Material and Method: A total of 110 panic disorder patients diagnosed with the DSM-5 Structured Clinical Interview-Clinical Version who met the inclusion criteria were included in this study. After the informed consent forms were obtained; Sociodemographic and Clinical Data Form, Panic and Agoraphobia Scale (PAS), Hamilton Depression Rating Scale (HAM-D), Hamilton Anxiety Rating Scale (HAM-A), Eysenck Personality Questionnaire Revised-Short Form (EPQR-S), Separation Anxiety Symptom Inventory, Anxiety Sensitivity Index-3 (ASI-3), and Childhood Trauma Questionnaire (CTQ) were applied to all participants. Univariate analysis, cluster analysis and discriminant analysis were performed with the collected data.

Results: As a result of multivariate analysis, we found two clusters in panic disorder. According to clustering characteristics, we identified 2 subtypes as traumatic panic disorder and nontraumatic panic disorder. Depression, anxiety levels, panic and agoraphobia scale, separation anxiety, anxiety sensitivity, neuroticism and psychoticism personality dimensions, childhood traumas, especially emotional abuse and sexual abuse, were found to be higher in the traumatic subtype. Eysenck extraversion personality dimension was higher in In the nontraumatic panic disorder subtype.

Conclusion: Characteristics in panic disorder patients are related to each other, and there are two subtypes: panic disorder that occurs with traumatic effect and Nontraumatic Panic Disorder.

In recent years, the number of studies aiming to define subtypes of panic disorder has increased. In different studies, Starcevic et al. and Cox et al.^2,3 made classifications based on symptoms. Massana et al.⁴

examined the biological and psychological differences of panic disorder patients. Fidaner et al., Bovasso and Eaton^5,6 made subtyping based on physical symptoms. There is a need for phenomenological reevaluation for future classifications of panic disorders. Cluster analysis is one of the multivariate statistical analysis methods that helps to separate units and variables into similar subsets (group, class) for groups that are not clearly known. The main goal of cluster analysis is to group units based on their characteristic features⁷. Although there are panic disorder subtyping studies conducted in terms of symptoms, cluster analysis has not been yet performed. We planned to carry out this study with an analysis based on the characteristics of the symptoms. Determination of subtypes by cluster analysis will also contribute to studies on the prognosi and treatment of the disease. The objectives of this study were to determine the distribution of symptoms in panic disorder, to define the clinical features of the subtypes of the disease and the factors that determine this subtyping, and to define possible symptom clusters that can define the subtypes of the disease by cluster analysis method.

All tests and scales were administered to the participants in this study by the same psychiatrist. The tests were placed in the test battery and the participants were asked to answer and fill them in a single session. The study type was determined as a crosssectional and explanatory hierarchical cluster analysis. In the hierarchical clustering method, first it is assumed that there are n clusters first. As a first step, the two closest clusters are combined. In the second step, the renewed distance matrix is found by reducing the number of clusters by one. These two steps are repeated n-1 times. There are six different techniques based on this algorithm, including single correlation, full correlation, group mean, center, median and minimum variance techniques.

In the study, the number of clusters was determined as 2 using hierarchical methods. In our study, whether comorbidities, panic agoraphobia levels, depression degree, anxiety assessment, personality dimensions and characteristics, separation anxiety, anxiety sensitivity and childhood psychological traumas indicate a clustering was investigated in panic disorder patients. When we look at the clustering characteristics, one of them was determined as a panic disorder subtype developed with traumatic effect due to the high trauma characteristics, and the other was determined as a nontraumatic panic disorder subtype.

Inclusion criteria were being in the age range of 18-65, being diagnosed with panic disorder as a result of SCID 5 (DSM-5 Structured Clinical Interview), having the physical and mental capacity to fulfill the research instructions, accepting to participate in the study and to give written consent, in its current state, experiencing panic disorder symptoms actively and having had at least one panic attack in the last month. Exclusion criteria in the study were having delirium, dementia, or similar diagnosis and diagnosis groups in which the ability to evaluate is impaired, being under the age of 18 or over the age of 65 (since cognitive problems are more common and difficult to distinguish over the age of 65, and the necessity to obtain the consent of the guardian and the difficulty in establishing a clear diagnosis under the age of 18), having a chronic disease such as uremic encephalopathy, presence of alcohol and substance-related encephalopathy, and being under the influence of alcohol and substance, as well as being unable to complete the clinical interview and self-assessment scales in a healthy way.

After the informed consent forms were obtained; Structured Clinical Interview for DSM-5 (SCID-5CV), Sociodemographic and Clinical Data Form, Panic and Agoraphobia Scale (PAS), Hamilton Depression Rating Scale (HAM-D), Hamilton Anxiety Rating Scale (HAM-A), Eysenck Personality Questionnaire Revised-Short Form (EPQR-S), Anxiety Sensitivity Index-3 (ASI-3), Separation Anxiety Symptom Inventory (SASI), and Childhood Trauma Questionnaire (CTQ) were applied to all participants.

Structured Clinical Interview for DSM-5 (SCID-5CV): Structured Clinical Interview for DSM-5 Disorders was developed by First et al. The Turkish validity and reliability of the Structured Clinical Interview for DSM-5 Disorders was demonstrated in the study of Elbir M. et al. Psychiatric diagnoses were made by applying SCID-5 CV to our patients⁸.

Panic and Agoraphobia Scale (PAS): It is used to rank severity in subsections by considering panic attacks, phobic avoidance, anticipatory anxiety, impairment in social relationships, and health concerns in patients. In a study conducted in Turkey, the highest specificity (92.94%) and sensitivity (92.00%) values are obtained with a cut-off value between 11-12 points. Turkish validity and reliability study was performed by Tural Ü et al.⁹.

Hamilton Depression Rating Scale (HAM-D): It measures the level and severity of depression in the patient: 0-7 points no depression, 8-15 mild depression, 16-28 moderate depression, ≥ 29 severe depression. In the study conducted in Turkey, the cut-off score was calculated as 7. The Turkish validity and reliability study was performed by Akdemir A et al.¹⁰.

Hamilton Anxiety Rating Scale (HAM-A): It is used to determine the anxiety level and symptom distribution to measure the change in anxiety severity of patients. It consists of 14 questions questioning both mental and physical symptoms. Turkish validity and reliability study was conducted by Yazici MK et al.¹¹.

Eysenck Personality Questionnaire Revised-Short Form (EPQR-S): Within the framework of Eysenck personality theory, it provides the evaluation and measurement of dimensions of personality such as neuroticism-stability, extroversion-introversion and psychoticism. Turkish validity and reliability study was made by Karancı AN et al.¹².

Anxiety Sensitivity Index-3 (ASI-3): It enables us to measure anxiety sensitivity with its physical, social and cognitive dimensions. It is a self-reporting scale. A high score indicates increased anxiety sensitivity. The Turkish validity and reliability study was performed by Mantar A et al.¹³.

Separation Anxiety Symptom Inventory (SASI): This scale is an inventory that questions childhood separation anxiety symptoms in adults. The people who filled out the scale are asked to try to remember the fears they might have experienced when they were younger than 18 years old and fill in that period. The Turkish validity and reliability study was performed by Diriöz M et al.¹⁴.

Childhood Trauma Questionnaire (CTQ): This scale is used to evaluate the experiences of abuse and neglect before the age of 20, retrospectively and quantitatively. It is a self-assessment scale, and patients are asked questions about the events they experienced before the age of 20, and to answer as truthfully as possible because it concerns their private lives. The Turkish validity and reliability study was performed by Şar V et al.¹⁵.

Statistical Analysis
The analysis of the data in this study was made with the SPSS 15 program and studied with a confidence interval of 95%. Study data were evaluated by basic frequency analysis, followed by univariate analysis (Chi-square and Student's t test, ANOVA) and multivariate analysis by cluster analysis and discriminant analysis. Analysis assumptions were examined. Since the total number of subjects in the study was less than 300, the hierarchical (staged) clustering method was used as the clustering method. The "mean correlation between groups" method was used as the similarity method, and two clusters were formed in the resulting dendrogram.

When the clustering analysis was repeated with the k-means clustering technique, the clusters in which the cases were placed with the hierarchical method (2 clusters as nontraumatic panic disorder and the traumatic panic disorder) were reconstructed.

The income level of 68.2% of the participants was medium, 53.6% were married, and 33.6% were housewives. Of all patients included in the study, 30.0% experienced permanent loss of parents in their childhood (Table 1).

Click Here to Zoom

Table 1: Distribution of demographic characteristics.

When the clinical characteristics of the patients in our study were examined; 47.3% of the patients had other anxiety disorders, 45.5% were using antidepressants actively and 51.8% were using antidepressants previously. A history of psychiatric treatment was found in 1st and 2nd degree relatives in 67.3% of the sample. The most common comorbidities in the study participants was major depressive disorder (31.8%), followed by agoraphobia (50.0%), and generalized anxiety disorder (20.0%).

As a result of the cluster analysis, 61 of the subjects were in the first cluster and 49 were in the second cluster. The 1st cluster was named as “the cluster formed by the traumatic effect” and the 2nd cluster as the “nontraumatic cluster”.

Panic agoraphobia scale, Hamilton depression rating scale, Hamilton anxiety rating scale, Eysenck Personality Questionnaire neuroticism and psychoticism subscale total scores, separation anxiety symptom questionnaire total scores, anxiety sensitivity index scores, Childhood Trauma Questionnaire emotional abuse, physical abuse, physical neglect, emotional neglect and sexual abuse subscale scores were found to be higher in the first cluster. Table 2 shows the mean values of the scale variables in the clusters.

Click Here to Zoom

Table 2: Distribution of mean statistics of scale variables by cluster.

In the ANOVA analysis performed to determine whether the variables differ in terms of clusters, the scales with significant differences between the groups are shown in table 3.

Click Here to Zoom

Table 3: Significance of variables for clusters*.

Accordingly; Panic and Agoraphobia Scale, Hamilton Depression Rating Scale, Hamilton Anxiety Rating Scale, Eysenck Personality Questionnaire Neuroticism and Extraversion subscales, Anxiety Sensitivity Index-3, Separation Anxiety Symptom Inventory, Childhood Trauma Questionnaire Emotional Abuse and Sexual Abuse subscales were found to be significant.

Table 4 shows that the mean scores of the scales included in the discriminant analysis were successful in separating the traumatic and non-traumatic groups.

Click Here to Zoom

Table 4: The success of the mean scores of the scales included in the discriminant analysis in separating traumatic and non-traumatic groups*.

(Canonical correlation 0.8192=0.67 and Eigen value >0.4). Table 5 shows that mean group values are sufficient to provide discrimination (Wilks’ Lambda: 0.33 and p <0.001).

Click Here to Zoom

Table 5: Adequacy of average group values to ensure discrimination*.

Table 6 shows which scale means are important in distinguishing between traumatic and nontraumatic groups.

Click Here to Zoom

Table 6: The importance of scale averages in differentiating traumatic and non-traumatic situations.

Accordingly; the Hamilton Depression Rating Scale and the Hamilton Anxiety Rating Scale were the most important scales and Panic - Agoraphobia Scale and Neuroticism subscale of the Eysenck Personality Questionnaire were the secondary scales in distinguishing between the groups.

Table 7 shows the classification results.

Click Here to Zoom

Table 7: The classification results.

Accordingly, 96.4% of the originally grouped cases and 95.5% of the cases grouped by cross-validation were correctly classified.

It has been demonstrated that anxiety disorders, including panic disorder, are associated with a history of childhood sexual and physical trauma^16-18. While a history of trauma is not a prerequisite for the development of panic disorder, a history of childhood trauma has been observed at a higher rate in clinical cases of panic disorder^17,19.

A study conducted with 200 patients diagnosed with panic disorder revealed that individuals who experienced childhood abuse exhibited more severe panic disorder and agoraphobia, a higher prevalence of comorbid depression, and more adverse psychosocial experiences²¹. In a study comprising 539 patients with panic disorder, it was reported that 54.5% of the patients had a history of childhood trauma (22). In our study, this rate was found to be 55.5%, exhibiting a course of development comparable to that observed in the existing literature. In a study conducted by Goodwin et al., it was demonstrated that the lifetime prevalence of numerous psychiatric disorders, including panic disorder and generalised anxiety disorder, increased markedly in individuals who had experienced abuse during childhood¹⁷. The results of our study indicated that the subscales measuring sexual abuse and physical abuse were significantly higher in the traumatic subtype. A recent meta-analysis, which employed a similar methodology to our study, found that early physical abuse and sexual abuse experiences were risk factors for panic disorder. These findings suggest that a reduction in childhood traumas could potentially prevent a significant number of psychiatric disorders, including panic disorder.

In the course of our study, the majority of participants were women, and this was the case in both subtypes. However, the proportion of women was higher in the traumarelated panic disorder subtype (75.4%) than in the non-traumatic subtype (65.3%). A review of the literature reveals that women represent the majority of patients diagnosed with panic disorder. Similarly, as observed in our study, the majority of women with panic disorder had experienced childhood trauma^20,22.

In our study, parental loss in childhood was observed with greater frequency in the trauma-related subtype (32.8%) than in the other subtype (26.5%). Prior research has demonstrated a correlation between parental loss or separation during childhood and the subsequent development of panic disorder in adulthood^23,24. It can be posited that psychiatric issues experienced by children who have lost their parents in childhood can be averted by enhancing social support.

Furthermore, patients with panic disorder exhibit a higher prevalence of additional physical disorders, including metabolic syndrome and cardiovascular pathologies, compared to the general population^25,26. Approximately half of the patients who participated in the study exhibited additional physical pathologies, particularly cardiovascular diseases.

The prevalence of depression was found to be significantly higher (51.8%) in the traumatic panic disorder subgroup. In a study investigating childhood traumas in panic disorder patients, a significantly higher rate of depression was observed in those with a history of trauma²¹. In another study, the prevalence of emotional neglect, psychological and physical abuse in patients with anxiety and depression was found to be twofold higher than in control groups and approximately threefold higher in comorbid conditions²⁷. The presence of comorbid psychiatric disorders has been shown to have a detrimental impact on treatment response and compliance with treatment.

A study of the literature revealed that patients with panic disorder accompanied by major depressive disorder exhibited significantly higher rates of severe agoraphobia, nocturnal panic attacks, and a greater number of panic attacks in the previous month, in addition to a higher prevalence of social phobia²⁸. This indicates that comorbidities and treatment resistance may be more prevalent in individuals with a traumatic subtype, while treatment response may be less robust. It is widely accepted that early childhood losses constitute a risk factor for both panic disorder and depressive disorders. In a study conducted with 157 patients diagnosed with panic disorder, 34% of the participants reported a history of loss or separation in the early periods of their lives (before the age of 15)²⁹. The relationships between depressive disorders, panic disorder and traumatic experiences, comorbidities and other characteristics identified in the studies are parallel with those observed in our own study. It is important to consider these comorbidities when developing a treatment plan.

The neuroticism subscale of the Eysenck Personality Questionnaire was found to exhibit a statistically significant elevation in the traumatic panic disorder subgroup of patients included in the study. The neuroticism subscale of the Eysenck Personality Questionnaire examines reactivity or emotional consistency. It has been reported that patients with high scores on this subscale may exhibit low self-confidence, excessive emotionality, depressive features, anxious features, nervousness and shyness³⁰. In a study conducted with 70 patients diagnosed with panic disorder, a significant relationship was identified between trauma and neuroticism, with neuroticism being observed to be elevated in this group³¹. In a study conducted with 314 patients with panic disorder, it was found that trauma was highly associated with neuroticism³². Our findings align with those of previous studies, indicating that a higher neuroticism score in the traumatic panic disorder subtype is associated with anxiety and depression.

The extraversion subscale of the Eysenck Personality Questionnaire was found to exhibit a significantly higher mean score in individuals diagnosed with non-traumatic panic disorder compared to those diagnosed with other subtypes. Extraversion is a personality dimension that encompasses sociability, enjoyment of communication, sociability, and enjoyment of social interaction. Those who obtain high scores on this dimension are more likely to be sociable, to engage in joking behaviour and to have a larger number of friends. Therefore, the higher scores observed in the non-traumatic subtype are a predictable finding and are consistent with the results of other studies.

The prevalence of separation anxiety was found to be statistically significantly higher in the subgroup of individuals with panic disorder that developed with a traumatic effect. Previous studies have demonstrated that individuals with a history of separation anxiety during childhood exhibit an elevated risk of developing panic disorder³³. Once more, childhood separation anxiety disorder has been identified as a significant risk factor for the development of various mental health disorders, particularly panic disorder and major depressive disorder³⁴.

The ADI-3 scores were found to be statistically significantly higher in the traumatic panic disorder subgroup. A substantial body of evidence has demonstrated that elevated anxiety sensitivity is associated with more intense and frequent panic attacks^35,36. A correlation was identified between elevated ADI-3 physical sub-dimension scores and the presence of panic disorder in individuals who have experienced trauma.

The experience of trauma during childhood has been identified as a contributing factor in the development of anxiety sensitivity, which in turn represents a significant risk factor for the onset of panic disorder^37,38. The markedly elevated prevalence of ADI-3 in the traumarelated PD subtype aligns with the existing literature. In light of these findings, it is imperative that childhood traumas be subjected to rigorous examination and treatment in order to prevent the development of psychiatric disorders in adulthood.

Upon examination of the clusters of panic disorder subtypes, it was found that the level of panic agoraphobia was statistically significantly higher in the traumatic panic disorder subtype. A study conducted in Turkey reported that the co-occurrence of panic disorder and agoraphobia was associated with a history of childhood sexual and physical abuse. Similarly, the same study revealed a strong correlation between childhood and adolescent traumas and the development of agoraphobia and suicidal tendencies³⁹.

A further statistically significant finding was that the level of anxiety was higher in the trauma-induced panic disorder subtype than in the non-traumatic panic disorder subtype. A substantial body of evidence indicates that exposure to traumatic experiences is associated with an increased risk of suicidal behaviour, more severe anxiety symptoms, depression and self-harm^40-42. The presence of trauma is associated with a heightened severity of anxiety symptoms in individuals diagnosed with panic disorder.

In our study, the prevalence of antidepressant use and combined psychiatric treatment was higher in patients with a traumarelated panic disorder subtype than in patients with a non-traumatic panic disorder subtype. It is postulated that the rate of receiving psychiatric treatment is higher in the traumarelated panic disorder subtype due to higher rates of anxiety sensitivity, separation anxiety, depression and trauma. As comorbidities are more common in this group, multidisciplinary clinical approaches may be appropriate in selected patients.

As a consequence, the Panic-Agoraphobia Scale, Hamilton Depression Rating Scale, Hamilton Anxiety Rating Scale, Eysenck Personality Questionnaire Neuroticism and Psychoticism dimension scores, Separation Anxiety and Anxiety Sensitivity Index, and Childhood Psychological Traumas Emotional Abuse, Emotional Neglect, Physical Abuse, and Sexual Abuse subscale scores were observed to be elevated in individuals diagnosed with traumatic panic disorder relative to those diagnosed with non-traumatic panic disorder.

Conclusion and Recommendations
The determination of panic disorder subtypes is of significant importance with regard to the course of the disease and the selection of appropriate treatment options. For this reason, studies aimed at subtyping have been conducted previously. However, our study is distinctive in that it is one of the few to examine the clustering of symptoms and personal characteristics of panic disorder patients and the relationship of these clusters with trauma.

In our study, we found that patients with traumatic panic disorder exhibited higher levels of depression and anxiety, separation anxiety, and anxiety sensitivity, as well as higher scores on neuroticism and psychoticism personality dimensions. Additionally, they reported a higher prevalence of childhood traumas, particularly emotional abuse and sexual abuse. In contrast, patients with non-traumatic panic disorder demonstrated higher levels of extroversion.

The findings of our study indicate that subtyping in panic disorder is essential for accurately assessing the severity of the disease, identifying medical and psychological comorbidities, evaluating the efficacy of pharmacological treatments, and differentiating between therapeutic responses. Given that the subtype of panic disorder caused by traumatic effects is more severe than the other subtype, with higher rates of comorbidities and resistance to treatment, it was hypothesised that the underlying traumatic issues should be addressed. It is therefore important to bear in mind that this subtype requires more intensive therapy, that follow-up examinations should be conducted at regular intervals, that comorbidities should be monitored closely and that organic diseases should be subject to rigorous surveillance. Furthermore, the findings of our study may prove useful in identifying risk groups in panic disorder and in guiding the implementation of appropriate measures.

It should be noted that this study is not without limitations. The sample size is relatively limited. A larger sample size will facilitate the identification of more distinct clusters and the discovery of additional clusters. It will thus become evident that certain characteristics, which are not deemed statistically significant, are in fact significant. Furthermore, due to the extensive number of scales utilized in the study, each patient was allotted an hour to complete the assessments. While completing the scales, patients occasionally expressed boredom and encountered difficulties, which may have resulted in distraction. Some patients indicated a preference for not self-reporting and reported difficulty in understanding the scales. Finally, medication use was not restricted in our panic disorder patients, as it was thought that medication use might negatively affect some scale scores, particularly those related to panic-agoraphobia. In conclusion, the existing literature on panic disorder is relatively limited with regard to studies investigating all of the aforementioned characteristics in terms of panic disorder and determining subtypes by cluster analysis. Furthermore, additional studies with a larger number of patients are required in order to investigate other potential clusters.

1) Davidoff J, Christensen S, Khalili DN et al. Quality of life in panic disorder: looking beyond symptom remission. Qual Life Res 2012; 21: 945-59.

2) Starcevic V, Kellner R, Uhlenhuth EH, Pathak D. The phenomenology of panic attacks in panic disorder with and without agoraphobia. Comp Psychiatry 1993; 34: 36-41

3) Cox BJ, Swinson RP, Endler NS et al. The symptom structure of panic attacks. Compr Psychiatry 1994; 35: 349-53.

4) Massana J, Lopez Risueno JA, Massana G et al. Subtyping of panic disorder patients with bradycardia. Eur Psychiatry 2001; 16: 109-14

5) Fidaner H, Tunca Z, Cimilli C. Panik bozukluğunda atakların klinik özellikleri: Q yöntemi ile bir çalışma. Türk Psikiyatri Dergisi 1998; 9: 277-83.

6) Bovasso G, Eaton W. Types of panic attacks and their association with psychiatric disorder and physical illness. Compr Psychiatry 1999; 40: 469-77.

7) Karabulut E, Alpar R, Özyar E. Hastalıkların Yere Göre Kümelenmesinde Kullanılan Yöntemler. İnönü Üniversitesi Tıp Fakültesi Dergisi 2006; 13: 37-43

8) Muge E, Ozge ALP T, Serkan B ve ark. DSM-5 Bozuklukları icin Yapılandırılmıs Klinik Gorusmenin Klinisyen Versiyonunun Turkceye Uyarlanması ve Guvenilirlik Calısması. Türk Psikiyatri Dergisi 2019; 30: 51-6

9) Tural Ü, Fidaner H, Alkın T, Bandelow B. Panik Agorafobi Ölçeği Türkçe Geçerlik ve Güvenilirlik Çalışması. Türk Psikiyatri Dergisi 2000; 11: 1.

10) Akdemir A, Örsel S, Dağ İ ve ark. Hamilton Depresyon Derecelendirme Ölçeği (HDDÖ)’nin geçerliği, güvenirliği ve klinikte kullanımı. Psikiyatri Psikoloji Psikofarmakoloji Dergisi 1996, 4: 251-9.

11) Yazıcı MK, Demir B, Tanrıverdi N, Karaağaoğlu E, Yolaç P. Hamilton Anksiyete Değerlendirme Ölçeği Türkçe Geçerlik ve Güvenilirlik Çalışması. Türk Psikiyatri Dergisi 1998; 9: 114-7.

12) Karancı AN, Dirik G, Yorulmaz O. Eysenck Kişilik Anketi-Gözden Geçirilmiş Kısaltılmış Formu’nun Türkiye’de Geçerlik ve Güvenilirlik Çalışması. Türk Psikiyatri Dergisi 2007; 18: 254-61.

13) Mantar A, Yemez B, Alkın T. Anksiyete Duyarlılığı İndeksi-3’ün Türkçe Formunun Geçerlik ve Güvenilirlik Çalışması. Türk Psikiyatri Dergisi 2010; 21: 225-34

14) Diriöz M, Alkın T, Yemez B ve ark. Ayrılma anksiyetesi belirti envanteri ile yetişkin ayrılma anksiyetesi anketinin Türkçe versiyonunun geçerlik ve güvenirliği. Turk Psikiyatri Dergisi 2012; 22: 108-16.

15) Şar V, Öztürk E, ve İkikardeş E. Çocukluk çağı ruhsal travma ölçeğinin Türkçe uyarlamasının geçerlilik ve güvenilirliği. Türkiye Klinikleri J Med Sci 2012: 1054-63.

16) Zou Z, Huang Y, Wang J et al. Association of childhood trauma and panic symptom severity in panic disorder: Exploring the mediating role of alexithymia. J Affect Disord 2016; 206: 133-9.

17) Goodwin RD, Fergusson DM, Horwood LJ. Childhood abuse and familial violence and the risk of panic attacks and panic disorder in young adulthood. Psychol Med 2005; 35: 881-90.

18) Tibi L, van Oppen P, Aderka IM et al. Examining determinants of early and late age at onset in panic disorder: an admixture analysis. J Psychiatr Res 2013; 47: 1870-75.

19) Zhang J, Wiecaszek P, Sami S, Meiser-Stedman R. Association between panic disorder and childhood adversities: a systematic review and meta-analysis. Psychol Med 2023; 53: 2585-95.

20) Avcı UK ve Çakmak S. Panik Bozuklukta Çocukluk Çağı Travmaları ile Disosiyatif Belirtiler Arasındaki İlişkide Aleksitiminin Aracı Rolü. AYNA Klinik Psikoloji Dergisi 2024; 11: 347-72.

21) Kay SJ, Keefe JR, Milrod BL, Barber JP. Childhood Trauma and Panic Disorder: The Impact of History of Child Abuse on Illness Severity and Treatment Response. Am J Psychother 2024; 77: 112-8.

22) De Venter M, Van Den Eede F, Pattyn T, Wouters K, Veltman DJ, Penninx BWJH, Sabbe BG. Impact of childhood trauma on course of panic disorder: contribution of clinical and personality characteristics. Acta Psychiatr Scand 2017; 135: 554-63.

23) Kessler RC, McLaughlin KA, Green JG et al: Childhood adversities and adult psychopathology in the WHO World Mental Health Surveys. Br J Psychiatry 2010; 197: 378-85.

24) Klauke B, Deckert J, Reif A et al: Life events in panic disorder: an update on “candidate stressors”. Depress Anxiety 2010; 27: 716-30.

25) Kaya Y, Demir EY, Keskin H ve ark. Metabolik sendrom ile panik bozukluk arasındaki ilişki. Abant Med J 2016; 5: 91-100.

26) Fikret Poyraz Ç. Panik Bozukluğu Hastaların Serum miRNA Ekspresyon Profilinin Değerlendirilmesi: Uzmanlık Tezi 2016.

27) Orsel S, Karadag H, Kahilogulları KA ve Akgun E. Psikiyatri hastalarında cocukluk cagı travmalarının sıklıgı ve psikopatoloji ile iliskisi. Anatolian J Psychiatry/Anadolu Psikiyatri Dergisi 2011; 12: 130-13.

28) Kara H, Ağargün M Y, Bılgın H ve Kıncır F. Panik Bozukluklarda Majör Depresyon Prevalans ı ve Majör Depresyonun Eşlik Ettiği Hastaların Klinik Özellikleri. Düşünen Adam 1994; 7: 23-9.

29) Servant D et al Parquet PJ. Early life events and panic disorder: course of illness and comorbidity. Progress in Neuro-Psychopharmacology and Biological Psychiatry 1994; 18: 373-9.

30) Eysenck HJ, Eysenck SB. Manual of the Eysenck Personality Questionnaire (adult and junior). Hodder & Stoughton, London 1975.

31) Kim HJ, Song C, Bang M, Lee SH. Early sexual trauma is related with the tapetum in patients with panic disorder. J Affec Dis 2020; 267: 107-13.

32) Lee SA, Bottomley JS. Personality and grief-related panic attacks: analysis of neuroticism, worry, and anxiety sensitivity. OMEGA-J Death Dying 2023; 86: 849-61.

33) Henin A, Faraone SV, Fraire M et al. Developmental trajectories of anxiety disorders in offspring at high risk for panic disorder and major depression. Psychiatry Res 2007; 153: 245-52.

34) Lewinsohn PM, Holm-Denoma JM, Small JW, Seeley JR, Joiner TE Jr. Separation anxiety disorder in childhood as a risk factor for future mental illness. J Am Acad Child Adolesc Psychiatry 2008;47:549-56.

35) McNally RJ. Anxiety sensitivity and panic disorder. Biolog Psych 2002; 52: 938-46.

36) Schmidt NB, Mitchell MA, Richey JA. Anxiety sensitivity as an incremental predictor of later anxiety symptoms and syndromes. Comprehensive Psych 2008; 49: 407-12.

37) Vujanovic AA, Zvolensky MJ, Bernstein A. Incremental associations between facets of anxiety sensitivity and posttraumatic stress and panic symptoms among trauma‐exposed adults. Cognitive Behaviour Therapy 2008; 37: 76-89.

38) Li W, Zinbarg RE. Anxiety sensitivity and panic attacks: a 1-year longitudinal study. Behav Modif 2007; 31: 145-61.

39) Özkan M, Özen Ş, Ertunç N. Panik Bozukluğuna Komorbid Eksen-1 Bozukluklarının Oluşumunda Travmatik Yaşam Olaylarının Rolü. Klinik Psikiyatri 2005; 8: 53-9.

40) Lippard ETC, Nemeroff CB. The devastating clinical consequences of child abuse and neglect: Increased disease vulnerability and poor treatment response in mood disorders. Am J Psych 2020; 177: 20-36.

41) Williams LM, Debattista C, Duchemin AM, Schatzberg AF, Nemeroff CB. Childhood trauma predicts antidepressant response in adults with major depression: data from the randomized international study to predict optimized treatment for depression. Translational Psych 2016; 6: e799.

42) Hovens JG, Giltay EJ, Wiersma JE et al. Impact of childhood life events and trauma on the course of depressive and anxiety disorders. Acta Psychiatrica Scandinavica 2012; 126: 198-207.