Community Mental Health Status Six Months After the Sewol Ferry Disaster in Ansan, Korea

Home , Ansan, Guri, Namyangju

Epidemiology and Health Volume: 37, Article ID: e2015046, 14 pages Epidemiology and Health http://dx.doi.org/10.4178/epih/e2015046

ORIGINAL ARTICLE Open Access Community mental health status six months after the Sewol ferry disaster in Ansan, Korea

Hee Jung Yang1, Hae Kwan Cheong2, Bo Youl Choi3, Min-Ho Shin4, Hyeon Woo Yim5, Dong-Hyun Kim6, Gawon Kim1, Soon Young Lee1

1Department of Preventive Medicine and Public Health, Ajou University School of Medicine, Suwon; 2Department of Social and Preventive Medicine, Sungkyunkwan University School of Medicine, Suwon; 3Department of Preventive Medicine, Hanyang University College of Medicine, Seoul; 4Department of Preventive Medicine, Chonnam National University Medical School, Gwangju; 5Department of Preventive Medicine, The Catholic University College of Medicine, Seoul; 6Department of Social and Preventive Medicine, Hallym University College of Medicine, Chuncheon, Korea

OBJECTIVES: The disaster of the Sewol ferry that sank at sea off Korea’s southern coast of the Yellow Sea on April 16, 2014 was a tragedy that brought grief and despair to the whole country. The aim of this study was to evaluate the mental health effects of this disaster on the community of Ansan, where most victims and survivors resided. METHODS: The self-administered questionnaire survey was conducted 4 to 6 months after the accident using the Korean Community Health Survey system, an annual nationwide cross-sectional survey. Subjects were 7,076 adults (≥ 19 years) living in two victimized communities in Ansan, four control communities from Gyeonggi- do, Jindo and Haenam near the accident site. Depression, stress, somatic symptoms, anxiety, and suicidal ideation were measured using the Center for Epidemiologic Studies-Depression Scale, Brief Encounter Psychoso- cial Instrument, Patient Health Questionnaire-15, and Generalized Anxiety Disorder 7-Item Scale, respectively. RESULTS: The depression rate among the respondents from Ansan was 11.8%, and 18.4% reported suicidal ideation. Prevalence of other psychiatric disturbances was also higher compared with the other areas. A multiple logistic regression analysis revealed significantly higher odds ratios (ORs) in depression (1.66; 95% confidence interval [CI], 1.36 to 2.04), stress (1.37; 95% CI, 1.10 to 1.71), somatic symptoms (1.31; 95% CI, 1.08 to 1.58), anxiety (1.82; 95% CI, 1.39 to 2.39), and suicidal ideation (1.33; 95% CI, 1.13 to 1.56) compared with Gyeonggi-do. In contrast, the accident areas of Jindo and Haenam showed the lowest prevalence and ORs. CONCLUSIONS: Residents in the victimized area of Ansan had a significantly higher prevalence of psychiatric disturbances than in the control communities.

KEY WORDS: Disasters, Community surveys, Mental health, Screening, Posttraumatic stress

INTRODUCTION Correspondence: Soon Young Lee Department of Preventive Medicine and Public Health, Ajou University School of Medicine, 206 World cup-ro, Yeongtong-gu, Suwon 16499, Korea On April 16, 2014, the Sewol ferry with 476 passengers sank Tel: +82-31-219-5301, Fax: +82-31-219-5084, E-mail: [email protected] at sea off South Korea (hereafter Korea)’s southern coast of the Yellow Sea. The number of dead and missing from this accident Received: Oct 13, 2015, Accepted: Oct 23, 2015, Published: Oct 23, 2015 This article is available from: http://e-epih.org/ was 304, of whom 250 were high school students on a field 2015, Korean Society of Epidemiology trip. This disaster was a tremendous shock to the entire country, This is an open-access article distributed under the terms of the Creative Commons leaving indelible emotional scars especially in the community Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original of Ansan, where most victims and survivors as well as their fam- work is properly cited. ilies resided.

1 Epidemiology and Health 2015;37:e2015046

Many studies have reported that such national disasters can MATERIALS AND METHODS trigger adverse outcomes not only among victims and their families, relatives, and friends, but also among a broad spec- Study population trum of the population exposed to the disaster situation [1-3], The study was conducted in 2014 using the national commu- including voluntary support workers and community residents nity health survey system. The Korean Community Health Sur- who indirectly experienced the disaster through the media [4,5]. vey (KCHS) is an annual national cross-sectional survey system In particular, human disasters tend to occur in places with high based on a standardized questionnaire carried out by the Korea concentrations of human and material resources, thus involving Centers for Disease Control and Prevention at the community a high death toll and affecting the entire society [6]. For the level to produce regional health indicators that serve as a basis purpose of this study, it was hypothesized that accidents involv- for establishing community health plans [12]. The KCHS con- ing specific local residents and students would have a greater ducted in 254 local districts, and the target population in each effect on the community concerned. area was about 900 adult residents (≥ 19 years). The KCHS The effect of accidents and disasters on health status mani- used a two-stage sampling process. The first stage was to apply fests in various forms, not only psychiatric symptoms such as a probability proportional to size sampling strategy (to select posttraumatic stress disorder and depression, but also somatic primary sampling units) and the second stage was to apply sys- symptoms, substance abuse, interpersonal relationship difficul- tematic sampling (selecting households). The KCHS collects ties, and loss of social network [7]. The mental health response various information on demographic and socioeconomic char- to disaster is an important research area, as shown by a large acteristics, health-related problems and past medical histories, number of studies on the 9/11 terrorist attack. Higher preva- administered by trained interviewers as face-to-face interviews. lence of posttraumatic stress was found in New York compared Eight survey locations were selected: two victimized commu- with Washington and other metropolitan areas [4], and the clos- nities in Ansan (Danwon and Sangnok), four control communi- er to the terrorist attack site the higher the prevalence [1]. In ties in Gyeonggi-do (Paldal in Suwon, Gunpo, Guri, and Namy- Korea, in the wake of the Hebei Spirit oil spill in 2007, a high angju), Jindo and Haenam near the accident site involved in prevalence of depression, stress, and suicidal ideation was ob- salvage works. A supplementary self-administered question- served among the residents of the victimized area [8]. A one- naire was distributed to the residents of the communities who year follow-up study of community health status in the affected completed the KCHS after receiving a separate informed con- areas reported high burdens of disease through mental disor- sent from each respondent (Figure 1). ders [9]. The survey was conducted from August 16 to October 31. Population groups exposed to a disaster undergo a generally The total sample was 7,310 participants: 918 and 923 from observed process of showing strong emotional and psychologi- Danwon and Sangnok in Ansan city, 916 and 917 from Paldal cal reactions for up to one year, and begin to recover after the (Suwon) and Gunpo in the southern Gyeonggi province, 912 anniversary reaction [10]. In this process, adequate interven- and 917 from Guri and Namyangju of the northern Gyeonggi tions can shorten the recovery period or mitigate psychological province, and 911 and 896 from Jindo and Haenam, respec- harm, whereas health problems can persist without such inter- tively. Of the 7,310, a total of 7,153 individuals (97.9%) signed ventions. Although the Trauma Center built in Ansan after the the informed consent form. Another 77 respondents that could disaster has provided counseling and monitoring for the be- not be identified due to errors in survey numbers resulted in reaved families and related individuals [11], the level of support 7,076 participants (96.8%) that were included in this study. and research provided for community residents is not up to the This study was approved by the institutional review board (IRB) magnitude of shock suffered by them. The effects of emotional of Ajou University (IRB no. SBR-SUR-14-252), and participants and psychological harm suffered by the community residents received an explanation pointing to their rights concerning re- as a whole should be analyzed to establish adequate disaster fraining from providing responses and withdrawing from the mental health response plan. study at any time. The purpose of this study was to determine the effect of a disaster on a local community by evaluating the mental health Measurements status of community residents in terms of depression, stress, so- The data used for the current study were those concerning matic symptoms, anxiety, and suicidal ideation. Evaluation took the general characteristics included in the KCHS survey and place six months after the disaster using a survey-based appro the mental health screening results from the mental health sta- ach comparing different geographical areas. tus questionnaire (Appendix 1). General characteristics collect- ed in the KCHS were location, sex, age, education level, and monthly household income. Depression, stress, somatic symp-

2 Yang HJ et al.: Community mental health status after disaster

Gyeonggi-do Figure 1. Map of the study area and sampling locations in Guri Gyeonggi-do, Jeollanam-do Namyangju

Ansan (Danwon/ Suwon Sangnok) Gunpo A

Jeollanam-do

Sewol ferry

Jindo

Haenam B

Figure 1. Map of the study area and sampling locations in (A) Gyeonggi-do, (B) Jeollanam-do.

toms, anxiety, and suicidal ideation were evaluated using self-re- general anxiety measurement tool, it is used as a comprehensive port screening tools. screening tool for anxiety disorders, such as panic disorder, so- The Center for Epidemiologic Studies-Depression Scale (CES- cial phobia, and posttraumatic stress disorder [19]. The measure D) is a non-diagnostic screening measure of depression consist- consists of 7 items that are responded to on a 4-point scale. Scores ing of 20 items that are responded to on a 4-point scale. Score of 5 to 9, 10 to 14, and 15 or higher indicate low-level, moder- of 21 or higher is epidemiologically defined as depression [13]. ate-level, and high-level severity of symptoms, respectively. The Korean version of the CES-D [14] was used in this study. Suicidal ideation was evaluated based on an affirmative or To measure stress, the Brief Encounter Psychosocial Instru- negative answer to the question “Have you ever thought of tak- ment [15] (the Korean version translated by Lim et al. [16]) was ing your own life?” that is included in the KCHS in every even- used. Participants respond to 5 items on a 5-point scale, and av- numbered year. erage scores up to 1.6, 1.6 to 2.8, and higher than 2.8 are defin ed as low-level, moderate-level, and high-level stress, respectively. Statistical analysis The Patient Health Questionnaire-15 (PHQ-15) is a measure Sociodemographic characteristics of the participants such as of somatic symptoms based on the Patient Health Question- sex and age distributions and education and income levels in naire (PHQ) domains of somatic, anxiety, and depressive symp- the three study areas were processed on an interval scale. The toms as simplified by Kroenke et al. [17]. It is used for evaluat- mental health screening results for each area are expressed in ing somatic symptoms and estimating the degree of somatiza- percentages and average scores, and the inter-area comparison tion. The measure consists of 15 items that are responded to on of mental health status is presented in standardized prevalence a 3-point scale. Scores of 5 to 9, 10 to 14, and 15 or higher are rates. The standardized prevalence rates were calculated using indicative of low-level, moderate-level, and high-level severity the direct standardization method for sex and age on the basis of somatic symptoms, respectively. In this study, the Korean of the standard population defined in the Census conducted in version of the PHQ-15 translated by Han et al. [18] was used. 2005 by Statistics Korea. Regional differences in prevalence were The Generalized Anxiety Disorder 7-Item Scale (GAD-7) was estimated using Pearson’s chi-square test. A multiple logistic re- used to measure generalized anxiety disorder. Developed as a gression analysis was performed to control for sociodemogra

3 Epidemiology and Health 2015;37:e2015046

Table 1. Sociodemographic characteristics of the study population (n = 7,076)

Ansan (n= 1,773) Gyeonggi (n= 3,507) Jindo & Haenam (n= 1,796) Variables p-value2 n1 % n % n % Sex Male 821 46.3 1,591 45.4 762 42.4 0.046 Female 952 53.7 1,916 54.6 1,034 57.6 Age (yr) 19–29 306 17.3 512 14.6 62 3.5 < 0.001 30–39 338 19.1 775 22.1 127 7.1 40–49 500 28.2 818 23.3 280 15.6 50–59 342 19.3 674 19.2 322 17.9 60–69 140 7.9 422 12.0 339 18.9 ≥ 70 147 8.3 306 8.7 666 37.1 Education Middle school or lower 359 20.2 688 19.7 1,155 64.4 < 0.001 High school 854 48.2 1389 39.7 412 23.0 College or higher 560 31.6 1,424 40.7 227 12.7 Income (104 Korean won/mo) ≤ 100 163 9.2 302 8.9 795 44.4 < 0.001 101–200 229 12.9 433 12.8 412 23.0 201–300 398 22.4 657 19.5 222 12.4 301–400 368 20.8 677 20.0 143 8.0 ≥ 401 615 34.7 1,308 38.7 217 12.1

1Actual (unweighted) sample size. Missing values on education (n= 8); income (n= 137). 2p-values were calculated by chi-square tests. phic variables, and the estimated mental health prevalence rates other two areas for all age groups (Figure 2). Females showed a in victimized and accident areas are presented as odds ratios higher prevalence of all psychiatric symptoms irrespective of (OR) with 95% confidence intervals (CI) relative to those in area. In general, upward trends in depression, anxiety, and sui- the areas used as a control. Statistical analysis was performed cidal ideation were observed with increasing age. The preva- using SPSS version 24.0 (IBM Corp., Armonk, NY, USA), and lence of stress in the young and middle-aged groups in urban the values were considered statistically significant at p< 0.05 areas (Ansan and Gyeonggi) were generally high compared for a two-tailed test. with Jindo and Haenam. A multiple logistic regression analysis was performed to ad- just for age, sex, education, and income, followed by inter-area RESULTS ORs for psychiatric symptoms (Table 3). Compared with the control area (Gyeonggi), Ansan showed significantly higher ad- The mean age of the subjects (n= 7,076) was 49.7 years, with justed ORs for depression (1.66; 95% CI, 1.36 to 2.04), stress males accounting for 44.9% of the sample. The rural areas of (1.37; 95% CI, 1.10 to 1.71), somatic symptoms (1.31; 95% Jindo and Haenam had a higher mean age than the urban ar- CI, 1.08 to 1.58), anxiety (1.82; 95% CI, 1.39 to 2.39), and eas in Ansan and control cities, as well as higher percentages of suicidal ideation (1.33; 95% CI, 1.13 to 1.56) compared with female residents and lower education and income levels (Table 1). the other areas. In contrast, Jindo and Haenam showed low Prevalence in Ansan for depression, moderate-to-severe stress, ORs in all domains. somatic symptoms, anxiety, and suicidal ideation were 11.4%, 8.8%, 11.9%, 5.9%, and 17.7% (crude rate), respectively. The highest levels among the three areas and the regional differenc- DISCUSSION es proved statistically significant (p< 0.05) (Table 2). After age and sex standardization as well, Ansan’s prevalence rates were In this study, the short-term to mid-term effects of the Sewol highest. Jindo and Haenam showed the lowest post-standard- ferry disaster on the affected community were evaluated by ization prevalence rates. conducting a survey on the mental health status of residents 4 The results for different sex and age groups confirmed the to 6 months after the disaster. Compared to the control area overall tendency of higher prevalence rates in Ansan than the (Gyeonggi), the Ansan area most heavily affected by the disas-

4 Yang HJ et al.: Community mental health status after disaster

Table 2. Prevalence of psychological distress detected by screening tests

Ansan (n= 1,773) Gyeonggi (n= 3,507) Jindo & Haenam (n= 1,796) Screening instruments p-value2 n1 % n % n % Depression (CES-D) Normal 1,515 88.2 3,179 92.8 1,608 91.9 < 0.001 Depression 203 11.8 245 7.2 141 8.1 Score (mean± SD) 10.54± 9.00 8.45± 8.02 8.14± 8.57 Age-sex standardized rate (depression)3 11.4 6.9 5.8 Stress (BEPSI) Low 1,608 91.1 3,272 93.6 1,681 94.0 0.001 Moderate 130 7.4 177 5.1 80 4.5 High 28 1.6 45 1.3 28 1.6 Score (mean± SD) 0.81± 0.67 0.72± 0.62 0.51± 0.66 Age-sex standardized rate (moderate-high) 8.8 6.5 4.4 Somatic symptoms (PHQ-15) Minimal 1,012 58.3 2,190 63.4 1,148 65.0 0.001 Low 511 29.4 930 26.9 439 24.8 Moderate 148 8.5 246 7.1 127 7.2 High 65 3.7 89 2.6 53 3.0 Score (mean± SD) 5.36± 4.38 4.87± 4.06 4.81± 4.10 Age-sex standardized rate (moderate-high) 11.9 9.4 6.4 Anxiety (GAD-7) None 1,343 76.1 2,839 81.4 1,548 86.6 < 0.001 Mild 309 17.5 532 15.3 165 9.2 Moderate 79 4.5 85 2.4 46 2.6 Severe 33 1.9 30 0.9 29 1.6 Score (mean± SD) 2.87± 3.69 2.28± 3.10 1.72± 3.36 Age-sex standardized rate (moderate-severe) 5.9 3.2 3.2 Suicidal ideation No 1,445 81.6 3,004 85.8 1,565 87.3 Yes 326 18.4 498 14.2 228 12.7 < 0.001 Age-sex standardized rate (“yes”) 17.7 13.5 8.5

CES-D, Center for Epidemiologic Studies Depression Scale; BEPSI, Brief Encounter Psychosocial Instrument; PHQ-15, Patient Health Questionnaire-15; GAD-7, Generalized Anxiety Disorder 7-Item Scale; SD, standard deviation. 1Missing values on depressive disorder (n= 185); stress (n= 27); somatic symptoms (n= 118); anxiety (n= 38); suicidal ideation (n= 10). 2p-values were calculated by chi-square tests. 3Age-sex standardized rates were calculated using direct standardization with the 2005 Census performed by the Statistics Korea as the reference.

Table 3. Adjusted odds ratios and 95% confidence interval for the logistic regression of psychological distress by region

Depression Stress Somatic symptoms Anxiety Suicidal ideation (CES-D) (BEPSI, moderate-high) (PHQ-15, moderate-high) (GAD-7, moderate-severe) Ansan 1.66 (1.36, 2.04)*** 1.37 (1.10, 1.71)** 1.31 (1.08, 1.58)** 1.82 (1.39, 2.39)*** 1.33 (1.13, 1.56)*** Gyeonggi 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference) 1.00 (reference) Jindo & Haenam 0.54 (0.42, 0.69)*** 0.62 (0.48, 0.82)** 0.50 (0.40, 0.63)*** 0.66 (0.47, 0.92)* 0.52 (0.43, 0.63)***

Adjusted for sex, age, education, income. CES-D, Center for Epidemiologic Studies Depression Scale; BEPSI, Brief Encounter Psychosocial Instrument; PHQ-15, Patient Health Questionnaire-15; GAD-7, Generalized Anxiety Disorder 7-Item Scale. *p< 0.05, **p< 0.01, ***p< 0.001. ter exhibited higher prevalence rates of the psychiatric symp- contrary to expectation, presumably due to its character as a toms assessed, and similarly high ORs were yielded by the mul- rural area. tiple logistic regression analysis adjusting for sociodemographic The relevance of the Sewol ferry disaster as it relates to the features. Furthermore, Jindo and Haenam near the accident significantly higher prevalence of psychiatric symptoms in the site showed the lowest prevalence of psychiatric symptoms, Ansan area may be estimated by investigating existing data. The

5 Figure 2. Regional distribution of mental health screening results presented by sex/age group Figure 2. Regional distribution of mental health screening results presented by sex/age group Figure 2. Regional distribution of mental health screening results presented by sex/age group A. Depression (CES-D) A. Depression (CES-D) B. Stress (BEPSI) A. Depression (CES-D) 60%0% 70% 80%B. 90%Stress (BEPSI) 100% 60%0% 70% 80% 90% 100% Epidemiology and Health 2015;37:e2015046 80%0% 85% 90% 95% 100% 60%0% Male 70% 80%92.6 90% 100% 7.4 Male 92.6 7.4 Figure 2. Regional distribution of mental health screening80%0% results presented85% by sex/age90% group 95% 100% Sex

Sex Male Female 84.3 15.7 Female92.7 84.3 6.0 1.3 15.7 Male 92.6 7.4 Sex Female 19-29 87.1 12.9 89.719-29 87.18.5 1.8 12.9Low Normal Sex Male Normal 92.7 6.0 1.3 Female 84.3 A. DepressionDepression (CES-D) (CES-D,15.7 %) 19-29 87.9 B. StressStress (BEPSI) (BEPSI,10.8 %) 1.3 Ansan Sex Ansan 30-39 92.7 7.3Ansan Female 89.7 30-39 8.5 1.892.7 7.3 19-29 87.1 12.9 Normal Depression30-39 92.5 6.9 LowLow 0.6 ModerateModerate DepressionHigh 40-49 90.1 9.9 19-29 40-49 90.1 9.9 Ansan 30-39 92.7 7.3 87.9 10.8 1.3

40-49 Age Age Ansan 93.0 5.8 1.2 50-59 86.1 60%0% 13.9Depression 70% 30-39 80% 90%92.5 100%50-59 6.9 86.1 0.680%0% Moderate85% 13.9High 90% 95% 100% 40-49 90.1 9.9 Age 50-59 88.2 8.5 3.2 60-69 85.9 14.1 40-49 60-69 85.9 14.1

Age 93.0 5.8 1.2 50-59 86.1 13.9MaleMale 60-69 94.2 MaleMale 5.1 High0.7 70- 79.9 20.1 Age 92.650-59 88.2 7.4 70- 8.5 79.9 3.2 92.720.1 6.0 1.3 Sex Sex 60-69 85.9 Sex Female14.1Female 70- 91.2 Sex FemaleFemale6.1 2.7 84.3 60-69 15.7 94.2 5.1 0.7 89.7 8.5 1.8 Low 70- 79.9 20.1 19-2919-29 19-2919-29 87.9 10.8 1.3 0% 87.1 70- 91.212.9 Normal0% 6.1 2.7 60%Ansan 70%Ansan 80% 90% 100% Ansan60%Ansan 70% 80% 90% 100% Moderate 30-3930-39 92.7 80%0% 7.3 85% 90%30-3930-39 95% 100%92.5 6.9 0.6 0% Depression 40-49 60%Male 70% 80%93.7 90% 40-4940-49 100% 6.3 90.1 9.9 Male 40-4993.7 93.0 6.3 5.8 1.2 High Age Age Sex

Age 80%0% 85% 90% 95%Age 100% Sex Male Female 92.2 50-5950-59 7.886.1 13.9 Female94.1 50-5950-5992.2 4.5 88.21.4 7.8 8.5 3.2 Male 93.7 6.3 Sex Female 19-29 93.3 60-6960-69 6.785.9 14.1 19-2993.3 60-6960-6993.35.5 1.2 94.2 6.7 5.1 0.7 Sex Male Gyeonggi 94.1 4.5 1.4 GyeonggiFemale 92.2 70-7.8 19-29 91.8 70- 8.0 0.2 70- 79.9 Sex 20.1 70- 91.2 6.1 2.7 30-39 94.2 Gyeonggi5.8 Female 93.3 30-39 5.5 1.294.2 5.8 19-29 93.3 6.7 30-39 93.9 4.9 1.2 Gyeonggi 40-49 94.2 5.8 19-29 91.8 40-49 8.0 0.294.2 5.8 30-39 94.2 5.8 Gyeonggi 40-49 Age

Age 95.2 4.1 0.7 50-59 91.5 60%0% 70%8.5 30-39 80% 90%93.9 100%50-59 4.9 1.291.580% 0% 85%8.5 90% 95% 100% 40-49 94.2 5.80% 70% 80% Age 50-5990% 100% 93.2 0%5.1 1.8 85% 90% 95% 100% 60-69 91.4 8.6 40-49 60-69 91.4 8.6

Age 95.2 4.1 0.7 50-59 91.5 Male8.5 60-69 92.4 Male4.7 2.8 70- 89.9 Male 10.1 Age 93.750-59 93.26.3 70- 5.1 Male1.889.9 94.1 10.1 4.5 1.4 Sex Sex

60-69 Sex 91.4 Sex 8.6 70- 94.7 3.6 1.6 FemaleFemale 92.260-69 7.8 FemaleFemale 93.3 5.5 1.2 70- 92.4 4.7 2.8 89.9 19-2919-2910.1 93.3 6.7 19-2919-29 91.8 8.0 0.2 0% GyeonggiGyeonggi 70- 94.7 Gyeonggi60%0%Gyeonggi 3.6 70% 1.6 80% 90% 100% 60% 70% 80% 90% 100% 30-39 30-3930-39 94.2 80%0% 5.8 85% 90%30-39 95% 100% 93.9 4.9 1.2 0% 60%Male 70% 80%93.1 90% 40-4940-49 100% 6.9 94.2 5.8 Male 40-4940-4993.1 95.2 6.9 4.1 0.7 Age Age Sex

80% 85% 90% 95%Age 100% Age Sex 0% Male Female 91.0 50-5950-59 9.0 91.5 8.5 Female 95.0 50-5950-5991.0 3.7 1.3 93.2 9.0 5.1 1.8 Male 93.1 6.9 Sex Female Jindo 19-29 95.1 60-6960-69 4.9 91.4 Jindo 8.6 19-2993.2 60-6960-69 95.15.1 1.7 92.4 4.9 4.7 2.8 Sex Male 95.0 3.7 1.3 Female 91.0 70-9.0 Jindo 19-29 96.7 70-70- 1.6 1.6 89.9Sex 10.1 94.7 3.6 1.6 & Haenam 30-39 95.9 70- 4.1 Female & Haenam93.2 30-39 5.1 1.7 95.9 4.1 Jindo 19-29 95.1 4.9 & Haenam 30-39 96.8 2.4 0.8 40-49 93.5 Jindo 6.5 19-29 96.7 40-49 1.6 1.6 93.5 6.5 & Haenam 30-39 95.9 4.1 40-49 Age Age 0% 95.4 4.3 0.4 50-59 94.6 60%0%& Haenam 70% 5.4 30-39 80% 90% 96.8 100% 50-59 2.4 0.880%0%94.6 85%85% 90%90%5.4 95%95% 100% 40-49 93.5 6.5 Age 50-59 95.0 4.0 0.9 60-69 92.1 7.9 40-49 60-69 92.1 7.9

Age 95.4 4.3 0.4 50-59 94.6 Male 5.4 60-69 94.1 Male 3.6 2.4 70- 88.9 Male 11.1 Age 93.150-59 95.06.9 70- 4.0 88.9Male0.9 95.0 11.1 3.7 1.3 Sex 60-69 Sex Sex 92.1 Sex 7.9 70- 92.0 5.9 2.1 FemaleFemale 91.0 60-69 94.19.0 3.6 FemaleFemale2.4 93.2 5.1 1.7 70- 88.9 11.1 Jindo JindoJindo & 19-2919-29 95.170- 92.0 4.9 Jindo5.9 & 19-2919-292.1 96.7 1.6 1.6 & Haenam 30-3930-39 95.9 4.1 & Haenam 30-3930-39 96.8 2.4 0.8 C. Somatic symptomsHaenam (PHQ-15) C. Somatic symptomsHaenam (PHQ-15) 40-4940-49 93.5 6.5 40-4940-49 95.4 4.3 0.4 Age Age Age C. Somatic symptoms (PHQ-15) Age 50-5950-59 94.6 C. Somatic symptoms5.4 (PHQ-15) 50-5950-59 95.0 4.0 0.9 60-6960-69 92.1 7.9 60-6960-69 94.1 3.6 2.4 C. Somatic symptoms (PHQ-15) 0% 20% 40% 60%C. Somatic 80%symptoms 100% (PHQ-15) 0% 20% 40% 60% 80% 100% 70-70- 88.9 11.1 70-70- 92.0 5.9 2.1 C. Somatic symptoms (PHQ-15) 0% 20%Male 40% 60%C. Somatic73.2 80% symptoms 100% (PHQ-15)21.4 3.7 1.6 0% 20%Male 40% 60%73.2 80% 100% 21.4 3.7 1.6

Sex Female 45.3 36.4 12.7 5.6 MinimalA Sex Female 45.3 36.4 12.7 5.6 MinimalB 0%Male 20% 40% 60%73.2 80% 100% 21.4 3.7 1.6 0%Male 20% 40% 60%73.2 80% 100% 21.4 3.7 1.6 19-29 19-29

Sex 59.9 27.8 8.3 4.0 Sex 59.9 27.8 8.3 4.0 Female 45.3 36.4 12.7 5.6 Minimal FemaleMild 45.3 36.4 12.7 5.6 Minimal Mild 0% 20%Male 40%Ansan 60%73.2 30-39 80% 100% 21.457.2 3.7 1.6 0%31.7 20%7.8Male 3.3 40%Ansan 60%73.2 30-39 80% 100% 21.457.2 3.7 1.6 31.7 7.8 3.3 19-29 C. Somatic symptoms (PHQ-15) 19-29 Sex 59.9 27.8 Minimal8.3 4.0 MildSex Moderate 59.9 27.8 Minimal8.3 4.0 Mild Moderate Female 45.3 Somatic40-49 symptoms36.4 (PHQ-15,58.812.7 5.6 %) 29.7 Female8.2 3.3 45.3 Anxiety40-49 (GAD-7,36.4 %) 58.812.7 5.6 29.7 8.2 3.3 Male Ansan 73.2 30-39 57.221.4 3.7 1.6 31.7 7.8Male 3.3 Ansan 73.2 30-39 D. Anxiety (GAD-7)57.221.4 3.7 1.6 31.7 7.8 3.3 19-29 Age 50-59 19-29 Age 50-59 Sex 59.9 27.8 60.5Minimal8.3 4.0 Sex MildMild 30.4 ModerateModerate6.6 2.4 SevereSevere 59.9 27.8 60.5Minimal8.3 4.0 MildMild 30.4 ModerateModerate6.6 2.4 SevereSevere Female 45.3 40-49 36.4 58.812.7 5.6 29.7 Female8.2 3.3 45.3 40-49 36.4 58.812.7 5.6 29.7 8.2 3.3 Ansan 30-39 57.2 60-69 31.7 0%63.67.8 3.3 20% Ansan 40%24.3 60%30-397.9 4.3 80% 100%57.2 60-69 31.7 63.67.8 3.3 24.3 7.9 4.3

Age Age 0% 19-29 59.9 50-59 27.8 60.5 8.3 4.0 0% Mild 30.420% Moderate6.619-2940% 2.4 60%Severe 59.980% 50-59100% 27.8 60.5 8.3 4.0 0%60% Mild 30.470% 70% Moderate6.6 2.4 80% 80% Severe 90% 90% 100% 100% 40-49 58.8 70- 29.745.0 8.2 3.3 29.3 17.140-49 8.6 58.8 70- 29.745.0 8.2 3.3 29.3 17.1 8.6 Ansan 30-39 57.2 60-69 31.7 63.6 7.8 3.3 Ansan24.3 7.930-39 4.3 57.2 60-69 31.7 63.6 7.8 3.3 24.3 7.9 4.3 Age 50-59 60.5 Male30.4Male Moderate6.6 2.4 Severe73.2 Age 50-59 21.4 3.7 1.6 60.5 Male30.4Male Moderate6.6 2.4 80.2 Severe 14.9 3.7 1.2

40-49 Sex 40-49 Sex Minimal 58.8 70- 29.7 8.2 3.3 58.8 70- 29.7 8.2Sex 3.3 45.0 29.3 17.1 8.6 45.0 Sex 29.3 17.1 8.6 Minimal 60-69 63.6 FemaleFemale24.3 7.9 4.3 45.3 36.460-69 12.7 5.6 63.6 FemaleFemale24.3 7.972.7 4.3 19.7 5.2 2.4 Age 50-59 60.5 30.4 0%6.6 2.4 20%Severe 40%Age 60%50-59 80% 100%60.5 30.4 0%6.6 2.4 20%Severe 40% 60% 80% 100% 70- 45.0 29.3 19-2919-29 17.1 8.6 59.9 70-27.8 8.3 45.04.0 Mild 29.3 19-2919-29 17.1 8.6 77.6 17.1 3.6 1.6 Mild 60-69 63.6 24.3 7.9 4.3 60-69 63.6 Ansan24.3 7.9 4.3 0%Ansan 20%MaleAnsan 40%30-3930-39 60%76.7 80%57.2 100% 19.0 3.031.71.3 7.8 3.30% 20%AnsanMale 40%30-3930-39 60%76.7 78.3 80% 100% 19.017.63.0 1.3 3.0 1.2 70- 70- Moderate

45.0 29.3 Sex 17.1 8.6 45.0 29.3 Sex 17.1 8.6 Moderate Female 40-4940-4952.3 58.8 33.5 10.5 29.73.6 8.2 3.3 Female 40-4940-4952.3 77.7 33.5 15.910.5 3.6 4.4 2.0 0%Male 20% 40% 60%76.7 80% 100% 19.0 3.01.3 0%Male 20% 40% 60%76.7 80% 100% 19.0 3.01.3 Age Age Age 19-29 50-5950-59 60.5 30.4 6.6 2.4 Severe19-29 Age 50-5950-59

Sex 63.9 25.7 7.8 2.5 Sex 63.9 72.6 25.720.6 7.8 2.5 4.7 2.1 Severe FemaleGyeonggi 52.3 33.5 10.5 3.6 FemaleGyeonggi 52.3 33.5 10.5 3.6 0% 20%Male 40% 60%76.7 30-39 80% 100%60-6960-69 19.061.5 3.01.3 63.6 0%29.9 20%Male6.9 1.724.3 40%7.9 4.3 60%76.7 30-39 80% 100%60-6960-69 19.061.5 3.074.81.3 29.916.5 6.9 1.7 6.5 2.2 19-29 19-29

Sex 63.9 25.7 7.8 2.5 Sex 63.9 25.7 7.8 2.5 GyeonggiFemale 52.3 40-49 33.5 70-70- 65.410.5 3.6 45.0 26.9Gyeonggi29.3 Female 6.0 1.8 17.1 8.6 52.3 40-49 33.570-70- 65.410.572.1 3.6 17.726.9 6.0 1.8 7.5 2.7 Male 76.7 30-39 61.519.0 3.01.3 29.9 6.9Male 1.7 76.7 30-39 61.519.0 3.01.3 29.9 6.9 1.7 19-29 Age 50-59 19-29 Age 50-59 Sex 63.9 25.7 71.27.8 2.5 Sex 22.2 4.9 1.8 63.9 25.7 71.27.8 2.5 22.2 4.9 1.8 GyeonggiFemale 52.3 40-49 33.5 65.410.5 3.6 26.9 GyeonggiFemale6.0 1.8 52.3 40-49 33.5 65.410.5 3.6 26.9 6.0 1.8 30-39 61.5 60-69 29.90% 61.86.9 1.7 20% 40%26.3 60%7.030-39 4.8 80% 100%61.5 60-69 29.9 0%61.86.9 1.7 26.3 7.0 4.8 19-29 63.9 Age 50-59 25.7 71.27.8 2.5 0% 20%22.2 4.919-2940% 1.8 60% 63.980% Age 50-59100% 25.7 71.27.8 2.5 60%0% 70% 70%22.2 4.9 1.8 80% 80% 90% 90% 100% 100% Gyeonggi 40-49 65.4 70- 26.947.2 Gyeonggi6.0 1.8 32.7 14.540-49 5.6 65.4 70- 26.947.2 6.0 1.8 32.7 14.5 5.6 30-39 61.5 60-69 29.9 61.8 6.9 1.7 26.3 7.030-39 4.8 61.5 60-69 29.9 61.8 6.9 1.7 26.3 7.0 4.8 Age 50-59 71.2 MaleMale22.2 4.9 1.8 76.7 Age 50-59 19.0 3.01.3 71.2 MaleMale22.2 4.9 1.8 83.5 13.9 1.8 0.8

40-49 Sex 40-49

65.4 70- 26.9 6.0 1.8 65.4 70- 26.9 6.0Sex 1.8 47.2 Sex 32.7 14.5 5.6 47.2 Sex 32.7 14.5 5.6 60-69 61.8 FemaleFemale26.3 7.0 4.8 52.3 60-6933.5 10.5 3.6 61.8 FemaleFemale26.3 7.0 4.8 79.7 16.4 2.9 0.9 Age 50-59 71.2 22.20% 4.9 1.8 20% 40%Age 60%50-59 80% 100%71.2 22.20% 4.9 1.8 20% 40% 60% 80% 100% 70- 47.2 32.7 19-2919-29 14.5 5.6 63.9 70- 25.7 7.8 47.22.5 Gyeonggi 32.7 19-2919-29 14.5 5.6 77.7 18.4 3.3 0.6 60-69 61.8 GyeonggiGyeonggi26.3 7.0 4.8 60-69 61.8 Gyeonggi26.3 7.0 4.8 0% 20%Male 40%30-3930-39 60%79.1 80%61.5 100% 15.4 4.0 1.529.9 6.9 1.70% 20%Male 40%30-3930-39 60%79.1 80%80.7 100% 15.4 4.016.81.5 1.9 0.5 70- 70-

47.2 32.7 Sex 14.5 5.6 47.2 32.7 Sex 14.5 5.6 Female 40-4940-49 54.7 65.4 31.7 9.5 4.126.9 6.0 1.8 Female 40-4940-49 54.7 81.1 31.7 9.5 16.34.1 2.1 0.5 0%Male 20% 40% 60%79.1 80% 100% 15.4 0%Male 20% 40% 60%79.1 80% 100% 15.4 Age 4.0 1.5 4.0 1.5 Age Age 19-29 50-5950-59 71.2 22.2 4.9 1.8 19-29 Age 50-5950-59 84.0 12.9 2.2 0.9 Sex Jindo 77.4 16.1 4.8 1.6 Sex Jindo 77.4 16.1 4.8 1.6 Female 54.7 31.7 9.5 4.1 Female 54.7 31.7 9.5 4.1 0% 20%Male 40% 60%79.1 30-39 80% 100%60-6960-69 15.467.74.0 1.5 61.8 0%26.8 20%Male3.9 26.31.6 40%7.0 4.8 60%79.1 30-39 80% 100%60-6960-69 15.467.74.0 1.5 83.3 26.8 3.911.9 1.6 2.4 2.4 &19-29 Haenam &19-29 Haenam

Sex Jindo 77.4 16.1 4.8 1.6 Sex Jindo 77.4 16.1 4.8 1.6 Female 54.7 40-49 31.770-70- 9.575.34.1 47.2 19.632.7Female 3.3 1.8 14.5 5.6 54.7 40-49 31.770-70- 9.575.34.1 82.3 19.6 3.313.1 1.8 3.6 1.0 Male & Haenam 79.130-39 15.467.74.0 1.5 26.8 3.9Male 1.6 & Haenam 79.130-39 15.467.74.0 1.5 26.8 3.9 1.6 19-29 Age 19-29 Age Sex Jindo 77.4 50-59 16.1 4.880.6 1.6 Sex Jindo 15.91.9 1.6 77.4 50-59 16.1 4.880.6 1.6 15.91.9 1.6 Female 54.7 40-49 31.7 75.39.5 4.1 19.6 Female3.3 1.8 54.7 40-49 31.7 75.39.5 4.1 19.6 3.3 1.8 & Haenam 30-39 67.7 60-69 26.80% 66.03.9 1.6 20% & Haenam 40%23.8 60%30-396.6 3.6 80% 100% 67.7 60-69 26.80% 66.03.9 1.6 23.8 6.6 3.6 Jindo 19-29 77.4Age 50-59 16.1 80.64.8 1.6 0% Jindo 20% 15.91.919-29 40% 1.6 60% 80%77.4Age 50-59100% 16.1 80.64.8 1.6 60%0% 70% 70% 15.91.9 1.6 80% 80% 90% 90% 100% 100% 40-49 75.3 70- 51.019.6 3.3 1.8 32.3 12.540-49 4.3 75.3 70- 51.019.6 3.3 1.8 32.3 12.5 4.3 & Haenam 30-39 67.7 60-69 26.866.0 3.9 1.6 & Haenam23.8 6.630-39 3.6 67.7 60-69 26.866.0 3.9 1.6 23.8 6.6 3.6 Age 50-59 80.6 MaleMale 15.91.9 1.6 79.1 Age 50-59 15.4 4.0 1.5 80.6 MaleMale 15.91.9 1.6 88.7 7.9 1.7 1.7

40-49 Sex 40-49

75.3 70- 19.6 3.3 1.8 75.3 70- 19.6 3.3Sex 1.8 Sex

51.0 Sex 32.3 12.5 4.3 51.0 32.3 12.5 4.3 60-69 66.0 FemaleFemale23.8 6.6 3.6 54.7 60-6931.7 9.5 4.1 66.0 FemaleFemale23.8 6.6 3.6 85.0 10.2 3.2 1.6 Age 50-59 80.6 15.91.9 1.6 Age 50-59 80.6 Jindo 15.91.9 1.6 70- 51.0 JindoJindo & 32.319-29 19-29 12.5 4.3 77.4 70- 16.1 4.8 1.651.0 Jindo & 32.319-29 19-29 12.5 4.3 85.5 11.3 3.2 60-69 66.0 23.8 6.6 3.6 60-69 66.0 & Haenam23.8 6.6 3.6 & HaenamHaenam 30-3930-39 67.7 26.8 3.9 1.6 Haenam 30-3930-39 89.0 9.4 1.6 70- 51.0 32.3 12.5 4.3 70- 51.0 32.3 12.5 4.3 40-4940-49 75.3 19.6 3.3 1.8 40-4940-49 88.2 8.6 2.9 0.4 Age Age Age 50-5950-59 80.6 15.91.9 1.6 Age 50-5950-59 89.7 6.9 1.6 1.9 60-6960-69 66.0 23.8 6.6 3.6 60-6960-69 85.9 8.4 2.7 3.0 70-70- 51.0 32.3 12.5 4.3 70-70- 84.4 10.8 3.3 1.5 C D

Figure 2. Regional distribution of mental health screening results (A: depression, B: stress, C: somatic symptoms, D: anxiety, E: suicidal ideation) presented by sex/age group. CES-D, Center for Epidemiologic Studies Depression Scale; BEPSI, Brief Encounter Psychosocial In- strument; PHQ-15, Patient Health Questionnaire-15; GAD-7, Generalized Anxiety Disorder 7-Item Scale. (continued to the next page)

6 Figure 2. Regional distribution of mental health screening results presented by sex/age group Figure 2. Regional distribution of mental health screening results presented by sex/age group A. Depression (CES-D) A. Depression (CES-D) 60%0% 70% 80% 90% 100% Yang HJ et al.: Community mental health status after disaster 60%0% Male 70% 80%92.6 90% 100% 7.4

Sex Female 84.3 15.7 Male 92.6 7.4 19-29 87.1 12.9 Sex Normal Female 84.3 15.7 Ansan 30-39 Suicidal ideation92.7 (%) 7.3 As indicators allowing for time-series comparison, experience 19-29 87.1 12.9 NormalNo DepressionYe s 40-49 E. Suicidal ideation90.1 9.9 Ansan 30-39 92.7 7.3 of depressive symptoms and perceived stress questions, one Age 50-59 86.1 50%0% 60%13.9Depression 70% 80% 90% 100% 40-49 90.1 9.90% 70% 80% 90% 100% 60-69 85.9 14.1 Age 50-59 86.1 13.9 item each, are annually assessed in the KCHS. In Ansan, the 70- 79.9 MaleMale 20.1 85.7 14.3 Sex 60-69 85.9 Sex Female14.1Female 78.0 22.0 age-standardized rate of depressive symptoms increased from 70- 79.9 20.1 19-2919-29 84.3 15.7 No 60%0% 70% 80% 90% 100% AnsanAnsan 30-3930-39 84.6 15.4 9.8% in 2013 to 13.0% in 2014 in Danwon, and decreased Yes 60%0% Male 70% 80%93.7 90% 40-4940-49 100% 81.06.3 19.0 from 12.7% to 8.1% in Sangnok, although higher than the av- Age Age Sex Female 92.2 50-5950-59 80.97.8 19.1 Male 93.7 6.3 19-29 93.3 60-6960-69 75.7 6.7 24.3 erage of 7.0% in Gyeonggi [20]. Danwon showed the highest Sex GyeonggiFemale 92.2 7.8 30-39 94.2 70-70- 78.25.8 21.8 19-29 93.3 6.7 depression rate of the 254 communities across the country. This Gyeonggi 40-49 94.2 5.8 30-39 94.2 5.8 Age 50-59 91.5 0%50% 60%8.5 70% 80% 90% 100% 40-49 94.2 0%5.8 70% 80% 90% 100% is consistent with our survey results using the CES-D and indic- 60-69 91.4 8.6 Age 50-59 91.5 Male8.5 70- 89.9 Male 10.1 89.0 11.0 ative of the effects of the Sewol ferry disaster. However, it Sex 60-69 91.4 Sex FemaleFemale8.6 83.1 16.9 70- 89.9 19-2919-2910.1 88.7 11.3 should be taken into account that the depression rate in Ansan 60%0% GyeonggiGyeonggi 70% 80% 90% 100% 30-3930-39 88.4 11.6 0% was also high in 2013 and over the 75% quantile. Perceived 60%Male 70% 80%93.1 90% 40-4940-49 100% 6.9 86.3 13.7 Age Age Sex Female 91.0 50-5950-59 9.0 85.5 14.5 stress slightly decreased from 33.3% to 31.7% in Danwon and Male 93.1 6.9 Jindo 19-29 95.1 60-6960-69 82.24.9 17.8 Sex Female 91.0 9.0 from 35.6% to 26.2% in Sangnok, without showing any no- & Haenam 30-39 95.9 70-70- 78.74.1 21.3 Jindo 19-29 95.1 4.9 40-49 93.5 6.5 & Haenam 30-39 95.9 4.1 ticeable differences from the Gyeonggi average of 30.2% (95% Age 50-59 94.6 50%0%0% 60%70% 5.4 70%80% 80%90% 90% 100%100% 40-49 93.5 6.5 60-69 92.1 7.9 CI, 29.7% to 30.7%) [20]. These findings deviate slightly from Age 50-59 94.6 MaleMale5.4 70- 88.9 11.1 90.7 9.3 Sex 60-69 92.1 Sex FemaleFemale7.9 84.8 15.2 what was found in this study for stress. 70- 88.9 JindoJindo & 19-2911.119-29 98.4 1.6 Somatic symptoms and anxiety were also measured in this & HaenamHaenam 30-3930-39 92.9 7.1 40-4940-49 87.5 12.5 study. The ORs in the Ansan area were higher than in Gyeonggi. Age Age 50-5950-59 91.6 8.4 60-6960-69 87.2 12.8 This result is consistent with the literature that indicates unex- 70-70- 83.0 17.0 pected disaster gives rise to anxiety through uncertainty about E danger [22], and triggers various nonspecific symptoms inexpli- Figure 2. Continued. cable with physical disorders [23]. Even though anxiety is a normal reaction to disaster in the short term, a prolonged peri- 2009 KCHS measured depression rates using the CES-D, and od of anxiety can develop into a chronic state of anxiety and suicidal ideation is assessed every other year. In 2009, the age- trigger psychosomatic problems [24]. standardized depression rates in Danwon and Sangnok of the The age-dependent prevalence of depression shown in this Ansan area were 4.3% and 4.8%, respectively. The rates for study is similar to the study of Oh et al. [25] conducted in 2009, Gunpo, Paldal (Suwon), Guri, and Namyangju in Gyeonggi in which the CES-D results from the community health survey were 6.2%, 8.3%, 7.1%, and 5.9%, respectively; and for Jindo were analyzed and mapped by age. Subjects in their 30s showed and Haenam rates were 1.7% and 3.5%, respectively [20]. The the lowest depression prevalence, which increased with age, 2014 depression data gathered in this survey demonstrated whereas those aged 19 to 29 showed a slightly higher depres- considerably higher depression rates in the Ansan area (11.8%) sion rate than people in their 30s. The effect of the disaster on as well as in Jindo and Haenam (8.1%). With a time gap of five children and adolescents could not be assessed because this years, this increase cannot be exclusively attributed to the ef- study only targeted adults. This aspect should be investigated in fects of the Sewol ferry disaster, but it may be assumed to have future research, given that early detection of and intervention certain effects, given that no significant changes took place in for psychological injury in children and adolescents still in the the control area (6.9% vs. 5.9 to 8.3%). formative years of their lives are of vital importance [26]. With regard to suicidal ideation, after age standardization the Most of the mental health research on large-scale disasters affirmative answer accounted for 9.6% (Danwon) and 17.2% has been conducted focusing on survivors or bereaved families, (Sangnok) in Ansan in 2013, 7.2% (Gunpo), 10.6% (Paldal/Su- and only a small number of studies have examined the effects won), 8.1% (Guri), and 13.5%, (Namyangju) in Gyeonggi, and of disaster on the general population at the community level. 3.3% and 4.9% in Jindo and Haenam, respectively [20]. This There are a few community-based studies in relation to the ocean indicates that in one year’s time suicidal ideation increased in contamination by the 2007 Hebei Spirit oil spill, but most par- all surveyed communities except for Namyangju, with Danwon ticipants were individuals directly involved through cleanup in Ansan demonstrating the highest increase of 7.6%. However, works or residents who suffered financial losses [8,9,27]. In oth- the higher baseline value in the Ansan area compared with er countries, too, most studies targeted eyewitnesses or resi- other areas should be taken into account. This may be inter- dents within the areas affected by natural disasters, such as ty- preted as a higher vulnerability of already affected subjects to phoons and earthquakes [5,28]. In contrast, the Sewol ferry di- the effects of the disaster [21]. saster did not take place in a physical space of residence of the

7 Epidemiology and Health 2015;37:e2015046 victims, and residents experienced the accident only indirectly directly involved in the disaster, such as bereaved families and via media or acquaintances. It is also distinct from other disas- relatives, could not be targeted for sampling due to the nature ters in that the main victims were adolescents from the commu- of the community-based sampling system of this study. Fur- nities concerned, thus bringing a tremendous shock and severe thermore, items evaluating the direct relevance of the Sewol emotional and psychological trauma to community members of ferry disaster were excluded from the questionnaire to forestall all ages. Taking into account such distinct features and the impli- prejudices related to the disaster. This did not allow for the ap- cations of this study, comprehensive mental health monitoring plication of detailed risk classification models in accordance and interventions are necessary for the community members. with the degree of exposure. Nor could we apply the survey in- Psychological and emotional tension caused by a disaster es- strument for posttraumatic stress disorder because most com- calates the collective stress of the whole community. Fears of munity samples did not directly experience the disaster. How- other disasters and worries about impending crises are ignited. ever, some studies showed that residents who had vicariously Such a changed atmosphere and loss of human and material re- experienced a national disaster developed posttraumatic stress sources caused by the disaster can create long-term stress [29]. disorder [3]. It is therefore considered necessary to develop a Although psychological damage may mitigate over time, if left screening tool for evaluating disaster-related stress in the gener- uncared for or complicated by other negative factors long-term al population. problems can develop. This study was conducted 4 to 6 months Despite these limitations, various aspects of mental health after the disaster, and a Taiwanese study conducted at the same were evaluated using screening tools with recognized efficien- point in time similarly reported symptoms of posttraumatic cies. A salient feature of this study is its representativeness, in stress disorder and severe depression after a major disaster [28]. which systematically extracted samples were used from the ex- Such symptoms have also been reported as much as seven years isting community health survey system. It is also significant that after the disaster [5,29]. A follow-up survey in relation to the this study could evaluate the mental health status of communi- 9/11 terrorist attack noted that persisting long-term psychologi- ties within six months after the disaster. Given the results of cal symptoms were more associated with absence of timely in- this cross-sectional survey, it is considered necessary to provide terventions than with direct exposure to the disaster [2]. There- adequate interventions to actively help residents of victimized fore, interventions should be administered timely. communities recover their mental health status. A follow-up One of the limitations of this study is the selection of control study to observe the change and recovery process in the affect- group. Four regions in Gyeonggi, the province where Ansan is ed communities will be significant from both academic and so- located, were selected as control communities to minimize re- cial perspectives. gional deviations, but even after adjusting for sociodemographic features there were inter-region deviations. Although Ansan showed a higher prevalence rate in 2014, the health index re- ACKNOWLEDGEMENTS lated to mental health status showed an elevated level in a prior survey. Jindo and Haenam had low prevalence rates in the pres- This study was supported by a grant from the Korea Centers ent and prior surveys but their geographical features are distinct for Disease Control and Prevention (no. 2014-P33002-00), Ko- from Gyeonggi, which did not allow for direct comparison with rean Society of Epidemiology. We would like to express appre- the control communities. Moreover, the Sewol ferry disaster ciation of all those who helped to conduct this study, including was a national tragedy that was reported in real time via media, the interviewers, researchers, the Korean Society for Preventive leaving strong emotional imprints in all Koreans. As such, there Medicine and the Korean Society of Epidemiology. may have been an underestimation of the effects of the disaster on the accident region. Another limitation was the lack of data and the impossibility CONFLICT OF INTEREST of time-series observation. However, meaningful observations could be made using different forms of variables and past data. The authors have no conflicts of interest to declare for this As for depression and suicidal ideation, different reactions of study. respondents between a face-to-face interview and self-administered questionnaire should also be taken into account, because sensitive questions may have been answered more honestly in SUPPLEMENTARY MATERIAL the latter such that underestimation of past prevalence rates cannot be ruled out. Supplementary material (Korean version) is available at http: As a third limitation, it should be pointed out that individuals //www.e-epih.org/.

8 Yang HJ et al.: Community mental health status after disaster

ORCID 2012;55:74-83 (Korean). 13. Radloff LS. The CES-D scale: a self-report depression scale for re- Hee Jung Yang http://orcid.org/0000-0001-5610-7075 search in the general population. Appl Psychol Meas 1977;1:385-401. 14. Cho MJ, Kim KH. Diagnostic validity of the CES-D (Korean ver- Hae Kwan Cheong http://orcid.org/0000-0003-2758-9399 sion) in the assessment of DSM-III-R major depression. J Korean Bo Youl Choi http://orcid.org/0000-0003-0115-5736 Neuropsychiatr Assoc 1993;32:381-399 (Korean). Min-Ho Shin http://orcid.org/0000-0002-2217-5624 15. Frank SH, Zyzanski SJ. Stress in the clinical setting: the Brief En- Hyeon Woo Yim http://orcid.org/0000-0002-3646-8161 counter Psychosocial Instrument. J Fam Pract 1988;26:533-539. 16. Yim J, Bae J, Choi S, Kim S, Hwang H, Huh B. The validity of mod- Dong-Hyun Kim http://orcid.org/0000-0002-1492-5253 ified Korean-translated BEPSI (Brief Encounter Psychosocial Instru- Gawon Kim http://orcid.org/0000-0002-4486-3428 ment) as instrument of stress measurement in outpatient clinic. J Ko- Soon Young Lee http://orcid.org/0000-0002-3160-577X rean Acad Fam Med 1996;17:42-53. 17. Kroenke K, Spitzer RL, Williams JB. The PHQ-15: validity of a new measure for evaluating the severity of somatic symptoms. Psycho- som Med 2002;64:258-266. REFERENCES 18. Han C, Pae CU, Patkar AA, Masand PS, Kim KW, Joe SH, et al. Psy- chometric properties of the Patient Health Questionnaire-15 (PHQ- 1. Galea S, Ahern J, Resnick H, Kilpatrick D, Bucuvalas M, Gold J, et 15) for measuring the somatic symptoms of psychiatric outpatients. al. Psychological sequelae of the September 11 terrorist attacks in Psychosomatics 2009;50:580-585. New York City. N Engl J Med 2002;346:982-987. 19. Spitzer RL, Kroenke K, Williams JB, Löwe B. A brief measure for 2. Silver RC, Holman EA, McIntosh DN, Poulin M, Gil-Rivas V. Na- assessing generalized anxiety disorder: the GAD-7. Arch Intern Med tionwide longitudinal study of psychological responses to September 2006;166:1092-1097. 11. JAMA 2002;288:1235-1244. 20. Korea Centers for Disease Control and Prevention. Community Health 3. Galea S, Vlahov D, Resnick H, Ahern J, Susser E, Gold J, et al. Trends Survey [cited 2015 Oct 12]. Available from: https://chs.cdc.go.kr/ of probable post-traumatic stress disorder in New York City after the chs/index.do (Korean). September 11 terrorist attacks. Am J Epidemiol 2003;158:514-524. 21. Nolen-Hoeksema S, Morrow J. A prospective study of depression 4. Schlenger WE, Caddell JM, Ebert L, Jordan BK, Rourke KM, Wil- and posttraumatic stress symptoms after a natural disaster: the 1989 son D, et al. Psychological reactions to terrorist attacks: findings from Loma Prieta Earthquake. J Pers Soc Psychol 1991;61:115-121. the National Study of Americans’ Reactions to September 11. JAMA 22. Norris FH, Stevens SP, Pfefferbaum B, Wyche KF, Pfefferbaum RL. 2002;288:581-588. Community resilience as a metaphor, theory, set of capacities, and 5. Kessler RC, Galea S, Gruber MJ, Sampson NA, Ursano RJ, Wessely strategy for disaster readiness. Am J Community Psychol 2008;41: S. Trends in mental illness and suicidality after Hurricane Katrina. 127-150. Mol Psychiatry 2008;13:374-384. 23. Escobar JI, Canino G, Rubio-Stipec M, Bravo M. Somatic symptoms 6. Lee DY, Na J, Sim M. Psychological reactions and physical trauma after a natural disaster: a prospective study. Am J Psychiatry 1992; by types of disasters: view from man-made disaster. J Korean Neuro- 149:965-967. psychiatr Assoc 2015;54:261-268 (Korean). 24. Norris FH, Stevens SP. Community resilience and the principles of 7. Norris FH, Friedman MJ, Watson PJ, Byrne CM, Diaz E, Kaniasty K. mass trauma intervention. Psychiatry 2007;70:320-328. 60,000 disaster victims speak: Part I. An empirical review of the em- 25. Oh DH, Kim SA, Lee HY, Seo JY, Choi BY, Nam JH. Prevalence pirical literature, 1981-2001. Psychiatry 2002;65:207-239. and correlates of depressive symptoms in korean adults: results of a 8. Song M, Hong YC, Cheong HK, Ha M, Kwon H, Ha EH, et al. Psy- 2009 Korean Community Health Survey. J Korean Med Sci 2013;28: chological health in residents participating in clean-up works of He- 128-135. bei Spirit oil spill. J Prev Med Public Health 2009;42:82-88 (Korean). 26. Reijneveld SA, Crone MR, Verhulst FC, Verloove-Vanhorick SP. The 9. Kim YM, Park JH, Choi K, Noh SR, Choi YH, Cheong HK. Burden effect of a severe disaster on the mental health of adolescents: a con- of disease attributable to the Hebei Spirit oil spill in Taean, Korea. trolled study. Lancet 2003;362:691-696. BMJ Open 2013;3:e003334. 27. Eum JH, Cheong HK, Ha M, Kwon HJ, Ha EH, Hong YC, et al. He- 10. Federal Emergency Management Agency. Crisis counseling assis- bei Spirit oil spill exposure and acute neuropsychiatric effects on res- tance and training program guidance; 2013 [cited 2015 Nov 6]. idents participating in clean-up work. Korean J Epidemiol 2008;30: Available from: http://www.dhhr.wv.gov/healthprep/programs/be- 239-251 (Korean). havioralhealth/Documents/FEMA%20CCP%20Toolkit.pdf. 28. Chou FH, Su TT, Chou P, Ou-Yang WC, Lu MK, Chien IC. Survey 11. Gwon MC. First report about actual condition of Sewol Ferry victims’ of psychiatric disorders in a Taiwanese village population six months families. CBS Nocut news; 2015 Apr 6 [cited 2015 Oct 12]. Avail- after a major earthquake. J Formos Med Assoc 2005;104:308-317. able from: http://www.nocutnews.co.kr/news/4393074 (Korean). 29. Bland SH, O’Leary ES, Farinaro E, Jossa F, Trevisan M. Long-term 12. Kim YT, Choi BY, Lee KO, Kim H, Chun JH, Kim SY, et al. Over- psychological effects of natural disasters. Psychosom Med 1996;58: view of Korean Community Health Survey. J Korean Med Assoc 18-24.

9 Epidemiology and Health 2015;37:e2015046

Appendix 1.

지역사회 건강조사 기반 사회심리 및 안전인식도 조사

조 사 원 : ______

◈ 아래의 칸은 조사원이 기입하는 부분입니다.

일자 관리 시도 보건소 표본지점 가구 가구원 번호

◈ 아래의 칸은 책임대학에서 기입하는 부분입니다.

0 0 0 0 0 0 0 0 0 0 0

1 군포시보건소 1 1 1 1 1 1 1 1 1 1 1

2 수원시 팔달구보건소 2 2 2 2 2 2 2 2 2 2

3 안산시 단원보건소 3 3 3 3 3 3 3 3 3 3

4 안산시 상록수보건소 4 4 4 4 4 4 4 4 4 4

5 남양주시보건소 5 5 5 5 5 5 5 5 5 5

6 구리시보건소 6 6 6 6 6 6 6 6 6 6

7 진도군보건소 7 7 7 7 7 7 7 7 7 7

8 해남군보건소 8 8 8 8 8 8 8 8 8 8

9 9 9 9 9 9 9 9 9 9

2014 지역사회건강조사

10 Yang HJ et al.: Community mental health status after disaster

지역사회 건강조사 기반 사회심리 및 안전인식도 조사 <자기기입식>

※ 다음 문항을 읽고, 해당되는 “□”를 → “■”로 까맣게 칠해주세요 ※ 컴퓨터로 자동 입력되는 용지이오니 반드시 위와 같이 표기하여 주시기 바랍니다.

Ⅰ. 지난 4주 동안 다음 신체 증상 때문에 얼마나 자주 방해를 받았습니까?

전혀 약간 대단히 항 목 시달리지 시달림 시달림 않음

1. 위통 □ □ □

2. 허리 통증 □ □ □

3. 팔, 다리, 관절 (무릎, 고관절 등)의 통증 □ □ □

4. 생리기간 동안 생리통 등의 문제 [여성만 응답함] □ □ □

5. 두통 □ □ □

6. 가슴 통증, 흉통 □ □ □

7. 어지러움 □ □ □

8. 기절할 것 같음 □ □ □

9. 심장이 빨리 뜀 □ □ □

10. 숨이 참 □ □ □

11. 성교 통증 등의 문제 □ □ □

12. 변비, 묽은 변이나 설사 □ □ □

13. 메슥거림, 방귀, 소화불량 □ □ □

14. 피로감, 기운 없음 □ □ □

15. 수면의 어려움 □ □ □

2014 지역사회건강조사

11 Epidemiology and Health 2015;37:e2015046

Ⅱ. 다음 각 항목에 대한 귀하의 생각은 어떻습니까?

항 목 예 아니오

1. 우리 동네 사람들은 서로 믿고 신뢰할 수 있다 □ □

2. 이웃에 경조사가 있을 때, 주민 사이에 서로 도움을 주고받는 전통이 있다 □ □

3. 우리 동네의 전반적 안전수준(자연재해, 교통사고, 농작업 사고, 범죄)에 대해 만족한다 □ □

4. 우리 동네의 자연환경(공기질, 수질 등)에 대해 만족한다 □ □

5. 우리 동네의 생활환경(전기, 상하수도, 쓰레기 수거, 스포츠시설 등)에 대해 만족한다 □ □

6. 우리 동네의 대중교통 여건(버스, 택시, 지하철, 기차 등)에 대해 만족한다 □ □

7. 우리 동네의 의료서비스 여건(보건소, 병의원, 한방병의원, 약국 등)에 대해 만족한다 □ □

Ⅲ. 다음 각 분야마다 우리 사회가 어느 정도 안전하다고 생각하십니까?

매우 비교적 비교적 매우 분 야 보통 안전 안전 불안 불안 1. 국가 안보(전쟁 가능성, 북핵문제 등) □ □ □ □ □

2. 자연재해(태풍, 지진 등) □ □ □ □ □

3. 건축물 및 시설물(주택, 교량 등) □ □ □ □ □

4. 교통사고 □ □ □ □ □

5. 화재(산불 포함) □ □ □ □ □

6. 식량 안보(곡물가 폭등, 식량 부족 등) □ □ □ □ □

7. 정보 보안(컴퓨터바이러스, 기타 해킹 등) □ □ □ □ □

8. 신종 전염병(신종 바이러스, 조류독감, 사스 등) □ □ □ □ □

9. 범죄 위험 □ □ □ □ □

10. 전반적인 사회 안전 □ □ □ □ □

2014 지역사회건강조사

12 Yang HJ et al.: Community mental health status after disaster

Ⅳ. 지난 1주일 동안 다음과 같은 일들이 얼마나 자주 일어났습니까?

1일 1~2 3~4 5일 항 목 미만 일 일 이상 1. 평소에는 아무렇지도 않던 일들이 괴롭고 귀찮게 느껴졌다 □ □ □ □ 2. 먹고 싶지 않고 식욕이 없었다 □ □ □ □ 3. 어느 누가 도와준다 하더라도 나의 울적한 기분을 떨쳐 버 □ □ □ □ 릴 수 없을 것 같았다 4. 무슨 일을 하든 정신을 집중하기가 힘들었다 □ □ □ □ 5. 비교적 잘 지냈다 □ □ □ □ 6. 상당히 우울했다 □ □ □ □ 7. 모든 일들이 힘들게 느껴졌다 □ □ □ □ 8. 앞일이 암담하게 느껴졌다 □ □ □ □ 9. 지금까지의 내 인생은 실패작이라는 생각이 들었다 □ □ □ □ 10. 적어도 보통 사람들만큼의 능력은 있다고 생각했다 □ □ □ □ 11. 잠을 설쳤다(잠을 잘 이루지 못했다) □ □ □ □ 12. 두려움을 느꼈다 □ □ □ □ 13. 평소에 비해 말수가 적었다 □ □ □ □ 14. 세상에 홀로 있는 듯한 외로움을 느꼈다 □ □ □ □ 15. 큰 불만 없이 생활했다 □ □ □ □ 16. 사람들이 나에게 차갑게 대하는 것 같았다 □ □ □ □ 17. 갑자기 울음이 나왔다 □ □ □ □ 18. 마음이 슬펐다 □ □ □ □ 19. 사람들이 나를 싫어하는 것 같았다 □ □ □ □ 20. 도무지 뭘 해 나갈 엄두가 나지 않았다 □ □ □ □

Ⅴ. 지난 1년 동안 다음과 같은 일들이 있었습니까?

항 목 예 아니오 1. 최근 1년 동안 죽고 싶다는 생각을 해 본 적이 있습니까? □ □ 2. 최근 1년 동안 실제로 자살시도를 해 본 적이 있습니까? □ □

2014 지역사회건강조사

13 Epidemiology and Health 2015;37:e2015046

Ⅵ. 지난 1달 동안 다음과 같은 일들이 얼마나 있었습니까?

종종 거의 전혀 간혹 항상 항 목 여러번 언제나 없었다 있었다 있었다 있었다 있었다

1. 살아가는데, 정신적·신체적으로 감당하기 힘들다고 느낀적이 있습니까? □ □ □ □ □

2. 자신의 생활 신념에 따라 살려고 애쓰다가 좌절을 느낀적이 있습니까? □ □ □ □ □

3. 한 인간으로서 기본적인 요구가 충족되지 않았다고 느낀적이 있습니까? □ □ □ □ □

4. 미래에 대해 불확실하게 느끼거나 불안해 한 적이 있습니까? □ □ □ □ □

5. 할 일이 너무 많아 정말 중요한 일들을 잊은 적이 있습니까? □ □ □ □ □

Ⅶ. 지난 2주 동안, 다음의 문제들로 인해서 얼마나 자주 방해를 받았습니까? 전혀 며칠 동안 절반 이상 거의 매일 항 목 방해 받지 방해 방해 방해 않았다 받았다 받았다 받았다 1. 초조하거나 불안하거나 조마조마하게 느낀다 □ □ □ □

2. 걱정하는 것을 멈추거나 조절할 수가 없다 □ □ □ □

3. 여러 가지 것들에 대해 걱정을 너무 많이 한다 □ □ □ □

4. 편하게 있기가 어렵다 □ □ □ □

5. 너무 안절부절못해서 가만히 있기가 힘들다 □ □ □ □

6. 쉽게 짜증이 나거나 쉽게 성을 내게 된다 □ □ □ □

7. 마치 끔찍한 일이 생길 것처럼 두렵게 느껴진다 □ □ □ □

Ⅷ. 모든 사항을 고려할 때, 최근 귀하의 삶에 (대체로) 어느 정도 만족합니까? 해당되는 “□”를 → “■”로 칠해주세요.

1 2 3 4 5 6 7 8 9 10

□‒□‒□‒□‒□‒□‒□‒□‒□‒□ 매우 불만족 매우 만족

* 설문에 응답해 주셔서 대단히 감사합니다 *

2014 지역사회건강조사