Determinants of Posttraumatic Stress Disorder (PTSD) among children and adolescents in the subacute stage of Kahramanmaras earthquake, Turkey

Abstract

Background

Earthquakes are potentially traumatic natural disasters due to their destructive nature, and huge impacts, producing scenes of horror, undesirable and uncontrollable results. Children are affected mainly by earthquakes not only physically but also psychologically. We aimed to evaluate the rates of probable PTSD and related factors in children and adolescents after the February 6 Kahramanmaras earthquake.

Methods

In this cross-sectional study, 246 children and adolescents were included. The research was conducted in Hatay, one of the most affected provinces, between 06/07/2023-06/08/2023. Sociodemographic form, Children’s Depression Inventory, Children’s Posttraumatic Response Reaction Index, Screen for Child Anxiety Related Emotional Disorders, Child and Youth Resilience Measure and Social Support Appraisals Scale for Children were applied by face-to-face survey method.

Results

The median age of the participants was 12 (8-18). 133 (54%) of the participants were girls. Probable depression rate was 98%, and probable anxiety disorder rate was 63%. Probable mild, moderate, severe and very severe PTSD rates were 18%, 29%, 43% and 8%, respectively. The probable severe PTSD rate was higher in children who were injured in the earthquake (p = 0.032), who received outpatient treatment (p = 0.016), and who saw people trapped under the collapse (p = 0.012). Also, the children whose sleep routines have changed post-earthquake were found to have more probable severe PTSD (p < 0.001). Access time to the potable water supply (p = 0.045), toilet facilities (p = 0.045), shelter facilities (p = 0.004), heating facilities (p = 0.001), clothing supply (p < 0.001) and healthcare services (p = 0.009) were found to be associated with probable severe PTSD right after the earthquake. In the fifth month of the earthquake, inadequate meeting of clothing needs (p = 0.018), educational services (p = 0.028) and social activity opportunities (p < 0.001) were significantly associated with probable severe PTSD. In the multivariate analysis, the risk of probable severe PTSD was increased 2.9 times in those with post-earthquake sleep changes (p = 0.001) and 11.1 times in those with probable anxiety disorders (p < 0.001). Also, it has been shown that each unit increase in the APP-family score reduces probable severe PTSD risk by 2% (p = 0.002).

Conclusion

In the current study, the rates of probable PTSD, depression and anxiety disorders were considerably high. Probable anxiety disorder and sleep disturbance were significant predictors of probable severe PTSD. On the other hand, social support from family was found to be a significant protective factor for severe PTSD risk. We conclude that the lack of meeting needs that increase the risk of probable severe PTSD differs in the acute and subacute phases.

Text box 1. Contributions to the literature
• Limited studies reported from Turkey, where the earthquake risk is extremely high throughout history, on post-earthquake mental health of children and adolescents, and the current study provides novel and specific results regarding the probable psychopathology rates in the subacute phase of the earthquake for children and adolescents.
• In the current study, earthquake-related trauma factors were evaluated detailly. It has been shown that traumatic events having particularly individual effects stand out among probable PTSD risk factors.
• Determinants of probable PTSD differ in the acute and subacute phases after the earthquake. These findings will guide policymakers to develop preventive and therapeutic measures.

Introduction

Exposure to major disasters increases the risk of various mental health problems [1]. Natural disasters are potentially traumatic events due to their destructive nature, massive impacts, producing scenes of horror, undesirable and uncontrollable results, and long-term changes in socioeconomic facilities. The trauma type, exposure severity, and previous traumatic experiences significantly predict various post-disaster mental health problems [2]. Earthquake is a seriously traumatic experience due to the threat to life, loss of property and witnessing horrific scenes. Earthquakes have the most significant destructive impact of all natural disasters [3]. Because earthquakes cause not only physical disorders but also psychological stress in victims. Studies have shown that exposure to earthquake-related traumatic events increases the risk of psychiatric disorders, especially Posttraumatic Stress Disorder (PTSD) [4,5,6]. Trauma is a common risk factor for PTSD and other psychiatric disorders, and PTSD is one of the most studied disorders [7].

A review showed that disaster victims in developing countries and children are mostly affected groups by traumatic events [8]. Children are more vulnerable to psychological problems caused by earthquakes, including PTSD, because they have not yet completed their physical and emotional development, are dependent on adults, and have inadequate coping skills [9, 10]. Studies on children exposed to earthquakes show that the prevalence of PTSD is 2.5–95% [11, 12]. The risk of post-earthquake PTSD may vary by many factors, such as demographic variables, individual factors, the severity of the earthquake, distance to the epicentre of the earthquake, degree of exposure, post-earthquake environmental factors, psychosocial support, and rehabilitation opportunities. Additionally, there is evidence that PTSD also co-occurs with other psychiatric disorders. Studies conducted after disasters show a correlation between depression, anxiety and PTSD [13, 14].

Earthquakes can affect developed and developing countries in different ways and levels. Major destructions resulting from unplanned urbanization, poverty, increasing settlement rate on fault systems and inadequate earthquake preparedness bring psychological problems. Although the post-earthquake PTSD prevalence rates reported from developed countries are quite low [15], higher rates have been consistently reported regarding PTSD and depression from developing countries persisting even years after the disaster [16,17,18,19].

The consecutive earthquakes of 7.7 and 7.6 (Mw) magnitude occurred on February 6, 2023, centred in Kahramanmaraş, affecting 11 provinces (Kahramanmaras, Gaziantep, Sanlıurfa, Diyarbakir, Adana, Adıyaman, Osmaniye, Hatay, Kilis, Malatya and Elazıg) in Turkey. Approximately 14 million people were affected, and more than 50 thousand deaths were reported [20]. The number of children affected by this earthquake is estimated to be approximately 5 million [21]. Millions of people lost their homes as more than half a million buildings were damaged due to the earthquakes [16]. Moreover, tens of thousands of aftershocks continued traumatizing people after the two major earthquakes [22]. The fact that the earthquake affected a wide region of Turkey caused official and civil facilities to be divided and inadequate in some places.

In the literature review, we noticed few published studies from Turkey on mental health in children and adolescents after disasters such as earthquakes. However, there is a need for studies reporting prevalence rates and focusing on risk factors to prevent, detect and intervene in psychiatric problems such as post-traumatic stress symptoms in children after the earthquake. The current study is the first study reported from the region examining the psychological effects of the Kahramanmaras earthquake on children and adolescents. We aimed to evaluate the rates of probable PTSD and related factors in children and adolescents in Hatay, one of the most intensely destructed provinces in the earthquake. We assumed that the probable rate of PTSD would be high due to the magnitude of the earthquake, the extent of the damage, the large area of the country being affected, and the inadequacy of official and civil facilities.

Materials and methods

This cross-sectional study was conducted in Hatay, one of the most affected provinces by the Kahramanmaras earthquake (Fig. 1), with children and adolescents who applied to Hatay Field Hospital Pediatrics Polyclinic on 06/07/2023-06/08/2023. Before the data collection, Ethics Committee permission (Amasya University Non-Interventional Ethics Committee, decision no: 2023/75, date: 04/05/2023) and Institutional Permission were provided. The purpose of the study and the methods to be followed were first explained to children and parents verbally before their participation in the study. Then, written informed consent was obtained.

Fig. 1

Map showing the location of Hatay province and its distances to 2 major earthquake epicentres (map data © Google Maps, 2023)

Participants

While planning the research, assuming that the confidence level is 95%, the probable PTSD prevalence rate is 20%, and the margin of error is 5%, the minimum sample size was calculated as 246. Children and adolescents who applied to the Hatay Field Hospital Pediatrics outpatient clinic, aged between 7 and 18, experienced the earthquake in Hatay and continued to live in Hatay after the earthquake, were literate and volunteered to participate in the study, were included in the study. The data collection was stopped when the targeted sample number was reached. Children and adolescents who were illiterate in Turkish and had neurological, mental or physical disabilities were excluded from the study.

Measures

The parents of the children and adolescents in the study completed a questionnaire that included sociodemographic variables, sleep changes and trauma characteristics related to the earthquake. Then, the self-report questionnaire method was used to administer the Children’s Depression Inventory (CDI), Children’s posttraumatic response reaction index (CPTS-RI), Screen for Child Anxiety Related Emotional Disorders (SCARED), Child and Youth Resilience Measure (CYRM), and Social Support Appraisals Scale for Children (APP) to the children and adolescents.

Children’s depression ınventory (CDI)

The scale was developed by Kovacs in 1981 [23]. It was based on the Beck depression scale, consisting of 27 items and can be applied to children and adolescents between 6 and 17. It is a self-assessment scale and evaluates the last two weeks. Each question is scored between 0 and 2 points. The highest score on the scale is 54. The higher score indicates the severity of the depression level. The validity and reliability study in Turkish was conducted by Öy, and the cut-off point for pathology was determined as 19 [24].

Children’s posttraumatic response reaction index (CPTS-RI)

It was developed by Pynoos et al. to investigate the severity of PTSD symptoms [25]. It is a 20-item scale developed to evaluate the stress reactions in children and adolescents after various traumatic experiences. A validity and reliability study was conducted by translating the scale into Turkish [26]. CPTS-RI is a five-point Likert-type semi-structured scale. A total score of 12–24 mild level, 25–39 moderate level, 40–59 severe level, and above 60 indicates a very severe PTSD reaction.

Screen for child anxiety-related emotional disorders (SCARED)

Birmaher et al. developed the scale to screen for childhood anxiety disorders [27], and its Turkish validity and reliability were tested by Çakmakçı et al. [28]. The parent and child forms of the SCARED are available. The scale consists of 41 items, and a score of 25 or above is considered a warning for anxiety disorder.

Child and youth resilience measure (CYRM)

Liebenberg et al. initially developed a scale consisting of 28 items [29], then reconstructed as a short form with 12 items. The Turkish adaptation was carried out by Arslan [30]. It has a five-point Likert structure and is rated between 1 and 5. As the scale score increases, psychological resilience also increases.

Social support appraisals scale for children (APP)

The scale was developed by Dubow et al. [31] and was adapted into Turkish by Gökler [32]. It consists of 41 items and three sub-dimensions: perceived friend social support, perceived family social support and perceived teacher social support. Children respond to each item on a five-point scale (1 = never, 2 = rarely/very rarely, 3 = sometimes, 4 = most of the time, 5 = always). The highest score on the scale is 205.

Statistical analysis

In statistical analyses, the distribution characteristics of the continuous variables were evaluated with the Kolmogorov-Smirnov test. Continuous variables were presented with median, minimum and maximum; categorical variables were presented with numbers and percentages. Mann Whitney U Test to compare continuous variables and Pearson Chi-square test to compare categorical variables were used. The correlation of two continuous variables was evaluated with the Spearman Correlation test. Binary Logistic Regression test was used as the multivariate analysis method. The statistical significance level was accepted as p < 0.05 for all tests.

Results

The median age of the participants was 12 (8-18). 133 (54%) of the participants were girls. Two hundred twenty-five participants (91%) currently reside in the container (Table 1). 3 (1%) participants lost one of their parents in the earthquake, and 33 (13%) participants had one of their parents injured. The rate of those whose houses completely collapsed was 28.5% (n = 70), and the rate of those whose houses were severely damaged was 66.3% (n = 163). 13 children (5%) were trapped under debris during the earthquake, and 12% (n = 30) were injured in the earthquake (Table 2). Analysis regarding the psychopathology risks showed that depression risk was 98% and anxiety disorder risk was 63%. The rates of probable mild, moderate, severe and very severe PTSD were found to be 18%, 29%, 43% and 8%, respectively (Table 3). The scale score distributions of the participants are shown in detail in Table 3.

As a result of the analysis of PTSD-associated factors after the earthquake, the rate of probable severe PTSD did not differ according to age and gender (p > 0.05) (Table 1). Probable severe PTSD was higher in children who were injured in the earthquake (70%) compared to those noninjured (49%) (p = 0.032), who received outpatient treatment (100%) compared to those not received (50%) (p = 0.016), who saw people trapped under the collapse (59%) compared to those did not see (43%) (p = 0.012). The probable severe PTSD was higher in children whose sleep routines have changed compared to those whose not changed (p < 0.001) (Table 2).

Table 1 Sociodemographic characteristics of the participants and the relationship with probable PTSD severity

Table 2 Trauma characteristics of the participants and the relationship with probable PTSD severity

Table 3 Participants’ probable psychopathology rates and distribution of scale scores

38% of post-earthquake search and rescue efforts, 30% of food supply, 23% of drinking water supply, 40% of potable water supply, 35% of toilet facilities, 48% of shelter facilities, 44% of heating facilities, 52% of clothing supply and 58% of healthcare services were provided after the first 72 h of the earthquake (Supp. Table 1). In the fifth month of the earthquake, it was concluded that clothing supply (17%), healthcare services (10%), education services (34%) and social activity facilities (30%) were inadequate (Supp. Table 2). Access time to the potable water supply (p = 0.045), toilet facilities (p = 0.045), shelter facilities (p = 0.004), heating facilities (p = 0.001), clothing supply (p < 0.001) and healthcare services (p = 0.009) were found to be associated with probable severe PTSD right after the earthquake. The highest rates were observed for the groups whose needs were met after the first 72 h (Table 4). In the analyses, access time to drinking water supply (p = 0.043), potable water supply (p = 0.015), toilet facilities (p = 0.015), shelter facilities (p = 0.002), heating facilities (p < 0.001), clothing supply (p < 0.001) and healthcare services (p = 0.001) were showed a significant linear relationship with probable severe PTSD. Accordingly, as access to these needs was delayed, the risk of probable severe PTSD increased (Table 4).

Table 4 Comparison of the participants’ access time to the needs post-earthquake and probable severe PTSD severity

In the analysis of the relationship between the meeting of the needs and probable severe PTSD in the fifth month after the earthquake, inadequate meeting of clothing needs (p = 0.018), educational services (p = 0.028) and social activity opportunities (p < 0.001) were significantly associated with the probable severe PTSD (Table 5). Additionally, inadequate meeting of food (p = 0.020), potable water (p = 0.044), toilet (p = 0.021), clothing (p = 0.003), educational services (p = 0.004) and social activity (p < 0.001) needs showed a significant linear relationship with probable severe PTSD. Accordingly, we can conclude that as the adequacy of meeting these needs increases, the risk for probable severe PTSD decreases (Table 5).

Table 5 Comparison of participants’ meeting of the needs and probable PTSD severity in the subacute stage of the earthquake

The participants’ CPTS-RI total score was positively correlated with the SCARED total score at a moderate level (r = 0.683, p < 0.001), negatively correlated with the APP-friend score (r=-0.187, p = 0.003), APP-family score (r=-0.143, p = 0.025) and APP-teacher score (r=-0.154, p = 0.016) at a low level. Additionally, the SCARED total score was negatively correlated with CYRM total score (r=-0.213, p = 0.001), APP-friend score (r=-0.131, p = 0.040), APP-family score (r=-0.156, p = 0.014) and APP-teacher score (r=-0.164, p = 0.010) at low level (Table 6).

Table 6 Correlation analysis of participants’ CDI, SCARED, CPTS-RI, CYRM total scores and APP subscale scores

According to the multivariate analysis with a model including the sociodemographic and earthquake trauma-related variables having a p-value of < 0.100 in univariate analyses, the risk of probable severe PTSD was increased 2.9 times in those with post-earthquake sleep changes (p = 0.001) and 11.1 times in those with probable anxiety disorders (p < 0.001). Also, it has been shown that each unit increase in the APP-family score reduces probable severe PTSD risk by 2% (p = 0.002) (Table 7).

Table 7 Multivariate analysis of related factors of the probable severe PTSD

Discussion

The current study evaluated the probable severe PTSD rate and associated risk factors in children and adolescents 5–6 months after the earthquake in a province where earthquake destruction was intense. Accordingly, the participants’ probable moderate PTSD rate was 29.3%, and the severe-to-very severe PTSD rate was 51.6%. In literature, it has been reported that between 4.5% and 95% of children and adolescents who experienced the earthquake developed PTSD [6, 12, 33,34,35,36]. Methodological differences between the studies and post-earthquake evaluation time frame may widen this prevalence range. In a prospective study conducted with children and adolescents after the Lushan earthquake in China, the probable PTSD rate in the third and sixth months was found to be 37% and 24%, respectively [37]. A study conducted three years after the 1999 Marmara earthquake in Turkey showed that 31% of adolescents and children had moderate and 28% had severe/very severe posttraumatic stress reactions [18]. In the third month of the 2011 Van earthquake, it was shown that 8.6% of children aged between 7 and 12 developed mild PTSD symptoms, 19.7% moderate PTSD symptoms, 47.7% severe PTSD symptoms, and 24% very severe PTSD symptoms [38]. The rates in this study, whose research time frame and region are close to our study, are parallel to our findings. The fact that the participants in our study live in a province close to the earthquake epicentre and where the most destruction occurred may be the reason why the probable PTSD rate is higher compared to some of the literature. In a study, PTSD rates were determined as 95%, 71% and 26% in three cities gradually moving away from the epicentre of the earthquake [12]. It is known that there is a dose-effect relationship between the severity of exposure to earthquake-related trauma and psychiatric disorders. In Armenia, high levels of PTSD [39] and major depression [17] were observed in people living in heavily affected areas.

In our study, some earthquake-related trauma factors were evaluated. One of these is the death of loved ones. While some studies show evidence that the death of family members is associated with particularly severe PTSD [40], others have found no association between casualties and PTSD [36]. After the earthquake in Greece in 1999, PTSD in adolescents was positively associated with home damage but not with the death or injury of family members [35]. Similarly, the current study found no relationship between parental loss, parental injury, loss of a relative or friend, and probable severe PTSD. The other factor we evaluated is the house damage level, and no significant relationship was found between house damage level and probable severe PTSD. Besides various results on this subject in the literature [35, 38], in the study in the third month of the 2011 Van earthquake, no significant relationship was found between the damage status of the house and PTSD scores, similar to our study [38]. The houses of almost all of the children and adolescents evaluated in our study were either moderately or severely damaged or completely collapsed. Therefore, we could not compare the effect of whether the house was damaged or not. Whether the house was moderately or severely damaged or completely collapsed did not differ in possible severe PTSD symptoms.

The literature has reported that physical injury is a risk factor for PTSD [34, 41, 42]. Additionally, witnessing the injury or death of another person has been identified as another risk factor for PTSD among child survivors of an earthquake [42]. A study from China showed that the risk of PTSD increased 1.5 times after being injured in an earthquake and 1.7 times after seeing people trapped under rubble during an earthquake [43]. In the current study, the probable severe PTSD rate was higher among the children and adolescents who were injured themselves or who saw people being pulled out through the rubble. Children are more susceptible to traumatic images than adults. Adults’ accumulation of life experience makes them more harmonious and calm against dramatic scenes [44]. However, children often relive the moment when they witnessed people being crushed and injured under collapsed buildings a few months or even a few years later through their lives [45].

It has been shown in the literature that separation from family members, lack of employment and housing, crowded living conditions, loss of social services, shortage of food, gasoline, medical supplies, and collapsed buildings increase PTSD symptoms [12]. In the current study, regarding the just aftermath of the earthquake, It was determined that the probable severe PTSD rate was higher in children who reported that their needs for drinking water, toilet, shelter, heating, clothing and health services were met later. In the subacute stage of the earthquake, the probable severe PTSD rate was higher in those whose clothing, education and social activity needs were not adequately met. Findings show that it is crucial to meet essential needs such as shelter, heating, toilet and water, which are necessary to keep living, especially in the acute phase, and that the inadequacy in meeting educational needs and social needs in the subacute phase of the earthquake increases the probable severe PTSD rate. Accordingly, essential needs must be met as quickly as possible in the first 72 h after the earthquake. It should be emphasized that needs may vary seasonally, and the need for shelter and heating after disasters occur in the winter, such as the Kahramanmaraş earthquake.

Subjective sleep problems are common posttraumatic psychological symptoms in children and adolescents [46,47,48]. Such problems may continue later in life and may be accompanied by symptoms of PTSD, depression and anxiety [46, 47]. However, only a few studies have reported a strong association between sleep disturbances, PTSD, and depressive symptoms in adolescent trauma survivors [46, 49]. In a follow-up study conducted after the earthquake in Lushan City, China, in 2013, insomnia symptoms were found to be 52% and 40% in the third and sixth months, respectively. It was found that insomnia symptoms in the third month increased PTSD symptoms 1.5 times in the sixth month of the earthquake [49]. Similar to the literature, we found that 56% of the participants’ sleep time decreased five months after the earthquake. Multivariate analyses showed that the probable severe PTSD rate was 2.9 times higher in those with sleep changes (increase or decrease) after the earthquake.

PTSD is found to be comorbid with other psychiatric conditions in children and adolescents in post-disaster studies [14]. In a study with 2081 adolescent survivors were evaluated after six months after the Wenchuan earthquake; of the 329 individuals screened for clinical PTSD, 84% had anxiety, 55% had depression, and 50% had anxiety and depression comorbidity [50, 51].

In a prospective study conducted after the Wenchuan earthquake, a bidirectional relationship between Generalized Anxiety Disorder and PTSD was detected [46]. Findings have been reported that depression and anxiety predict later PTSD in adolescent survivors [50, 51]. We found that 98% of the children and adolescents had depressive symptoms, and 63.4% had anxiety symptoms. The CPTS-RI total score was positively correlated with the SCARED score at a moderate level. Additionally, multivariate analyses showed that the probable severe PTSD rate was 11 times higher in those with probable anxiety disorder post-earthquake. Accordingly, we concluded that anxiety disorder after the earthquake is a prominent risk factor for PTSD.

Social support is a protective variable for preventing or reducing post-earthquake PTSD symptoms [52,53,54]. The current study found that the CPTS-RI total score was negatively correlated with the APP-family, APP-friends and APP-teacher scores. In multivariate analysis, we detected that each unit increase in the APP-family score reduced the probable severe PTSD rate by 2%. Lack of social support after natural disasters and traumatic events was associated with PTSD symptoms because insufficient support can increase fear, terror and helplessness [36, 55, 56]. A study conducted after Hurricane Katrina in Mexico reported that those who experienced more posttraumatic stress symptoms were the ones who either avoided social support sources, perceived the available social support as less, or remained outside of supportive relationships [57]. In a case-control study examining the effect of psychosocial support on the development of PTSD and depression in young after the Haiti earthquake in 2010, the rates of PTSD and severe depression in the group receiving psychosocial support were lower (50% and 20%, respectively) than in the group that did not receive support (68% and 41% respectively) [58]. As known, while facing the effects of an earthquake, social relationships and moral encouragement from family and friends positively balance the adverse effects of the trauma [42]. Social support may be helpful to prevent and decrease PTSD symptoms. Social support provided by related government institutions and other civil organizations should also be encouraged.

Limitation and strengths

The current study has several limitations. First, the psychopathology risk was evaluated with scales; they indicate symptom occurrence and probable diagnosis, not a definitive diagnosis. Secondly, because of the study’s cross-sectional design, psychopathology risks before and after the earthquake could not be compared. That is why it is difficult to determine the direction of the causal relationships found in the study. Since no sample selection was applied, we can not generalize the findings to all children survivors post-earthquake. Third, the participants were children and adolescents exposed to a quite destructive earthquake, indicating that the current findings may not be generalizable to other types of trauma. Fourth, the study was conducted in the subacute period of the earthquake and did not indicate the long-term results and risk. Therefore, the generalizability of the findings over a more extended period should be discussed. Despite these limitations, this is a comprehensive study evaluating the probable PTSD rate and related risk factors among child adolescent survivors after the earthquake. The current findings will guide health professionals and policymakers while planning interventions needed to help child and adolescent survivors of the earthquake cope with posttraumatic stress.

Conclusion

In conclusion, we determined that the probable rates of PTSD, depression and anxiety disorder were high among children and adolescents affected by the earthquake. The significant relationship between probable anxiety disorder and severe PTSD has been demonstrated. We concluded that meeting essential needs in the acute stage of the earthquake reduced the risk of PTSD, and social support, especially from the family, is a protective factor for PTSD. These findings will guide policymakers in creating prevention, intervention and improvement programs for at-risk groups in Turkey, where numerous earthquakes have been experienced and the earthquake risk is considerably high in the future. Additionally, these results may help understand the risk factors associated with PTSD and provide treatment with an early diagnosis before it becomes chronic. In this context, there is a need for post-disaster mental health recovery programs, including ongoing psychosocial support for children and adolescents who survived the earthquake.