Post-traumatic Stress Disorder and Social Isolation among North Korean Refugee Women in South Korea: The Moderating Role of Formal and Informal Support

: While there is abundant literature on the positive correlation between refugee post-traumatic stress disorder (PTSD) and social isolation, few studies have examined the moderating effects of formal and informal interpersonal support. This study on 154 refugee women from North Korea examines the moderating effects of formal and informal interpersonal support on the relationship between PTSD and social isolation. Data from the 2012 survey of the Ministry of Gender Equality and Family in South Korea Study on Support for Independent Living Tailored to North Korean Refugee Women Exposed to Violence were included in the present study. The study results conﬁrmed the positive correlation between PTSD and social isolation and veriﬁed the protective effects of formal support. These ﬁndings will contribute to policy changes and early proactive interventions to reduce the social isolation of North Korean refugee women in South Korea.


Introduction
Refugee resettlement and adaptation is an important issue for many countries, including the United States and the E.U. [1][2][3]. The year 2017 witnessed unprecedented growth in the number of migrants across the world due to war and hunger in countries such as Syria, Afghanistan, and Libya [4]. This refugee influx has aggravated social problems and conflicts in many countries [5].
The end of the Cold War eventuated the political and economic isolation of North Korea. This isolation led to massive defections beginning in the 1990s, largely due to famine and pervasive malnutrition [6]. South Korea is presently home to 31,000 North Korean refugees and approximately 1000 people arrive in South Korea each year [7]. The South Korean government has been making efforts to promote integration in the resettlement process. The South Korean constitution stipulates that the territory of the nations of South Korea and North Korea is one whole, wherein the migrants from North Korean are accepted as South Korean citizens rather than as refugees [8]. Several comprehensive measures, such as economic, medical, and educational support, for resettlement in South Korea, are in place [7] (p. 2). As a result, numerous North Korean refugees have been successfully resettled as members of South Korean society, although some have experienced difficulty. There have been instances of depression and suicide among refugees owing to severe social isolation [9].
Social isolation faced by refugees and migrants is an important factor related to survival [10]. It is often inevitable when integrating within a new society [11]. Several qualitative studies on social isolation of North Korean refugees concluded that social factors such as discrimination, economic hardship, and cultural heterogeneity can affect social isolation [12,13]. Studies in Europe and the United

Background: Traumatic Events and PTSD among North Korean Refugee Women
People in North Korea experience massive trauma before fleeing to South Korea [31]. The communist regime of North Korea has for more than half a century exercised absolute internal control and enforced isolation policies on its people [32], resulting in human rights violations, extra-judicial executions, forced disappearances, torture, famine as an instrument of political control, internment in political prisons that have been compared to the concentration camps of the Nazis, and other violations too numerous and heinous to list [33,34]. North Korea is characterised by a pervasive patriarchal family environment [35], leading to an oppressive domestic environment typified by violent husbands and fathers [36]. The North Korean regime uses patriarchal and neo-Confucian values as a means to secure its own legitimacy and success [37]. North Korean refugee women are thus vulnerable to PTSD because of victimization by the state and in private life inside North Korea. Such victimization may involve witnessing death, beatings, surveillance, life-threatening food shortages, and numerous traumatic events such as domestic violence. To date, there is lack of research on PTSD in North Korean society. A comparison, however, with Sudan, which has suffered civil war and dictatorship, shows that 48% of the 3323 subjects surveyed in 2004 showed symptoms of PTSD [38].
Fleeing from North Korea can be traumatic and is sometimes deadly for refugees. The escape, even when successful, leads to instability among refugees ranging from few months to over 5 years depending on whether they attempt to stay in China. The refugees living along the Chinese border of North Korea are not allowed to hold refugee status in China and are forcibly repatriated to North Korea if detected by authorities [39]. Leaving the North Korean state without permission is considered treason [40]. Hence, those forcibly repatriated are subject to torture, sexual assault, forced abortion, and infanticide by border guards, malnourishment and internment in disease ridden labour camps, internment in political prisons, and summary execution [41]. Therefore, North Korean refugees are under constant fear of being arrested, of facing forced marriage via human traffickers, and of violence and/or death. A study of 170 North Korean refugees living in China reported that they each experienced an average of 12 different traumas, such as the 'unnatural death of family or friend' and 'being close to death', and 56% of the respondents showed at least one symptom of PTSD [42].
Such serious traumatic events faced by North Korean refugees have led to high prevalence of PTSD before entry into South Korea (nearly 25-50% among those surveyed in South Korea [43,44]). Moreover, North Korean refugee women show two or more times more severe PTSD symptoms than men [45], probably because refugee women are more likely to be traumatised due to widespread sexual violence. Even 10 years after resettlement, 86% of the refugees surveyed are reported to show significant PTSD symptoms [46].

Relationship between PTSD and Social Isolation
PTSD is a major cause of social isolation [47]. Risk factors for refugees or migrants regarding social isolation include problems in the resettlement process such as social discrimination [48] and economic deprivation [49]. Unlike migrants, however, refugees are likely to face emotional crises such as unresolved PTSD [50,51]. Although North Korean refugee women can lead better lives in South Korea due to stable jobs and improved economic status, these women may suffer from psychological conditions such as PTSD. The most common responses to PTSD symptoms are interpersonal avoidance, self-critical thinking [52], and emotional adjustment problems [53,54]. These symptoms affect North Korean refugees' relational and emotional ability to settle into South Korean society, resulting in social isolation. PTSD symptoms lead to symptoms of avoidance [18] (p. 2) or fear [19] (p. 2), and reduced trust in others, which can lead to dysfunctional choices in interpersonal relationships [55]. Hofmann, Litz, and Weathers [47] (p. 3) reported that soldiers with post-war PTSD experienced numerous interpersonal conflicts and isolation issues in the process of reintegration into their family and society. Self-critical thinking can result in extremely passive attitudes in primary relationships, because the individual feels helpless and inferior to others [56,57]. Emotional regulation is another important risk factor for relational isolation [58]. As noted above, symptoms of PTSD can significantly exacerbate the isolation of North Korean refugees.

Moderation Effect of Formal and Informal Social Support
Social support is an important protective factor that can help reduce the social isolation of refugees suffering from emotional crises [20,21] (p. 2). Typically, social support not only alleviates stress but also improves the individual's coping ability [59] and provides psychological compensation for social conflicts or lack of relationships [60]. Neurobiological factors corroborate these results. Individuals suffering from PTSD or isolation exhibit a high level of the oxidative stress hormone called cortisol [61]. Social support plays a significant role in lowering cortisol levels [62], and it can thus play a role in mitigating social isolation by improving psychological stability [16] (p. 2) among refugees with symptoms of PTSD. Such support may be particularly more important for refugee women than for refugee men. For example, a study of victims of violence reported that satisfaction with social support can positively affect their mental health, and the effect is higher for women than in men [63]. Previous studies on social support provide an integrated view of the environmental support system for individuals. As North Korean refugees are generally viewed as a group that hesitates to forge new relationships South Korean society, the social support system is limited, and its influence may vary.
This study examines the formal support extended by the support staff to North Korean refugees and the informal support extended by their neighbours. In the context of language barriers and cultural differences, as well as lack of economic and cultural information, formal support becomes crucial for North Korean refugee women. Formal support is limited to providing information alone and does not provide active psychological support [64]. However, previous research suggests informal support plays a complementary role in overcoming the limitations of formal support [65], and numerous refugee women rely on local community support to offset the limitations of formal support [66]. A few studies have reported that formal support was highly effective [67]. On the other hand, some studies reported that informal support is more effective than formal support in stressful situations [68,69]. These contradictory studies suggest that the usefulness of social support may vary by purpose and group [28] (p. 2). Therefore, it is necessary to ascertain the roles played by both formal and informal support in the relationship between PTSD and social isolation among North Korean refugees.

Covariates
Various causes other than PTSD may also lead to social isolation among North Korean refugee women. Refugees may find resettling in a new society difficult due to limitation of occupational opportunities [70]. Moreover, language and cultural differences between the two Koreas [71] may be important factors that aggravate social isolation. North Korean refugees consider economic independence as the most important condition for living in South Korean society [72]. However, their greatest desire is to engage in meaningful occupational activity in South Korean society. Many North Korean refugees experience difficulty in finding a job, and the average length of employment is only 25.2 months', even when employment is found [73]. Insecurity in employment may result from workplace discrimination and prejudice, as well as a lack of education or adaptive orientation to social interactions [74].
In addition to employment issues, North Korean refugees face language and cultural barriers [71] (p. 4). In the past 50 years, the two Koreas have seen significant changes in their respective cultures and languages. Differences in vocabulary, intonation, grammar, and neologisms have widened [75]. For example, in South Korea, mobile phones, ice cream, and knocking are used as direct loanwords, while North Korea translates these into Korean words. Therefore, although North Korean refugees can manage basic communication in South Korea, they have trouble engaging in meaningful conversations and understanding the context [76]. In addition, the distinct North Korean dialect and intonation reveals the origin of the speaker to members of South Korean society and may result in discrimination [77]. Furthermore, age [78], length of sojourn in South Korea [79], and income are important factors in determining social isolation [80].

Sample and Procedures
For the purpose of this study, data were collected from 154 North Korean refugee women living in South Korea. A 2012 study by the Ministry of Gender Equality and Family in South Korea titled 'A Study on Support Plans for Independent Living Tailored to North Korean Refugee Women who are Victims to Violence' formed the basis of secondary data. Primary data was collected through a survey conducted from April to June 2012. Despite limitation of generalisability, the snowball sampling method was used. Residential information of North Korean refugees remains strictly confidential. Moreover, most North Korean refugees did not want to reveal their identity because of threat of terrorism or social discrimination. Therefore, several studies on North Korean refugees have relied on the snowball sampling method instead of unrealistic random sampling. In this study, experts with excellent understanding of North Korean refugees were introduced to each family by visiting them. Survey participants were remunerated 10,000 Korean Won ($10) for their participation in the study.
The sample size required was calculated by power analysis using G * power 3.1 [81]. The effective size was assumed to be 0.15, at a significance level of 0.05; 0.80 was the power level used, and the number of predictors was 8. Consequently, the reliable sample size in this study was determined to be approximately 109 [82], which confirmed that the sample size of 156 subjects in this study was appropriate. Table 1 shows the demographic characteristics and the main variables of the subjects. The age range of the subjects was 21-77 years (M = 42.78, SD = 9.23), and the resettlement duration in South Korea was 2-16 years (M = 6.49, SD = 2.72). About 4.5% of the subjects had no income, and the average monthly household income of 35.7% was at the low level of $1000 or less. The mean value of social isolation, which is a dependent variable in this study, was reported to be 2.09 (SD = 0.39), corresponding to a moderate social isolation score. Of the subjects, 59.7% had been diagnosed with PTSD, and over half of the subjects showed PTSD symptoms. Finally, it was seen that North Korean refugee women better appreciated formal support (M = 2.56, SD = 0.60) than informal (M = 2.58, SD = 0.57).

Social Isolation
Social isolation is adopted from the scale used in a study by Lee [83], who adapted and translated the UCLA Social Loneliness (UCLA-SL) Scale. This scale was developed by Russell [84] for use in Korean subjects. The UCLA-SL has been used in many studies to measure social isolation. The scale includes a total of 10 items. Six items are as follows: 'I am no longer close to anyone', 'I often feel isolated from others', 'There is no one I can turn to when I am in trouble', 'People are around me but not with me', 'I feel left out', and 'My social relationships are superficial'. Four reverse items are as follows: 'I am an outgoing person', 'I can find companionship when I want it', 'I feel in tune with the people around me', and 'There are people I can talk to'. Likert scale was used to measure: 1 (never), 2 (rarely), 3 (sometimes), and 4 (often). Mean values were calculated after recoding reverse items for the final analysis ( Table 2). The Cronbach's alpha of this scale was 0.802.

PTSD
The PTSD diagnostic scale of the Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV) was used in this study. This scale consists of 19 items in total, including five for re-experiencing symptoms after the trauma, seven for symptoms of avoidance and paralysis, five for symptoms of hyper arousal, one for dysfunction lasting 1 month or more, and one item for dysfunction-related difficulties in daily life. The rating scale consists of 'Never (0)', 'Insignificant (1)', and 'Severe (2)', and answers were obtained as 'less than one year after entry' and 'past one week'. In this study, based on temporal causality with dependent variables, data of 'past one week' were used. As shown in Table 2, the mean value was used for the final analysis. The method of diagnosis specified in DSM-IV was used in Table 1, which shows the presence of PTSD symptoms. The Cronbach's alpha of this scale was 0.973.

Formal Support
This study adopted the social relationship scale used in the 2012 Ministry of Gender Equality and Family survey in South Korea to measure formal support for North Korean refugee women. This scale is adapted for the characteristics of North Korean refugees. It measures the social relations that North Korean defectors have with special people and staff. In terms of formal support, only four items on the relationship with special people and staff in charge for North Korean refugees were used from the above scales. However, since Cronbach's alpha for the four items was very low at 0.677, only three items were used for the analysis, excluding the one with the lowest reliability. The excluded item is 'I can talk to the special people and staff for North Korean refugees about my problems.' The final analysis included the following three items: 'The special people and staff for North Korean refugees really try to help me', 'I get the help and support from the special people and staff for North Korean refugees in South Korea', and 'Special people and staff for North Korean refugees are willing to help you making my own decisions'. The average value was included in the final analysis. The rating scale consists of 'Not at all', 'Do not agree', 'Agree', and 'Highly agree'. The Cronbach's alpha of this scale was 0.859.

Informal Support
In this study, we used the neighbourhood scale in the 2012 Ministry of Gender Equality and Family Survey to measure informal support for North Korean refugees. The scale includes four questions: 'Do you have friendly relationships with your neighbours?', 'Are there local residents to help you when your family faces an important problem?', 'Do you think your neighbours trust each other?', and 'Do you think your local residents are willing to help their neighbours?' The average value was used for the final analysis, and the rating scale consisted of 'Not at all', 'Do not agree', 'Agree', and 'Highly agree'. The Cronbach's alpha of this scale was 0.861.

Covariates
From previous studies, the present study used control variables for the factors affecting social isolation in addition to the main variables. Demographic characteristics included age, duration of resettlement, and income as control variables, in addition to 'work-related problems' and 'difficulties in language and cultural adjustment' as factors of social isolation faced during the process of settling in South Korean society. Age

Data Analysis Plan
In this study, data were analysed using SPSS 24.0. Analyses were conducted in three stages. First, frequency analysis and descriptive statistics were conducted to identify the demographic characteristics, as well as PTSD status, social isolation, and formal and informal support. Control variables included age, resettlement duration, average monthly household income, employment and work difficulties, and language and culture difficulties. Second, to examine the relationship between PTSD and social isolation for North Korean refugee women, as well as to study the moderating effect of formal and informal support, multiple regression analysis was conducted using interaction terms.
Multi-collinearity was examined for every regression analysis, and there was no problematic multi-collinearity in the analyses; the highest VIF was 1.685, and the lowest tolerance was 0.605. Table 2 presents the results of the multivariable regression models. This model verified the relationship between PTSD and social isolation, as well as the moderating effects of formal and informal supports. The variables used in this study were analysed as continuous variables after mean centering. The explanatory power of the model was 28.6%. Moreover, the results showed that PTSD and social isolation had a positive correlation (β = 0.96, p < 0.001) and that informal support showed a negative correlation (β = −0.277, p < 0.01). The results indicate that social isolation was high in case of higher symptoms of PTSD and low informal support. In the validation of the moderating effect, formal support was verified as a factor controlling PTSD and social isolation (β = −0.258, p < 0.01). This result verified that formal support influences the relationship between PTSD and social isolation. For the control variables, the greater the difficulty at work (β = −0.24, p < 0.05), and in language and culture (β = −0.24, p < 0.05), the greater the social isolation. Figure 1 shows a graph to determine whether formal support is statistically significant in the influence of PTSD on social isolation based on the regression analysis above. Consequently, social isolation increased as PTSD increased, confirming the positive correlation, and the inclination was less steep in the group with higher formal support than it was in the group with lower formal support.

Results
The results indicate that formal support plays a role as a control variable in protecting the relationship between social isolation and PTSD. Figure 1. Moderation effect of formal support on the association between PTSD and social isolation. Range of PTSD and formal support are from (mean − 1SD) to (mean + 1SD). These variables before being centered to 0.

Discussion
Studies, so far, on the social isolation of refugees have focused on the problems arising from migration. However, because refugees characteristically have a high prevalence of PTSD resulting from serious trauma, it may be that PTSD interferes with the formation of social relations and plays an important role in exacerbating isolation; this is supported by the relationship between PTSD and isolation verified by this study. First, 59.7% of subjects showed PTSD symptoms similar to the rates of PTSD symptoms among North Korean refugee women surveyed in other studies [86]. Furthermore, PTSD symptoms and isolation showed a positive correlation. In addition, work-related problems experienced in South Korean society as well, as difficulties with language and cultural adaptation, were also shown to be risk factors for isolation. Comprehensively, the results suggest that multidimensional intervention is required to alleviate social isolation in North Korean refugee women. At present, worldwide, initial intensive assistance is being provided for early resettlement of refugees. The South Korean government also provides economic, housing, medical, and employment support services to help North Korean refugees settle in society [7] (p. 2). Refugees receive information to help them understand South Korean society, as well as educational services during the first 14 weeks of entry (12 weeks in resettlement support facilities called 'Hanawon', 2 weeks in residence). However, this is too short a duration to understand South Korean society and to alleviate isolation. The time to gain emotional stability and promote health is only 50 h of 392 [87]. The relationship between PTSD and isolation examined in this study is not a temporary phenomenon. Therefore, various long-term interventions such as emotion therapy, language and cultural education, and vocational education should be undertaken, rather than focusing on the initial education alone.
The roles of formal support and informal support verified in this study are also important. In this study, informal support was directly related to social isolation, although it was not verified as a factor that could control PTSD and social isolation. By contrast, formal support was not directly related to social isolation, but it was verified as a factor controlling PTSD and social isolation. The results that previously described it have some implications for the roles of formal and informal support.
First, informal support, which is directly related to daily life, is important, because it involves a closer relationship than formal support. These results support the findings that informal support is crucial for refugee women to settle into a new society. Residents in the neighbourhoods [88] who have daily contact with refugee women can be an important resource for alleviating refugees' social isolation. While refugee women can use these relationships and activities to create a measure of balance and gain greater autonomy in their lives, they are also the sites for waging battles over their past and present identities [89]. Thus, it is necessary to link and activate local communities with refugee women so that these women can form positive neighbourhood identities within South Korean society. Second, high-quality service is essential for North Korean refugee women suffering such serious emotional crises as PTSD. The study results verified that formal support influences the relationship between PTSD and social isolation. This result could provide important intervention directions for North Korean refugees. It is thus necessary to provide pre-screening and information before they face critical maladjustment such as isolation due to emotional crisis. North Korean refugees may not have sufficient understanding of the health and counselling systems available in South Korean society, as well as the importance of health and counselling during initial resettlement. However, as mentioned above, the counselling service offered immediately upon arrival in South Korea is qualitatively and quantitatively insufficient. It is necessary to increase professional counselling hours at Hanawon and other interventions in the case of refugees identified as having problems through pre-screening. Furthermore, it is necessary to consider Korean embassies in third, transit countries such as China and Thailand. Most North Korean refugees stay in embassies of third countries for a short duration before entering South Korea [39] (p. 2). As numerous studies suggest the importance of early intensive interventions in PTSD [90,91], it is necessary to amend policies in which the Korean Embassy in the third countries would not only protect the refugees but would also offer expert crisis interventions to refugees exposed to traumatic events. In addition, there is a need to strengthen the quantitative and qualitative formal support systems provided after resettlement. Although refugee women have some social capital, they do not have a strong connection with social capital [92]. Therefore, the role of the public workers that provides formal support is very important. Currently, protection officers are deployed for regional resettlement support, administrative support, and the personal protection of North Korean refugees. Protection officers are highly valued as total formal supporters, who are responsible for coordinating various services such as government subsidies, housing, and medical support services, as well as security services, for North Korean refugees. However, the number of protection officers is insufficient to handle 31,000 North Korean refugees. Currently, there are 800 protection officers in total; in other words, one officer for every 30 refugees, on average [93]. Moreover, some protection officers lack expertise. Thus, improvement in the number and quality of the current protection officers is warranted.

Limitations and Suggestions for Future Research
There are some limitations to this study. First, the study results cannot be generalised, because the data were collected through snowball sampling. It is unclear whether its results accurately represent the entire population. The research on the North Korean refugee group, for which the residence and personal information are thoroughly confidential, inevitably relies on snowball sampling. Snowball sampling, while effective in estimating a hidden population, did not fully reveal this population's problems despite its exposure to serious crises. For more accurate research results, we suggest unbiased estimates through respondent-driven sampling.
Second, the scope of the support system examined was limited. The formal support system used in this study corresponds to the support for North Korean refugees by staff, and informal support corresponds to neighbourhood relationships. Although there could be various support systems through which North Korean refugees create relationships, there are limitations to characterising the full range of variety in these research data. The two representative support systems that form the closest relationship with North Korean refugees were considered, however, and their relevance was based on previous studies.
Third, in the neighbor relationship scale for informal support used in this study, it was not easy to accurately determine whether the neighbor was a North Korean refugee or a native South Korean resident. In the case of refugees, the neighbor could play different roles even in the same situation, depending on whether the neighbor is a North Korean refugee or a South Korean. Therefore, further study is required to explore in detail the determinants of informal support.
Fourth, future studies must examine how subjective satisfaction with social support differs from the quantity and frequency of social contacts. Many existing studies [94][95][96] propose that an individual's perception of social networks can play an important role in determining the degree to which individuals are willing to find and utilize necessary resources. Such studies can provide important guidance for intervention with North Korean refugees lacking or disoriented in social networks.

Conclusions
The current worldwide increase in population migration has repercussions on the social discourse surrounding refugee issues in host countries [97][98][99]. Some countries have strictly limited the recognition of refugee status or have reduced migration inflow because of social cost, increased crime rates, or social conflicts. However, aside from humanitarian concerns and costs, the influx of refugees could also be positive with regard to securing an additional workforce from other countries that are suffering from war and hunger and allowing host countries to develop new sources of social improvement [98][99][100][101]. Identifying causes of refugee social isolation and early intervention methods is essential, because preventing isolation in a new society is an important determinant of a success social integration. From this viewpoint, the relationship between PTSD and social isolation, the protective effect of formal support, and the main effect of informal support examined in this study offer valuable directions for interventions to alleviate the social isolation of refugees.