Summary: Post Traumatic Stress Disorder
Post Traumatic Stress Disorder in Kurdish Iranians
Mehdi Ghazinour1, Naser Mofidi1, Nader Esmail-Nasab2, Jörg Richter3
1 Department of Clinical Sciences, Division of Psychiatry, Umeå University, S-90187 Umeå, Sweden
2 Kurdistan University of Medical Sciences, Sanandaj, Iran
3 Centre of Child and Adolescent Mental Health, Regions South and East, P.O.Box 23 Tåsen, N-0801 Oslo, Norway
Corresponding author:
Prof. Dr. Jörg Richter
Centre for Child and Adolescent Mental Health, Eastern and Southern Norway,
P.O.Box 23 Tåsen
N-0801 Oslo
Norway
Running head: PTSD in Kurdish Iranians
Key words: PTSD, general mental health, life events, Kurdish Iranian, general population
Abstract
Posttraumatic stress disorder (PTSD) represents …show more content…
a major public health problem world-wide. In order to derive preliminary prevalence figures of PTSD and general mental health in Kurdish Iranians the applicability of the PTSD Checklist (PCL) was investigated combined with the 12- item version of the General Health Questionnaire (GHQ-12) and a slightly modified version of the Life event Checklist (LEC). 1000 randomly selected individuals were investigated. About 40.1 % of the subjects reported that they have been affected by one of the possible traumatic events. A prevalence of PTSD based on the PCL of 10.9 % was found which was higher than that reported in other countries, but lower than expected. With decreasing emotional involvement in the event, the probability of suffering from a PTSD decreased. Women suffered significantly more often from PTSD then men especially from re-experiencing and arousal symptoms. Particularly women seem to be in need for culturally adapted intervention procedures related to PTSD. The results support the construct and differential validity of the Persian version of the PCL.
Introduction
Post Traumatic Stress Disorder (PTSD), one type of anxiety disorder, represents a major public health problem world-wide [1]. The initial focus on trauma-related consequences historically changed from an exclusive somatic perspective to a much more comprehensive understanding that incorporated a major psychological perspective. Investigations of PTSD as a public health problem were mainly initiated in the U.S.A. and led to the inclusion of PTSD as a distinct psychiatric entity in the DSM-III. By accepting PTSD as a psychiatric disorder, its nature and origin became subject of comprehensive discussions. It is argued that PTSD represents a timeless condition, which existed before it was codified in modern diagnostic systems, but was described in different terms such as ‘railway spine’ and ‘shellshock’. Others suggest that PTSD is a novel presentation that is caused by modern interaction between trauma and culture [2].
PTSD represents a development occurring in individuals after they have experienced a traumatic event. A 'traumatic' event is characterized by its capacity to provoke fear, helplessness, or horror in response to threat of injury or death. Symptoms include distressing 'flash backs' (re-experiencing the event), avoidance of situations reminding the individual of the event, increased arousal, and psychological numbing.
The fist epidemiologic report from the U.S.A. conducted by Helzer and colleagues [3] estimated a prevalence of PTSD of about 1 % in the general population and identified a higher rate of PTSD in women than in men. However one criticism of this study was a probability of underestimation of PTSD [4,5]. Another prevalence study of high validity was conducted in the general population found an overall prevalence rate of PTSD of 7.8 % with a rate of 10.4 % among women, twice as high as that found among men (5.0 %) [6]. Estimates of trauma exposure showed that about 60% of men have been the victim of one or more traumatic events, while 51% of women were exposed to such events [4].
Epidemiological investigations of traumatic life events and PTSD provide important guidance for researchers and clinicians in determining who may possibly develop a PTSD following certain traumatic life events.
Against this background, epidemiological research on exposure to traumatic events and PTSD in different areas of life increased dramatically with a major focus on PTSD in soldiers; for example Persian Gulf War veterans [7] or soldiers on active duty in Somalia [4,8]. In addition, PTSD caused by catastrophes and disasters like bush fires or air crashes [9-11]; following domestic violence, rape or sexual abuse [12-14]; or after ware related traumatic life events, torture and terrorist attacks [15-19] has been investigated in many …show more content…
countries.
Since PTSD represents a global problem, and epidemiological data provides important knowledge for health care planning, our research group conducted an investigation of general mental health and the prevalence of posttraumatic stress disorder among Kurds living in Iran. It represents the first study of mental health problems in this particular population. The research was conducted in accordance with World Health Organization’s programs [20] in collaboration with the Medical University School of Kurdistan in the West of Iran.
About Kurdistan
Kurdistan is a traditional ethnographic region and covers the regions of Northern and Northeastern Mesopotamia. It is a mountainous region politically shared between several nations: Turkey, Iran, Iraq and Syria with overall estimated 27-28 million people living on an area of 190,000 km² (74,000 square miles). One part of Kurdistan (28,817 km²) is located in the West of Iran, with borders to Iraq in the West and to West Azerbaijan in the North. Sanandaj, the capital of Iranian Krdistan, had an estimated population of 358,084 inhabitants in 2006. The population of Sanandaj is mainly Kurdish with Armenian and Jewish minorities. The Kurds have been fighting for democracy and autonomy since centuries which represents one of the reasons for Kurdistan’s engagement in civil war. During the Iran-Iraq war (1980 -1988) Kurdistan was attacked by Iraq and heavily bombed [21]. During that time, trauma and loss were daily life experiences for many individuals [22]. Today the Kurdish population encounters many stresses including low income, employment difficulties, and ethnic repression.
The aims of the present study were; (a) To investigate the applicability of the PTSD Checklist (PCL) [23] to a population from a Middle-Eastern country and to derive preliminary prevalence figures of PTSD in Iranian Kurdistan; (b) To analyze psychometric properties of the Persian version of the PCL.
Methods
The presented findings represent part of a more comprehensive investigation relating to attitudes towards suicide, general mental health, quality of life and posttraumatic stress disorders among Kurdish people in Iran [24,25]. The research plan was approved by the ethics committee at the University of Umeå, Sweden. Before the project started, it was presented to the deputy of research of the Kurdistan University of Medical Sciences for evaluation and approval.
Sample
The sample under study consisted of 1000 Kurds randomly selected from 1000 of 68000 households in Sannandaj city.
The data collection was performed during April and May 2006 in collaboration with University of Medical Science in Kurdistan. Households were selected by a cluster random sampling process based on the 24 health care districts of Sannandaj city and dependent on their size. 100 clusters were derived in which ten households each were approached. Each data collector continued to approach households until they obtained ten respondents per cluster. The person that opened the door was invited to participate in the study if he or she was older than eighteen and none of the approached individuals refused to participate. Notwithstanding the random sampling procedure, individuals with higher education are substantially overrepresented in our sample and among women in
particular.
Every health care centre in Sannandaj city participated in this investigation, providing two health care professionals who were experienced in data collection. Data collectors received an one day training by the first author concerning the questionnaires, data collection procedures and how to deal with illiterate people. If the participant was illiterate, the data collector read the items aloud for the person.
Insert Table 1 about here
Measures
The PTSD Checklist, civil version, was chosen to dimensionally assess strain caused by traumata and to facilitate a related PTSD diagnosis [23]. It consists of seventeen items which correspond to the DSM-IV symptoms of PTSD. Where an individual reported a major life incident, they were instructed to indicate how much they had been bothered by each symptom in the past month using a five point scale from 1 = not at all to 5 = extremely. Scores related to the three PTSD symptom dimensions (re-experiencing, numbing and avoidance, and increased arousal) and a total score indicating the PTSD symptom severity were calculated by summing scores across the respective items. The Cronbach’s alpha coefficients (between 0.92 and 0.97) and the item-intercorrelation (between 0.62 and 0.87) related to the 3 symptom-cluster scales and the total score are reported to be highly satisfying [23]. In our study, we found the psychometric properties to be somewhat lower, but still acceptable, taking the low number of items into account (Re-Experiencing scale – cluster B symptoms: alpha = 0.79; corrected item-intercorrelation: 0.47 – 0.63; Numbing and Avoidance scale – cluster C symptoms: alpha = 0.70; corrected item-intercorrelation: 0.29 – 0.47; Increased arousal scale - cluster D symptoms: alpha = 0.77; corrected item-intercorrelation: 0.46 – 0.61; Total score: alpha = 0.88; corrected item-intercorrelation: 0.33 – 0.67).
A reduced and adapted version of the Life Event Checklist (LEC [26]) was used to indicate the occurrence of traumatic events experienced by the individuals. Fourteen possibly traumatic events were included (car accident; threat; sudden death; injury; maltreatment; disease; drowning accident; burning (scalding) oneself/suicide; fall; explosion/fire accident; shock; bombardment; and prison). The participants were instructed to provide information about whether they had experienced the event themselves, were a witness to the event, had simply heard about it, were not sure about the source of information, or had no experience of such a type of event. Due to very few reports of traumatic events (1 – 5) in six categories we decided to add these experiences to analogous categories.
General mental health was assessed by means of the 12-item General Health Questionnaire (GHQ-12 - [27]). This screening instrument was developed in order to detect psychiatric disorders in community settings and non-psychiatric clinical settings. It has been used in many investigations in various cultural regions including Iran [28] and its psychometric properties have been extensively investigated. Our findings are based on the original scoring used by Goldberg with response categories score ‘not at all’ and ‘no more than usual’ as 0 and ‘rather more than usual’ and ‘much more than usual’ as 1 providing a possible range from 0 to 12. By means of an exploratory factor analysis we derived two factors explaining 40 per cent of the variance. In accordance with Goldberg, Oldehinkel and Ormel’s [29] p. 921) recommendation “…, if the mean is below 1.85 then the threshold of ½, from 1.85 to 2.7 a threshold of 2/3, and above 2.7 a threshold of ¾ seems to work best for the GHQ-12.”, we decided to use a score of 5 (individuals with a total GHQ-12 score > 5 are considered as cases), the ¾ score in our sample, as the cut-off point for the differentiation between individuals with and without psychiatric morbidity.
Following the investigation of Montazeri and co-workers in Iran [28], two additional items related to overall quality of life were added to the GHQ (‘How do you evaluate your quality of life?’ and ‘How do you evaluate your general health?’ with possible answers: Very good; good; moderate; bad; very bad).
Statistics
Means and standard deviations and percentages are provided. T-tests for independent samples, chi-square-tests and one-way analyses of variance with Bonferroni controlled post-hoc comparisons were used for searching for differences between subgroups under study dependent on the scale-level in univariate analyses. Pearson correlation coefficients, partial correlations and linear multiple regressions were calculated for the identification of relationships between variables.
Results
Four hundred and eight of the 1000 individuals approached reported that they had once experienced, witnessed, been told about, or were not sure where they had got the information from, but had been affected by one of the possible traumatic events set out in the reduced life events checklist (Table 2). Most of them had witnessed these events. There was a significant relationship between the type of event and the relationship to the event with injuries, diseases and maltreatments most often being self-experienced; sudden deaths and drowning most often reported as witnessed; drowning, burns/scalds and suicides were most often reported as being heard about from somebody (χ2 = 48.37; p = 0.001).
Table 2 about here
According to the PCL, 10.9 % of the individuals in the investigated sample have to be classified as suffering from a PTSD. With decreasing emotional involvement in the event, the probability of suffering from a PTSD decreased (PTSD when it was ‘one’s own experience’: 25 %, when it was ‘witnessed’: 22 %, when ‘told about by somebody’: 14 %, when they were ‘not sure of the source’: 7 % - Fisher’s exact test = 127.01; p < 0.001). There was no substantial relationship between PTSD and particular types of events (χ2 = 4.61; p = 0.708). Women suffered significantly more often from PTSD than men (12.7 % vs. 8.8 % - χ2 = 4 .02; p = 0.045); whereas there was no difference in the number of men and women who had to be regarded as case based on the applied GHQ-12 cut-off score (χ2 = 0.17; p = 0.680).
Taking the GHQ-12 total score as reference, the PCL diagnosis had a sensitivity of 16 % and a specificity of 9 % in our sample (Table 3).
Table 3 about here
Based on dimensional analyses, we could not find any gender related differences on either the GHQ-12 total score or on the factors. However, women reported more severe re-experiencing (F = 9.87; p = 0.002), more arousal symptoms (F = 5.71; p = 0.017), and a higher PCL total score than (F = 6.58; p = 0.011) (see Table 3). The employment situation and the level of education were found unrelated to the PCL scores, whereas an one-way ANOVA procedure yielded a significant result for the GHQ-12 total score (F = 3.18; p = 0.002 and 5.43; p < 0.001 respectively) with the first relationship not based on particular differences between groups indicated by significant differences in Bonferroni controlled post-hoc comparisons. Whereas the marital status was found to be related to GHQ-12 total score (F = 14.82; p < 0.001), it was unrelated to PTSD scores.
Table 4 about here
There was no relationship between the type of event and the PCL scores. However, the severity of re-experience symptoms differed significantly depending upon the type of traumatic event (F = 2.85; p = 0.007), a finding mainly based on the fact that those who experienced sudden death reported that they had been more bothered by re-experience symptoms than those who reported car-accidents (p = 0.006).
Age and the number of children were found to be unrelated to PTSD symptom severity scores, but were significant negatively related to the GHQ-12 total score with lower scores correlating with higher age and number of children (r = - 0.14; p < 0.001 and r = - 0.13; p
Mehdi Ghazinour1, Naser Mofidi1, Nader Esmail-Nasab2, Jörg Richter3
1 Department of Clinical Sciences, Division of Psychiatry, Umeå University, S-90187 Umeå, Sweden
2 Kurdistan University of Medical Sciences, Sanandaj, Iran
3 Centre of Child and Adolescent Mental Health, Regions South and East, P.O.Box 23 Tåsen, N-0801 Oslo, Norway
Corresponding author:
Prof. Dr. Jörg Richter
Centre for Child and Adolescent Mental Health, Eastern and Southern Norway,
P.O.Box 23 Tåsen
N-0801 Oslo
Norway
Running head: PTSD in Kurdish Iranians
Key words: PTSD, general mental health, life events, Kurdish Iranian, general population
Abstract
Posttraumatic stress disorder (PTSD) represents …show more content…
a major public health problem world-wide. In order to derive preliminary prevalence figures of PTSD and general mental health in Kurdish Iranians the applicability of the PTSD Checklist (PCL) was investigated combined with the 12- item version of the General Health Questionnaire (GHQ-12) and a slightly modified version of the Life event Checklist (LEC). 1000 randomly selected individuals were investigated. About 40.1 % of the subjects reported that they have been affected by one of the possible traumatic events. A prevalence of PTSD based on the PCL of 10.9 % was found which was higher than that reported in other countries, but lower than expected. With decreasing emotional involvement in the event, the probability of suffering from a PTSD decreased. Women suffered significantly more often from PTSD then men especially from re-experiencing and arousal symptoms. Particularly women seem to be in need for culturally adapted intervention procedures related to PTSD. The results support the construct and differential validity of the Persian version of the PCL.
Introduction
Post Traumatic Stress Disorder (PTSD), one type of anxiety disorder, represents a major public health problem world-wide [1]. The initial focus on trauma-related consequences historically changed from an exclusive somatic perspective to a much more comprehensive understanding that incorporated a major psychological perspective. Investigations of PTSD as a public health problem were mainly initiated in the U.S.A. and led to the inclusion of PTSD as a distinct psychiatric entity in the DSM-III. By accepting PTSD as a psychiatric disorder, its nature and origin became subject of comprehensive discussions. It is argued that PTSD represents a timeless condition, which existed before it was codified in modern diagnostic systems, but was described in different terms such as ‘railway spine’ and ‘shellshock’. Others suggest that PTSD is a novel presentation that is caused by modern interaction between trauma and culture [2].
PTSD represents a development occurring in individuals after they have experienced a traumatic event. A 'traumatic' event is characterized by its capacity to provoke fear, helplessness, or horror in response to threat of injury or death. Symptoms include distressing 'flash backs' (re-experiencing the event), avoidance of situations reminding the individual of the event, increased arousal, and psychological numbing.
The fist epidemiologic report from the U.S.A. conducted by Helzer and colleagues [3] estimated a prevalence of PTSD of about 1 % in the general population and identified a higher rate of PTSD in women than in men. However one criticism of this study was a probability of underestimation of PTSD [4,5]. Another prevalence study of high validity was conducted in the general population found an overall prevalence rate of PTSD of 7.8 % with a rate of 10.4 % among women, twice as high as that found among men (5.0 %) [6]. Estimates of trauma exposure showed that about 60% of men have been the victim of one or more traumatic events, while 51% of women were exposed to such events [4].
Epidemiological investigations of traumatic life events and PTSD provide important guidance for researchers and clinicians in determining who may possibly develop a PTSD following certain traumatic life events.
Against this background, epidemiological research on exposure to traumatic events and PTSD in different areas of life increased dramatically with a major focus on PTSD in soldiers; for example Persian Gulf War veterans [7] or soldiers on active duty in Somalia [4,8]. In addition, PTSD caused by catastrophes and disasters like bush fires or air crashes [9-11]; following domestic violence, rape or sexual abuse [12-14]; or after ware related traumatic life events, torture and terrorist attacks [15-19] has been investigated in many …show more content…
countries.
Since PTSD represents a global problem, and epidemiological data provides important knowledge for health care planning, our research group conducted an investigation of general mental health and the prevalence of posttraumatic stress disorder among Kurds living in Iran. It represents the first study of mental health problems in this particular population. The research was conducted in accordance with World Health Organization’s programs [20] in collaboration with the Medical University School of Kurdistan in the West of Iran.
About Kurdistan
Kurdistan is a traditional ethnographic region and covers the regions of Northern and Northeastern Mesopotamia. It is a mountainous region politically shared between several nations: Turkey, Iran, Iraq and Syria with overall estimated 27-28 million people living on an area of 190,000 km² (74,000 square miles). One part of Kurdistan (28,817 km²) is located in the West of Iran, with borders to Iraq in the West and to West Azerbaijan in the North. Sanandaj, the capital of Iranian Krdistan, had an estimated population of 358,084 inhabitants in 2006. The population of Sanandaj is mainly Kurdish with Armenian and Jewish minorities. The Kurds have been fighting for democracy and autonomy since centuries which represents one of the reasons for Kurdistan’s engagement in civil war. During the Iran-Iraq war (1980 -1988) Kurdistan was attacked by Iraq and heavily bombed [21]. During that time, trauma and loss were daily life experiences for many individuals [22]. Today the Kurdish population encounters many stresses including low income, employment difficulties, and ethnic repression.
The aims of the present study were; (a) To investigate the applicability of the PTSD Checklist (PCL) [23] to a population from a Middle-Eastern country and to derive preliminary prevalence figures of PTSD in Iranian Kurdistan; (b) To analyze psychometric properties of the Persian version of the PCL.
Methods
The presented findings represent part of a more comprehensive investigation relating to attitudes towards suicide, general mental health, quality of life and posttraumatic stress disorders among Kurdish people in Iran [24,25]. The research plan was approved by the ethics committee at the University of Umeå, Sweden. Before the project started, it was presented to the deputy of research of the Kurdistan University of Medical Sciences for evaluation and approval.
Sample
The sample under study consisted of 1000 Kurds randomly selected from 1000 of 68000 households in Sannandaj city.
The data collection was performed during April and May 2006 in collaboration with University of Medical Science in Kurdistan. Households were selected by a cluster random sampling process based on the 24 health care districts of Sannandaj city and dependent on their size. 100 clusters were derived in which ten households each were approached. Each data collector continued to approach households until they obtained ten respondents per cluster. The person that opened the door was invited to participate in the study if he or she was older than eighteen and none of the approached individuals refused to participate. Notwithstanding the random sampling procedure, individuals with higher education are substantially overrepresented in our sample and among women in
particular.
Every health care centre in Sannandaj city participated in this investigation, providing two health care professionals who were experienced in data collection. Data collectors received an one day training by the first author concerning the questionnaires, data collection procedures and how to deal with illiterate people. If the participant was illiterate, the data collector read the items aloud for the person.
Insert Table 1 about here
Measures
The PTSD Checklist, civil version, was chosen to dimensionally assess strain caused by traumata and to facilitate a related PTSD diagnosis [23]. It consists of seventeen items which correspond to the DSM-IV symptoms of PTSD. Where an individual reported a major life incident, they were instructed to indicate how much they had been bothered by each symptom in the past month using a five point scale from 1 = not at all to 5 = extremely. Scores related to the three PTSD symptom dimensions (re-experiencing, numbing and avoidance, and increased arousal) and a total score indicating the PTSD symptom severity were calculated by summing scores across the respective items. The Cronbach’s alpha coefficients (between 0.92 and 0.97) and the item-intercorrelation (between 0.62 and 0.87) related to the 3 symptom-cluster scales and the total score are reported to be highly satisfying [23]. In our study, we found the psychometric properties to be somewhat lower, but still acceptable, taking the low number of items into account (Re-Experiencing scale – cluster B symptoms: alpha = 0.79; corrected item-intercorrelation: 0.47 – 0.63; Numbing and Avoidance scale – cluster C symptoms: alpha = 0.70; corrected item-intercorrelation: 0.29 – 0.47; Increased arousal scale - cluster D symptoms: alpha = 0.77; corrected item-intercorrelation: 0.46 – 0.61; Total score: alpha = 0.88; corrected item-intercorrelation: 0.33 – 0.67).
A reduced and adapted version of the Life Event Checklist (LEC [26]) was used to indicate the occurrence of traumatic events experienced by the individuals. Fourteen possibly traumatic events were included (car accident; threat; sudden death; injury; maltreatment; disease; drowning accident; burning (scalding) oneself/suicide; fall; explosion/fire accident; shock; bombardment; and prison). The participants were instructed to provide information about whether they had experienced the event themselves, were a witness to the event, had simply heard about it, were not sure about the source of information, or had no experience of such a type of event. Due to very few reports of traumatic events (1 – 5) in six categories we decided to add these experiences to analogous categories.
General mental health was assessed by means of the 12-item General Health Questionnaire (GHQ-12 - [27]). This screening instrument was developed in order to detect psychiatric disorders in community settings and non-psychiatric clinical settings. It has been used in many investigations in various cultural regions including Iran [28] and its psychometric properties have been extensively investigated. Our findings are based on the original scoring used by Goldberg with response categories score ‘not at all’ and ‘no more than usual’ as 0 and ‘rather more than usual’ and ‘much more than usual’ as 1 providing a possible range from 0 to 12. By means of an exploratory factor analysis we derived two factors explaining 40 per cent of the variance. In accordance with Goldberg, Oldehinkel and Ormel’s [29] p. 921) recommendation “…, if the mean is below 1.85 then the threshold of ½, from 1.85 to 2.7 a threshold of 2/3, and above 2.7 a threshold of ¾ seems to work best for the GHQ-12.”, we decided to use a score of 5 (individuals with a total GHQ-12 score > 5 are considered as cases), the ¾ score in our sample, as the cut-off point for the differentiation between individuals with and without psychiatric morbidity.
Following the investigation of Montazeri and co-workers in Iran [28], two additional items related to overall quality of life were added to the GHQ (‘How do you evaluate your quality of life?’ and ‘How do you evaluate your general health?’ with possible answers: Very good; good; moderate; bad; very bad).
Statistics
Means and standard deviations and percentages are provided. T-tests for independent samples, chi-square-tests and one-way analyses of variance with Bonferroni controlled post-hoc comparisons were used for searching for differences between subgroups under study dependent on the scale-level in univariate analyses. Pearson correlation coefficients, partial correlations and linear multiple regressions were calculated for the identification of relationships between variables.
Results
Four hundred and eight of the 1000 individuals approached reported that they had once experienced, witnessed, been told about, or were not sure where they had got the information from, but had been affected by one of the possible traumatic events set out in the reduced life events checklist (Table 2). Most of them had witnessed these events. There was a significant relationship between the type of event and the relationship to the event with injuries, diseases and maltreatments most often being self-experienced; sudden deaths and drowning most often reported as witnessed; drowning, burns/scalds and suicides were most often reported as being heard about from somebody (χ2 = 48.37; p = 0.001).
Table 2 about here
According to the PCL, 10.9 % of the individuals in the investigated sample have to be classified as suffering from a PTSD. With decreasing emotional involvement in the event, the probability of suffering from a PTSD decreased (PTSD when it was ‘one’s own experience’: 25 %, when it was ‘witnessed’: 22 %, when ‘told about by somebody’: 14 %, when they were ‘not sure of the source’: 7 % - Fisher’s exact test = 127.01; p < 0.001). There was no substantial relationship between PTSD and particular types of events (χ2 = 4.61; p = 0.708). Women suffered significantly more often from PTSD than men (12.7 % vs. 8.8 % - χ2 = 4 .02; p = 0.045); whereas there was no difference in the number of men and women who had to be regarded as case based on the applied GHQ-12 cut-off score (χ2 = 0.17; p = 0.680).
Taking the GHQ-12 total score as reference, the PCL diagnosis had a sensitivity of 16 % and a specificity of 9 % in our sample (Table 3).
Table 3 about here
Based on dimensional analyses, we could not find any gender related differences on either the GHQ-12 total score or on the factors. However, women reported more severe re-experiencing (F = 9.87; p = 0.002), more arousal symptoms (F = 5.71; p = 0.017), and a higher PCL total score than (F = 6.58; p = 0.011) (see Table 3). The employment situation and the level of education were found unrelated to the PCL scores, whereas an one-way ANOVA procedure yielded a significant result for the GHQ-12 total score (F = 3.18; p = 0.002 and 5.43; p < 0.001 respectively) with the first relationship not based on particular differences between groups indicated by significant differences in Bonferroni controlled post-hoc comparisons. Whereas the marital status was found to be related to GHQ-12 total score (F = 14.82; p < 0.001), it was unrelated to PTSD scores.
Table 4 about here
There was no relationship between the type of event and the PCL scores. However, the severity of re-experience symptoms differed significantly depending upon the type of traumatic event (F = 2.85; p = 0.007), a finding mainly based on the fact that those who experienced sudden death reported that they had been more bothered by re-experience symptoms than those who reported car-accidents (p = 0.006).
Age and the number of children were found to be unrelated to PTSD symptom severity scores, but were significant negatively related to the GHQ-12 total score with lower scores correlating with higher age and number of children (r = - 0.14; p < 0.001 and r = - 0.13; p