On April 16, 2014, the Sewol ferry sank in the West Sea off the coast of South Korea. Only 181 of the 476 passengers survived. Among these survivors was a group of students from Danwon High School who were on a school trip at the time of the accident; 75 of these individuals, who came to be known as the “Danwon High School survivors,” survived, whereas 250 died or were never found. The school's vice principal, who had been rescued from the ferry, committed suicide a few days after the disaster,1 and a female Danwon High School survivor tried to kill herself by overdosing on medication and was hospitalized in a state of unconsciousness.2 This student had written a letter explaining that she missed her friend, who had died in the disaster.

An inability to find meaning in life and a general lack of a sense of self-efficacy have been identified as significant predictors of depressive symptomatology after combat trauma.3 Significant correlations have also been observed among one dimension of meaning (the search for meaning), perceived social support, and posttraumatic growth in survivors of the Colorado floods.4 A limited sense of meaning in life is associated with depression, hopelessness and suicide, as well as substance abuse and emotional dysregulation, whereas perceiving life to be meaningful is thought to facilitate optimal recovery from adverse events.5

The Danwon High School survivors were disappointed in and resentful about the failure of the national disaster response, and they refer to their survival in terms of “escape” instead of “rescue.” Intrusive media attention increased this mistrust and resentment. Furthermore, distorted news reports led to malicious comments and cyberbullying, which elicited feelings of hurt and confusion.6 One consequence of a negative life event, even if it is neither life-threatening and nor repeated, is embitterment, which can impair mental health.

Embitterment results from persistent experiences of being let down or insulted, feeling like a “loser,” or wishing for revenge but feeling helpless.7 Embitterment has been observed in East Germans who experienced uncertainty or sudden, uncontrollable events, such as the abrupt and unexpected loss of a job during the unification of Germany.7 Historically, embitterment is a familiar concept to Koreans, as the Korean culture-bound syndrome, Hwa-Byung (literally, fire disease), has been understood as deriving from anger.8

It is important that survivors of traumatic events reduce feelings of embitterment and receive increased social support. One study of former political prisoners found that embitterment was correlated with overgeneral autobiographical memory, and that survivors with social support networks were better able to recall specific memories.9

To the best of our knowledge, scant research has focused on the relationship among embitterment, social support, and the perception of meaning in life among survivors of disasters. However, as noted above, embitterment has been a common affective response to traumatic events among South Koreans, and the perception of meaning in life has been shown to enhance recovery from negative events. Our research, therefore, was designed to test and explore the relationships among embitterment, social support, and perceptions of meaning in life in the Danwon High School survivors of the Sewol ferry disaster. Specifically, we hypothesized that 1) embitterment, social support, and the perception of meaning in life are significantly correlated and 2) social support mediates the relationship between posttraumatic embitterment (PTE) and the perception of meaning in life.

Table 1 presents the demographic characteristics of the participants, and their MLQ scores according to demographic characteristics. Of the 48 participants included in our study, 24 were male and 24 were female. More than four-fifths (85.42%) were born in 1997. There were no significant differences in MLQ scores by demographic characteristics (p>0.05).

Table 2 shows the bivariate correlations among the variables of interest included in the hypothetical model. As expected, PTED scores were negatively correlated with scores on the FSSQ and MLQ-P (r=-0.43 and -0.40, respectively). However, PTED scores were not associated with scores on the MLQ-S (r=0.03, p>0.05). On the other hand, FSSQ scores were significantly positively associated with MLQ-P, but not with MLQ-S scores (r=-0.44 and -0.03, respectively). Additionally, MLQ-P scores were significantly correlated with MLQ-S scores (r= 0.61, p<0.001).

Table 3 presents the analyses of the models used to determine whether FSSQ scores mediated the significant relationship between PTED and MLQ-P scores. We conducted a three-step hierarchical regression analysis for two models. In the first step of model 1, PTED scores were significantly related to MLQ-P scores (B=-0.63, SE=0.22). In the second step, the linear regression analysis revealed that PTED scores were significantly related to FSSQ scores (B=-0.45, SE=0.14). In the third step, the multiple regression analysis indicated that FSSQ scores were significantly related to MLQ-P scores when PTED scores were held constant (B=0.50, SE=0.22); this step also revealed that PTED scores were not significantly related to MLQ-P scores when FSSQ scores were held constant (B=-0.41, SE=0.23). Additionally, analyses relying on the bootstrapping method revealed a significant indirect effect of PTED scores on MLQ-P scores via FSSQ scores when 1000 replications and a 95% CI were used (-0.5912:-0.0365). In summary, the hierarchical regression analysis showed that FSSQ scores may fully mediate the effects of PTED scores on MLQ-P scores, given that the indirect effect was significant whereas the direct effect was not (Fig. 1). However, in model 2, PTED scores were not significantly related to MLQ-S scores. Accordingly, model 2 was not subjected to the subsequent steps because the total effect was not significant.

This study demonstrated that PTE was negatively correlated with the perception of meaning in life among Danwon High School survivors. However, the level of PTE was not significantly related to searching for meaning in life. The “presence of meaning in life” refers to the extent to which people comprehend, make sense of, or perceive significance in their lives, as well as the degree to which they view themselves as having a purpose or mission.13 The “search for meaning in life” refers to their motivation to find goals and a purpose in life. Our results indicate that, although embitterment might have temporarily interfered with the ability of Danwon High School survivors to perceive meaning in life, it did not prevent them from searching for life goals that would eventually imbue their lives with meaning.

Consistent with previous research, embitterment was negatively correlated with perceived social support. A cross-sectional study on chronic embitterment among occupational health professionals showed that embittered staff perceived their organization as being unsupportive and characterized by low levels of procedural justice.17 People who are embittered typically feel victimized, experience resentment and injustice, resist help, and have difficulty coping.18 Decreasing embitterment encourages positive beliefs, assists survivors in perceiving others as being supportive, and promotes help-seeking.

In this research, hierarchical regression analysis showed that perceived social support may fully mediate the effect of PTE on the perception of meaning in life. Our findings imply that clinical interventions to reduce PTE may help disaster survivors find meaning in life by increasing their perceived level of social support. However, data on the treatment of embitterment remain inadequate. One pilot study reported significant and clinically meaningful reductions in Symptom Checklist scores after wisdom therapy compared to routine treatment, and suggested that cognitive-behavior therapy based on wisdom psychology might constitute a valid approach to treating PTED.10 Further studies about the determinants of and treatments for embitterment are needed.

The present study had several limitations. First, the cross-sectional design renders it difficult to identify causal relationship among factors. As a result, a use of a longitudinal design is recommended. Second, sampling bias may have occurred, as only 64% of the surviving students participated in the study; therefore, the results should be interpreted with caution. Third, although the analysis accounted for demographic characteristics and psychological variables, some additional factors, such as quality of parental relationships, self-efficacy, and coping style, should be included in future studies. Fourth, although the MLQ underwent translation and back-translation by a native speaker, this scale has yet to be standardized for a Korean population.

In conclusion, despite these limitations, our study provides important insights into PTE and the perception of meaning in life. Our study analyzed the mediating effect of perceived social support on the relationship between PTE and the perception of meaning in life, and proposed a new approach to mental health interventions. Indeed, it is necessary to develop clinical interventions to reduce PTE by increasing perceived social support and resilience in response to emotional difficulties. Such interventions will reduce stress and increase the perception of life having meaning. Therefore, therapies aimed at reducing embitterment may play an important role in increasing perceived social support and, eventually, a sense of meaning in life.