Classifications of Trauma Exposure and Their Associations With 3-Year Follow-up Psychological Symptoms Among Korean Victims of Disaster: A Latent Class Analysis

Abstract

This study aimed to classify patterns of trauma exposure among disaster victims using latent class analysis (LCA) and to examine group differences in post-traumatic stress disorder (PTSD), anxiety, and depressive symptoms over three years. Data were obtained from a 3-year panel survey of Korean disaster victims (male = 461, female = 513) who responded to four types of trauma experiences: threatened death, injury or disease, witnessing injury or death, and loss of family or relatives in 2017. The LCA yielded three classes: the low-threat (66.1%), life-threat (29.4%), and multi-threat (4.5%) groups. One-way analysis of variance showed that the multi-threat group had the highest levels of PTSD, depressive, and anxiety symptoms in 2017, 2018, and 2019. Post-traumatic stress disorder symptoms significantly decreased between 2017 and 2018 in the life-threat and multi-threat groups. However, there were no differences between 2018 and 2019 in any groups. These results highlight the importance of early intervention during the proximal period after trauma, particularly in individuals exposed to multiple types of traumatic experiences.

Keywords

disaster latent class analysis post-traumatic stress disorder symptoms depressive symptoms anxiety symptoms

What We Already Know

Trauma can lead to negative psychological outcomes, such as post-traumatic stress disorder (PTSD) and depression.

Individuals who are directly exposed to a disaster are more likely than those with indirect trauma to have severe PTSD symptoms.

Given the possibility of co-occurrence of traumatic events, latent class analysis can provide information on which combinations of trauma exposure are more vulnerable to mental health.

What This Article Adds

This is the first study identifying phenotypes of various types of trauma and their long-term (3 years) relationships with psychological symptoms among disaster victims in South Korea.

Exposure to multiple types of trauma predicts more severe PTSD, depressive, and anxiety symptoms.

Post-traumatic stress disorder symptoms decline during the proximal period, which indicates the importance of early intervention by service providers.

Introduction

Considerable evidence has shown that trauma exposure is significantly associated with negative psychological symptoms, such as post-traumatic stress disorder (PTSD), depression, or anxiety.^1,2 Considering that trauma can be drawn from various types of events, researchers have generally focused on finding the types of traumatic events that have a stronger or independent impact on mental health outcomes. For example, research on veterans has shown that moral injury and exposure to malevolent environments are more likely to predict PTSD symptoms than general combat exposure.³ In a systematic review, individuals who were directly exposed to a disaster, rather than indirect victims, showed higher levels of PTSD symptoms.⁴

However, traumatic events are more likely to occur simultaneously than in isolation.⁵ In addition, findings on the differential effects of traumatic events are somewhat mixed; some demonstrated a unique impact of killing on PTSD symptoms, independent of general combat exposure, such as perceived danger and exposure to the death of others.⁶ By contrast, others found no unique association between killing and PTSD and anxiety symptoms, with which witnessing the death or injury of others was significantly associated.⁷ In this regard, an additive approach, in which variables to assess traumatic events are included in a regression model, may be insufficient to understand the relationship between co-occurring trauma and the consequent psychological distress.

An effective alternative for identifying patterns of various types of co-occurring trauma exposure is to use latent class analysis (LCA), which is a person-centered analytic approach. Using LCA, researchers can classify multiple dimensions of traumatic events into groups.⁵ In addition, LCA can provide information on which combinations of trauma exposure are more vulnerable to mental health, which helps health-related professionals identify top priority groups for intervention.

Despite these advantages, little is known about the patterns of trauma exposure using LCA. Although a few studies examined group classifications using various disaster-related variables, the patterns of traumatic events may have been influenced by the probabilities of other nontraumatic experiences, such as work-related problems or relationships with neighbors.⁸ This study focused on specific types of trauma exposure based on the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5). According to the DSM-5, traumatic experiences include direct or indirect exposure to death, threatened death, and actual or threatened serious injury. Witnessing trauma and learning that a relative or close friend was exposed to trauma also pertain to the type of trauma.⁹

In addition, it is important to examine longitudinal associations between trauma and mental health. Psychological symptoms appear to vary depending on the type of trauma. For example, veterans who were directly exposed to combat showed chronic or relapse patterns of PTSD and depressive symptoms over time, and reported higher PTSD prevalence rates than those who were traumatized by indirect exposure to war-related trauma.^10,11 In addition, the prevalence of PTSD in individuals with intentional injury or trauma, such as assault or war, increased over 12 months, whereas those who were exposed to nonintentional traumatic events, such as natural disasters or vehicle accidents, showed a decreasing prevalence of PTSD.¹² These findings indicate that longitudinal changes in mental health following trauma can vary depending on the severity or intensity of the traumatic events. However, little is known about long-term changes across groups with different patterns of trauma exposure. Thus, the aim of this study was twofold: to classify patterns of trauma exposure among disaster victims using LCA and to examine group differences in subsequent psychological (PTSD, anxiety, and depressive) symptoms.

Methods

Sample and Procedure

This study used 3-year panel data drawn from 974 disaster victims collected by the National Disaster Management Research Institute, Republic of Korea. This panel survey was designed to trace mental health outcomes and determine specific difficulties in life among disaster victims who were exposed to building fires or natural disasters, such as typhoons.¹³ The type of trauma experienced during disaster exposure was measured in 2017 and mental health outcomes (including PTSD, depressive, and anxiety symptoms) were assessed annually from 2017 to 2019. Our sample consisted of 461 men and 513 women with a mean age of 57 years (SD = 17.6) in 2017. The number of samples from 2017 to 2018 did not change, but 33 cases (3.4%) were lost in the 2019 survey. Thus, we compared demographic variables between respondents (941) and nonrespondents (33) in 2019, using age, sex, education, and marital status from the 2017 survey, and found no significant differences. As of the 2017 survey, 70.4% were currently married and 29.6% were not currently married. In terms of educational level, 47.9% were middle school or lower graduates, 38.3% were high school graduates, and 13.8% were college or higher graduates. All study procedures were approved by the institutional review board of the Korea National Institute for Bioethics Policy (P01-202306-01-001).

Measures

Demographics

Age, sex, education, and marital status were included as demographic variables. Chronological age was coded as 1 = 29 years or below, 2 = 30 to 59 years, and 3 = 60 years or above. Sex was coded as 1 = male and 2 = female, and education level was categorized as 1 = middle school or below, 2 = high school, and 3 = college or above. Marital status was categorized as 0 (not currently married) and 1 (currently married).

Types of disaster-related trauma

Disaster-related trauma was assessed using four items: “Were you threatened with death?,” “Were you injured or infected by disease?,” “Did you witness people who were injured or dead?,” and “Were there any casualties (injury, disease, death, or missing) from the disaster among your family members or relatives?” The items were rated on a dichotomous scale (1 = yes, 0 = no).

Mental health outcomes

Mental health outcomes included PTSD, depressive, and anxiety symptoms. PTSD symptoms were measured using a 22-item Korean version of the Impact of Event Scale–Revised (IES-R), which was rated on a five-point Likert scale (1 = not at all to 5 = extremely).¹⁴ The IES-R assesses the frequency with which individuals have felt subjective distress in the past 7 days in response to a disaster. The total sum of the item responses was used in this study. Thus, higher scores reflected higher levels of PTSD symptoms (Cronbach’s α = .98, .98, and .98 in 2017, 2018, and 2019, respectively).

The Patient Health Questionnaire 9 (PHQ-9) and Generalized Anxiety Disorder 7 (GAD-7) questionnaire are validated and widely used to assess depressive and anxiety symptoms, respectively.¹⁵ This study used the Korean versions of the PHQ-9 and GAD-7, which are freely downloadable from the PHQ website.¹⁶ Participants were asked to rate how frequently they had been disturbed by each depressive (nine items) and anxiety (seven items) symptoms over the past 2 weeks on a four-point Likert scale (1 = not at all to 4 = nearly every day). The total sum of the scores for each scale was used. Thus, higher scores indicated higher levels of depressive or anxiety symptoms (Cronbach’s α = .92, .93, and .92 for PHQ-9 in 2017, 2018, and 2019, respectively; .94, .95, and .94 for GAD-7 in 2017, 2018, and 2019, respectively).

Analyses

Latent class analysis is a model-based technique used to identify the optimal number of latent classes within a population, based on individuals’ responses to categorical indicators.¹⁷ In this study, three different LCAs, from a two-class to a four-class model, were conducted to classify the patterns of trauma exposure using Mplus 7.0. The model fit indices considered for selecting the optimal number of classes were the Akaike information criterion (AIC), Bayesian information criterion (BIC), sample size-adjusted Bayesian information criterion (SABIC), Vong-Lo-Mendell-Rubin likelihood ratio test (VLMR), Lo-Mendell-Rubin adjusted likelihood ratio test (adjusted LMR), bootstrap likelihood ratio test (BLRT), and entropy.

Information-theoretic criteria, including AIC, BIC, and SABIC, indicate that the model with the lowest values is the best-fitting model for the given data.¹⁷ Likelihood ratio tests, including VLMR, adjusted LMR, and BLRT, provide a p value to examine whether the k class model has a statistically significant improvement over the k-1 class model.¹⁸ That is, a probability value of less than .05 between two models (e.g., three-class model vs two-class model) indicates that the three-class model (k class model) is preferred over the two-class model (k-1 class model). Finally, entropy ranges from 0 to 1, where a value greater than 0.8 means “good” classification accuracy.¹⁷

After determining the optimal number of classes based on the seven model fit indices, χ² tests and one-way analysis of variance (ANOVA) with Scheffé’s test for post hoc analysis were conducted to examine group differences in demographic variables and mental health outcomes in 2017, 2018, and 2019. In addition, paired t tests were used to examine changes in psychological symptoms between 2017 and 2018, and 2018 and 2019. Eta squared and Cohen’s d were also considered for the effect size. An effect size is interpreted following benchmarks: small (d = 0.2), medium (d = 0.5), and large (d = 0.8) for Cohen’s d, and small (η² = 0.01), medium (η² = 0.06), and large (η² = 0.14) for eta squared.¹⁹ There were no missing values in the study variables from the 2017 and 2018 survey, but the missing data of psychological symptoms in the 2019 survey were 3.4%. Thus, the missing values were imputed by single imputation using expectation maximization, which enables the missing data to be replaced by maximum likelihood values. The results of one-way ANOVA and paired t tests were computed using SPSS 27.

Results

Latent Class Analysis

Table 1 provides the model fit indices of three LCAs, indicating that a three-class model was the best classification of traumatic experiences in this sample. Although BIC was the lowest in the two-class models, other information-theoretic criteria, including AIC and SABIC, showed that the three-class model had the lowest values with the highest entropy. In addition, likelihood ratio tests consisting of the VLMR, adjusted LMR, and BLRT demonstrated that the three-class model was preferred.

Table 1.

Fit Indices for Different Latent-Class Solutions (N = 974).

Model	Log-likelihood	AIC	BIC	SABIC	Entropy	LMR	Adjusted LMR	BLRT
2. Classes	-1178.86	2375.72	2419.65	2391.07	.776	<.001	<.001	<.001
3. Classes	-1163.28	2354.56	2422.90	2378.44	.891	<.001	<.001	<.001
4. Classes	-1162.17	2362.35	2455.09	2394.75	.842	.26	.27	.67

Abbreviations: Adjusted LMR, Lo-Mendell-Rubin adjusted likelihood ratio test; AIC, Akaike information criterion; BIC, Bayesian information criterion; BLRT, bootstrap likelihood ratio test; LMR, Vong-Lo-Mendell-Rubin test; SABIC, sample size-adjusted Bayesian information criterion.

The conditional response probabilities of the three-class model are listed in Supplemental Table 1. Class 1 (low-threat) had the lowest response probabilities for all four types of traumatic experiences, whereas class 3 (multi-threat) had the highest response probability for being injured or having a disease, witnessing acquaintances being injured or dead, and losing family members or relatives. Class 2 (life-threat) had the highest probability of threatened death, but the probabilities of other types of traumatic experiences were very low. An illustration of the classification of traumatic experiences is shown in Figure 1.

Group Differences in Demographic Variables and Mental Health Outcomes

As indicated in Table 2, χ² tests indicated that there were no group differences in age, sex, or marital status, but the education level was found to be significantly different across the classes. Additional analyses using Spearman’s rank correlation showed a positive association between the levels of education and exposure to multiple types of trauma (r_s = .11, p < .001), whereas other demographic variables did not have significant relations. Regarding mental health, ANOVAs showed that class 3 (multi-threat) had the highest levels of PTSD, depressive, and anxiety symptoms in 2017, 2018, and 2019. Class 2 (life-threat) had higher scores for the three mental health outcomes than class 1 (low-threat) in 2017 and 2018, but their significant differences disappeared in 2019. Given the significant group differences in education level, analyses of covariance (ANCOVAs) were additionally conducted to ascertain the differences in PTSD, depressive, and anxiety symptoms among the classes, controlling for the effect of education. Post hoc (Bonferroni) analyses based on estimated marginal means showed that the group differences in the psychological symptoms were the same as those from ANOVAs. In sum, victims of disasters with multiple types of trauma showed the most negative psychological symptoms over 3 years.

Table 2.

Differences in Demographics and Psychopathological Symptoms Among the Classes.

Variables		Class 1, low-threat, n = 644 (66.1%)	Class 2, life-threat, n = 286 (29.4%)	Class 3, multi-threat, n = 44 (4.5%)	χ² test (df)	ANOVA, F (df)	ANOVA, post hoc (Scheffé)	η²
Variables		M (SD) or n(%)	M (SD) or n(%)	M (SD) or n(%)	χ² test (df)	ANOVA, F (df)	ANOVA, post hoc (Scheffé)	η²
Demographics
Age	29 years or below	78 (12.1%)	25 (8.7%)	5 (11.4%)	χ² (4) = 5.07
	30 years to 59 years	237 (36.8%)	125 (43.7%)	16 (36.4%)
	60 years or above	329 (51.1%)	136 (47.6%)	23 (52.3%)
Sex	Male	306 (47.5%)	130 (45.5%)	25 (56.8%)	χ² (2) = 2.001
Sex	Female	338 (52.5%)	156 (54.5%)	19 (43.2%)
Education	Middle school or below	336 (52.2%)	113 (39.5%)	18 (40.9%)	χ² (4) = 16.10^**
	High school	227 (35.2%)	130 (45.5%)	16 (36.4%)
	College or above	81 (12.6%)	43 (15.0%)	10 (22.7%)
Marital status	Not married	196 (30.4%)	82 (28.7%)	10 (22.7%)	χ² (2) = 1.33
Marital status	Married	448 (69.6%)	204 (71.3%)	34 (77.3%)
2017 PTSD symptoms		34.22 (14.80)	41.24 (17.87)	49.45 (21.00)		F (2,971) = 32.81^***	3 > 2 > 1	0.06
2017 Depressive symptoms		11.43 (3.55)	13.04 (5.29)	16.16 (7.85)		F (2,971) = 32.97^***	3 > 2 > 1	0.06
2017 Anxiety symptoms		8.56 (2.82)	9.43 (3.88)	11.86 (6.14)		F (2,971) = 23.68^***	3 > 2 > 1	0.05
2018 PTSD symptoms		32.25 (14.62)	36.26 (16.94)	40.43 (18.87)		F (2,971) = 10.80^***	3, 2 > 1	0.02
2018 Depressive symptoms		11.81 (4.25)	12.61 (4.89)	14.55 (6.51)		F (2,971) = 9.33^***	3 > 2 > 1	0.02
2018 Anxiety symptoms		8.76 (3.27)	9.73 (4.39)	10.23 (4.40)		F (2,971) = 9.03^***	3, 2 > 1	0.02
2019 PTSD symptoms		34.17 (14.67)	35.69 (14.70)	40.88 (16.03)		F (2,971) = 4.82^**	3 > 1	0.01
2019 Depressive symptoms		12.08 (4.46)	12.44 (4.66)	14.24 (5.82)		F (2,971) = 4.80^**	3 > 1	0.01
2019 Anxiety symptoms		9.04 (3.44)	9.27 (3.51)	10.56 (4.44)		F (2,971) = 3.99^*	3 > 1	0.01

Abbreviations: ANOVA, analysis of variance; PTSD, post-traumatic stress disorder.

p < .05. ^**p < .01. ^***p < .001.

Supplemental analyses using linear regression were conducted to examine the differential impacts of each class on mental health outcomes (see Supplemental Table 2). Similar to the results from ANOVAs, class 3 (multi-threat) had the highest impacts on the PTSD, depressive, and anxiety symptoms across the 3 years (β = 0.09-0.22, p < .01), controlling for the effect of education. Class 2 (life-threat) also showed unique impacts on mental health outcomes in 2017 and 2018 (β = 0.10-0.20, p < .01), but the significant effect did not remain in 2019 (β = 0.04-0.06, p > .05).

Correlations and Changes in Mental Health Outcomes

Post-traumatic stress disorder, depressive, and anxiety symptoms in 2017, 2018, and 2019 were all significantly and positively correlated with each other, ranging from r = .30 to .81, p < .001. In general, age, sex, and education levels were significantly, but weakly, associated with mental health outcomes. Specifically, individuals who were older and female, and had lower levels of education, showed higher levels of PTSD, depressive, and anxiety symptoms (for correlations among study variables, see Supplemental Table 3). Given the minimum effect size for Cohen’s d, PTSD symptoms significantly decreased from 2017 to 2018 in class 2, t(285) = 4.77, p < .001, Cohen’s d = 0.28, and class 3, t(43) = 2.78, p = .008, Cohen’s d = 0.42. However, there were no differences between 2018 and 2019 in any class. Depressive and anxiety symptoms did not significantly change over the years in all the classes, except for anxiety symptoms in class 3 between 2017 and 2018, t(43) = 2.11, p = .041, Cohen’s d = 0.32 (for paired t tests and Cohen’s ds among the classes, see Supplemental Table 4).

Discussion

Given the significant association between trauma and mental health outcomes,^1,2 this study sought to identify how types of traumatic events are grouped and their long-term relationship with psychological symptoms among disaster victims, about which relatively little is known. The LCA used in this study yielded three different classes for this sample: class 1 (low-threat), class 2 (life-threat), and class 3 (multiple-threat). Individuals who were exposed to multiple threats showed the highest levels of PTSD, depressive, and anxiety symptoms over time, whereas those who were not or were less exposed to traumatic events reported relatively low levels of these symptoms.

These findings are similar to a prior study that showed the highest PTSD, depressive, and anxiety symptoms were found in individuals who endorsed items related to trauma exposure with a probability above 30%.⁸ Psychological symptoms can be proportional to the number of trauma exposures. For example, cumulative childhood maltreatment has been associated with increased PTSD, anxiety, and depressive symptoms.²⁰ In addition, civilians who were exposed to a large number of war trauma showed higher levels of PTSD severity.²¹ In line with studies demonstrating a dose-response relationship between trauma and negative mental health outcomes,^21,22 this study revealed that exposure to multiple types of traumatic experiences predicted more severe PTSD, depressive, and anxiety symptoms.

In addition to the number of trauma exposure episodes, the severity of exposure can be related to mental health outcomes. Prior research has shown that individuals who directly experienced a disaster, war trauma, or terror-related trauma showed higher levels of PTSD symptoms compared with those who experienced indirect trauma.^4,10 This study also showed that class 2 (life-threat), which was characterized by direct exposure to death, had higher levels of PTSD, depressive, and anxiety symptoms in 2017 and 2018 than class 1 (low-threat) that was characterized by less direct exposure to trauma.

Post-traumatic stress disorder symptoms in classes 2 and 3 significantly decreased from 2017 to 2018, but did not decrease from 2018 to 2019. Anxiety symptoms in class 3 also declined from 2017 to 2018, but not from 2018 to 2019. These results indicate that trauma or anxiety related to negative psychological symptoms following a disaster may be mitigated over time, especially in the period proximal to the traumatic event. Despite exposure to severe trauma, individuals may be resilient enough to adjust to their lives to some degree.²³ A review research also yielded that a substantial proportion of trauma survivors showed a steep decline in PTSD rates without treatment in the first year.²⁴ Nonetheless, prior research has demonstrated that the trajectories of psychological symptoms are represented by various patterns, such as chronic or recovery groups.²⁵ Inasmuch as this study focused on examining differences between paired values, it would be worth ascertaining patterns of trajectories of PTSD symptoms among disaster victims with severe traumatic experiences in future studies using a large sample.

However, the decline in PTSD symptoms during the proximal period reflects the importance of early intervention for disaster victims. Several studies have demonstrated the effectiveness of early treatment (such as psychological debriefing, cognitive behavioral therapy, or eye movement desensitization and reprocessing) in reducing PTSD symptoms.^24,26 Although some individuals can recover without psychological intervention, others who are vulnerable to trauma may require help to alleviate consequential distress.

This study had several limitations. First, the classifications of trauma exposure used in this study were drawn from disaster victims, which limits their generalizability to other trauma-exposed groups. Disaster-related trauma may be a one-off event with a relatively short duration, whereas trauma events due to motor vehicle accidents or combat exposure can occur simultaneously and last for a long time. For example, physical wounds or injuries are highly likely to co-occur with a perceived threat to life in individuals involved in motor vehicle accidents, which may increase the proportion of multi-threat groups.²⁷ Combat exposure is more likely to include various trauma, such as direct or indirect threats to one’s life and moral injury; its frequency can also vary depending on the combat mission.³ Thus, future studies should consider the characteristics of trauma and its duration, frequency, and intensity to classify the patterns of trauma exposure.

Second, we analyzed changes in PTSD, depressive, and anxiety symptoms based on the classes drawn from the LCA. However, changes in psychological symptoms may have been influenced by social support or therapeutic interventions, which were not considered in this study. For example, a meta-analysis revealed that social support significantly decreased PTSD symptoms.²⁸ As described above, trauma-focused clinical interventions are also effective in alleviating psychological symptoms.²⁶ Thus, these protective or therapeutic variables should be considered in future studies to examine the relationship between patterns of trauma exposure and mental health outcomes.

Third, PTSD, depressive, and anxiety symptoms were compared over just 3 years in this study, which may limit our understanding of the longitudinal trajectories of the symptoms following disaster-related trauma. Prior research has revealed that some individuals who have been exposed to trauma show a delayed-onset trajectory of psychological symptoms, indicating that PTSD or anxiety symptoms begin to increase at a certain point.²⁹ Thus, future studies need to examine the long-term trajectories of dysfunctional psychological symptoms to extend our understanding of the impact of disaster-related trauma.

Conclusion

Despite these limitations, our results demonstrate that patterns of trauma exposure can vary depending on its intensity and number. In particular, more multiple threats to life can lead to higher levels of PTSD, depressive, and anxiety symptoms among disaster victims. Therefore, public health professionals or service providers should be alert to victims who are exposed to multiple types of trauma. In addition, the rapid decrease in PTSD symptoms during the proximal period indicates the importance of early intervention. It is notable that the number of natural disasters increased from 1990 to 2021 by and large, and the largest number of those occurred in Asia-Pacific area as of 2021.³⁰ Given the continuous global warming and the consequent climate change, disaster victims in Asia-Pacific area are highly likely to increase. In this sense, the findings of this study focusing on multiple types of trauma and early intervention can be used as important information for public health-related programs or policy.

Supplemental Material

sj-docx-1-aph-10.1177_10105395241240965 – Supplemental material for Classifications of Trauma Exposure and Their Associations With 3-Year Follow-up Psychological Symptoms Among Korean Victims of Disaster: A Latent Class Analysis

Supplemental material, sj-docx-1-aph-10.1177_10105395241240965 for Classifications of Trauma Exposure and Their Associations With 3-Year Follow-up Psychological Symptoms Among Korean Victims of Disaster: A Latent Class Analysis by Hyunyup Lee and Sungrok Kang in Asia Pacific Journal of Public Health

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by 2023 research fund of Korea Military Academy (Hwarangdae Research Institute, RN:2023B1026) and Establishment of Relief Service for disaster victims, National Disaster Management Research Institute, Republic of Korea.

ORCID iD

Hyunyup Lee

Supplemental Material

Supplemental material for this article is available online.

References

Furr

Comer

Edmunds

Kendall

PC.

Disasters and youth: a meta-analytic examination of posttraumatic stress. J Consult Clin Psychol. 2010;78(6):765-780. doi:10.1037/a0021482.

Rubens

Felix

Hambrick

EP.

A meta-analysis of the impact of natural disasters on internalizing and externalizing problems in youth. J Trauma Stress. 2018;31(3):332-341. doi:10.1002/jts.22292.

Lee

Aldwin

Kang

Do different types of war stressors have independent relations with mental health? Findings from the Korean Vietnam Veterans Study. Psychol Trauma. 2022;14(5):840-848. doi:10.1037/tra0000557.

Neria

Nandi

Galea

Post-traumatic stress disorder following disasters: a systematic review. Psychol Med. 2008;38(4):467-480. doi:10.1017/S0033291707001353.

O’Donnell

Schaefer

Varker

, et al. A systematic review of person-centered approaches to investigating patterns of trauma exposure. Clin Psychol Rev. 2017;57:208-225. doi:10.1016/j.cpr.2017.08.009.

Maguen

Vogt

King

, et al. The impact of killing on mental health symptoms in Gulf War veterans. Psychol Trauma: Theory Res Pract Policy. 2011;3(1): 21-26. doi:10.1037/a0019897.

Shea

Presseau

Finley

Reddy

Spofford

Different types of combat experiences and associated symptoms in OEF and OIF National Guard and reserve veterans. Psychol Trauma. 2017;9(suppl 1):19-24. doi:10.1037/tra0000240.

Felix

Binmoeller

Nylund-Gibson

Benight

Benner

Terzieva

Addressing disaster exposure measurement issues with latent class analysis. J Trauma Stress. 2019;32(1):56-66. doi:10.1002/jts.22368.

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Publishing; 2013.

10.

Armenta

Walter

Geronimo-Hara

, et al. Longitudinal trajectories of comorbid PTSD and depression symptoms among U.S. Service Members and Veterans. BMC Psychiatry. 2019;19:396. doi:10.1186/s12888-019-2375-1.

11.

Solomon

Horesh

Ginzburg

Trajectories of PTSD and secondary traumatization: a longitudinal study. J Psychiatr Res. 2021;138:354-359. doi:10.1016/j.jpsychires.2021.03.027.

12.

Santiago

Ursano

Gray

, et al. A systematic review of PTSD prevalence and trajectories in DSM-5 defined trauma exposed populations: intentional and non-intentional traumatic events. PLoS One. 2013;8(4):e59236. doi:10.1371/journal.pone.0059236.

13.

National Disaster Management Research Institute. An introduction to panel data for disaster victims. Accessed August 1, 2022. https://www.ndmi.go.kr/research/migration/migration.jsp?link=7#.

14.

Lim

Woo

Kim

, et al. Reliability and validity of the Korean version of the Impact of Event Scale-Revised. Compr Psychiatry. 2009;50(4):385-390. doi:10.1016/j.comppsych.2008.09.011.

15.

Kroenke

Spitzer

Williams

Löwe

The patient health questionnaire somatic, anxiety, and depressive symptom scales: a systematic review. Gen Hosp Psychiatry. 2010;32(4):345-359. doi:10.1016/j.genhosppsych.2010.03.006.

16.

Pfizer Inc. PHQ and GAD-7 screeners. Accessed August 1, 2022. http://www.phqscreeners.com.

17.

Nylund-Gibson

Choi

AY.

Ten frequently asked questions about latent class analysis. Transl Issues Psychol Sci. 2018;4(4):440-461. doi:10.1037/tps0000176.

18.

Nylund

Asparouhov

Muthén

BO.

Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Struct Equ Modeling. 2007;14(4):535-569. doi:10.1080/10705510701575396.

19.

Lakens

Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs. Front Psychol. 2013;4:863. doi:10.3389/fpsyg.2013.00863.

20.

Steine

Winje

Krystal

, et al. Cumulative childhood maltreatment and its dose-response relation with adult symptomatology: findings in a sample of adult survivors of sexual abuse. Child Abuse Negl. 2017;65:99-111. doi:10.1016/j.chiabu.2017.01.008.

21.

Johnson

Thompson

The development and maintenance of post-traumatic stress disorder (PTSD) in civilian adult survivors of war trauma and torture: a review. Clin Psychol Rev. 2008;28(1):36-47. doi:10.1016/j.cpr.2007.01.017.

22.

Kaysen

Rosen

Bowman

Resick

PA.

Duration of exposure and the dose-response model of PTSD. J Interpers Violence. 2010;25(1):63-74. doi:10.1177/08862605083291.

23.

Bonanno

GA.

Loss, trauma, and human resilience: have we underestimated the human capacity to thrive after extremely aversive events?

Am Psychol. 2004;59(1):20-28. doi:10.1037/0003-066X.59.1.20.

24.

Ehlers

Clark

Early psychological interventions for adult survivors of trauma: a review. Biol Psychiatry. 2003;53(9):817-826. doi:10.1016/S0006-3223(02)01812-7.

25.

Steenkamp

Dickstein

Salters Pedneault

-K

Hofmann

Litz

BT.

Trajectories of PTSD symptoms following sexual assault: is resilience the modal outcome? J. Trauma. Stress. 2012;25(4):469-474. doi:10.1002/jts.21718.

26.

Roberts

Kitchiner

Kenardy

Lewis

Bisson

JI.

Early psychological intervention following recent trauma: a systematic review and meta-analysis. Eur J Psychotraumatol. 2019;10(1):1695486. doi:10.1080/20008198.2019.1695486.

27.

Marasini

Caleffi

Machado

Pereira

BM.

Psychological consequences of motor vehicle accidents: a systematic review. Transp Res F: Traffic Psychol Behav. 2022;89:249-264. doi:10.1016/j.trf.2022.06.017.

28.

Wang

Chung

Wang

Kenardy

Social support and posttraumatic stress disorder: a meta-analysis of longitudinal studies. Clin Psychol Rev. 2021;85:101998. doi:10.1016/j.cpr.2021.101998.29.

29.

Galatzer-Levy

Huang

Bonanno

GA.

Trajectories of resilience and dysfunction following potential trauma: a review and statistical evaluation. Clin Psychol Rev. 2018;63:41-55. doi:10.1016/j.cpr.2018.05.008.

30.

Institute for Economics Peace. Ecological Threat Report 2022: Analysing Ecological Threats, Resilience & Peace. Sydney. Accessed December 26, 2023. http://visionofhumanity.org/resources.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB