated
with the mental health directly upon returning from Bosnia (post-deployment
dichotomised GHQ-28 variable) was the pre-deployment GHQ-28 variable.
29
The R statistic indicates that SOC also made a small partial contribution to
the likelihood of obtaining a poor mental health score at the post-deployment
assessment.
In order to find out what combination of stressors was connected with
poor mental health we divided the participants into four groups according to
reported stressors: (1) no trauma in Bosnia or stressors post-deployment, (2)
stressors only post-deployment, (3) trauma only in Bosnia, and (4) trauma in
Bosnia and stressors post-deployment. An overall difference in general
mental health (GHQ-28) was found among these four groups on all
assessment occasions, except pre-deployment. The ranking order between
these groups in means on GHQ-28 was as follows: lowest was the group
without any stressor, followed by the group with only traumatic experience
in Bosnia, the third group was the one consisting of those who reported
stressors only after returning home, and finally the highest mean value on
GHQ-28 was found in the group that experienced traumatic events in Bosnia
as well as stressors back home (paper II).
At the 1-year follow up (paper II) another logistic regression analysis was
performed to discover which of the measured factors contributed to poor
mental health. The dependent variable was the GHQ-28 after 1 year
(dichotomous scoring, 0-4 vs. 5-28). The model chi-square with five degrees
of freedom was significant and the proportion of correctly predicted cases
was 90%. The result was that ”Stressful life events post-deployment” was
associated with the 1-year GHQ-28 variable. The odds ratios indicated that
“prior family problems” and “suicide thoughts” could also make a
contribution to the likelihood of registering a poor mental health score after
one year.
Effect of different kinds of support (paper I)
Among the 181 individuals who reported traumatic experiences four kinds of
support were noted after the event. One subgroup (n = 56) reported that they
had not received any support at all. A second subgroup (n = 29) received
peer support only. A third subgroup (n = 60) received peer support and in
addition a defusing session. A final subgroup (n = 36) received peer support,
a defusing session and also a debriefing session. These four groups were
compared. We found that the subgroup that reported that they had not
received any support also reported to a lesser extent that they had “Been
involved in a serious accident” and “Other situations”. They also reported a
lower number of events than those receiving any support.
There was no difference between the subgroups on any pre-deployment
assessments. However, on the post-deployment assessment a mean
difference was noted on the overall GHQ-28 scale as well as on two
subscales, Social Dysfunction and Severe Depression. The group receiving
30
peer support and in addition a defusing session had the most favourable
score and the group receiving no support had the least favourable score. The
addition of psychological debriefing sessions did not lead to a more
favourable result.
Personality disorders and their impact on general
mental health (paper III)
The rate of personality disorders in the studied cohort was 9.5% (n = 49).
The frequency of the most common DSM-IV personality disorders we found
in our study was as follows: obsessive 6.5% (n = 33); schizotypal 5.5% (n =
28); paranoid 3.1% (n = 16); avoidant 3.1% (n = 16); borderline 2.8% (n =
14) and antisocial 2.6% (n = 13). An almost identical order was found in
ICD-10 personality disorders. Of the 49 peacekeepers with DSM-IV
personality disorders, 3% (n = 15) had only one personality disorder, 2.2%
(n = 11) had two, and 1.2% (n = 6) had three. The remaining 3.4% (n = 17)
had up to 9 personality disorders. The distribution of ICD-10 personality
disorders followed the same pattern. There was a high degree of
concordance between DSM-IV and ICD-10 in this study. Of the 49 with
personality disorders, 46 overlapped the ICD-10 and DSM-IV (kappa .932)
There were more individuals with personality disorders in the group with
impaired general mental health than in the group without. This applies to
DSM-IV before deployment (14 out of 49 = 28.6% vs. 25 out of 458 = 5.5%
and at the one-year follow up (8 out of 33 = 24.2% vs. 27 out of 322 = 8.4%.
This is also the case regarding ICD-10 personality disorders. We found no
such difference after return and at the six-month follow-up.
There were also more individuals with personality disorders among those
who reported traumatic experiences during deployment than among those
who did not: DSM-IV: 24 out of 49 = 49.0% vs. 155 out of 458 = 33.8%.
The same results were obtained in the ICD-10 personality disorders.
However, those few individuals, numbering seven, who were at risk of
developing post-traumatic stress disorders, defined by IES-15 scores > 31,
were not more frequent in the personality disorder group than in the nonpersonality disorder group. This includes all measurement occasions.
We also studied what effect personality disorders would have on general
health after one year. We found that the means in general mental health, as
measured by the GHQ-28, differed at all measurement occasions between
those with and those without personality disorders. The same difference was
also obtained among those who answered at all four measurement occasions
(personality disorders, DSM-IV n = 30, no personality disorders n = 286). A
logistic regression analysis also suggests that DSM-IV personality disorders
contribute to the odds of reporting poor mental health at the one-year follow
up. This also applies to ICD-10.
31
There was no difference in personality disorders between different
occupational groups.
Furthermore, we found no difference in personality disorders between
professional military officers and other personnel.
Suicide among former peacekeepers (paper IV)
The total number of observed deaths in the cohort was almost half of the
expected deaths (1140 vs.1991.4). The total number of suicides was 182
whereas the expected number was 272.5 with an SMR value of .67 and a
95% CI of .58 - .77. The 182 suicides correspond to a suicide rate
(suicides/100 000 inhabitants/year) of 11.8. Unclear cases, accidents,
murders and all other causes of death were also lower than the expected
cases.
The number of suicides in males was 179 and the expected number was
266.7. The SMR was .7 with a 95% CI of .58 - .78. The same difference was
found in the other categories as well, with lower observed cases than
expected.
Among females there were only 3 cases of suicide where the expected
rate was 5.9. The SMR was .5 with a 95% CI of .11 – 1.5. The rate of
observed female accidents was also 3 with an expected rate of 3.9 (SMR =
.8, 95% CI = .16 –2.27). With regard to other causes of death in females the
observed rate was 22 and the expected rate 36.6 (SMR = .6, 95% CI = .38 .91). There were no unclear cases and no murder cases among females
during the studied period.
When the SMR was calculated for suicide in five-year groups we found
differences between observed and expected suicides in all these five-year
groups. Three suicides after the age of 60 were excluded. The closest gap
between observed and expected was in the years 1975-1979, with 27 and
32.2 suicides respectively and an SMR of .83.
In some of the five-year groups suicide was more prominent, although not
significantly, in certain age groups, for instance in 1975-1979 in the 30-34
age group; in 1980-1984 and 1985 –1989 in the 40-44 age group and in
1990-1994 in the 35-39 age group.
Dropouts (papers I-III)
The dropouts (n = 198), on one or more assessment occasions, were
compared with the participants with complete responses (n = 316) on sex,
age, and each of the four GHQ-28 measurements. One difference in means
emerged: the dropouts had a more favourable post-deployment mean score
on the GHQ-28 (0.95 vs. 1.37), (papers I, II, III).
32
General discussion
Taking part in a peacekeeping operation in a low-intensity conflict led to a
low degree of general mental health problems and post-traumatic stress
reactions. Poor general mental health one year after returning from service
was correlated with factors that existed prior to service or problems
encountered after the mission. Peer support and defusing sessions led by
ordinary platoon leaders could be valuable instruments after traumatic
experiences. The levels of personality disorders were the same as or lower
than in the civilian population and were associated with poorer general
mental health after one year. During the studied period the observed suicides
and other violent deaths among former Swedish voluntary peacekeepers
were lower than in a comparable group of the general population. This was
also the case when the studied period was split into five-year groups and
suicides alone were studied.
The low levels of general mental health problems found in this work were
in agreement with the findings among American troops in Kosovo studied by
Maguen et al. (2004). These results from low-intensity conflicts could be
compared to the relatively higher incidence of post-traumatic stress reactions
in other studies from more intense conflicts (Deahl et al. 1994; Baggaley et
al. 1999; Litz et al. 1997; Orsillo et al. 1998). We can note that those
reporting poor mental health before deployment not surprisingly are to a
great extent the same ones who also report poor mental health after
deployment.
About one third of the participants had experienced traumatic situations
during their service. As hypothesised, traumatic experiences during service
in Bosnia only appeared to make a transient difference (the post-deployment
and the 6-month assessments). Traumatic experiences during deployment
only seemed to have an influence on poor mental health if combined with
post-deployment stressors. Thus, taking part in a peacekeeping mission in a
low-intensity conflict did not appear to result in lasting mental health
problems. However, a selection hypothesis cannot be ruled out. It should be
noted that most participants were young, voluntarily recruited men, who
reported favourable psychological status before departing for Bosnia.
Relevant reference data exist on the SOC scale, in which the same
instrument was used in a nation-wide sample selected to represent the
Swedish population (Larsson & Kallenberg 1996). The mean score obtained
on the SOC scale in our sample (69.63) approximately corresponds to the
33
65th percentile in the nation-wide sample as a whole as well as in a
subsample consisting of 21- to 35-year-old men.
One aim of this thesis was to try to identify some possible predictors for
poor mental health after serving in a peacekeeping mission. In order to
achieve this, different approaches were used, altogether indicating that some
factors before deployment, during deployment and after deployment could
contribute to poor mental health. Before deployment, those reporting lower
mean scores on the SOC questionnaire and on the Emotional Stability scale
of the FFPI reported poorer mental health. This is also the case for those
reporting “prior family problems” or “suicide thoughts” which is consistent
with other findings (Weisaeth et al. 1993). Regression analysis indicated that
the main contributors to poor mental health one year after returning from
duty abroad were problems experienced post-deployment, such as “relation
problems”, “financial problems” and “severe illness of a close relative”.
Personality disorders in general were related to reported impaired general
mental health. Personality disorders also seem to contribute to poor mental
health one year after returning home from a mission abroad in conjunction
with ”prior family problems” and ”problems after returning.” An explanation
might be that personality disorders and ”prior family problems” could be an
expression of vulnerability and, ultimately, might have a role in causing
”problems after returning”.
It was found that the combination of peer support and a defusing session
led by the platoon commander had a positive effect on post-deployment
mental health in this group. These findings are promising from a practical
point of view. All officers had received fairly structured training on how to
lead a defusing session before leaving for Bosnia. The results indicate that
this kind of training is valuable and that the lowest level of command plays
an important role in the mental health of troops in a stressful context. This
would be consistent with, for instance, the view of Gal (1986, p 219), “...the
recognition that the only available preventive measure against excessive
psychiatric casualties involve enhancing unit morale, interpersonal
relationships and effective leadership”. On the other hand we did not find
that the addition of psychological debriefing as a support tool resulted in a
more favourable outcome in traumatised individuals. One explanation could
be that it seems that those who also received psychological debriefing
belonged to the group that had been exposed to more severe traumatic events
than those who did not.
The rate of personality disorders in this study seems to be at the same
level as, or at a slightly lower level than, the general population. The lack of
studies on personality disorders in military settings prevents good
comparisons. The most common personality disorders were obsessivecompulsive, schizotypal and paranoid. This means that there is a striking
similarity in the prevalence of the different personality disorders among the
peacekeepers in our study and what is found in the general population
34
specifically compared to the study of Ekselius et al. (2001). However,
compared to most other studies there might be a slight over-representation of
paranoid and schizotypal personality disorders in our study. The presence of
obsessive-compulsive and paranoid personality disorders might possibly be
explained by both what is expected of and what could attract a presumptive
peacekeeper, but one wonders what makes a schizotypal individual prone to
seek such an assignment. In the absence of additional data, we can only
speculate that it is the feeling of suspicion, the lack of friends back home,
and certain conceptual references that eventually could drive such an
individual away to an armed mission abroad. Although we find that
personality disorders in general may cause problems for peacekeepers,
further studies are needed to find out what kinds of personality disorders are
the most problematic in this sense. This might help us to make the selection
processes even better and we would perhaps in the future find even fewer
personality disorders among peacekeepers.
The studied cohort in paper IV is made up mostly of men and is fairly
normally distributed in age at the end of the study, except for a small dip in
the 50-54 age group. Furthermore, there is no accumulation of deaths in any
age group at the end of the study. An explanation for the dip referred to
could be that 30 years earlier (1967-1971), when these men would have
considered taking part in peacekeeping operations not many Swedish
peacekeepers were needed. This was due to the fact that the only mission in
which Sweden participated at this time, and had done so since 1964, was in
Cyprus. However, the demand for peacekeepers increased again after the
Yom Kippur War in 1973 and the Turkish invasion of Cyprus in 1974. It is
also interesting to note that around this time period criticism of the Vietnam
War was intense and an anti-authoritarian and anti-military movement (“the
1968 student movement”) was challenging societies in the Western
Hemisphere.
The results in our study are in line with the hypothesis that there would be
a lower suicide rate among former Swedish peacekeepers. Since earlier
studies show that Swedish peacekeepers tend to report more favourable
psychological status than a comparable sample from the general population,
the selection process could contribute to the differences found in this study.
We cannot find any simple explanation as to why the closest gap between
observed and expected suicides is found in the five-year group 1975-1979.
The age groups in which suicides were most common were 30-34 and 40-44
between 1975 and 1989, which would also be expected. There is no
straightforward explanation for suicides being most common in the 35-39
age group between 1990 and 1994. Since the number of suicides was only 10
in this group during those five years the increase might just be accidental.
Care should be taken in comparing these results with other studies on
suicide. Different outcome could probably be explained partly by differences
in methodology, for instance the lack of information about the start time in
35
our study, which is discussed below. Having considered that, differences
might also relate to likely dissimilarities in the selection processes in
different countries that will not be elaborated here. Another theoretical, but
perhaps more likely explanation could be that countries have sent
peacekeepers to different missions, with perhaps different levels of conflict.
There are also other differences: the Swedish sample is larger (39 768) than
the Norwegian one (22 275), for instance, and the studied period is also more
prolonged in this study (1960-1999) than in the Norwegian one (1978-1995).
Compared to previous investigations in this field, this work has some
strengths: it uses a longitudinal design (papers I-III) that tells us about the
situation over time; control for a high number of potential confounders (e.g.
sex, age, education, pre-trauma psychological status, etc); sample size: there
was a high response rate from the start (more than 70%), and after one year
still more than 50% returned their questionnaires. This drop could, of course,
cause a sampling bias, but the dropouts did not differ from those completing
the study except that they reported somewhat more favourable postdeployment mental health. The strength in the register study (paper IV) was
that the studied cohort was so big, the studied time period was so long,
almost 40 years, and that it covered many different missions.
One main weakness (papers I-III) is related to the fact that we do not have
any records of previous lifetime traumatic experiences prior to deployment.
Such experiences could confound with effects of trauma experience during
military duty and might explain why experiencing trauma in Bosnia did not
affect the 1-year GHQ-28 outcome. On the other hand, the general low
degree of mental health problems found in this study could also indicate that
any confounder, be it any other lifetime or recent occurrence of interpersonal
violence or other stressors, has not had a great impact on the responders over
time. Other methodological weaknesses include the following: (a) since it
was hard to find suitable control groups, individuals were categorised into
comparison groups depending on traumatic experiences; (b) we could not
fully study the impact of traumatic events since this was a low-intensity
conflict; (c) it would have been a contribution to use an instrument that
would also measure hyperarousal; (d) this study used self-report
questionnaires only. Another weakness is that the personality disorders were
assessed by the use of a self-report instrument only. Self-report instruments
generally tend to overdiagnose personality disorders, which, on the other
hand, would support our hypothesis that we would find a lower rate of
personality disorders in peacekeepers than in the general population. Yet
another weakness is that the numbers of the different personality disorders
were too low to allow us to compare the impact of different personality
disorders on general mental health.
There were also some weaknesses in the register study of suicides among
the peacekeepers (paper IV). Firstly, we did not have any information about
the start date for when every individual went on his or her tour of service.
36
We therefore set the start time at 20 years or 1960. Secondly, there was no
information about individuals emigrating. This means that the risk period
was probably too long. For instance, if an individual went on his or her tour
at the age of 25 the risk period time would be prolonged by 5 years when he
or she in reality was not at risk as a peacekeeper. This could lead to the
difference between our studied cohort and the general population being
overestimated. However, the difference between the studied cohort and the
general population was so great that even a correct start time and
information about emigrants would probably not be enough to explain it.
Third, it would have been better to compare our cohort with a control group
that had also been selected just like the studied group, but this was not
possible. Instead we used the general population with comparable
distributions of age and sex.
37
Conclusions
The general level of mental health problems and post-traumatic reactions in
a low intensity conflict was low and did not change significantly over time.
Traumatic experiences during service appeared only to make a transient
difference to general mental health.
Possible predictors for poor mental health in peacekeepers might be:
lower mean scores on the SOC questionnaire and on the Emotional Stability
scale of the FFPI; personality disorders in general; family problems or
psychiatric problems expressed through suicide thoughts before deployment;
problems experienced post-deployment, such as “relationship problems”,
“death of a close relative”, or “financial problems”.
It was found that the combination of peer support and a defusing session
led by the platoon leader had some positive effects on post-deployment
mental health.
The rate of personality disorders appears to be at the same level as or at a
slightly lower level than in the general population.
Swedish personnel serving in international peacekeeping operations
appear not to have higher suicide rates than the general population.
The selection procedure involving psychological interviews, well
performed conscription training and training for the actual mission seemed
to work with few psychiatric problems among the individuals.
38
Acknowledgements
I wish to express my sincere gratitude to everyone who has made this work
possible and especially to:
All the personnel in the 6th Bosnian Battalion for agreeing to take part in this
work and thereby make its fulfilment possible.
Assistant Professor Tom Lundin, my main supervisor, friend and tutor for
sharing his professional knowledge in the field, his never-ending optimism
and all his support.
Professor Gerry Larsson for excellent co-supervision, for friendship and for
supporting me with his deep knowledge of methodology and statistics.
Dean and Professor Fritz-Axel Wiesel for his enthusiasm and constructive
criticism.
Charlotte Björkenstam, at the Epidemiological Centre of the Swedish
National Board of Health and Welfare, for her excellent co-operation in
comparing our material with the general population and the national deaths
registry.
Lottie Söder at the office of the Department of Neuroscience and to AnnMari Lundin for all their support.
Göran Engemar at the Uppsala University Library for his excellent
contribution in turning my papers into this book.
My second home, the Swedish Armed Forces, and my superiors there,
Surgeon General Sture Andersson and Colonel Eskil Dalenius, for their
support.
My colleagues at the Armed Forces Medical Centre, Hammarö, and at the
Defence College/ILM in Karlstad for their cheerful support.
39
And finally, the people who are closest to me, Sara, my brothers and sisters
and their families: Gudrun, Sylvia, Eva M., Wolfgang, Giesela, Morgan,
Torleif and also Ulla, Edith and Eva C. for their love and for trusting in me.
The accomplishment of this thesis has been made possible by the Swedish
Armed Forces having generously given me the opportunity to conduct
research as part of my work and by grants from the Swedish Military
Medical Officers Association. I am very grateful for this support.
40
References
Ahrenfeldt, R.H. (1958). Psychiatry in the British Army in the Second World War.
New York: Columbia University Press.
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. Revised.) Washington DC: Author.
American Psychiatric Association. (1994). Diagnostic and statistical manual of
mental disorders (4th Ed.) Washington DC: Author.
Antonovsky, A. (1987). Unraveling the mystery of health: How people manage
stress and stay well. San Francisco: Jossey-Bass.
Antonovsky, A. (1993). The structure and properties of the Sense of Coherence
Scale. Social Science and Medicine, 36, 725-733.
Armstrong, K., O´Callahan, W., Marmar, C.R. (1991). Debriefing Red Cross
disaster personnel: The Multiple Stressor debriefing Model. Journal of
Traumatic Stress, 4, 581-593.
Baggaley, M.R., Piper, M.E., Cumming, P. & Murphy, G. (1999). Trauma related
symptoms in British soldiers 36 months following a tour in the former
Yugoslavia. Journal of the Royal Army Medical Corps, 145, 13-14.
Bank P, & Silk K. (2001). Axis I and Axis II interactions. Current opinion in
Psychiatry, 14, 137-142.
Bisson, J.I., Jenkins, P.L., Alexander, J., Bannister, C. (1997). Randomised
controlled trial of psychological debriefing for victims of acute burn trauma.
British Journal of Psychiatry, 171, 78-81.
Bonhoeffer, K. (1907). Klinische Beiträge zur Lehre von den
Degenerationspsychosen. Halle: Marhold.
Bordow, S. & Porrit, D. (1975). An experimental evaluation of crisis intervention.
Social Science and Medicine, 13a, 251-256.
Bowling, A. (1995). Measuring disease: A review of disease-specific quality of life
measurement scales. Philadelphia: Open University Press.
Bramsen, I., Dirkzwager, A.J.E., Van der Ploeg, H.M. (200). Predeployment
personality traits and exposure to trauma as predictors of posttraumatic stress
symptoms A prospective study of former peacekeepers. American Journal of
Psychiatry, 157, 1115-1119.
Breslau, N., Davis, G.C., Andreski, P., Peterson, E. (1991). Traumatic events and
posttraumatic stress disorder in an urban population of young adults. Archives of
General Psychiatry, 48, 216-222.
Brickenstein, R. (1967). Beiträge zur Wehrpsychiatrie (Contribution to military
psychiatry), (pp.122-151). Bonn: Bundesminis. der Verteidigung
Britt, T.W., Bliese, P.D. (2003). Testing the stress-buffering effects of self
engagement among soldiers on a military operation. J Pers, 71, 245-265.
Brock, A.J. (1918). The re-education of the adult: The neurasthenic in war and
peace. Sociological Review, 10, 25-40.
41
Bullman, T.A., Kang, H.K. (1994). Post-traumatic stress disorder and the risk of
traumatic death among Vietnam veterans. Journal of Nervous and Mental
Disease, 182, 604-610.
Campion, E.W. (1996). Disease and suspicion after the Persian Gulf war. The New
England Journal of Medicine, 14, 1525-1527.
Caplan, G. (1964). Principles of preventive psychiatry. New York: Basic Books.
Carlier, I.V.E., Lamberts, R.D., Gersons, B.P.R. (1997). Risk factors for
posttraumatic stress symptomatology in police officers: A prospective analysis.
Journal of Nervous and Mental Disease, 185, 498-506.
Carlström, A., Lundin, T., & Otto, U. (1990). Mental adjustment of Swedish UN
soldiers in south Lebanon 1988. Stress Medicine, 6, 305-310.
Charcot, J.M. (1887). Lecons sur les maladies du système nerveux faites à la
Salpêtrière. Paris: Progrés Medical en A. Delahaye & E. Lecrosnie.
Costa, P.T.Jr., & McCrae, R.R. (1985). The NEO Personality Inventory Manual.
Odessa, FL: Psychological Assessment Resources.
DaCosta, J.M. (1871). On irritable heart: A clinical study of a form of functional
disorder and its consequences. American Journal of Medical Science, 61, 17-52.
Davidson, J.R., Hughes, D., Blazer, D.G., Georges, L.K. (1991). Post-traumatic
stress disorder in the community: An epidemiological study. Psychological
Medicine, 21, 713-721.
Davidson, J.R.T., & Fairbank, J.A. (1992). The epidemiology of posttraumatic stress
disorder. In Davidson J.R.T., & Foa E.B. (Eds.) Posttraumatic Stress Disorder:
DSM-IV and Beyond (pp. 147-172). Washington, DC: American Psychiatric
Press.
Deahl, M.P., Gillham, A.B., Thomas, J., Searle, M.M., & Srinivasan, M. (1994).
Psychological sequels following the Gulf war: factors associated with
subsequent morbidity and the effectiveness of psychological debriefing. British
Journal of Psychiatry, 165, 60-65.
Deahl, M.P., Srinivasan, M., Jones, N., Thomas, J., Neblett, C., & Jolly, A. (2000).
Preventing psychological trauma in soldiers: The role of operational stress
training and psychological debriefing. British Journal of Medical Psychology,
73, 77-85.
Dyregrov, A. (1989). Caring for helpers in disaster situations: Psychological
debriefing. Disaster Management, 2, 25-30.
Ekselius, L. (1994). Personality disorders in the DSM-III-R. Comprehensive
summaries of Uppsala dissertations from the faculty of medicine 484.
Stockholm: Almqvist & Wiksell.
Ekselius, L., Lindström, E., von Knorring, L., Bodlund, O., Kullgren, G. (1994).
SCID II interviews and the SCID screen questionnaire as diagnostic tools for
personality disorders in DSM-III-R. Acta Psychiatr Scand, 90, 120-123.
Ekselius, L., Tillfors, M., Furmark, T., Fredriksson, M. (2001). Personality disorders
in the general population: DSM-IV and ICD-10 defined prevalences as related
to sociodemographic profile. Pers Individ Diff, 30, 311-320.
Elklit, A. (1998). UN-soldiers serving in peacekeeping missions: A review of the
psychological after-effects. International Review of the Armed Forces Medical
Services, LXXI 7/8/9, 197-207.
Engelstad, J. (1968). Suicides and attempted suicides in the Norwegian armed forces
during peacetime. Military medicine, 133, 437-449.
Erichsen, J. E. (1866). On railway and other injuries of the nervous system. London:
Walton & Moberly.
42
Erichsen, J. E. (1886). On concussion of the spine, nervous shock and other obscure
injuries to the nervous system in their clinical and medico-legal system. New
York: William Wood.
Fauerbach, J.A., Lawrence, J.W., Schmidt, C.W., Munster, A.M., Costa, P.T.
(2000). Personality predictors of injury-related posttraumatic stress disorder.
Journal of Nervous and Mental Disease, 188, 510-517.
Fontana, A., Rosenheck, R. (1995). Attempted suicide among Vietnam veterans: A
model of aetiology in a community sample. American Journal of Psychiatry,
152, 102-109.
Ford, J.D., Shaw, D., Sennhauser, S., Greaves, D., Thacker, B., Chandler, P.,
Schwartz, L., McClain, V. (1993). Psychosocial debriefing after Operation
Desert Storm: Marital and family assessment and intervention. J Soc Issues, 49,
73-102
Freud, S. (1964a). Beyond the pleasure principle. In J. Strachey (Ed & Trans.), the
standard edition of the complete psychological works of Sigmund Freud (pp. 764). London: Hogarth Press. (Original work published in 1920).
Gal, R. (1986). A portrait of the Israeli soldier. New York: Greenwood
Gal, R. (1995). Personality and intelligence in the military: The case of war heroes.
In D.H. Saklofske & M. Zeidner (Eds.), International handbook of personality
and intelligence (pp. 727-737). New York: Plenum.
Glass, A.J. (1954). Psychotherapy in the combat zone. American Journal of
Psychiatry, vol. 110, p 727.
Goldberg, D. P., & Hillier, V. F. (1979). A scaled version of the General Health
Questionnaire. Psychological Medicine, 9, 139-145.
Goodale, P.A. (1999). Anger profile of suicidal inpatient Vietnam veterans.
Dissertation Abstracts International, 59, 5577-B.
Gundersen, E.K., Hourani, L.L. (2003). The epidemiology of personality disorders
in the U.S. navy. Mil Med, 168:575-585.
Helzer, J.E., Robins, L.N., McEvoy, L. (1987). Posttraumatic stress disorder in the
general population. N Engl J Med, 317, 1630-1634.
Hendin, H., Polliger Haas, A. (1991). Suicide and guilt as manifestations of PTSD in
Vietnam combat veterans. American Journal of Psychiatry, 148, 586-591.
Hendriks, A.A.J. (1997). The construction of the Five-Factor Personality Inventory
(FFPI). Groningen, The Netherlands: Rijksuniversiteit Groningen.
Hewitt, P.L. & Flett, G.L. (1996). Personality traits and the coping process. In M.
Zeidner & N.S. Endler (Eds.), Handbook of coping: Theory, research,
applications (pp.410-433). New York: Wiley.
Hobbs, M., Mayou, R., Harrison, B. & Warlock, P. (1996). A randomised controlled
trial of psychological debriefing for victims of road traffic accidents. British
Medical Journal, 313, 1438-1439.
Hofer, J. (1678). Diss de nostalgia. Basel.
Horowitz, M.J., Wilner, M., & Alvarez, W. (1979). Impact of Event Scale: A
measure of subjective distress. Psychosomatic Medicine, 41, 209-218.
Hotopf, M., David, A.S., Hull, L., Ismail, K., Palmer, I., Unwin, C., Wessely, S.
(2003). The health effects of peacekeeping in the UK Armed Forces: Bosnia
1992-1996. Predictors of psychological symptoms. Psychol Med, 33, 155-162.
Hourani, L.L., Warrack, A.G., Coben, P.A. (1999). Suicide in U. S Marine Corps,
1990 to 1996. Military Medicine, 164, 551-555.
Huffman, A. H., Ed, A., Adler, A. B. & Castro, C. A. (1999) The impact of
deployment history on the well being of military personnel. Paper presented at
the American Psychological Association (APA) Convention, 24 August 1999,
Boston.
43
Hytten, K., Weisaeth, L. (1989). Suicide among soldiers and young men in the
Nordic countries, 1977-1984. Acta Psychiatrica Scandinavica, 79, 224-228.
Janet, P. (1889). L´automatisme psychologique. Paris: Alcan.
Janet, P. (1894). Histoire d`une idée fixe. Revue Philosophique, 37, 121-163.
Jessen, G., Hansen-Schwartz, J., Andersen, K., Jörgensen, H.O. (2002). Suicide after
deployment in UN peacekeeping missions - A Danish pilot study. Crisis, 23, 14.
Johansson, E. (2001). The unknown soldier –A portrait of the Swedish peacekeeper
at the threshold of the 21st century. Karlstad: Karlstad University Studies.
Kang, H.K., Bullman, T.A., (1996). Mortality among U.S. Veterans of the Persian
Gulf War. New England Journal of Medicine, 335, 1498-1504.
Kettner, B. (1972). Combat strain and subsequent mental health. A follow-up study
of Swedish soldiers serving in the UN-forces 1961-62. Acta Psychiatrica
Scandinavica Suppl, 230, 1-112.
Kilpatrick, D.G. & Resnick, H.S. (1993). Posttraumatic stress disorder associated
with exposure to criminal victimisation in clinical and community populations,
In Davidson J.R.T., & Foa E.B. (Eds.) Posttraumatic Stress Disorder: DSM-IV
and Beyond. (pp. 113-143), Washington, DC: American Psychiatric Press,.
Kramer, T.L., Lindy J.D., Green B.L., Grace, M.C., Leonard, A.C. (1994). The comorbidity of post-traumatic stress disorder and suicidality in Vietnam veterans.
Suicide and life-Threatening Behaviour, 24, 58-67.
Kulka, R. A., Schlenger, W.E., Fairbank, J.A. et al. (1990). Trauma and the Vietnam
War generation. New York: Brunner/Mazel.
Kulka, R. A., Schlenger, W.E., Fairbank, J.A. et al. (1991). Assessment of
posttraumatic stress disorder in the community: Prospects and pitfalls from
recent studies of Vietnam War veterans. Psychological Assessment. Journal of
Consulting and Clinical Psychology, 3, 547-560.
Larsson, G. & Kallenberg, K. (1996). Sense of coherence, socio-economic
conditions and health: Interrelationships in a nation-wide Swedish sample.
European Journal of Public Health, 6, 175-180.
Larsson, G. & Österdahl, L. (1996). Crisis Support: Handbook for ordinary people.
Karlstad, Sweden: The Swedish Rescue Services Agency.
Larsson, G. Kallenberg, K., Setterlind, S. & Starrin, B. (1994). Health and loss of a
family member: Impact of sense of coherence. International Journal of Health,
5, 3-9.
Lee, C., Slade, P. & Lygo, V. (1996). The influence of psychological debriefing on
emotional adaptation in females following early miscarriage. British Journal of
Medical Psychology, 69, 47-58.
Lester, D. (1994). The military participation rate and suicide rates in Austria, 18731913. Psychological Reports, 75, 894.
Lester, D. & Yang, B. (1991). The effect of war on personal violence. Medicine &
War, 7, 215-218.
Lim, L.C., Ang, Y.G. (1992). Parasuicide in male conscripts: A Singapore
experience. Military Medicine, 157, 401-403.
Litz, B.T., Orsillo, S.M., Friedman, M., Elich, P. & Batres, A. (1997). Post traumatic
stress associated with peacekeeping duty in Somalia for United States military
personnel. American Journal of Psychiatry, 154, 178-184.
Litz, B. (2004). Paper presented at ISTSS annual meeting in New Orleans.
Longato-Stadler, E., von Knorring, L., Hallman, J. (2002). Mental and personality
disorders as well as personality traits in a Swedish male criminal population.
Nord J Psychiatry, 56, 137-144.
44
Lundin, T. (1995). Stressreaktioner och psykiskt trauma: Diagnostik och behandling
[Stress reactions and mental trauma: Diagnostics and treatment]. Helsingborg,
Sweden: Rhône-Poulenc RorerAB.
Lundin, T. (1995). Collision at sea between two Navy vessels. Military Medicine,
160, 323-325.
Lundin, T. & Otto, U. (1989). Stress reactions among Swedish health care personnel
in Unifil, South Lebanon 1982-1984. Stress Medicine, 5, 237-246.
Lundin, T. & Otto, U. (1992). Swedish UN soldiers in Cyprus, UNFICYP: Their
psychological and social situation. Psychotherapy and Psychosomatics, 57, 187193.
MacDonald, C., Chamberlain, K., Long, N., Pereira-Laird, J., & Mirfin, K. (1998).
Mental health, physical health, and stressors reported by New Zealand Defence
Force Peacekeepers: A longitudinal study. Military Medicine, 163, 477-481.
Maguen, S., Litz, B.T., Wang, J.L., Cook, M. (2004) The stressors and demands of
peacekeeping in Kosovo: predictors for mental health response. Military
Medicine. 169, 198-206.
Maier, W., Lichterman, D., Klingler, T., Heun, R. (1992). Prevalence of personality
disorders (DSM-III-R) in the community. J Pers Disord, 6, 187-96.
Marshall, S.L.A. (1944). Island Victory. New York: Penguin Books
Marshall, S.L.A. (1947). Men under fire: The problem of battle command in the
future war. New York: William Morrow.
McDowell, I. & Newell, C. (1987). Measuring health: A guide to rating scales and
questionnaires. New York: Oxford University Press.
McFarlane, A.C. (1988b). The phenomenology of posttraumatic stress disorders
following natural disasters. Journal of Nervous and Mental Disease, 176, 22-29.
McFarlane, A.C. (1992). Avoidance and intrusion in posttraumatic stress disorder.
Journal of Nervous and Mental Disease, 180, 439-445.
Mehlum, L. & Weisaeth, L. (2002). Predictors for posttraumatic stress reactions in
Norwegian U.N. peacekeepers 7 years after service. Journal of Traumatic
Stress, 15, 16-26.
Milgram, N.A. (1986). Stress and coping in time of war: Generalising from the
Israeli experience. New York: Brunner/Mazel.
Mitchell, J. (1983). When disaster strikes: The critical incident stress debriefing
process. Journal of Emergency Medical Services, 8, 36-39.
Mitchell, J.T. & Bray, G.P. (1990). Emergency services stress: Guidelines for
preserving the health and careers of emergency personnel. Englewood Cliffs,
NJ: Brady Prentice Hall Career & Technology.
Mitchell, J.T. & Everly, G.S. (1995). Critical incident stress debriefing: An
operations manual for the prevention of traumatic stress among emergency
services and disaster workers. Elicott City, MD: Chevron.
Myers, C.S. (1915). A contribution to the study of shell shock. Lancet I, 316-320.
Narud, K. & Dahl, A. (2002). Quality of life in personality and personality
disorders. Current Opinion in Psychiatry, 15, 131-133.
National Institute of Mental Health (2002). Mental Health and Mass Violence:
Evidence Based Early Psychological Interventions for Victims /Survivors of
Mass Violence. A Workshop to Reach Consensus on Best Practices. NIH
Publication No. 02-5138, Washington D.C.: US Government Printing Office.
Oppenheim, H. (1889). Die traumatischen Neurosen. Berlin: Hirschwald.
Orsillo, S.M., Roemer, L., Litz, B.T., Ehlich, P., & Friedman, M.J. (1998).
Psychiatric symptomatology associated with contemporary peacekeeping: An
examination of post-mission functioning among peacekeepers in Somalia.
Journal of Traumatic Stress, 11, 611-624.
45
Ottosson, H., Bodlund, O., Ekselius, L., von Knorring, L., Kullgren, G., Lindström,
E, et al. (1995). The DSM-IV and ICD-10 Personality Questionnaire (DIP-Q):
Construction and preliminary validation. Nord J. Psychiatry, 49, 285-291.
Ottoson, H., Bodlund, O., Ekselius, L., Grann, M., von Knorring, L., Kullgren, G.,
Lindström, E., Söderberg, S. (1998). DSM-IV and ICD-10 Personality
Disorders: A comparison of a self-report questionnaire (DIP-Q) with a
structured interview. European Psychiatry, 13, 246-253.
Page, H. W. (1883) Injuries of the spine and spinal cord without apparent
mechanical lesion and nervous shock in their surgical and medico-legal aspects.
London: J & A Churchill.
Parkes, C.M. (1980). Bereavement counselling: Does it work? British Medical
Journal, 281, 3-6.
Paykel, E.S., Prusoff, B.A., Myers, J.K. (1975). Suicide attempts and recent life
events. Archives of General Psychiatry, 32, 327-333.
Pirkis, J., Burgess, P., Jolley, D. (1999). Suicide attempts by psychiatric patients in
acute treatment, long-stay in-patient and community care. Soc Psychiatry
Epidemiol, 34, 634-644.
Polak, P.R., Egan, D., Vandebergh, R. & Williams, W.V. (1975). Prevention in
mental health A controlled study. American Journal of Psychiatry, 132, 146149.
Ponteva, M., Jormanainen, V., Nurro, S., Lehesjoki, M. (2000). Mortality after the
UN service. Follow-up study of the Finnish peacekeeping contingents in the
years 1969-1996. Int Rev Armed Forces Med Serv, 73, 235-239.
Raphael, B. (1977). Preventive interventions with the recently bereaved. Archives of
General Psychiatry, 34(12), 1450-1454.
Raphael, B., Singh, B., Bardbury, L., Lambert, F. (1983). Who helps the helpers:
The effect of a disaster on rescue workers. Omega, 14, 9-20.
Raphael, B. (1986). When disaster strikes (pp. 222-244). New York: Basic Books.
Reich, J., Yates, W., Nduaguba, M. (1989). Prevalence of DSM-III personality
disorders in the community. Soc Psychiatr Epidem, 24, 12-16.
Rose, S., Bisson, J., Wessely, S. (2002). Psychological debriefing for preventing
posttraumatic stress disorder (PTSD). In The Cochrane Library, Issue 1.
Oxford: Update software.
Rosenbaum, M. (1988). Learned resourcefulness, stress and self-regulation. In S.
Fisher & J. Reason (Eds.), Handbook of life stress, cognition and health (pp.
483-496). New York: Wiley.
Rosenheck, R., Becnel, H-, Blank, A.S. et al. (1992). Returning Persian Gulf
troops: First year findings. Report of the department of Veterans affairs to the
United States Congress on the Psychological Effects of the Persian Gulf War.
Rothberg, J., Rock, N., Shaw, J., Jones, F. (1988). Suicide in United States army
personnel, 1983-1984. Military Medicine, 153, 61-64.
Rikhye, I.J., Harbottle, M., & Egge, B. (1974). The thin blue line. New Haven and
London: Yale University Press
Salmon, T. (1917). The care and treatment of mental diseases and war neuroses
(“shell shock”) in the British Army. Mental Hygiene, 1, 509-547.
Samuels, J., Eaton, W.W., Bienvenu, O.J.3rd., Brown, C.H., Costa, P.T.Jr., Nestadt,
G. (2002). Prevalence and correlates of personality disorders in a community
sample Br J Psychiatry, 180, 536-542.
Schnurr, P.P., Friedman, M.J., Rosenberg, S.D. (1993). Preliminary MMPI scores as
predictors for combat related PTSD. Psychotherapy and Psychosomatics, 72,
333-342.
46
Schroderus, M., Lönnqvist, J.K., Aro, H.M. (1992). Trends in suicide rates among
military conscripts. Acta Psychiatrica Scandinavica, 86, 233-235.
Segal, D.R., Furukawa, T.P. & Lindh, J.C. (1990). Light infantry as peacekeepers in
the Sinai. Armed Forces & Society, 16, 385-403 (1990).
Shalev, A.Y. (1994). Debriefing following traumatic exposure. In R.J. Ursano, B.G.
McCaughey & C. S. Fullerton (Eds.), Individual and community responses to
trauma and disaster: The structure of human chaos. (pp. 201-219). Cambridge:
Cambridge University Press.
Shalev, A.Y., Peri, T., Rogel-Fuchs, Y., Ursano, R.J., Marlowe, D. (1998).
Historical group debriefing after combat exposure. Military Medicine, 163, 494498
Shay, J. (1994). Achilles in Vietnam. Combat trauma and the undoing of character.
New York: Atheneum.
Shea, M., Zlotnik, C., Weisberg, R. (1999). Commonality and specificity of
personality disorder profiles in subjects with trauma history. J Personal Disord,
13, 1733-1739.
Shepard, B. (2000). A war of nerves. Soldiers and psychiatrists 1914 –1994.
London: Jonathan Cape.
Shore, J.H., Tatum, E.L., Vollmer, V.M. (1986). Psychiatric reactions to disaster:
The Mount St. Helens experience. Am J Psychiatry, 143, 590-595.
Shore, J.H., Vollmer, V.M., Tatum, E.L. (1989). Community patterns of
posttraumatic stress disorders. J Nerv Ment Dis, 177, 681-685.
Smith, E. & North, C. (1993). Posttraumatic stress disorder in natural disasters and
technological accidents. In Wilson J.P. & Raphael B. (Eds.), International
Handbook of Traumatic Stress Syndromes (pp. 405-420). New York: Plenum,
Soloff, P.H., Lynch, K., Kelly, T.M, et al. (2000). Characteristics of suicide attempts
of patients with major depressive episodes and borderline personality disorders:
a comparative study. Am J Psychiatry, 157, 601-607.
Solomon, Z. (1993). Combat stress reaction: The enduring toll of war. New York:
Plenum.
Solomon, Z., Benbenishty, R. (1986). The role of proximity, immediacy and
expectancy in frontline treatment of combat stress reactions among Israelis in
the Lebanon War. Am J Psychiatry, 143, 613-617.
Spiller, R.J. (1980). S.L.A. Marshall at forth Leavenworth 1952-1962. Fort
Leavenwort: KS, The Combined Arms Center.
Spiller, R.J. (1988). S.L.A. Marshall and the ratio of fire. The Royal United Service
Institute for Defence Studies Journal. 133, 63-71
Steinglass, P. & Gerrity, E. (1990). Natural disasters and posttraumatic stress
disorder: Short-term versus long-term recovery in two disaster-affected
communities. Journal of Applied Social Psychology, 20, 1746-1765.
Suominen, K., Isometsa, E., Henriksson, M. (2000). Suicide attempts and
personality disorder. Acta Pychiatr Scand, 102, 118-125.
Sutker, P.B., Winstead, D.K., Galina, Z.H. (1991). Cognitive deficits and
psychopathology among former prisoners of war and combat veterans of the
Korean conflict. Am J Psychiatry, 148, 67-72.
Thoresen, S., Mehlum, L., Moller, B. (2003). Suicide in peacekeepers--a cohort
study of mortality from suicide in 22,275 Norwegian veterans from international
peacekeeping operations. Soc Psychiatry Psychiatr Epidemiol, 11, 605-10.
Torgersen, S., Kringlen, E., Cramer, V. (2001). The prevalence of personality
disorders in a community sample. Arch Gen Psychiatry, 58, 59-596.
47
Ursano, R.J., Boydstun, J.A., Wheatley, R.D. (1981). Psychiatric illness In U.S. Air
force Vietnam prisoners of war: A five-year follow up. Am J Psychiatry, 138,
310-314.
Wasserman, D. (2003). General aspects on poor mental health (Allmänt om psykisk
ohälsa). In Theorell T. (Ed.), Psychosocial environment and stress. (Psykosocial
miljö och stress). Lund: Studentlitteratur (p141).
Weisaeth, L., Aarhaug, P., Mehlum, L., & Larsen, S. (1993). The UNIFIL study
(1991-1992). Report I: Results and recommendations. Headquarters Defence
Command Norway - The Joint Medical Service.
Weisaeth, L. (2002). The European history of Psychotraumatology. Journal of
Traumatic Stress, 6, 443-452.
Weisaeth, L. (2003). The psychological challenge of peacekeeping operations. In
Britt T.W. & Adler A.B. (Eds.), The Psychology of the Peacekeeper - Lessons
from the field (pp. 207-222). London: Praeger
Wessely, S., Rose, S. & Bisson, J. (2000). Brief psychological interventions
(“debriefing”) for trauma-related symptoms and the prevention of
posttraumatic stress disorder. Cochrane review. Oxford: The Cochrane Library.
Veterans Health Administration, Department of Veterans Affairs and Health Affairs,
Department of Defense, (2003). Management of Post-Traumatic Stress.
Washington, DC: VA/DoD Clinical Practice Guideline Working Group, Office
of Quality and Performance Publication 10Q-CPG/PTSD-04.
Wong; A., Escobar, M., Lesage, A., Loyer, M., Vanier, C., Sakinofsky, I. (2001).
Are UN peacekeepers at risk for suicide? Suicide Life Threat Behav, 31, 103112.
World Health Organisation. (1957). International Classification of Diseases
(Revision 7). Geneva: WHO Library.
World Health Organisation. (1967). International Classification of Diseases
(Revision 8). Geneva: WHO Library.
World Health Organisation. (1977). International Classification of Diseases
(Revision 9). Geneva: WHO Library.
World Health Organisation. (1992). International Classification of Diseases
(Revision 10). Geneva: WHO Library.
Zimmermann, M. & Coryell. W. (1990). Diagnosing personality disorders in the
community. Arch Gen Psychiatry, 47, 527-531.
Zlotnik, C., Franklin, L. & Zimmerman M. (2002). Is co-morbidity of post-traumatic
stress disorder and borderline personality disorder related to greater pathology
and impairment? Am J Psychiatry, 159, 1940-1943.
48
2 3 3
/ 4 5 .6 5 2 57 .6 5 8 3
368 ..6 .776 5 .70 5 9 2 5
5 7 7: 08 .776 0.
6 . 4 $7
.77 5 3 4 57 .6 5
9 ; <.68 ())'8 .0
. =$7 .77 5 3
4 57 .6 5 =9>
40./ .09..9
./0//../!,1('