Journal International de Victimologie

n December 26^th 2003 a major earthquake (6.6 on the Richter scale) devastated Bam in south Iran. This city is famous for its 2500 year-old historical site “Argue Bam”. In seconds the whole city and its historical monuments were destroyed; more than 40,000 of the 100,000 residents were dead and 30,000 were injured. In the days following the earthquake, the number of psychosomatic complaints increased. Surviving patients were consulted for what looked like functional disorders (presenting as headache, sore stomach, muscle aches, etc.). The psychosomatic complaints likely represent a manifestation of a socially more "acceptable" moral pain in this traumatized population with often little access to psychiatric care. The study of Escobar, Canino, Rubio-Stipec, and Bravo (1992) had previously noted an increase in psychosomatic complaints (gastrointestinal and pseudo-neurological) following a disaster in Nicaragua. Several studies showed that psychosomatic disorders and PTSD are significantly related and that psychosomatic symptoms are more frequent among subjects suffering from PTSD (Jakupcak, Osborne, Michael, Cook, Albrizio & Mcfall, 2006, Tagay, Herpertz, Langkafel & Senf, 2004, Zatzick, Russo& Katon, 2003 and  Perkonigg, Pfister, Stein, Höfler, Lieb, Maercker & Wittchen, 2005).

Sack, Lahmann, Jaeger and Henningsen (2007) found that psychosomatic symptoms were notably more prevalent in traumatized patients when compared with non traumatized patients. Because little studied, we have therefore chosen as the main goal description and analysis of these psychosomatic complaints, particularly in relation to stress related consequences of the trauma. 

According to psychiatric classifications, for example Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American Psychiatric Association ,1994), the psychopathologic consequences of traumatic events can be found in acute stress disorder (ESA), post traumatic stress disorder (PTSD) and adjustment difficulties. All categories of psychiatric disorders (anxiety, depression, psychosis, addiction etc) can be manifested post traumatically, with or without a direct link to the events. It is also important to explore categories of somatoform disorders and dissociative disorders among the victims of  traumatic event. 

The most frequent and typical chronic disorder following a traumatic event is PTSD. This syndrome can occur when the subject has been exposed to a traumatic event or has witnessed an event during which their life or health has been threatened (Yehuda, 2002). Thus, subjects who have experienced intense fear, helplessness or horror during an event may suffer persistent PTSD after a latency period, which can be a few days, weeks, months or years. 

The epidemiology of PTSD is complex to establish because it depends on the rate of exposure of the population to various types of traumatic events. The epidemiological studies in general populations indicate that these rates vary very substantially according to country and environment. In USA, the rate of exposure to an individual traumatic event is between 60 and 90 % (Breslau, Kessler, Chilcoat Schultz, Davis, & Andreski, 1998; Kessler, Sonnega, Bromet, Hughes, & Nelson, 1995). In Europe, a study was carried out in the region of Munich with a representative sample of adolescents and young adults: the prevalence of exposure to traumatic events was 17%, and the incidence over 3-4 years was 20% (Stein, Hofler, Perkonigg, Lieb, Pfister, Maercker, & Wittchen, 2002).These studies suggest that about 2 to10% of the general population in these countries may suffer PTSD. However, there are large differences in the risk of PTSD according to the nature of the events experienced. For example, the risk is 1.4% when only hearing of a traumatic event and 65% for subjects who are raped (Yehuda, 2002). PTSD is about twice as prevalent in women as men in some populations and its duration of evolution is on average 3 to 5 years. Comorbid mental disorders include the depressive disorder (risk enhanced 6 fold) and alcoholism (risk enhanced 3 fold). Although there have been several studies on the psychiatric consequences of earthquake, few have been well planned or rigorously conducted.

The studies by Bodvarsdottir and Elkit (2004) in Iceland and Bland, O'leary, Farinaro, Jossa and Trevisan (1996) in Taiwan were based upon dimensional assessments of anxiety and depression but without specific diagnosis. Other studies have explored the prevalence of PTSD in populations exposed to earthquake, and report high rates: 24% in Iceland (Bodvarsdottir, & Elkit, 2004), 25% in Italy (Catapano, Malafronte, Lepre, & Cozzolino, 2001), 30% in China (Wang, Gao, Shinfuku, Zhang, Zhao, & Shen, 2000) and 63% in Turkey (Livanou, Basoglu, & Kalender, 2002). In a study in Greece, Roussos, Goenjian and Steinberg (2005) found a prevalence of only 4.5%, but the evaluations were done only three months after the earthquake and the subjects were only children some distance away from centre of the earthquake. Each of these studies used different methods to establish the diagnosis, and different inclusion criteria; comparison of the results is therefore difficult. 

As we said before, psychosomatic complaints are important among the patients suffering from PTSD (Escobar and al.1992). An epidemiological study by Davidson, Hughes, Blazer and George (1991) in the general population in the United States found 90-fold higher risk of presenting “somatic complaint” among subjects with than without PTSD criteria. Psychosomatic complaints have also been reported in earthquake victims by Papadatos, Nikou and Potamianos (1990) in Greece and Wang et al. (2000) in China. 

Only two psychiatric studies concerning the Bam earthquake have been published (Montazeri, Baradaran, Omidvari, Azin, Ebadi, Garmroudi, Harirchi, & Shariati, 2005; Hagh-Shenas, Goodarzi, Farajpoor & Zamyad, 2006). The study by Montazeri and al. (2005) included 916 survivors of the earthquake zone selected in at random: 41 had lost 3 to 5 members of their family in the earthquake. Fifty-eight percent of the subjects had high scores on the General Health Questionnaire (GHQ-12) questionnaire measuring general distress, anxiety and depression symptoms. In a previous Iranian study, 21% of the general population gave similar high scores (Noorbala, Bagheri Yazdi, Yasamy, & Mohammad, 2004). Montazeri and al. (2005) showed that the factors among earthquake victims significantly associated with a high score on GHQ-12 scale (and thus with severe psychological distress) were female sex, lower education, unemployment, and loss of family members. Hagh-Shenas and al. (2006) evaluate that 81% of the participants (n=145) were eligible for PTSD diagnosis according to DSM-IV criteria in this population. No significant correlation was found between demographic variables or history of psychiatric illness and measures of psychological distress.

It was important for us to continue to study this population for a several reasons. First, human and material losses were massive: nearly 20,000 homes destroyed creating 45,000 homeless. Second, the psychiatric health situation in this country is very poor, with only 735 psychiatrists and 7850 hospital beds for the whole population of Iran; it has been estimated that 10 or 12 million people in the country require psychiatric care. There are numerous health problems in the region, including addiction, and particularly dependence on opiates. The links between PTSD and addiction have been clearly established. Third, the risk of earthquake in the region of Bam, and throughout Iran is high, and disasters of similar magnitude are likely in the future (there have been further, less severe earthquakes in the region in the last two years). Therefore, it would be valuable to have a better understanding to help prepare for the psycho-traumatic consequences of such events.

OBJECTIVES

The objectives of this study were: to describe trauma-related anxiety, depression and psychosomatic symptoms during 14 months in a sample of subjects exposed to an earthquake and referred initially for psychosomatic complaints and to search for links between intensity of the psychosomatic complaints and various psychopathologic variables related to traumatic event.

METHODOLOGY

Population

The plan was to study a sample of patients consulting a mental health centre in Bam, either on their own initiative or who were initially identified by physicians in the primary health care centre. The study was divided into two phases each of nearly one month, 7 months (T1) and 14 months (T2) after the earthquake of December 2003. This allowed analysis of the lag time (the time from onset of trauma, i.e. earthquake). 

The inclusion criteria were: patients of both sexes more than 15 years old, attending their first psychological consultation, and which were present during the earthquake.

  The exclusion criterion was impossibility to participate in the study, linguistic difficulties, known psychotic syndromes, and mental retardation.

Evaluation

After obtaining the patient’s consent and verifying inclusion criteria, clinical interviews were carried out using various evaluative scales by a trained clinical psychologist (author MA). Because validated translations into the local language (Farsi) were not available for many scales, we used hetero-evaluative scales and scales that could be read to the subjects (translated by us for the study). Socio-demographic variables and personal consequences of earthquake (number of deaths of family members, and degree of closeness with the dead, physical injury, taking drugs before and after the earthquake) were collected. The following psychometric instruments were used:

1-For post traumatic symptoms:

-The PTSD-Interview (Watson, Juba, Manifold, Kucala, Anderson, 1991), is a structured interview which allows study of the diagnostic criteria of PTSD according to DSM-IV and provides a global score of severity and 3 subscores of intensity of the symptomatic dimensions of PTSD (reexperiencing, avoidance and increased arousal).

-The IES-R scale (Impact of Event Scale-revised), developed by Horowitz, is a dimensional instrument for measuring the severity of PTSD between 0 and 88 (Creamer, Bell, & Failla, 2003). 

2-For the somatic symptoms: 

- The Clinical Global Impression (CGI) of severity for somatic complaints (from 1 “not ill” to 7   “very ill” patients) was estimated by the evaluator, based on the interview and medical files of patient (Guy, 1976); a qualitative description of types of manifestations and of their localisations was recorded. 

3- For anxious and depressive symptoms: 

- Covi and Raskin scales, which provide scores (0 to 12) of general anxiety and depression respectively (Covi, Lipman, Mcnair, & Crezlinsky, 1979); 

- The depressive mood scale developed by Jouvent et al. (1987) which provides a score for anhedonia and a score of impulsivity and irritability related to depressive mood.

Statistical analysis

SPSS software (version 10) was used. Due to the samples sizes and the non-normal distribution of different variables, non parametric tests were used for comparisons of means (Mann-Whitney test) and for correlations (Spearman). To assess links between psychosomatic complaints and psychopathologic variables, taking socio-demographic markers into account, a linear regression analysis was used, with step by step introduction of independent variables found by the CGI scale to be associated.  The threshold of significance was set at alpha=0.05 for all tests, two-sided.

RESULTS

Description of the population

We planed to include about 100 subjects, on the bases of power analyses for statistical tests. They have been included in two phases: June 2004 (T1) and February 2005 (T2). All analyses were conducted on the total sample, but also on the two sub-groups (T1 and T2) in order to confirm the homogeneity of the sample and to detect possible changes of symptoms with time. All eligible subjects accepted to participate, and we included a total of 105 subjects (30 men and 75 women), 50 at T1 and 55 at T2, mean age 33.5 +/- 13.7 years (16 to 68 years). More than half of the subjects lived with a partner (60%), and 59.1% of subjects had continued in education after the age 16. Nearly one third (28.6%) of the patients were physically injured during the earthquake (23.3% of the men and 30.7% of the women; p=0.31). The injury was considered to result in impairment in about 5% of subjects. All the subjects had lost one or more family members, and 51.4% had lost at least one first degree relative. Twenty-one percent (n=23) of included subjects declared regular opium use, and in most cases for more than one year (n=21). Many more women than men (57.7% vs. 9.3%; p<0.001) were opium users. No significant difference appeared between subjects included in the first phase and those included in the second phase for these socio-demographic features, except for sex-ratio (16.0 % of males in the first phase, vs 40.0 % in the second phase; p=0.007).  

Symptoms

The mean scores of the psychopathologic scales are summarized in table I. We have computed, for PTSD scales, the Cronbach alpha coefficient to estimate the internal consistency of these measures in this sample. The results are, globally, in favour of satisfactory coefficients: IES-R 0.88, PTSD-I total score 0.84, PTSD-I-B 0.64, PTSD-I-C 0.55, and PTSD-I-D 0.79. Applying DSM VI criteria with the PTSD interview, 87 patients (82.9%) were diagnosed to have PTSD. Only six patients (5.7%) did not fulfil the criteria A of the PTSD definition, indicating that they did not experience horror or intense helplessness during the earthquake. The patients who presented PTSD criteria had higher scores than those who did not for all of the PTSD scale (PTSD Interview score total 79.9 vs. 53.3 p<0.001, IES- R score 61.4 vs. 43, 7 p<0.001), anxiety scale (Covi 7.9 vs.; p<0.001) and depression (Raskin 8.8 vs.6.1 p<0.001; EDH anhedonia 14.5 vs. 9.9 p<0.001; EDH impulsiveness 13.0 vs. 10.3 p<0.001)

Most of the measures on PTSD, of anxiety and of depression were significantly correlated with each other (see Table II). 

Age was significantly correlated to Covi (r=0.30; p=0.002), Raskin (r=0.26; p=0.008). The PTSD total score was higher for women than men (78.5 vs. 68.8; p=0.007), as was IES-R scale score (60.6 vs. 52.6; p= 0.01). There were no significant differences for scores on other scales between men and women.

The total score for the PTSD-Interview was higher for subjects who suffered physical injury during the earthquake (83.1 vs. 72.7; p=0.01), had little education (80.7 vs. 72.1; p =0.009), were without professional activity (79.1 vs. 71.5; p=0.01) and who had lost a first degree member of their family (80.6 vs. 71.1; p=0.01) than other subjects. Marital status and drug abuse did not correlate with PTSD-interview scale scores. The IES-R scale revealed the same psychopathologic correlations and socio-demographics as PTSD-interview scale scores. 

 One hundred one subjects (96.2%) suffered from physical complaints, mostly pains (headaches, abdominal or limbs pains or the extremities), but also from respiratory complaints, palpitation, vertigo and trembling. The average score on the psychosomatic complaints CGI scale was 4.8 ± 1.6 (range 1 to 7). There was no significant difference between men and women. However, the CGI scale score correlated significantly with age (r=0.47; p< 0.001). 

Comparison score between the two phases

The comparison of the psychopathologic scores between the two phases (7 and 14 months) is presented in table III. The intensity of PTSD scores was significantly lower at T2 than T1, but with limited differences (15% for the total score for the PTSD-Interview, and 18% for IES-R). The number of patients presenting PTSD criteria was 43 of 50 (86%) at T1, and 44 of 55 (80%) at T2 (the difference was not significant). 

Concerning psychosomatic complaints, the mean scores for the CGI scales were 5.59 +/- 1.44 at T1 and 4.45 +/- 1.48 at T2 (p=0.01). 

Correlations with severity of psychosomatic complaints

Scores obtained for severity of psychosomatic complaints were significantly correlated with all scores obtained for psychopathologic scales and also for age (see Table IV). The separate analyses of these figures in T1 subjects and in T2 subjects showed the same trends for all variables, except for PTSD-subscore C (avoidance) which is not correlated to the CGI score in T2 subjects (r=0.19; p=0.17).

Significant differences were found in mean CGI scores for five dichotomised variables:

-subjects who had suffered physical injury as a consequence of the earthquake (5.8 vs. 4.4; p<0.001);

-subjects who had not been educated past age 16: 5.5 vs. 4.3 (p<0.001);

-subjects who had lost a family member (5.5 vs. 4.2; p<0.001); 

-subjects who were living alone (5.2 vs. 4.2; p=0.003);

-subjects who had no professional activity (5.2 vs. 4.4; p=0.01).

The average score of CGI scale differed significantly between subjects with substance abuse and other subjects. A linear regression (with step by step introduction) was performed, in which all variables appearing to be significantly correlated with CGI scale score in the bivariate analyses were included (age, sub score B, C and D of PTSD-Interview, IES-R, Covi, Raskin, EDH-anhedonia and impulsiveness, existing physical injury, extent of education, professional activity, marital status, loss of a member of family and evaluation at T1 or T2). Only four variables remained significantly linked to CGI scale in this analysis:

-the D subscore (sympathetic hyperactivity) of the PTSD-Interview (p<0.001);

-age (p=0.002);

-the impulsiveness subscore of EHD (p=0.037);

-and losing a first degree of family member during the earthquake (p=0.047).

The absence of a significant link between CGI score and time of inclusion (T1 or T2) confirmed that the total sample can be considered as a whole.

DISCUSSION

In this study, 105 victims of an earthquake in Bam were evaluated 7 or 14 months after the earthquake. Our analyses suggested that:

-A large proportion of the patients included in the study suffered from a post traumatic pathology, particularly PTSD (82.9% according to an estimation of DSM-IV diagnostic by the PTSD-I) 

-The mean score for the IES-R scale (58.3) indicated a high intensity of posttraumatic symptoms (the threshold usually used to suspect a diagnosis of PTSD is 33).

- Scores on Raskin and Covi scales, although not allowing definite diagnosis, indicated highly prevalent anxious and / or depression states. 

-Although the evolution during the seven months separating the two phases of evaluation appeared to be favorable, it was relatively modest and the prevalence of disorders remained high more than one year after the earthquake. 

It is difficult to compare these results with those published in the literature concerning other victims of earthquakes for methodological reasons (choice of studied population, types of measures, etc.). However, the prevalence of PTSD (and anxiety and depression) we report here is higher than in any previous study (Bodvarsdottir & Elklit, 2004; Catapano et al., 2001; Jacobsen, Southwick, & Kosten, 2001; Ozen & Sir, 2004; Roussos et al., 2005; Wang et al., 2000). There are various possible reasons for this. First, we studied subjects who were exposed to psycho-traumatic consequences because of geographic proximity to the earthquake epicenter. It has been reported that in Turkey (Kilic & Ulusoy, 2003) there can be a linear relationship between intensity of PTSD symptoms and the distance from the earthquake epicenter. Also, our sample was subject to numerous  factors of extreme traumatic stress for a prolonged period, including in particular major destruction of habitat and consequently persistence of very difficult living conditions, and in many cases bereavement, separations (more than 50% of subjects in our study lost a member of their family), and injury.

The region of Bam remains at high risks of a new earthquake. Indeed, there have been tremors since 2003 and other recent earthquakes in other regions of Iran. The fear of a new disaster and continued publicity of the risk contribute to maintain a climate of tension and anxiety.

The inclusion methods have consequences on the observed prevalence of symptoms. Our sample was constituted from patients consulting a medical center for physicals complaints. The sample was therefore not representative of the general population. Note that our aim was to describe and to analyze a particular group a high risk to elucidate the pathology and to identify health needs.

Our results confirm that these subjects require health care and that physicians satisfactorily identify psychiatric needs. Our results confirm the link between psychosomatic complaints and anxiety-depression and post-traumatic symptoms suggested by previous studies (Davidson et al., 1991; Escobar et al., 1992; Papadatos et al., 1990; Wang et al., 2000). According to the literature (Breslau, Davis, Andreski, Peterson, & Schultz, 1997; Livanou et al., 2002) more women than men contribute to increase the level of morbidity. Our study also shows that scores for anxiety, depression and the symptoms of PTSD are higher for women than men. 

Our study also raises the issue of the factors associated with severity of somatic symptoms. The average CGI score was 4.5, which is classified as between “moderately ill” and “severely ill”. In the bivariate analysis, this score was significantly correlated with all measures of anxious morbidity, depression and post traumatic stress, and with most of medico-socials disorders examined, whether or not associated with the earthquake (including physical injury during the earthquake, little education, no professional activity and marital status). This could be the consequence of a “halo” effect non-specific indices of physical, psychological and social suffering being linked by these tests. However, multivariate analysis failed to confirm several of these associations with psychosomatic complaints severity, and only four independent variables remained significantly correlated. PTSD-Interview subscore D was most significantly associated, possibly because of the superposition of the somatic symptoms, present in the cluster D of PTSD (neuro-vegetative signs linked at a phenomenon of general activity). Overall, our analysis suggests that there was no specific link between psychosomatic complaints and other specific dimensions of PTSD. Age was the second independent factor linked with severity of the psychosomatic complaints; this may be explained by adaptation and physical resistance to stress factors becoming more problematic with age. The two other independent factors were probably related to the depressive dimension: the first was the “impulsiveness-irritability” dimension of depressive mood in the EHD scale, and the second was the death of a close family member during the earthquake. Although not correlated to the depression score, these factors may cause a psychological vulnerability linked to bereavement and on a more “productive” (impulsiveness, intrinsic agitation, etc.) than negative dimension; this may then be expressed as psychosomatic complaints. Our analysis is in line with the concept of “psychosomatic” medicine: physical affections reflecting a psychological suffering in the absence of psychiatric manifestation (Haynal, Pasini, & Archinard, 1978). Our findings are consistent with what we know about clinical depression masked by somatic and functional complaints. Although these manifestations are usually found in the elderly, the population we studied mainly consists of young adults (average age 33 years).

Our study suggests that the psychosomatic complaints reflect a psychopathologic dimension different from the state of post traumatic stress (and not directly dependent on physical injury), and based on a depressive fragility and /or bereavement. Distinguishing complaints in this population would make a large contribution to detecting masked physical troubles, and revealing depressive disorders added to post-traumatic troubles. Because of the paucity of data in the literature, it remains unknown whether this phenomenon is specific to populations exposed to this type of disaster or whether it could in part be linked to the socio-cultural factors specific to the Iranian population.

The limitations of this study are, above all, methodological and consequences of the conditions imposed. This study is not an epidemiological analysis in the general population, because we worked only with a subgroup of victims strongly exposed and demanding care. The studies by Noorbala et al. (2004) and Montazeri et al. (2005) are the only epidemiologic studies of the population of Bam, but they did not evaluate post traumatic disorder or psychosomatic complaints. The second methodological limitation arises from the evaluating instruments used. We chose the tools both because they are the most widely used for evaluation of the disorders involved (PTSD-Interview, IES-R, Covi, Raskin, etc.), and also for feasibility reasons. Unfortunately, none of these tools were available and validated in the local language (Persian). They had to be translated rapidly from French or English, and the translated versions were not validated before the study. The results of this study resemble data in the literature, suggestive of satisfactory validity, but rigorous experimental validation is lacking. 

There are also various points to be considered concerning the evaluations. Measurements, mostly based on hetero-evaluations, were performed by the same clinician, favoring homogeneity of data collection. However, this may also have contributed to “halo” biases (see above) in the correlations among different measures. Psychiatric diagnoses other than PTSD were not performed in a categorical manner because the use of a structural diagnostic instrument would have been problematic. The analyses of the addictive problems are probably insufficient because it was based solely on the responses to a question about using drugs. This problem, which is very important in many cases of post traumatic syndrome, is probably underestimated in our study, because there was an absence of clarity in many of the answers to the relevant question. 

CONCLUSION

This study provides insights into the severity and chronic nature of psychological disorders affecting a group of subjects consulting a mental health center in an earthquake zone. The results were in accordance with those obtained in similar studies conducted in other countries world wide. However, we present an original observation: the analysis of the factors associated with psychosomatic complaints which seem to have a depressive psychosomatic dimension which differs from pure PTSD symptoms. These observations could help improve detection and treatment plans for this type of disaster. Appropriate preparation is especially important for this highly exposed Iranian population, and may also be valuable for other populations subject to natural disasters of other kinds (Tsunami for example).

We intend to extend this study by re-evaluating the same population two years and four years after the earthquake, and by validating and evaluating the instruments translated into the local language.  

Acknowledgments

The authors thank Dr Afrassiab Z. for his helpful review of the manuscript, and Dr Kia T. (Nejat) and Welfare in Iran for their support in local clinical work.

REFERENCES

American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders, fourth edition. Washington DC.

Bland, S. H., O'leary, E. S., Farinaro, E., Jossa, F., & Trevisan, M. (1996). Long-term psychological effects of natural disasters. Psychosom Med, 58, 18-24.

Bodvarsdottir, I., & Elklit, A. (2004). Psychological reactions in Icelandic earthquake survivors. Scand J Psychol, 45, 3-13.

Breslau, N., Davis, G. c., Andreski, P., Peterson, E. L., & Schultz, L. R. (1997). Sex differences in posttraumatic stress disorder. Arch Gen Psychiatry, 54:1044-8.

Breslau, N., Kessler, R. C., Chilcoat, Schultz, L. R., Davis, G. C., & Andreski, P. (1998). Trauma and posttraumatic stress disorder in the community: the 1996 Detroit Area Survey of Trauma. Arch Gen Psychiatry, 55, 626-32.

Catapano, F., Malafronte, R., Lepre, F., & Cozzolino, P., et al. (2001). Psychological consequences of the 1998 landslide in Sarno, Italy: a community study. Acta Psychiatr Scand,104, 438-42.

Covi, L., Lipman, R., Mcnair, D. M., & Crezlinsky, T. (1979). Symptomatic volunteers in multicenter drug trials. Progr Neuropsychopharmacol, 3, 521.

Creamer, M., Bell, R., & Failla, S. (2003). Psychometric properties of the Impact of Event Scale – Revised. Behav Res Ther, 41, 1489-96.

Davidson, J., Hughes, D., Blazer, D., & George, L. (1991). Post-traumatic disorder in the community: an epidemiological study. Psychol Med, 21, 7133-721.

Escobar,  J., Canino, G., Rubio-Stipec, M., & Bravo, M. (1992). Somatic symptoms after a natural disaster a prospective study. Am J Psychiatry, 149, 7.

Guy, W. (1996). Clinical Global Impression. In: ECDEU Assessment Manual for Psychopharmacology. Rockville: revised National Institute of Mental Health.

Hagh-Shenas, H., Goodarzi, M. A., Farajpoor, M., & Zamyad, A. (2006). Post-traumatic stress disorder among survivors of Bam earthquake 40 days afther the event. East Mediterr Health J. 12 Suppl 2:S118-25.

Jacobsen, L. K., Southwick, S. M., & Kosten, T. R. (2001). Substance use disorders in patients with posttraumatic stress disorder: a review of the literature. Am J Psychiatry, 158,1184-90. 

Jakupcak, M., Osborne, T., Michael, S., Cook, J., Albrizio, P., & Mcfall, M. (2006). Anxiety sensitivity and depression: mechanisms for understanding somatic complaints in veterans with posttraumatic stress disorder. J Trauma Stress, Aug; 19(4),471-9.

Jouvent, R., Hardy, P., Bouvard, M., Braconnier, A., Roumengous, V., Grasset, F., & Widlocher, D. (1987) Hétéro-évaluation de l’humeur depressive. Construction d’une échelle multidimensionnelle. Encéphale, 13, 233-7.

Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., & Nelson, C. B. (1995) Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry, 52,1048-60.

Kilic, C., & Ulusoy, M. (2003). Psychological effects of the November 1999 earthquake in Turkey: an epidemiological study. Acta Psychiatr Scand, 108, 232-8.

Livanou, M., Basoglu, M., & Kalender, D. (2002). Traumatic stress reponses in treatment-seeking earthquake survivors in Turkey. J Nerv Ment Dis, 190, 816-823.

Montazeri, A., Baradaran, H., Omidvari, S., Azin, S. A., Ebadi, M., Garmroudi, G., Harirchi, A. M., & Shariati, M. (2005). Psychological distress among Bam earthquake survivors in Iran: population-based study. BMC Public Health, 5, 4.

Noorbala, A., Bagheri, Yazdi, S. A., Yasamy, M. T., & Mohammad, K. (2004). Mental health survey of the adult population in Iran. Br J Psychiatry, 184, 70-73.

Ozen, S., & Sir, A. (2004). Frequency of PTSD in a group of search and rescue workers two months after 2003 Bingol (Turkey) earthquake. J Nerv Ment Dis, 192, 573-5.

Papadatos, Y., Nikou K., & Potamianos, G. (1990). Evaluation of psychiatric morbidity following an earthquake. International Journal of Social Psychiatry, 36, 131-136.16.

Perkonigg, A., Pfister, H., Stein, M. B., Höfler, M., Lieb, R., Maercker, A., & Wittchen, H. U. (2005). Longitudinal course of posttraumatic stress disorder and posttraumatic stress disorder symptoms in a community sample of adolescents and young adults. Am J Psychiatry, 162, 1320-1327.

Raskin, A., Schulterbrant, J., Reatig, N., & Mckeon, J. J. (1969). Replication of factors of psychopathology in interview, ward behaviour and self-report ratings of hospitalized depressives. J Nerv Ment Dis,148 , 87-98.

Roussos, A., Goenjian, A. K., & Steinberg, A. M. (2005). Posttraumatic stress and depressive reactions among children and adolescents after the 1999 earthquake in Ano Liosia, Greece. Am J Psychiatry, 162, 530-7.

Sack, M., Lahmann, C., Jaeger, B., & Henningsen, P. (2007) Trauma prevalence and somatoform symptoms: related to traumatic experiences? J Nerv Ment Dis.  Nov; 195 (11), 928-33.

Tagay, S., Herpertz, S., Langkafel, M., & Senf, W. (2004). Trauma, post-traumatic stress disorder and somatisation, Psychother Psychosom Med Psychol, May; 54(5), 198-205.

Wang, X., Gao, L., Shinfuku, N., Zhang, H., Zhao, C., & Shen, Y. (2000). Longitudinal study of earthquake-related PTSD in a randomly selected community sample in north China. Am J Psychiatry, 157, 1260-6.

Watson, C. G., Juba, M. P., Manifold, V., Kucala, T., & Anderson, P. E. (1991). The PTSD Interview: rationale, description, reliability, and concurrent validity of a DSM-III-based technique. J Clin Psychol, 47, 179-188.

Yehuda, R. (2002). Post-traumatic stress disorder. N Engl J Med., 346,108-14.

Zatzick, D. F., Russo, J., & Katon, W. (2003). Somatic, posttraumatic stress and depressive symptoms among injured patients treated in trauma surgery. Psychosomatics, December, 44, 479-484.