Background
Civilian populations exposed to trauma, such as military persecution and war, often suffer from a variety of psychological complaints including anxiety, depression, and posttraumatic stress symptoms. For instance, prevalence rates of 48% and 42% for mood and anxiety disorders, respectively, were found in Kosovar civilian adults having survived the war [
1].
Somatization symptoms, i.e., somatic symptoms for which no or no sufficient organic causes are found, are frequently found to be related with posttraumatic stress disorder (PTSD), regardless of the traumatic event [
2]. For example, in war veterans, combat stress reactions and PTSD were related to somatization symptoms, i.e. pain [
3,
4]. Similar findings were found among a sample of factory accident survivors, where PTSD severity was positively correlated with somatization [
5]. Disaster survivors suffering from PTSD reported more physical symptoms than those without PTSD [
6], a finding that was also found 14 years after the genocide in Ruanda [
7]. Another study found in a refugee population an association of somatization with PTSD severity [
8]. Andreski et al. [
9] found in their prospective study of PTSD and somatization, that PTSD increased the risk of developing somatization symptoms but that new PTSD cases were not elevated in people with a somatization disorder. Taken together, it seems that somehow PTSD may mediate the development of somatization following a traumatic event [
10‐
12].
Various studies have examined the role of the different PTSD clusters on somatic symptoms. As such, studies have found that specific PTSD symptoms, e.g., hyperarousal may cause repeated muscle tension that could result in somatization complaints [
13]. In other studies intrusive re-experiencing symptoms [
6] and numbing [
14] also predicted somatic symptoms in PTSD patients. There are various theoretical models to explain the association between posttraumatic stress and somatization health symptoms. For example, the shared vulnerability [
15,
16] and mutual maintenance model [
17], and a recent extension, the perpetual avoidance model [
18], propose that individuals with PTSD and somatization symptoms have shared vulnerability to both conditions, or alternatively, that pain symptoms and posttraumatic stress symptoms interact with each other.
While there is a considerable amount of literature addressing survivors of combat, motor vehicle accidents and interpersonal trauma, war and postwar civilian populations have been less studied to date [
19,
20]. To the best of our knowledge, no studies have examined war-traumatized civilians in terms of the differential relationship of each posttraumatic stress symptom cluster and somatic problems. This cross-sectional study aimed at investigating the relationship between trauma, posttraumatic stress symptoms and somatization symptoms in postwar civilians, i.e., Kosovar civilians more than a decade after the war. We hypothesized that despite the relatively long time lapse since the end of war, participants would suffer from clinically relevant symptoms of posttraumatic stress and somatization symptoms, and that specific PTSD symptom clusters would be associated with somatization symptoms.
Discussion
In this cross-sectional study, we examined the relationship between trauma exposure, posttraumatic stress symptoms and the level of somatization symptoms in a randomly selected sample of Kosovar civilian war survivors (N = 142). Substantial trauma exposure and high levels of clinically relevant symptoms of PTSD and somatization symptoms were found. Somatization symptoms correlated significantly with PTSD symptoms. In mediation analyses, the relationship between traumatic exposure and somatization symptoms was partly mediated by active avoidance and by posttraumatic hyperarousal, accounting for 37% of the variance in somatization symptoms.
As expected, compared to non-conflict countries, lifetime trauma exposure was substantial in the civilian population in Kosovo [
35‐
37]. This is unsurprising since the cohort was sampled from a war-affected region. Further, the high rates of probable PTSD (26%) and the high levels of somatization symptoms in the present sample are in line with earlier studies conducted in the post-war Balkans [
1,
38,
39].
In addition, consistent with findings in the previous literature, female gender was associated with higher scores on somatization symptoms [
40‐
42]. However, PTSD in the present study was equally found in women and men, which is in contrast to the epidemiological literature, but has been shown in samples of military personnel [
43]. There is evidence, though, that gender may play a mediating role in the relationship of trauma and somatization symptoms, with females developing more somatization complaints as compared to men [
9,
40]. This pattern overlaps with the robust finding of more females developing PTSD than males [
44].
One of the interesting findings in this study was the differential role of PTSD symptom clusters on somatization symptoms. Specifically, we found that arousal mediated the association between trauma exposure and somatization symptoms using the PTSD cluster’s classification by King and colleagues [
27]. A comparable result was found by McFarlane et al. [
6] in a sample of fire-fighters. The present study adds to the previous literature, however, by subdividing the avoidance cluster into active avoidance and numbing to the model. It is also conceivable that the numbing symptoms overlap with alexithymia and depressive symptoms, which have been shown to mediate the association of trauma exposure on somatization in major depression [
45]. In addition, it is possible that passive avoidance may contribute to somatization presentations, as shown in the Escalona and Colleagues study [
14], because these responses can include social and emotional withdrawal, which in turn might increase awareness of and focus on internal sensory perception. Avoidance behaviors in people suffering from pain can be attributed, in part, to catastrophizing about the severity of the pain as a predominant somatic symptom and the patient’s inability to function, which may contribute to fear of pain and lead to avoidance of activities [
46].
Active avoidance is a mediator of the relationship between trauma and somatic symptoms in a negative direction. This means the more one avoids situations and thoughts that remind him of the trauma, the less somatization symptoms one has. Further, this could mean, if someone is avoidant of thoughts and feelings, he is not very good at recognizing how he feels and so reports lower levels of subjective somatic symptoms. There has also been considerable attention given to the bidirectional relationship between PTSD and posttraumatic somatization symptoms such as pain. As Sharp and Harvey (Sharp & Harvey, 2001) have argued, PTSD patients can have their trauma memories triggered by episodes of pain because their pain is associated with the experience of the traumatic event. Conversely, posttraumatic stress can exacerbate pain by (a) increasing arousal, (b) elevating muscle tension, (c) promoting vigilance to pain sensations, and (d) exaggerating negative appraisals about the severity of the pain experience.
Additional war-related consequences, such as social problems, may complicate the understanding of etiology, assessment, and course of treatment. In a low-income country such as Kosovo, socio-economic hardship and health problems might additionally be associated. Beyond this, it has been suggested that among some non-western, collectivistic cultures, somatization symptoms are among the most frequent presentations of trauma survivors [
47,
48]. In Kosovo, it is conceivable that being psychologically ill might be socially unacceptable, whereas presenting somatization symptoms is more acceptable in Kosovarian people.
This study has several limitations. First, our results are based on self-reports assessed by questionnaires which usually have lower reliability than structured clinical interviews. However, because questionnaires were administered within a personal interview conducted by trained clinical psychology master students, we expect sufficient reliability. Another limitation is the use of instruments that have not yet been validated in Albanian. However, translations and back-translations were thoroughly performed by experienced and natively speaking interpreters in mental health and after consensus discussions adapted by mental health care providers, if necessary, in order to achieve maximum quality. The long period between the war and our assessment might have induced inaccurate recall; moreover, it is not clear whether the problems reported by participants were a consequence of the war, the current social situation and living conditions, or other adversities participants had been exposed to during the course of their lifetime. Finally, the cross-sectional approach of the present study does not allow to draw any conclusions regarding causal relationships between trauma exposure and the presence of somatization symptoms. However, our findings allow us to generate specific hypotheses regarding possible mechanisms and causal factors contributing to the prominent somatization symptoms in traumatized civilian war survivors. These precise assumptions about etiological mechanisms linking trauma exposure to somatization symptoms must be investigated in further research with longitudinal designs.