**1. Introduction**

According to a report by the Ministry of Unification (September 2020) [1], there are 33,718 North Korean refugees (NKRs) living in South Korea. Among them, a significant proportion, 40% (13,367), are adolescents and young adults aged 10–29. North Korean refugee youths (NKRYs) are likely to have experienced various physical and psychological traumas in North Korea and are known to experience various direct and indirect traumas from the process of defection to South Korea [2,3]. Additionally, one of the latest trends regarding NKRYs is the growing proportion of teenagers born in third countries to parents who are NKRs [4]. In 2020, the proportion of NKRYs born in a third country, including China, was 62.8% [5], and they have been shown to experience difficulties with both language communication and cultural differences [6]. Considering that NKRYs will play a socioeconomic role as members of South Korean society in the future, their stable adaptation is critical.

The cultural adaptation stress experienced by NKRYs in the process of settling is a significant factor that makes adapting to South Korean society difficult [7]. In particular, it is highly likely that they will experience additional stress due to the developmental transition from childhood to adulthood [8,9]. NKRYs are known to experience many psychological symptoms, such as posttraumatic stress disorder, anxiety and depression, delinquency, aggression, and hostile behavior [10–14]. Among them, depression is the most commonly observed psychological problem in NKRYs.

Depression, along with posttraumatic stress disorder, has been used as an indicator of the mental health of these groups, and is commonly experienced by immigrants and

**Citation:** Seong, Y.; Park, S. Factors Affecting Changes in the Mental Health of North Korean Refugee Youths: A Three-Year Follow-Up Study. *Int. J. Environ. Res. Public Health* **2021**, *18*, 1696. https:// doi.org/10.3390/ijerph18041696

Academic Editor: Lillian Mwanri Received: 24 December 2020 Accepted: 8 February 2021 Published: 10 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

refugees [15]. High levels of depression have been reported in NKRYs who perceive their daily stress to be high [11]. Low resilience and low self-esteem are known to be factors that maintain depression [16]. Particularly, resilience is one's capacity for adapting successfully in spite of adversity or overwhelming circumstances, and its importance as a protective factor for depression was highlighted in a South Australian study with refugee youths [17,18]. Furthermore, low levels of life satisfaction and low expectations of the future were also found to be factors related to depression in NKRYs [19]. Expressive suppression is one of the emotional regulation strategies in which one intentionally suppresses emotionally expressive behavior while experiencing emotional stimulation [20]. It has been shown to exacerbate the effects of early trauma on depressive symptoms [21]. In terms of external factors, low levels of emotional or practical social support contribute to depression in NKRYs [11,19]. On the other hand, several studies have found that an internal locus of control or family support protect against depression in NKRYs and have been reported to be able to help them adapt to living in a new society [8,9,22–24]. In addition, participation in social groups (e.g., religious and social organizations) was also found to reduce depression [25].

A two-year follow-up study with 1348 Southeast Asian refugees found that preimmigration stressors affected the initial mental health of migrants, but over time, the effects of post-settlement stressors, such as financial problem or cultural adaptation were greater [26]. A three-year follow-up study with 151 NKRs in South Korea also showed that experiencing stress after settling in South Korea, such as misunderstanding the language or law, and lack of basic knowledge needed in everyday life, has a stronger impact on depression than the psychological trauma experienced during the defection process [27]. These studies suggest that factors affecting depression of NKRYs may vary over time, and it is therefore necessary to conduct a longitudinal follow-up study taking into account temporal changes. According to a literature review by Lee, Lee, and Park [28], most studies on NKRs used cross-sectional research methods, with very few longitudinal studies to consider changes over time. Furthermore, previous studies have only identified the relationship to each individual factor, but no studies have so far been found to identify relative influences by simultaneously looking at internal and external factors affecting NKRYs' mental health. The specific purpose of this study is to answer the following questions: First, do the symptoms of depression in North Korean defectors change over time? Second, what are the risk factors or protective factors that affect changes in their symptoms of depression? Thus, the present study seeks to find a longer-term way to intervene in the psychological problems of NKRYs by clarifying the factors affecting changes over time, using three-year follow-up data. In addition, this study clearly identified intrapersonal and external factors, as well as the preventative and risk factors affecting depression.

### **2. Method**

### *2.1. Participants*

Participants were recruited from two alternative schools for NKRYs in Seoul. Both schools participated voluntarily in the study for the benefit of the mental health screening program. Our research team visited each school and asked for consent to participate in the study after explaining the contents of the questionnaire and the procedures of this research. Furthermore, in the case of participants under the age of 20, the consent of the participant and his/her parents was obtained together. All participants were firstly provided with questionnaires in Korean. However, for participants whose main language is Chinese, questionnaires with Chinese translations were provided, and questionnaires without Chinese versions were directly explained to and answered by Korean and Chinese bilingual speakers. Among all students who attended the two schools, 174 students enrolled in our study in 2017 and 2018 (baseline). A total of 108 completed the following year's questionnaire (T2), and only 64 were finally included in the study after completing a three-year follow-up questionnaire in 2018–2019 and 2019–2020 (attrition rate: 63.29%), because of changes such as graduation, suspension of study, and relocations.

Sociodemographic information and childhood trauma experience were included in the baseline questionnaire. In addition, depression, emotional regulation strategies, resilience, life satisfaction, psychological and practical support, family adaptation, and cohesion were reported (T1). One year later, follow-up data were collected using an identical questionnaire (T2), and the year after, the second set of follow-up data were accumulated (T3). The study was reviewed and approved by the institutional review board of the National Center for Mental Health (No. 116271-2017-11).

#### *2.2. Measurements*

#### 2.2.1. Sociodemographic Characteristics

Sociodemographic data were collected including age, gender, period of living in South Korea, country of birth, and residential type (i.e., with the immediate family, other relatives, friends, alone, in a dormitory, or in a facility).

#### 2.2.2. Childhood Trauma Experience

Early trauma experiences were assessed using the Adverse Childhood Experiences (ACE) questionnaire [29]. The ACE questionnaire consists of 17 items that assess whether participants had ever experienced various adverse childhood experiences and dysfunctional family relations, including child abuse (6 items), neglect (4 items), and household dysfunction (7 items) such as domestic violence and mental illness (1 = yes, 0 = no). The total score ranges from 0 to 17, with higher scores indicating more experiences of early trauma. Cronbach's α for the scale was 0.80.

### 2.2.3. Depression

Depression symptoms were measured using the Center for Epidemiologic Studies Depression Scale (CES-D), which was developed to measure depressive symptoms in the general population [30]. The Center for Epidemiology Studies-Depression Child Scale (CES-DC) [31] was used for individuals younger than 19 years. Both scales were translated into Korean and validated with the Korean population [32,33], and we utilized the Korean versions in this study. Also, Chinese validated version of CES-D and CES-DC were provided for participants whose main language is Chinese [34,35]. The CES-D and CES-DC are comprised of 20 items: four positive items and sixteen negative items. The scale was rated on four points: 0 = rarely or none of the time (less than one day per week), 1 = occasionally (one or two days per week), 2 = frequently (three–four days per week), and 3 = most or all of the time (five–seven days per week). Higher scores indicate greater depressive symptoms; total scores reflect depression [36]. In the present sample, Cronbach's α for the scale was 0.85.

#### 2.2.4. Protective and Risk Factors: Intrapersonal Factors

Emotion regulation strategies were assessed using the Emotion Regulation Questionnaire (ERQ) [37], which comprises 10 items measuring two emotion regulation strategies: cognitive reappraisal (6 items) and expressive suppression (4 items). The original version of the ERQ was rated using a seven-point Likert scale (1 = strongly disagree, 7 = strongly agree). In this study, we adopted the validated Korean version of the ERQ, which has been modified into a five-point Likert scale to make it easier to answer [38]. The total scores range from 6 to 30 for cognitive reappraisal and 4 to 20 for expressive suppression, with higher scores indicating greater use of the corresponding emotion regulation strategy. We used each sub-item separately, Cronbach's α for cognitive reappraisal was 0.76, and for expressive suppression was 0.58 in the present study.

Resilience was assessed using the Brief Resilience Scale [39], which is the self-perceived ability to recover from stress. The six items were scored on a five-point scale (1 = strongly disagree, 5 = strongly agree). The scale consists of three positive and three negative items. The range of total possible scores is from 6 to 30, and a higher score represents higher resilience. In this study, Cronbach's α for the scale was 0.79.

Life satisfaction was measured by the subjective well-being question from the Gallup World Poll [40], which was originally from the Cantril Self-Anchoring Striving Scale [41]. Participants describe their level of life satisfaction during the past, the present, and the future on a scale of 0 to 10 (0 = worst, 10 = best); we used only the response pertaining to the present.
