Latent Profile Analysis of Resilience and Its Association With Psychological Factors in Patients Diagnosed With Depression and Anxiety Disorders

Article information

Psychiatry Investig. 2025;22(8):949-959

Publication date (electronic) : 2025 August 7

doi : https://doi.org/10.30773/pi.2025.0018

Mi-Sun Lee ¹^,²^,^*

, Hyu Jung Huh ³^,⁴^,^*

, Jeong-Ho Chae^,⁴^,⁵

¹Faculty of Psychology, University of Warsaw, Warsaw, Poland

²Department of Preventive Medicine, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

³Department of Psychiatry, Incheon St. Mary’s Hospital, The Catholic University of Korea, Incheon, Republic of Korea

⁴Department of Psychiatry, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

⁵Department of Psychiatry, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

Correspondence: Jeong-Ho Chae, MD, PhD Department of Psychiatry, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 06591, Republic of Korea Tel: +82-2-2258-6083, Fax: +82-2-2258-3870, E-mail: alberto@catholic.ac.kr

*These authors contributed equally to this work.

Received 2025 January 23; Revised 2025 May 3; Accepted 2025 June 12.

Abstract

Objective

This study aimed to classify the level of resilience among outpatients and investigate the relationship of resilience with depression, state anxiety, and psychological factors.

Methods

A total of 1,498 outpatients were recruited from a university hospital in Korea. The latent profile analysis of the resilience factor was identified using the R-based Jamovi 2.3.24 software. One-way analysis of variance was used to compare the differences between depression, state anxiety, and psychological factors; Scheffe’s test was used to conduct multiple comparisons.

Results

Three latent profiles were identified, including the high level of resilience (Class 1, 19.7%), the moderate level of resilience (Class 2, 47.9%), and the low level of resilience (Class 3, 32.4%). Depression and state anxiety were higher in Class 3 than Class 1 and 2. In analyzing Class 1, individuals with depressive and anxiety symptoms scored higher on anger rumination compared with those without symptoms, but there were no differences in cognitive emotion regulation. Childhood emotional neglect was higher for individuals with depressive symptoms compared to those without symptoms in Class 1.

Conclusion

This study provides an in-depth understanding of resilience and insights into the association between resilience, depression, anxiety, and psychological factors. It is necessary to provide sufficient support and interventions to regulate anger rumination and emotional factors among outpatients in Class 1 with depression and state anxiety symptoms.

Keywords: Resilience; Depression; Anxiety; Anger rumination; Childhood trauma; Latent profile analysis

INTRODUCTION

Resilience is an adaptive and dynamic process that helps an individual maintain normal physiological and psychological functions. It enables individuals to effectively self-regulate and recover from severe stress or trauma, thereby preventing or reducing mental health problems [1,2]. Resilience may decrease depression and anxiety by promoting behaviors such eISSN 1976-3026 OPEN ACCESS as tolerance, responsibility, self-examination, adaptability to various circumstances, and an optimistic perspective on reality [3,4]. Resilience has been identified as the primary factor that contributes to a higher quality of life in patients [5,6].

There has been increasing interest in determining the psychological factors that may enhance resilience in patients [7]. Prior research has demonstrated that a high level of resilience can help patients face illnesses optimistically, actively participate in rehabilitation and treatment, and alleviate depression and anxiety symptoms [4,8,9]. To enhance the quality of life of patients with mental health problems, a deeper understanding of resilience, which is associated with various psychosocial factors, is needed [7].

Resilience is a very complex concept and has different definitions depending on how it is conceptually framed [10-12]. Some researchers define resilience as the absence of psychopathology after stress or trauma, while others focus on the character-istics and strengths of individuals that enable them to recover and grow after stress or trauma [12-16]. Research on the relationship between resilience and psychopathology may differ depending on how resilience is defined. Individuals diagnosed with depression or anxiety disorders may be considered to be less resilient in the sense that they have psychopathology [16,17]. However, from the perspective of resilience, which is based on individual traits and strengths that enable recovery and growth, it is possible to characterize people with psychopathology as having resilient traits. In other words, people with depression and anxiety disorders may not be objectively resilient due to their psychiatric symptoms, but may subjectively rate themselves as resilient. This aspect should be considered in the interpretation of research findings on resilience, especially in studies using self-report scales. Furthermore, when analyzing the relationship between resilience and psychiatric symptoms such as depression and anxiety, it would be valuable to explore the characteristics of individuals who subjectively report themselves as resilient, but who have psychiatric symptoms (and are not resilient by objective standards), given the complexity of the concept of resilience.

Most self-reported questionnaires on resilience factors focus on evaluating the general level of resilience [18-20]. Regarding outpatients, there is heterogeneity in their resilience and mental health. Since patients have different demographic characteristics and resilience statuses, their profiles may be not equal [20]. The precise mechanisms through which variations in resilience levels among patients impact psychological factors remain unclear and it is needed to classify the level of resilience specifically [1,18]. Further study is needed to examine the relationship between resilience and psychological factors in clinical samples.

Latent profile analysis (LPA) is a systematic method that focuses on individuals and can identify distinct groups of people with similar characteristics [21,22]. Individuals within the subgroups exhibit similar characteristics, while the characteristics are significantly different between subgroups [23,24]. Person-centered methods could identify the unobserved heterogeneity within the clinical population and generate categories and characteristics among outpatients [25]. It may provide valuable insights into such associations between resilience levels and various psychological factors.

This study conducted two main investigations. In the first study, we used LPA to categorize patients with depressive and anxiety disorders based on the subfactors of the self-reportbased resilience scale and compared the differences in depressive and anxiety symptoms following the resilience-based categorization. In the second study, we explored the characteristics of a group of individuals who subjectively reported themselves as highly resilient but were objectively not resilient due to psychopathology based on cluster analysis. Although individuals in this group subjectively reported high levels of resilience, the presence of psychiatric symptoms suggests that various factors may impede the effective manifestation of their resilience. Previous studies have identified such factors, including environmental adversities, tendencies to attribute problems externally, resulting anger rumination, and maladaptive forms of emotion regulation [26-30]. Based on these findings, the present study aimed to explore childhood trauma, maladaptive cognitive emotion regulation strategies, and anger rumination as potential psychological vulnerabilities that may hinder the functional expression of resilience. In this context, we examined differences in anger rumination, cognitive emotion regulation strategies, and childhood maltreatment experiences according to the severity of depressive and anxiety symptoms in the group that was subjectively resilient but not objectively resilient.

METHODS

Participants

We recruited participants who were treated at the Mood and Anxiety Disorders Unit at Seoul St. Mary’s Hospital, The Catholic University of Korea from September 2009 to November 2021.

A total of 1,498 patients were recruited. The inclusion criteria were as follows: 1) age ≥18 years, 2) diagnosed with depressive and/or anxiety disorder based on the Diagnostic and Statistical Manual of Mental Disorders, 4th ed (DSM-IV), and 3) provided informed consent for participation in the study. The exclusion criteria were as follows: 1) diagnosed with severe physical illnesses or lifetime psychiatric diagnosis (bipolar disorder, psychotic disorder, any mental disorder, or mental retardation), 2) engaged in other treatments or clinical trials that could potentially influence the study, and 3) exhibited communication difficulties.

After obtaining informed consent, the researchers provided questionnaires to the patients in person. Participants completed the questionnaires regarding their psychiatric status at their initial visit of Mood and anxiety disorders Unit.

Measurements

General demographic variables

We collected general demographic data using self-report questionnaires, which included age, sex, education level (≥college, ≤high school), marital status (unmarried, married, divorced/separated/others), employment status (employed, unemployed), monthly household income (<1,000,000, ≥1,000,000 to <3,000,000, ≥3,000,000 to <5,000,000, ≥5,000,000 KRW), current smoking (yes or no), and current drinking (yes or no).

Connor-Davidson resilience scale

Resilience was assessed using the Connor-Davidson resilience scale (CD-RISC). This scale is the most frequently used and widely accepted resilience scale and has been shown to have psychometric properties [7]. The CD-RISC consists of 25 items, each evaluated on a 5-point scale ranging from 0 to 4. Higher scores indicate higher levels of resilience. This scale consists of five subfactors: 1) notion of personal competence, high standards, and tenacity; 2) trust in one’s instincts, tolerance of negative affect, and strengthening effects of stress; 3) positive acceptance of change, and secure relationships; 4) control; and 5) spiritual influences. In this study, Cronbach’s α coefficient for CD-RISC was 0.944.

Beck Depression Inventory

Depression was assessed using the Beck Depression Inventory (BDI), and we used the standardized and validated Korean version scale [31]. The BDI includes 21 items, each rated on a 4-point Likert scale ranging from 0 to 3, with higher scores indicating higher depressive symptoms. The total score ranges from 0 to 63 and is classified into normal (0–9), mild (10–15), moderate (16–23), and severe depressive symptoms (24–63). This study showed a reliability coefficient of 0.839.

State Anxiety Inventory

State anxiety was assessed using the State Anxiety Inventory (SAI) [32,33]. The SAI consists of 40 items, developed to examine the dimensions of state anxiety. The scores range from 20 to 80, with higher scores indicating increased state anxiety levels [32]. The total SAI scores are classified into normal (<52), mild (52–56), moderate (57–61), and severe anxiety symptoms (≥62). This study showed a reliability coefficient of 0.947.

Korean version of the Anger Rumination Scale

Anger rumination was evaluated using the Korean version of the Anger Rumination Scale (K-ARS) [34]. The K-ARS includes 19 items, each rated on a 4-point Likert scale ranging from 1 to 4, with higher scores indicating higher anger rumination. The total score ranges from 19 to 76 [34]. This scale consists of three subfactors, which are as follows: 1) angry memories, 2) thoughts of revenge, and 3) understanding of causes. This study showed a reliability coefficient of 0.957.

Korean version of the Childhood Trauma Questionnaire

Childhood trauma was assessed using the Korean version of the Childhood Trauma Questionnaire (K-CTQ).35 The sub-categories of childhood trauma include five types: emotional abuse, emotional neglect, physical abuse, physical neglect, and sexual abuse. The K-CTQ includes 28 items, each rated on a 5-point scale. The scores for the five trauma subcategories range from 5 to 25, respectively [35]. Higher scores indicate higher levels of childhood trauma. This study showed a reliability coefficient of 0.847.

Korean version of the Cognitive Emotion Regulation Questionnaire

Cognitive emotion regulation was assessed using the Korean version of the Cognitive Emotion Regulation Questionnaire (K-CERQ) [36]. The K-CERQ comprises 36 items (5-point scale) and ranges each score for nine sub-factors from 4 to 20 [37]. This scale consists of nine subfactors, with higher scores indicating increased cognitive emotion regulation [36]. This study showed a reliability coefficient of 0.853.

Statistical analysis

Study 1

LPA is a statistical method used to examine the association between continuous-type indicators and potential categorical variables [38]. The latent profile statistical method, which uses multiple variables, can effectively identify subgroups characterized by similar psychological patterns [39].

LPA was analyzed using the R-based Jamovi 2.3.24 software (https://www.jamovi.org/). The models with one to five profiles were identified. The Akaike information criterion (AIC), Bayesian information criteria (BIC), likelihood-ratio statistic (G2), and entropy value (≥0.70 was acceptable) as the criteria was applied to identify the optimal number of LPA [22]. The classification accuracy was evaluated using the relative entropy, which was reported as values that ranged from 0.0 to 1.0 [39]. Higher values of relative entropy indicate greater accuracy [25]. Finally, the selection model took into consideration both the class distinction and the interpretability of the plot. According to the model fit statistics, a three-profile provided the most accurate fit (G2=-18,954, AIC=37,952, BIC=38,069, entropy=0.868, p<0.001). The identification of the latent profile and statistics of model fit statistics are shown in Supplementary Table 1.

Descriptive statistics and χ² test were used to examine the overall demographic characteristics of participants according to the three latent profiles. The BDI and SAI scores of the three latent profiles were compared using one-way analysis of variance (ANOVA). Scheffe’s test was used for conducting multiple comparisons.

Study 2

To explore the characteristics of patients who subjectively report themselves as having high levels of resilience, but who have psychiatric symptoms (subjectively resilient-objectively not resilient), Class 1 (high level of resilience) was re-classified into sub-groups by the presence or absence of depression and anxiety symptoms. Differences in childhood trauma (K-CTQ) and psychological factors such as anger rumination (K-ARS) and cognitive emotion regulation strategy (CERQ) for each sub-group were analyzed through an independent samples ttest. A p-value <0.05 was considered statistically significant.

Ethics

All participants provided informed consent. The study was performed following the ethical standards specified in the Declaration of Helsinki. This study was approved by the Institutional Review Board of the Ethics Committee of Seoul St. Mary’s Hospital at The Catholic University of Korea (KC09FZZZ0211).

RESULTS

Study 1

Table 1 shows the general characteristics of the 1,498 participants. The average age was 35.58 years (standard deviation [SD]=12.86), and 56.6% were female. In the total group, 58.7% reported a college graduate degree or higher, 39.7% were married, and 34.5% were currently working. Approximately 63.8% of participants had an income of more than 3 million won per month; 18.8% currently smoke and 64.4% currently drink alcohol. The mean BDI score was 24.34±12.29, the mean SAI score was 56.84±13.33, the mean K-CERQ score was 102.76± 17.44, the mean K-ARS score was 47.84±14.12, and the mean K-CTQ score was 54.60±14.42.

Table 1.

Demographic characteristics of participants (N=1,498)

Table 2 lists the estimated probabilities of CD-RISC by LPA. Class 1 (n=293, 19.7%) was the group with the highest level of probability of resilience. Class 2 (n=713, 47.9%) was the group with the moderate level of probability of resilience. Class 3 (n=483, 32.4%) was defined as the group with the lowest level of probability of resilience. The CD-RISC items for each class classified based on LPA are shown in Supplementary Table 2. Figure 1 shows the latent three-profile plot according to CD-RISC scores. Pearson correlation matrix showed that CD-RISC was significantly correlated with each sub-factor (p<0.05). Figure 2 presents the correlation plot for a mixture model.

Table 2.

Three-profiles model: estimated probabilities by latent profile analysis

Figure 1.

Latent profile plot of resilience (CD-RISC). CD-RISC_Sub factor_01, notion of personal competence, high standards, and tenacity; CD-RISC_Sub factor_02, trust in one’s instincts, tolerance of negative affect, and strengthening effects of stress; CD-RISC_Sub factor_03, positive acceptance of change, and secure relationships; CD-RISC_Sub factor_04, control; CD-RISC_Sub factor_05, spiritual influences; CD-RISC, Connor-Davidson resilience scale.

Figure 2.

Correlation plot for a mixture model. CD-RISC_Sub factor_01, notion of personal competence, high standards, and tenacity; CDRISC_ Sub factor_02, trust in one’s instincts, tolerance of negative affect, and strengthening effects of stress; CD-RISC_Sub factor_03, positive acceptance of change, and secure relationships; CD-RISC_Sub factor_04, control; CD-RISC_Sub factor_05, spiritual influences; CDRISC, Connor-Davidson resilience scale.

Table 3 presents the demographic characteristics and psychological factors by three latent profiles. The mean±SD of CD-RISC of Class 1, 2, and 3 were 76.72±9.05, 49.76±7.52, and 25.40±8.76, respectively. In Class 1, 54.1% were female (mean age=38.79 years, SD=12.14). In this group, 68.2% graduated from college or above, 54.4% were married, 51.4% were employed, 15.9% reported currently smoking, and 67.4% reported currently drinking. Additionally, 36.5% had moderate-severe depressive symptoms and 18.4% had moderate-severe anxiety symptoms. In Class 2, 57.6% were female (mean age=36.85 years, SD=13.17); 60.3% graduated from college or above, 42.9% were married, 34.2% were employed, 18.5% reported currently smoking, and 62.7% were currently drinking. Additionally, 73.1% had moderate-severe depressive symptoms and 49.3% had moderate-severe anxiety symptoms. In Class 3, 56.5% were female (mean age=31.77 years, SD=11.90). Additionally, 50.4% graduated from college or above, 25.9% were married, 24.6% were employed, 21.0% reported currently smoking, and 65.1% were currently drinking. Moreover, 96.7% had moderate-severe depressive symptoms and 81.0% had moderate-severe anxiety symptoms.

Table 3.

Demographic characteristics, depression, and anxiety by three latent profiles

Table 4 lists the results of one-way ANOVA according to LPA. One-way ANOVA showed significant differences in BDI and SAI scores between the three patterns. In a post hoc analysis (in Scheffe’s test), BDI and SAI scores showed the highest values in Class 3 and the differences was statistically significant (p<0.001).

Table 4.

The results of one-way analysis of variance according to the latent profile analysis

Study 2

Table 5 shows the comparison of psychological factors according to the presence or absence of depression and state anxiety symptoms in the Class 1 group (high level of resilience). Upon comparing psychological factors by reclassifying depression into normal-mild and moderate-severe levels in Class 1, there were statistically significant differences in all subfactors of K-ARS and emotional neglect and emotional abuse factors in K-CTQ. Upon comparing psychological factors by reclassifying anxiety into normal-mild and moderate-severe levels in Class 1, there were statistically significant differences in all subfactors of K-ARS. However, there were no statistical-ly significant differences in K-CTQ and K-CERQ.

Table 5.

Comparison of psychological factors according to the presence or absence of depression and state anxiety symptoms among the Class 1 group (high level of resilience)

DISCUSSION

This study identified three latent profiles of resilience and investigated the association between the latent profiles and psychological factors in clinical samples in Korea. Our results identified the three latent patterns of resilience: “high level of resilience,” “moderate level of resilience,” and “low level of resilience.” These findings have several implications and can be summarized as follows.

First, in the clinical population, resilience was associated with socioeconomic factors such as education, employment, and marital status. This is consistent with the findings of several previous studies [40-42]. The Connor-Davison scale used in this study reflects various forms of psychological resources, such as conscientiousness, self-regulation, emotional regulation, and trust and acceptance in interpersonal relationships. However, whether an individual’s psychological resources related to resilience are sequentially influenced by an individual’s higher education, economic well-being, and marital status or whether higher education, economic well-being, and maritalbased social support contribute to higher levels of resilience is unknown. Further research is needed to clarify the relationship between resilience and socioeconomic factors, including more in-depth analyses of the role of socioeconomic factors in the process of bouncing back from stress after adversity. However, the results of this study suggest that resilience is related to socioeconomic factors and thus interventions should take into account the impact of socioeconomic factors in the process of coping with various stresses and adversities. In particular, when aiming to enhance resilience among patients with depressive and anxiety disorders, it is important to consider strategies beyond merely strengthening personal coping resources. Practical support, such as counseling for education and employment, and facilitating access to social support networks or welfare systems, should also be considered. Interventions that incorporate socioeconomic factors in this manner may be especially effective in promoting resilience among individuals with socioeconomic vulnerabilities.

Additionally, levels of resilience were associated with the severity of depressive and anxiety symptoms: individuals who reported lower levels of resilience had more depressive and anxiety symptoms, while those who reported higher levels of resilience had relatively fewer depressive and anxiety symptoms. These findings are consistent with numerous previous studies that have explored the relationship between resilience and psychopathology, suggesting that resilience, the psychological resources that enable people to cope with stress and adversity, may be a protective factor against depressive and anxiety symptoms [43-47]. These psychological resources are thought to help regulate depressive and anxiety symptoms by contributing to emotion regulation: after an emotional response to a stressful stimulus, a series of emotion regulation strategies, such as situational change, attention deployment, reappraisal, and response modification, help regulate the emotional response and contribute to recovery from various adversities and stresses [10,48]. From this perspective, it can be inferred that part of resilience reflects emotional regulation, which may have contributed to the reduction of depressive and anxiety symptoms. These findings suggest that interventions that promote resilience in general will be beneficial in reducing depressive and anxiety symptoms in clinical populations.

Finally, this study explored the characteristics of the relatively high resilience group (Class 1) based on cluster analysis. The CD-RISC, a self-report measure of resilience, was used to measure resilience. Patients objectively considered with low resilience may actually rate themselves as resilient, i.e., they may be considered resilient from a subjective perspective. The mean CD-RISC score for the Class 1 group in this study was 76.72, which is above the top 25th percentile according to previous studies and is relatively high within the overall population [49]. However, these individuals may be objectively considered less resilient because of their depressive or anxiety symptoms. In this study, we aimed to characterize patients who subjectively report relatively high resilience but have low resilience in terms of psychiatric symptoms such as depression or anxiety. To explore this issue, we performed additional analyses to compare the characteristics of two groups within the relatively high resilience group based on the presence or absence of depressive and anxiety symptoms. The results showed that among patients who subjectively reported relatively high resilience, those with depressive and anxiety symptoms did not differ in cognitive emotion regulation compared with those without, but these individuals had significantly higher anger rumination. Additionally, those with depressive symptoms had higher childhood emotional neglect scores compared with individuals without these symptoms. A possible explanation for these findings is that although individuals report having many psychological capacities to bounce back from stress and adversity, those with psychiatric symptoms may have externalized rather than internalized attributions for their current difficulties, and the anger and rumination that result from externalizing attributions may have contributed to the severity of their psychiatric symptoms. It is possible to infer, with caution, that higher levels of childhood emotional neglect among individuals with depressive symptoms, despite their reporting relatively higher subjective resilience, may also be because they attribute their difficulties to environmental factors rather than personal characteristics. These findings suggest that patients who are subjectively resilient but not objectively resilient, and who exhibit psychiatric symptoms, may need interventions that go beyond simply building their personal capacity to cope with stress or adversity, to fostering an accepting attitude through a wiser and more balanced understanding of their situation. Future research will need to analyze the characteristics of these groups in greater depth.

This study has several limitations. First, this was a cross-sectional study that assessed measurements concurrently. There-fore, the relationship between socioeconomic factors, resilience, and psychopathology should be interpreted with caution as causal. More extensive and detailed longitudinal studies of socioeconomic and parenting environments, the development of resilience, and the occurrence of psychopathology in adulthood will be needed to clarify these causal relationships. Second, this study used a self-reported questionnaire, and there is a possibility for social desirability bias or retrospective recall bias. In addition, psychiatric symptoms such as depression and anxiety may have affected the results of self-reported assessments. For example, individuals experiencing depression may have rated themselves more negatively. Third, we could not apply a standardized instrument for evaluating and excluding comorbid personality disorders. The subjective impression of investigators may have influenced the exclusion process of patients with comorbid personality disorders, constituting a limitation of the present study. Fourth, depressive and anxiety symptoms as well as their relationship with resilience may have been influenced by socioeconomic factors such as income, educational level, employment status, and marital status. However, these influences were not fully controlled for, given the cross-sectional design of the study. Understanding how socioeconomic factors impact resilience is crucial for informing future interventions aimed at enhancing resilience. Therefore, future research using more sophisticated study designs and statistical methods, such as multinomial logistic regression with R3STEP to evaluate whether socioeconomic variables predict resilience profiles, is necessary to better elucidate the relationship between resilience and socioeconomic factors.

In conclusion, we classified the three levels of resilience factors among clinical samples in Korea. Our findings showed how the three latent patterns that emerged from resilience influenced depression, state anxiety symptoms, and psychological factors. Compared with Class 1 (high level of resilience), Class 2 (moderate level of resilience) and Class 3 (low level of resilience) were associated with depression and state anxiety. Furthermore, the present findings suggest that anger ruminations and childhood traumatic experiences may be important factors, particularly in populations that subjectively report high levels of resilience but have depressive and anxiety symptoms. Clinical populations that report high subjective resilience but have psychiatric symptoms may require interventions that take these characteristics into account.

When identifying high-risk groups, clinicians and professionals should take consideration data relating to patterns of resilience, depression, state anxiety, and other psychological factors. Given the limitations of subjectively measured resilience and the complex nature of resilience, interventions will need to be tailored to individual characteristics.

Supplementary Materials

The Supplement is available with this article at https://doi.org/10.30773/pi.2025.0018.

Supplementary Table 1.

Latent profile model identification and model fit statistics

pi-2025-0018-Supplementary-Table-1.pdf

Supplementary Table 2.

CD-RISC items for each class classified based on latent profile analysis

pi-2025-0018-Supplementary-Table-2.pdf

Notes

Availability of Data and Material

The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.

Conflicts of Interest

The authors have no potential conflicts of interest to disclose.

Author Contributions

Conceptualization: all authors. Data curation: all authors. Formal analysis: Mi-Sun Lee. Funding acquisition: Jeong-Ho Chae. Investigation: all authors. Methodology: Mi-Sun Lee. Project administration: Jeong-Ho Chae. Resources: Jeong-Ho Chae. Software: Mi-Sun Lee. Supervision: Jeong-Ho Chae. Validation: all authors. Visualization: Mi-Sun Lee. Writing—original draft: Mi-Sun Lee, Hyu Jung Huh. Writing—review & editing: all authors.

Funding Statement

None

Acknowledgments

None

References

1. Zhang X, Zhang H, Zhang Z, Fan H, Li S. The mediating effect of resilience on the relationship between symptom burden and anxiety/depression among Chinese patients with primary liver cancer after liver resection. Patient Prefer Adherence 2023;17:3033–3043.

2. Rutter M. Annual research review: resilience--clinical implications. J Child Psychol Psychiatry 2013;54:474–487.

3. Southwick SM, Charney DS. The science of resilience: implications for the prevention and treatment of depression. Science 2012;338:79–82.

4. González-Flores CJ, García-García G, Lerma A, Pérez-Grovas H, Meda-Lara RM, Guzmán-Saldaña RME, et al. Resilience: a protective factor from depression and anxiety in Mexican dialysis patients. Int J Environ Res Public Health 2021;18:11957.

5. Harms CA, Cohen L, Pooley JA, Chambers SK, Galvão DA, Newton RU. Quality of life and psychological distress in cancer survivors: the role of psycho-social resources for resilience. Psychooncology 2019;28:271–277.

6. Boškailo E, Franjić D, Jurić I, Kiseljaković E, Marijanović I, Babić D. Resilience and quality of life of patients with breast cancer. Psychiatr Danub 2021;33(Suppl 4):572–579.

7. Mohlin Å, Axelsson U, Bendahl PO, Borrebaeck C, Hegardt C, Johnsson P, et al. Psychological resilience and health-related quality of life in Swedish women with newly diagnosed breast cancer. Cancer Manag Res 2020;12:12041–12051.

8. Kim J, Jeong HG, Lee MS, Pae CU, Patkar AA, Jeon SW, et al. Effect of frailty on depression among patients with late-life depression: a test of anger, anxiety, and resilience as mediators. Clin Psychopharmacol Neurosci 2024;22:253–262.

9. Chung DXY, Loo YE, Kwan YH, Phang JK, Woon TH, Goh WR, et al. Association of anxiety, depression and resilience with overall health and functioning in axial spondyloarthritis (axSpA): a cross-sectional study. BMJ Open 2023;13:e071944.

10. Troy AS, Willroth EC, Shallcross AJ, Giuliani NR, Gross JJ, Mauss IB. Psychological resilience: an affect-regulation framework. Annu Rev Psychol 2023;74:547–576.

11. Bhatnagar S. Rethinking stress resilience. Trends Neurosci 2021;44:936–945.

12. Sisto A, Vicinanza F, Campanozzi LL, Ricci G, Tartaglini D, Tambone V. Towards a transversal definition of psychological resilience: a literature review. Medicina (Kaunas) 2019;55:745.

13. Block J, Kremen AM. IQ and ego-resiliency: conceptual and empirical connections and separateness. J Pers Soc Psychol 1996;70:349–361.

14. Kalisch R, Müller MB, Tüscher O. A conceptual framework for the neurobiological study of resilience. Behav Brain Sci 2015;38:e92.

15. Masten AS. Resilience theory and research on children and families: Past, present, and promise. J Fam Theory Rev 2018;10:12–31.

16. Bonanno GA. Loss, trauma, and human resilience: have we underestimated the human capacity to thrive after extremely aversive events? Am Psychol 2004;59:20–28.

17. Waugh CE, Fredrickson BL, Taylor SF. Adapting to life’s slings and arrows: Individual differences in resilience when recovering from an anticipated threat. J Res Pers 2008;42:1031–1046.

18. Griffin KH, Johnson JR, Kitzmann JP, Kolste AK, Dusek JA. Outcomes of a multimodal resilience training program in an outpatient integrative medicine clinic. J Altern Complement Med 2015;21:628–637.

19. Morete MC, Solano JPC, Boff MS, Filho WJ, Ashmawi HA. Resilience, depression, and quality of life in elderly individuals with chronic pain followed up in an outpatient clinic in the city of São Paulo, Brazil. J Pain Res 2018;11:2561–2566.

20. Victor P, Teismann T, Willutzki U. [Fostering resilience while waiting for psychotherapy: evaluation of a group intervention in an outpatient care setting]. Psychother Psychosom Med Psychol 2016;66:486–488. German.

21. Yalçın İ, Can N, Mançe Çalışır Ö, Yalçın S, Çolak B. Latent profile analysis of COVID-19 fear, depression, anxiety, stress, mindfulness, and resilience. Curr Psychol 2022;41:459–469.

22. Lee MS, Lee H. Chronic disease patterns and their relationship with health-related quality of life in South Korean older adults with the 2021 Korean national health and nutrition examination survey: latent class analysis. JMIR Public Health Surveill 2024;10:e49433.

23. Mai Y, Wu YJ, Huang Y. What type of social support is important for student resilience during COVID-19? A latent profile analysis. Front Psychol 2021;12:646145.

24. Li J, Yin YZ, Zhang J, Puts M, Li H, Lyu MM, et al. A latent profile analysis of resilience and their relation to differences in sleep quality in patients with lung cancer. Support Care Cancer 2024;32:155.

25. Băjenaru L, Balog A, Dobre C, Drăghici R, Prada GI. Latent profile analysis for quality of life in older patients. BMC Geriatr 2022;22:848.

26. du Pont A, Rhee SH, Corley RP, Hewitt JK, Friedman NP. Rumination and psychopathology: are anger and depressive rumination differentially associated with internalizing and externalizing psychopathology? Clin Psychol Sci 2018;6:18–31.

27. Dvir Y, Ford JD, Hill M, Frazier JA. Childhood maltreatment, emotional dysregulation, and psychiatric comorbidities. Harv Rev Psychiatry 2014;22:149–161.

28. McLaughlin KA, Colich NL, Rodman AM, Weissman DG. Mechanisms linking childhood trauma exposure and psychopathology: a transdiagnostic model of risk and resilience. BMC Med 2020;18:96.

29. Watters ER, Aloe AM, Wojciak AS. Examining the associations between childhood trauma, resilience, and depression: a multivariate meta-analysis. Trauma Violence Abuse 2023;24:231–244.

30. Zhou HY, Zhou L, Zheng TX, Ma LP, Fan MX, Liu L, et al. Unraveling the link between childhood maltreatment and depression: Insights from the role of ventral striatum and middle cingulate cortex in hedonic experience and emotion regulation. Dev Psychopathol 2025;37:292–302.

31. Lee YH, Song JY. A study of the reliability and the validity of the BDI, SDS, and MMPI-D Scales. Korean J Clin Psychol 1991;10:98–113.

32. Shah NN, Schwandt ML, Hobden B, Baldwin DS, Sinclair J, Agabio R, et al. The validity of the state-trait anxiety inventory and the brief scale for anxiety in an inpatient sample with alcohol use disorder. Addiction 2021;116:3055–3068.

33. Julian LJ. Measures of anxiety: State-Trait Anxiety Inventory (STAI), Beck Anxiety Inventory (BAI), and Hospital Anxiety and Depression Scale-Anxiety (HADS-A). Arthritis Care Res (Hoboken) 2011;63 Suppl 11:S467–S472.

34. Lee GB, Cho HC. Validation study of the Korean version of the Anger Rumination Scale. Journal of Emotional & Behavioral Disorders 2008;24:1–22.

35. Kim D, Park SC, Yang H, Oh DH. Reliability and validity of the Korean version of the childhood trauma questionnaire-short form for psychiatric outpatients. Psychiatry Investig 2011;8:305–311.

36. Ahn HN, Lee NB, Joo HS. Validation of the Cognitive Emotion Regulation Questionnaire in a Korean population. Koran Journal of Counseling 2013;14:1773–1794.

37. Garnefski N, Kraaija V, Spinhoven P. Negative life events, cognitive emotion regulation and emotional problems. Pers Individ Dif 2001;30:1311–1327.

38. The jamovi project (2022). Jamovi (Version 2.3) [Internet]. Available at: https://www.jamovi.org. Accessed March 6, 2024.

39. Lee MS, Lee H. Latent class analysis of health behaviors, anxiety, and suicidal behaviors among Korean adolescents. J Affect Disord 2024;354:339–346.

40. Logie CH, Wang Y, Kazemi M, Hawa R, Kaida A, Conway T, et al. Exploring social ecological pathways from resilience to quality of life among women living with HIV in Canada. AIDS Care 2018;30:S67–S75.

41. Ong HL, Vaingankar JA, Abdin E, Sambasivam R, Fauziana R, Tan ME, et al. Resilience and burden in caregivers of older adults: moderating and mediating effects of perceived social support. BMC Psychiatry 2018;18:27.

42. Wong R, DeGraff DS, Orozco-Rocha K. Economic resources and health: a bi-directional cycle for resilience in old age. J Aging Health 2023;35:767–780.

43. Min JA, Jung YE, Kim DJ, Yim HW, Kim JJ, Kim TS, et al. Characteristics associated with low resilience in patients with depression and/or anxiety disorders. Qual Life Res 2013;22:231–241.

44. Min JA, Lee CU, Chae JH. Resilience moderates the risk of depression and anxiety symptoms on suicidal ideation in patients with depression and/or anxiety disorders. Compr Psychiatry 2015;56:103–111.

45. Min JA, Lee CU, Hwang SI, Shin JI, Lee BS, Han SH, et al. The moderation of resilience on the negative effect of pain on depression and posttraumatic growth in individuals with spinal cord injury. Disabil Rehabil 2014;36:1196–1202.

46. Min JA, Yu JJ, Lee CU, Chae JH. Cognitive emotion regulation strategies contributing to resilience in patients with depression and/or anxiety disorders. Compr Psychiatry 2013;54:1190–1197.

47. Lee JS, Ahn YS, Jeong KS, Chae JH, Choi KS. Resilience buffers the impact of traumatic events on the development of PTSD symptoms in firefighters. J Affect Disord 2014;162:128–133.

48. Gross JJ. Emotion regulation: Current status and future prospects. Psychol Inq 2015;26:1–26.

49. Baek HS, Lee KU, Joo EJ, Lee MY, Choi KS. Reliability and validity of the Korean version of the connor-davidson resilience scale. Psychiatry Investig 2010;7:109–15.

Article information Continued

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Variables	Value
Age (yr)	35.58±12.86
Sex (female)	56.6
Education level (≥college)	58.7
Marital status (married)	39.7
Employment status (employment)	34.5
Household income (≥3,000,000 won/month)	63.8
Current smoking	18.8
Current drinking	64.4
Depression (BDI)	24.34±12.29
State anxiety (SAI)	56.84±13.33

	Class 1 (N=293, 19.7%)	Class 2 (N=713, 47.9%)	Class 3 (N=483, 32.4%)	p
Age (yr)	38.79±12.14	36.85±13.17	31.77±11.90
Sex				0.598
Male	134 (45.9)	301 (42.4)	210 (43.5)
Female	158 (54.1)	409 (57.6)	273 (56.5)
Education level				<0.001
≥College	199 (68.2)	426 (60.3)	238 (50.4)
≤High school	93 (31.8)	281 (39.7)	234 (49.6)
Marital status				<0.001
Unmarried	112 (39.3)	348 (51.1)	312 (67.2)
Married	155 (54.4)	292 (42.9)	120 (25.9)
Divorced/separated/others	18 (6.3)	41 (6.0)	32 (6.9)
Employment status				<0.001
Employed	145 (51.4)	231 (34.2)	114 (24.6)
Unemployed	137 (48.6)	444 (65.8)	349 (75.4)
Household income (won/month)				<0.001
<1,000,000	18 (6.5)	61 (9.7)	61 (15.2)
≥1,000,000 to <3,000,000	60 (21.7)	156 (24.9)	116 (28.9)
≥3,000,000 to <5,000,000	71 (25.6)	156 (24.9)	97 (24.2)
≥5,000,000	128 (46.2)	253 (40.4)	127 (31.7)
Current smoking				0.208
No	24 4 (84.1)	550 (81.5)	373 (79.0)
Yes	46 (15.9)	125 (18.5)	99 (21.0)
Current drinking				0.343
No	93 (32.6)	264 (37.3)	166 (34.9)
Yes	192 (67.4)	443 (62.7)	319 (65.1)
Resilience (CD-RISC)	76.72±9.05	49.76±7.52	25.40±8.76
Depression (BDI)	13.68±8.64	22.44±10.23	33.62±10.30	<0.001
Normal	101 (34.5)	65 (9.1)	2 (0.4)
Mild depressive symptom	85 (29.0)	127 (17.8)	14 (2.9)
Moderate depressive symptom	69 (23.5)	215 (30.2)	70 (14.5)
Severe depressive symptom	38 (13.0)	306 (42.9)	397 (82.2)
State anxiety (SAI)	46.27±11.36	55.79±12.37	64.80±10.54	<0.001
Normal	202 (68.9)	253 (35.5)	54 (11.2)
Mild state anxiety	37 (12.6)	108 (15.2)	38 (7.9)
Moderate state anxiety	20 (6.8)	95 (13.3)	68 (14.1)
Severe state anxiety	34 (11.6)	256 (36.0)	323 (66.9)

Variables	Profile	M±SD	F (p)	Post hoc (Scheffe)
Depression (BDI)	Class 1 (a)	13.68±8.64	390.409 (<0.001)	c>b>a
	Class 2 (b)	22.44±10.23
	Class 3 (c)	33.62±10.30
State anxiety (SAI)	Class 1 (a)	46.27±11.36	237.919 (<0.001)	c>b>a
	Class 2 (b)	55.79±12.37
	Class 3 (c)	64.80±10.54

	Depression (BDI)		State anxiety (SAI)
	M±SD	t (p)	M±SD	t (p)
K-CERQ_Self-blame		0.525 (0.375)		0.567 (0.699)
None-mild	12.07±3.56		12.04±3.73
Moderate-severe	11.80±3.91		11.69±3.64
K-CERQ_Acceptance		0.439 (0.205)		1.840 (0.342)
None-mild	12.39±3.15		12.51±3.09
Moderate-severe	12.20±2.88		11.61±2.73
K-CERQ_Focus on thought/rumination		0.148 (0.863)		1.033 (0.622)
None-mild	13.53±3.47		13.63±3.54
Moderate-severe	13.46±3.61		13.04±3.43
K-CERQ_Positive refocusing		0.448 (0.474)		1.432 (0.661)
None-mild	9.06±3.64		9.15±3.80
Moderate-severe	8.82±3.90		8.29±3.52
K-CERQ_Refocus on planning		0.926 (0.627)		0.120 (0.126)
None-mild	12.53±4.00		12.26±4.06
Moderate-severe	12.03±3.86		12.18±3.56
K-CERQ_Positive reappraisal		1.021 (0.576)		2.666 (0.272)
None-mild	11.02±4.09		11.16±4.18
Moderate-severe	10.44±4.16		9.41±3.62
K-CERQ_Putting into perspective		1.022 (0.584)		1.727 (0.593)
None-mild	11.18±3.53		11.20±3.60
Moderate-severe	10.68±3.59		10.20±3.32
K-CERQ_Catastrophizing		1.198 (0.444)		0.832 (0.553)
None-mild	11.61±4.15		11.18±4.28
Moderate-severe	10.91±4.34		11.76±4.09
K-CERQ_Blame others		0.895 (0.469)		0.419 (0.344)
None-mild	10.12±3.67		9.86±3.84
Moderate-severe	9.65±3.98		10.13±3.70
K-ARS_Angry memories		5.155 (<0.001)		2.957 (0.003)
None-mild	18.39±6.34		19.61±6.64
Moderate-severe	23.04±6.75		22.90±7.21
K-ARS_Thoughts of revenge		4.337 (<0.001)		2.496 (0.013)
None-mild	13.13±4.60		14.02±5.28
Moderate-severe	16.38±5.95		16.29±5.75
K-ARS_Understanding of causes		3.540 (0.001)		2.266 (0.024)
None-mild	6.69±2.37		7.00±2.51
Moderate-severe	8.92±2.64		8.03±2.62
K-CTQ_Emotional neglect		2.954 (0.003)		1.045 (0.297)
None-mild	9.72±4.75		10.41±5.19
Moderate-severe	11.84±5.76		11.31±5.62
K-CTQ_ Emotional abuse		2.210 (0.025)		1.299 (0.195)
None-mild	7.05±3.43		7.14±3.63
Moderate-severe	8.98±3.71		7.97±4.91
K-CTQ_ Physical neglect		0.875 (0.382)		0.207 (0.836)
None-mild	7.35±3.04		7.55±3.26
Moderate-severe	7.76±3.71		7.44±3.62
K-CTQ_ Physical abuse		0.736 (0.463)		0.807 (0.421)
None-mild	8.20±4.29		8.26±4.16
Moderate-severe	8.65±4.59		8.83±5.22
K-CTQ_ Sexual abuse		1.356 (0.54)		0.342 (0.733)
None-mild	5.89±2.46		6.17±2.72
Moderate-severe	6.51±3.23		6.32±3.13

Class	Parameter	Estimate	SE	p
1	CD-RISC_Sub factor_01	25.09	0.33	<0.001
	CD-RISC_Sub factor_02	20.34	0.29	<0.001
	CD-RISC_Sub factor_03	16.12	0.20	<0.001
	CD-RISC_Sub factor_04	9.22	0.14	<0.001
	CD-RISC_Sub factor_05	5.85	0.11	<0.001
2	CD-RISC_Sub factor_01	15.52	0.30	<0.001
	CD-RISC_Sub factor_02	13.03	0.20	<0.001
	CD-RISC_Sub factor_03	10.76	0.16	<0.001
	CD-RISC_Sub factor_04	5.82	0.11	<0.001
	CD-RISC_Sub factor_05	4.53	0.08	<0.001
3	CD-RISC_Sub factor_01	7.07	0.27	<0.001
	CD-RISC_Sub factor_02	7.23	0.24	<0.001
	CD-RISC_Sub factor_03	5.76	0.16	<0.001
	CD-RISC_Sub factor_04	2.51	0.11	<0.001
	CD-RISC_Sub factor_05	2.92	0.09	<0.001