The relationship between PTSD and suicidality among Wenchuan ...

Journal of Affective Disorders 235 (2018) 90–95

Contents lists available at ScienceDirect

Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad

Research paper

The relationship between PTSD and suicidality among Wenchuan earthquake survivors: The role of PTG and social support ⁎

T

⁎

Guo Jinga, , Liu Chengchenga, Kong Dexiab, , Phyllis Solomonb, Fu Mingqia a b

School of Sociology, Huazhong University of Science and Technology, Wuhan 430074, PR China School of Social Policy & Practice, University of Pennsylvania, Philadelphia, PA, USA

A R T I C LE I N FO

A B S T R A C T

Keywords: Suicidality PTSD PTG Social support Earthquake

Background: Previous studies have found that suicide rates are likely to increase after the occurrence of earthquakes. Most existing studies that examined the relationship between PTSD and suicidality focus on risk factors. It remains unclear whether protective factors, such as post-traumatic growth (PTG) and social support, play a role in this relationship. Objectives: The aim of this study was to investigate the role of PTG and social support in the association between PTSD and suicidality, using data from a cross-sectional study conducted in China. Methods: 1,369 participants were recruited from two study sites that were severely affected by the Wenchuan earthquake. Univariate and multivariate regression analyses were conducted to examine the relationship between PTSD, PTG, social support, and suicidality. Results: The results indicate that the prevalence of suicidal ideation, suicide plans, and suicide attempts among adult survivors were 9.06%, 2.97% and 3.31%, respectively, even after 8 years following the Wenchuan earthquake. Furthermore, the findings show that PTSD was significantly associated with suicidality (OR = 1.96; 95% CI = 1.53, 2.52). Additionally, individuals reporting lower levels of social support and the acquisition of PTSD (OR = 5.99; 95% CI = 1.66, 21.56) were significantly related to suicidality. Moreover, compared to individuals who reported no presence of PTSD and high PTG, those who lived with PTSD and lower levels of PTG (OR = 2.33; 95%CI = 1.00, 5.42) were more likely to report suicidality. Limitations: The cross-sectional design of this study limits our ability to determine causal relationships. Effects of other related factors, such as cultural and life events, were not examined in this study. Conclusions: Suicidality is a long-term health issue among survivors of the Wenchuan earthquake. PTG and social support play important roles on the association between PTSD and suicidality and are important contributing factors to understanding this relationship. These results contribute new knowledge of suicidal risk for a number of years after an earthquake and have implications for further mental health promotion following earthquakes.

1. Introduction Suicide is a major health problem worldwide (Juurlink et al., 2004), and has been found to be one of the primary health causes of mortality and morbidity (Yershova et al., 2016). Previous studies have found that suicide rates are likely to increase after the occurrence of natural or human traumatic events, such as earthquakes (Chen et al., 2016), terrorist attacks (De Lange and Neeleman, 2004) and wars (Sandman et al., 2017). However, this association has varied by geographical location. When it comes to the earthquake which occurred in Nantong, Taiwan, the overall suicide rate grew from 12.6% (per 100, 000) in 1998 to 22.9% (per 100, 000) in 2001 (Yip, 2009). On the contrary, data from L'Aquila (Italy) earthquake found that the number

⁎

Corresponding authors. E-mail addresses: [email protected] (J. Guo), [email protected] (D. Kong).

https://doi.org/10.1016/j.jad.2018.04.030 Received 10 December 2017; Received in revised form 4 March 2018; Accepted 4 April 2018 Available online 06 April 2018 0165-0327/ © 2018 Elsevier B.V. All rights reserved.

and rate of suicide behaviors was at its lowest point in 2009, the year of the earthquake, since 2004 (Stratta and Rossi, 2013). Unfortunately, due to the limits of public resources and attention distribution, few studies have examined the association between earthquake experience and suicide rate under Chinese social settings (Guo et al., 2017b). Of all of the natural disasters that have occurred in China since the beginning of the 21st century, the Ms 8.0 Wenchuan Earthquake on May 12, 2008 was the most devastating (Liu, 2014; Qiao et al., 2008). There was a confirmed accelerating trend of suicidal behaviors among survivors in the aftermath of Wenchuan earthquake (Ran et al., 2015). However, since suicidality is a complex psychological mechanism, reasons for the association between disaster exposure and suicidal risks have not been thoroughly investigated to date (Kilshaw et al., 2016;


J. Guo et al.

understanding of the needed direction for suicide prevention efforts in the after math of an earthquake. Based on a cross-sectional survey conducted among survivors of the Wenchuan earthquake in 2016, the objectives of this study were to (1) examine the relationship between PTSD and suicidality; and (2) investigate whether PTG and social support contribute to understanding this relationship.

Orui et al., 2014). According to one study of post disaster research by Kumar, specific risk factors tend to stimulate suicidal behaviors while certain protective factors help to insulate individuals against these stimulants (Kumar et al., 2012). Thus, unravelling the related factors associated with suicidality, and further verifying the risk and protective factors seem to warrant further research, as it may enhance the development of suicide prevention interventions and policies. As the possible mechanisms to understanding suicidal behavior post-disaster suggested by Kumar, suicidal related factors are divided into two categories: risk factors such as Posttraumatic Stress Disorder (PTSD) (Ying et al., 2015) and depression (Ran et al., 2011), as well as protective factors such as social support (Nakamura et al., 2014). PTSD is one of the most commonly reported psychological disorders following an earthquake (Cheng et al., 2014; Wu et al., 2014), which is not surprising given that PTSD is a psychiatric sequelae of traumatic exposure (Carmassi et al., 2016; Goenjian et al., 2009). Furthermore, PTSD has been found to be highly related to committing suicide; however, the level of the association has varied across studies. For those whose exposure to the traumatic event included, for example, being trapped in debris and not being recovered for a long period of time, the acquisition of PTSD likely means a greater chance of suicidality (Guo et al., 2017b). In contrast, individuals with less traumatic exposure may significantly decrease the probability of subsequent PTSD and consequently a decreased likelihood of suicidal thoughts and behavior post event (Ariapooran et al., 2016; Yu et al., 2010). After reviewing the literature, it seems that the majority of existing studies have focused on the relationship between suicidality and PTSD from a passive perspective, tending to take risk factors into consideration more often, with little being known regarding the role of protective factors. Consequently, knowledge about the relationship between risk factors resulting from exposure to an earthquake and suicidality is insufficient. Thus, portraying only a negative and destructive image of earthquakes on society, without a countervailing strength perspective or protective factors. However, an earthquake can be viewed as a double-edged sword. The repercussions of such events can induce not only passive effects among survivors, but also growth and other positive consequences, which may potentially benefit suicidal prevention efforts. Therefore, in order to enhance our understanding of the association between posttraumatic consequences and suicidality, it is necessary to examine the role of protective factors in the relationship between PTSD and suicidal thoughts and behavior among survivors after an earthquake. Among potential protective factors catalyzed by an event such as an earthquake, posttraumatic growth (PTG) and social support are the most common ones likely to reduce suicidal ideation (Bush et al., 2011; Yu et al., 2010). As one of the most discussed positive posttraumatic consequences (Zhou et al., 2015), PTG enables earthquake survivors to reframe their experience and perceive potential benefits from having endured such an event, resulting in improving relationships with others, creating new possibilities, advancing personal strength, bringing spiritual change or increasing the appreciation of life (Jin et al., 2014). Thus, PTG may provide disaster survivors with a more positive attitude towards life and the world in which they live, that may counteract thoughts of committing suicide. As Bush suggested in his study, the greater the degree of PTG individuals reported, the less suicidal ideation they subsequently espoused (Bush et al., 2011). Furthermore, social support also plays an independent role in alleviating stress reactions after stressful life events (Cohen and Syme, 1985), and can buffer the negative impact of disasters. Previous studies indicated that perceived social support may protect individuals from suicidal ideation as it offers tangible benefits which are available and beneficial in times of crisis (Hirsch and Barton, 2011) and may enhance self-esteem (Kleiman and Riskind, 2013). Although the majority of the research demonstrated the potential protective relationship of PTG and social support to suicidal ideation, there were no studies that examined the role of PTG and social support on the relationship between PTSD and suicidality. Understanding this relationship may provide enhanced

2. Methods 2.1. Participants and procedures This study was a community-based, cross-sectional study design with in-person interviews conducted in February 2016 with 1,369 adult survivors of the 2008 Wenchuan earthquake. Two different sites in Sichuan province, which were severely affected by the earthquake, were included in this study. Both were close to the epicenter but at somewhat different distances, yet they had similar socioeconomic and demographic characteristics prior to the earthquake. One was the township of Yongan (115.7 km from the epicenter), which is located in a mountainous area. Over 90% of the buildings in Yongan were damaged in the earthquake. The other was the township of Guangji (58.3 km from the epicenter) in Manzhu city, which is located on a plain. More than 96% of the buildings in Guangji were damaged in the earthquake. Participants were selected by using a combination of multistage random sampling and convenience sampling. In the first stage, 12 villages were randomly selected from the total of 29 villages in the two study sites. At the second stage, households were randomly selected from each village's registration listing of households. However, in a small number of villages, many former residents were no longer residing in their homes, therefore, where households were unoccupied, these households were replaced with a neighboring one. The third stage was to select the individual in the household to be interviewed. This was done by a convenience sample, with anyone within the household over age 16 who agreed to be interviewed (one adult was randomly selected if two or more members in a household agreed to be interviewed). Those with serious mental disorders or cognitive disorders were excluded. Of the 1,482 households approached in 2016, 1,369 adults gave consent and completed the interview, yielding a response rate of 92%. Data were collected by college-level research assistants (RAs) who conducted in person interviews with participants in Mandarin Chinese or the local dialect. All participants gave consent after being informed of the aim of the survey and their right to refuse to participate. RAs received extensive training on survey research prior to data collection. The training involved the following five sessions: introduction, sampling design and procedures, review of the questionnaire, methods and techniques of interviewing, and potential challenges in conducting the survey and strategies for handling them. The RAs read each question to the respondents, and then recorded their answers on the interview schedule. Interviews lasted an average of an hour. The study protocol was approved by the institutional review board of Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China. 2.2. Measurements 2.2.1. Confounding variables Sociodemographic characteristics: The following demographic and socioeconomic information was collected: location of residence (Yongan/Guangji); gender (female/male); ethnicity (Han/minority, the latter including Qiang, Tibetan, Hui, Mongol and other minority groups); religious belief (Yes/No); age (16–35 years/35–55 years/ > 55 years); marital status (married/divorced/widowed/unmarried); education (in years); employment status (farmer/unemployed/employed); 91


J. Guo et al.

the scale in the present study was 0.93. PTG: PTG was assessed using the post traumatic growth inventory (PTGI), which consists of 21 items within five domains: relating to others (seven items), new possibilities (five items), personal strength (four items), spiritual change (two items), and appreciation of life (three items). The participants responded on a 6-point scale ranging from 0 (no change) to 5 (complete change), with higher scores representing higher levels of PTG following the Wenchuan earthquake. The spiritual change domain was excluded in this study (“A better understanding of spiritual matters” and “I have a stronger religious faith”), as these two items were deemed not applicable to the local culture (Guo et al., 2017a). The final score was a sum of the scores for the 19 items, ranging from 0 to 95. This inventory was found to have good internal and test-retest reliability within a Chinese sample (Jin et al., 2014). In this study, the internal reliability of the scale was 0.92 (internal reliabilities for the four domains, as the spiritual domain was excluded) were 0.85, 0.76, 0.78, and 0.69, respectively).

perceived household economic status (poor/medium/rich). Depression: The Chinese version of the Center for Epidemiologic Studies Depression Scale (CES-D) (Wang, 1999) was used to assess probable major depression. The scale is one of the most commonly used instruments to measure depression in community-based studies. The Chinese version of the CES-D scale has demonstrated good reliability and validity across all age groups in urban populations (Zhang et al., 2010). In our study, we used 21 as the cutoff point, as this has been shown to be a good predictor for major depression in Chinese populations (Cheng and Chan, 2005). The internal consistency coefficient (Cronbach's alpha) of the scale in the present study was 0.88. 2.2.2. Dependent variables Suicidality: In this study, suicidality was conceptualized as having three dimensions including suicidal ideation, suicidal plan and suicidal attempts (Kessler et al., 2005). Suicidal ideation was assessed by the question, “Have you thought seriously about suicide since the earthquake?” Suicide plans was measured by “Have you ever made a plan to commit suicide since the earthquake?” Suicide attempts were measured by the question, “Since the earthquake, have you ever attempted suicide?” All three aspects of suicidality are widely used in China and have been found to have good validity (Sun et al., 2017b). Suicidality was defined as answering yes to any one of the three questions.

2.2.4. Statistical analyses Descriptive statistics were calculated for all of the socio-demographic characteristics. Univariate and multivariate logistic regression analyses (Model 1) were conducted to examine the relationship between suicidality and the socio-demographic variables, PTSD, PTG and social support. Then, according to the level of PTSD and social support, the whole sample was stratified into 4 groups to examine the interaction effects of PTSD and social support on suicidality. Furthermore, according to the level of PTSD and PTG, the whole sample was stratified into another 4 groups (see Table 3 for the 4 groupings) to examine the interaction effects of PTSD and PTG on suicidality. Two multivariate logistic regression analyses (Model 2 & Model 3) were conducted to examine the interaction effects. All P values were derived from adjusted Wald F tests in these regression models. These analyses enabled us to examine the relationship between PTSD and suicidality, as well as the role of PTG and social support, which was outlined in the introductory section of this paper. The data were analyzed using Stata Version 13.

2.2.3. Independent variables Social support: Social support was measured by the 10-item Social Support Rating Scale (SSRS), which was developed for a Chinese population and has been widely used in China (Sun et al., 2017a; Xiao, 1994). The scale consists of three domains, including objective support, subjective support, and support utilization. Objective support is actual or perceived support, including direct material assistance from social network members; subjective support refers to subjective experience or emotional support, which mainly refers to the individual's emotional experience and satisfaction of support from social network members; support utilization is the individual's frequency of soliciting help from his/her social network. Study participants rated the extent to which they endorsed each scale item using a Likert scale ranging from 1 to 4, except the sixth and seventh items. The sixth and seventh items were calculated by summing the number of social support sources listed by the respondent. The final score is the sum of the scores for all scale items, ranging from 12 to 66. The SSRS has shown good validity and reliability among the Wenchuan earthquake survivors (Ma et al., 2011), and the overall internal consistency coefficient (Cronbach's alpha) of the scale in the present study was 0.65 (Cronbach's alpha of the objective support, subjective support and support utilization was 0.67, 0.69, 0.65, respectively). We used a cutoff point of 30, which is the median of the total inventory scores in this study, to stratify the sample into high versus low levels of social support in the interaction effect analyses. PTSD: PTSD was assessed by the Impact of Event Scale-Revised (IES-R), a self-report measure of subjective distress in response to a specific traumatic event (Motlagh, 2010).The scale has been widely used in the field of traumatic stress. It includes 22 items which assess the three major symptom clusters of PTSD: intrusion, avoidance, and hyper-arousal. The scale has been reported to have sound psychometric properties (Creamer et al., 2003). The participants were asked to indicate the frequency of their distress using 4 response options (0 = not at all, 1 = seldom, 3 = sometimes, 5 = often) (Chan et al., 2011). Subscale scores were calculated as the means of the responses to all items in the specific sub-scale, and the total score was calculated as the mean response across all of the scale items. Since there is no recommended cutoff point for the IES-R, this study adopted a mean score of 2.0 across all items of the IES-R as the cutoff point for clinically significant PTSD symptoms, as was done in earlier studies (Chan et al., 2011; Guo et al., 2015). A prior study using this cutoff point reported a sensitivity of 0.89 and a specificity of 0.90 (Wohlfarth et al., 2003). Cronbach's alpha of

3. Result 3.1. Sample characteristics As shown in Table 1, among 1369 respondents, individuals over the age of 55 accounted for half (50.35%) of the sample, revealing that many young residents were away from the dwelling and were likely working as migrants. Three quarters of the study respondents were farmers (46.75%) or were unemployed (28.06%), and 28.08% of them perceived themselves as poor (low-income), which supports the idea that the community in which they lived was sustained in essentially abject poverty. What's more, just over half of the respondents were women (53.43%) and nearly all of whom belonged to the Han ethnicity (94.62%). In the survey area, about 82.58% of the sample were married, and had a significantly low education level (mean = 6.33, SD = 3.16). Only 13.10% of respondents had a religious belief, with the most common being Buddhism. Approximately 11% of the respondents reported that they experienced suicidality, including those that had thought seriously about suicide (9.06%), made plans to commit suicide (2.87%) or attempted suicide (3.31%) since the occurrence of earthquake. Further, the mean score of PTSD was 0.70 (SD = 0.80), while PTG and social support had a mean score at 39.63 (SD = 18.15) and 38.56 (SD = 6.86), respectively. About eleven percent (11.3%) of respondents reported no presence of PTSD (score = 0), and the prevalence of probable PTSD was 11.8% (score ≥ 2). Almost a quarter (24.84%) of respondents who had significant depressive symptoms had CESD scores equal to or greater than 21 (the cut-off point). Other social demographic characteristics are listed in Table 1. 92


J. Guo et al.

Table 1 Descriptive characteristic for adult survivors of Wenchuan earthquake in China (N = 1369). Variables Location Guangji Yongan Gendera Male Female Ageb 15–35 35–55 55+ Ethnicity Han c Minority Marital Statusd Single Married Divorced/widowed Religion beliefe No Yes Household economic statusf Poor Medium Rich Employmentg Farmer Employed Unemployed Depression Yes No Suicidal ideationh Yes No Suicidal plansi Yes No Suicidal attemptsj Yes No

Table 2 Logistic regression analysis for suicidality among adult survivors of Wenchuan Earthquake. (N = 1369).

N/Mean

Percentage/SD

677 692

49.45 50.55

632 725

46.57 53.43

Location (Ref:Guangji) Yongan

151 484 644

11.81 37.84 50.35

Gender (Ref:Male) Female

1283 73

94.62 5.38

Ethnicity (Ref:Minority) Han

94 1109 140

7.00 82.58 10.42

1181 178

86.90 13.10

381 861 115

28.08 63.45 8.47

633 341 380

46.75 25.18 28.06

340 1029

24.84 75.16

124 1244

9.06 90.94

39 1322

2.87 97.13

45 1316

3.31 96.69

144 1224 6.33 0.70 39.63 38.56

10.52 89.41 3.16 0.80 18.15 6.86

Age (Ref:16–35) 35–55 55+ Education year Marital status (Ref:Married) Single Divorced /widowed Religious belief (Ref:No) Yes Household economic status (Ref:Rich) Poor Medium Employment (Ref: Employed) Unemployed

k

Suicidality Yes No Education years (Mean, SD) PTSD (Mean, SD) PTG (Mean, SD) Social support (Mean, SD)

Farmer Depression (Ref:No) Yes PTSD PTG

Notes: a Means there is 12 missing data upon this variable. b Means there is 90 missing data upon this variable. c Means there is 13 missing data upon this variable. d Means there is 26 missing data upon this variable. e Means there is 10 missing data upon this variable. f Means there is 12 missing data upon this variable. g Means there is 12 missing data upon this variable. h Means there is 1 missing data upon this variable. i Means there is 8 missing data upon this variable. j Means there is 8 missing data upon this variable. k Means there is 1 missing data upon this variable.

Social support

Univariate regression Suicidility OR[95%CI]

Multivariate regression Suicidility OR[95%CI]

1.00 [0.71,1.41]

0.80 [0.51,1.25]

2.04*** [1.41,2.97]

1.79* [1.11,2.87]

0.95 [0.45,2.02]

0.73 [0.30,1.75]

4.18** [1.65,10.61] 3.48** [1.38,8.78] 0.87*** [0.82,0.93]

3.31 [0.89,12.30] 2.03 [0.52,7.88] 0.98 [0.89,1.08]

0.29* [0.09,0.93] 1.82* [1.13,2.95]

1.02 [0.20,5.24] 1.49 [0.82,2.72]

1.50 [0.94,2.38]

1.34 [0.77,2.34]

2.12* [1.05,4.28] 0.90 [0.45,1.81]

1.14 [0.49,2.65] 0.76 [0.33,1.74]

1.38 [0.81,2.37] 1.94** [1.21,3.13]

0.89 [0.44,1.79] 1.44 [0.79,2.65]

4.34*** [3.04,6.19] 2.10*** [1.76,2.51] 0.99 [0.98,1.00] 0.95*** [0.93,0.98]

2.08*** [1.33,3.26] 1.96*** [1.53,2.52] 0.99* [0.97,1.00] 0.96* [0.93,1.00]

Note: Odds Ratio were reported; 95% confidence intervals in brackets; * p < 0.05, ** p < 0.01, *** p < 0.001.

3.3. Multivariate logistic regression analyses Logistic regression was used to examine factors related to suicidality, which combines suicidal ideation, plans, and attempts. Location, gender, ethnicity, age, level of education, marital status, religious belief, economic status, employment status, and depression were included as confounding variables. After controlling for potential confounding variables, this study found that PTSD was significantly associated with suicidality (OR = 1.96; 95% CI = 1.53, 2.52). From the aspect of protective factors, individuals who received more social support were less likely to be at risk of suicidality (OR = 0.96; 95% CI = 0.93, 1.00), and similarly individuals with greater PTG were at lower risk for suicidality (OR = 0.99; 95% CI = 0.97, 1.00). Details are presented in Table 2.

3.2. Bivariate logistic regression analyses Table 2 displays the results of the bivariate correlation analyses of the factors associated with suicidality. The prevalence of suicidality was significantly higher among females, the aged, those with lower education, unmarried and divorced, being poor, a farmer, and being depressed. In addition, social support was significantly related to suicidality. Details are shown in Table 2. 93


J. Guo et al.

preventing PTSD and promoting PTG in relationship to the development of social supports did enhance the benefit of prevention of suicidality to a great extent. Moreover, this study found that PTSD, PTG, and social support following an earthquake worked in combination to influence the occurrence of suicidal ideation. Previous research has found that social supports were protective for suicide risk both in the presence or not of PTSD, but were weaker among survivors of traumatic events reporting the presence of PTSD (Matthew et al., 2010). In accordance with this finding, this study demonstrated that for individuals who had less social support, the presence of PTSD was significantly related to suicidality. Although numerous studies have confirmed that PTSD and PTG may coexist among traumatic survivors (Jin et al., 2014; Zhou et al., 2016), there was no relevant research confirming the combined effect of PTSD and PTG on suicidality until the present analyses. Thus, it is necessary to consider the existence of the combined role of PTSD, social support and PTG in the development of preventive interventions addressing suicidality for survivors of traumatic events. Finally, this study examined the relationship between demographic variables, such as gender, ethnicity, age, education, marriage, religious beliefs and economic status, with suicidality. Results demonstrated similar association with prior studies that gender differences did exist on the issue of suicidality, with women being at higher risk of suicidality (Allison et al., 2001; Yoshimasu et al., 2006). In addition we found that depression is a risk factor for suicide, which is consistent with previous studies (Horwitz et al., 2011; Ortíz-Gómez et al., 2014). Also, individuals who are relatively disadvantaged socioeconomically had a higher likelihood of developing suicidal risk, which corresponds with the findings of previous studies (Agerbo et al., 2006; Denney et al., 2009). Consequently, the prevention efforts for survivors of earthquakes need to focus more on females and individuals who are disadvantaged socioeconomically, as these individuals are at greater risk of committing suicide. The findings of this study are subject to several limitations. First, the cross-sectional design of this study limits our ability to determine causal relationships. Future longitudinal studies are needed to disentangle the causal relationships. Second, information on pre-earthquake suicidality of the participants was not available. Thus we were not able to assess changes in suicidality and make further comparisons. Additionally, the effects of other related variables, such as cultural and life events, were not examined in this study, resulting in potential bias in the study findings. Lastly, it remains unclear whether the findings could be generalizable to other rural populations in China or elsewhere. Despite these limitations, this study contributes to the literature by examining the relationship between protective factors and suicidality, exploring the combined effects of PTG and PTSD, social support and PTSD on suicidal ideation, and provides valuable suggestions for further development of suicide preventive interventions.

Table 3 Interaction effects of PTSD and PTG, PTSD and social support on suicidality after controlling for confounding variables. Model 4

PTSD*Social support (Ref: none PTSD * high Social support) None PTSD * low Social support PTSD * high Social support PTSD * low Social support

OR [95%CI]

1.66 [0.88,3.10] 2.14* [1.12,4.09] 5.99** [1.66,21.56]

Model 5

PTSD*PTG (Ref: none PTSD * high PTG) None PTSD * low PTG PTSD * high PTG PTSD * low PTG

OR [95%CI]

1.08 [0.61,1.91] 2.24 [0.87,5.76] 2.33* [1.00,5.42]

Note: Odds Ratio were reported; 95% confidence intervals in brackets; * p < 0.05, ** p < 0.01, *** p < 0.001.

3.4. Interaction effects analyses Table 3 shows the interaction effects of PTSD and protective factors on suicidality. Individuals reporting lower levels of social support and the acquisition of PTSD (OR = 5.99; 95% CI = 1.66, 21.56) were significantly related to the presence of suicidality. In addition, compared to individuals who reported no presence of PTSD and high PTG, those who lived with PTSD and lower levels of PTG (OR = 2.33; 95%CI = 1.00, 5.42) were more likely to report suicidality. 4. Discussion This study examined the relationship between PTSD and suicidality and whether PTG and social support contribute to the understanding of such a relationship. Based on data collected in 2016 from 1,369 adult survivors of Wenchuan earthquake, the findings suggest that survivors were at a high risk for suicidality even eight years after the earthquake. Further, PTSD was found to be a significant risk factor for developing suicidality, while PTG and social support played a protective role in this association, lowering the probability of suicidality. The results indicate that the prevalence of suicidal ideation, suicide plans and suicide attempts among adult survivors were 9.06%, 2.97% and 3.31%, respectively, even after 8 years following the Wenchuan earthquake. Consequently, approximately 11% of the respondents were at risk of suicidality. This estimated prevalence was significantly higher than that in rural population in Beijing, in which the prevalence of suicidal ideation, suicidal plans and suicidal attempts were 2.8%, 1.6% and 1.3%, respectively (Xin et al., 2009). A recent meta-analysis estimated that lifetime prevalence of suicidal ideation and suicide attempts were 3.9% (95% CI: 2.5−6.0%) and 0.8% (95% CI: 0.7−0.9%) in the general population of China, respectively (Cao et al., 2015). The high prevalence of suicidality of survivors of an earthquake underlies the pressing need to develop psychological interventions for survivors in order to prevent earthquake-related suicides. This study examined the relationships between both protective and risk factors and suicidality after the Wenchuan earthquake. A significant association between PTSD and suicidality was observed, which is consistent with previous research, revealing that the presence of PTSD was correlated with suicidality (Ran et al., 2011; Ying et al., 2015). Furthermore, suicidal risks from a protective perspective were proposed in the current study. Individuals reporting higher levels of PTG were less likely to have suicidal ideation, to make plans to commit suicide and to attempt suicide, which also corresponded with previous findings (Bush et al., 2011; Yu et al., 2010). In addition, prior studies have found that social support is an important strategy for suicide prevention (Park, 2017). The current study validated the positive contribution of social support for suicidality from Wenchuan earthquake survivors. Thus, from the perspective of post-earthquake suicidal prevention, this study suggests that psychological intervention in

5. Conclusion Our findings indicate that suicidality represents a long-term health issue among survivors of the Wenchuan earthquake. The findings further demonstrate the protective roles of PTG and social support on the association between PTSD and suicidality. PTG and social support are important contributing factors to understanding this relationship. These results add knowledge of suicidal risk a number of years after an earthquake and have implications for further mental health promotion following earthquakes. Contributors JG, CL, & MF drafted the manuscript. DK, PS, CL, JG, & MF was involved the data analysis and revised the manuscript. All authors were involved in writing the manuscript and approve of its final version. 94


J. Guo et al.

Acknowledgments

illness and the risk of suicide in the elderly. Arch. Intern. Med. 164, 1179–1184. Kessler, R.C., Berglund, P., Borges, G., Nock, M., Wang, P.S., 2005. Trends in suicide ideation, plans, gestures, and attempts in the United States, 1990-1992 to 2001-2003. JAMA J. Am. Med. Assoc. 293, 2487–2495. Kilshaw, M., Haddock, G., Pratt, D., Panagioti, M., 2016. The effects of emotion dysregulation on suicidality in people who have experienced trauma: a systematic review. Available from: http://www.crd.york.ac.uk/PROSPERO/display_record.php?ID= CRD42016036801. Kleiman, E.M., Riskind, J.H., 2013. Utilized social support and self-esteem mediate the relationship between perceived social support and suicide ideation: a test of a multiple mediator model. Crisis 34, 42–49. Kumar, M.B., Walls, M., Janz, T., Hutchinson, P., Turner, T., Graham, C., 2012. Suicidal ideation among Métis adult men and women – associated risk and protective factors: findings from a nationally representative survey. Int. J. Circumpolar Health 71, 18829. Liu, W., 2014. Forensic analyses on a compound disaster and its impacts following the 2008 Wenchuan earthquake in China. In: EGU General Assembly Conference. Ma, X., Liu, X., Hu, X., Qiu, C., Wang, Y., Huang, Y., Wang, Q., Zhang, W., Li, T., 2011. Risk indicators for post-traumatic stress disorder in adolescents exposed to the 5.12 Wenchuan earthquake in China. Psychiatry Res. 189, 385–391. Matthew, J.P.D., Steven, V.P.D., Zac, I.P.D., Jessica, W.C.P.D., Alan, F.P.D., Rosenheck, R., Miles, M.P.D., 2010. Does PTSD moderate the relationship between social support and suicide risk in Iraq and Afghanistan War veterans seeking mental health treatment? Depress. Anxiety 27, 1001–1005. Motlagh, H., 2010. Impact of event scale-revised. J. Physiother. 56, 203. Nakamura, K., Kitamura, K., Someya, T., 2014. Psychological recovery 5 years after the 2004 Niigata-Chuetsu earthquake in Yamakoshi, Japan. J. Epidemiol. 24, 125–131. Ortíz-Gómez, L.D., López-Canul, B., Arankowsky-Sandoval, G., 2014. Factors associated with depression and suicide attempts in patients undergoing rehabilitation for substance abuse. J. Affect. Disord. 169, 10–14. Orui, M., Harada, S., Hayashi, M., 2014. Changes in suicide rates in disaster-stricken areas following the Great East Japan earthquake and their effect on economic factors: an ecological study. Environ. Health Prev. Med. 19, 459–466. Park, Y., 2017. Effects of depression, self-esteem, and social support on suicidal ideation in college students. Child Health Nursing Res. 23, 111–116. Qiao, S., Zhang, D., Chen, Y., Xiao, C., Feng, C., 2008. Role of young scientists in the emergency response and scientific investigation related to the 2008 Wenchuan earthquake: performance evaluation and discussion on the implications to seismological education and training. In: AGU Fall Meeting. Ran, M., Zhang, Z., Fang, D., 2011. Suicidal ideation, PTSD and depression among adolescents after Wenchuan earthquake in China. J. Gerontol. Soc. Work 27, 45–54. Ran, M.S., Zhang, Z., Fan, M., Li, R.H., Li, Y.H., Ou, G.J., Jiang, Z., Tong, Y.Z., Fang, D.Z., 2015. Risk factors of suicidal ideation among adolescents after Wenchuan earthquake in China. Asian J. Psychiatry 13, 66–71. Sandman, N., Valli, K., Kronholm, E., Vartiainen, E., Laatikainen, T., Paunio, T., 2017. Nightmares as predictors of suicide: an extension study including war veterans. Sci. Rep. 7, 44756. Stratta, P., Rossi, A., 2013. Suicide in the aftermath of the L'Aquila (Italy) earthquake. Crisis 34, 142–144. Sun, J., Shi, B., Shi, J., Yue, X., Sun, L., Tian, Q., 2017a. The intervention effect assessment on social support condition of the first settlers in Dan Jiangkou reservoir area, China. In: MATEC Web of Conferences. 100. pp. 05084. Sun, L., Zhou, C., Xu, L., Li, S., Kong, F., Chu, J., 2017b. Suicidal ideation, plans and attempts among medical college students in China: the effect of their parental characteristics. Psychiatry Res. 247, 139–143. Wang, 1999. Psychological assessment scale manual. Chin. Mental Health J. 200–202. Wohlfarth, T.D., Van, d.B.W., Winkel, F.W., Ter, S.M., 2003. Screening for posttraumatic stress disorder: an evaluation of two self-report scales among crime victims. Psychol. Assess. 15, 101–109. Wu, Z., Xu, J., He, L., 2014. Psychological consequences and associated risk factors among adult survivors of the 2008 Wenchuan earthquake. BMC Psychiatry 14, 126. Xiao, S., 1994. Theoretical basis and application in research of social support rating scale. J. Clin. Psychol. Med. 4 (2), 98–99. Xin, M., Xiang, Y.T., Cai, Z.J., Li, S.R., Xiang, Y.Q., Guo, H.L., Hou, Y.Z., Li, Z.B., Li, Z.J., Tao, Y.F., 2009. Lifetime prevalence of suicidal ideation, suicide plans and attempts in rural and urban regions of Beijing, China. Aust. N.Z. J. Psychiatry 43, 158–166. Yershova, K., Lesser, A., Logan, K., Posner, K., 2016. Asking About Suicide as Suicide Prevention: The Columbia Suicide Severity Rating Scale (C-SSRS). Springer International Publishing. Ying, L., Chen, C., Lin, C., Greenberger, E., Wu, X., Jiang, L., 2015. The relationship between posttraumatic stress symptoms and suicide ideation among child survivors following the Wenchuan earthquake. Suicide Life-Threat. Behav. 45, 230–242. Yip, P., 2009. Effects of an earthquake on suicide rates in Nantou, Taiwan. Br. J. Psychiatry J. Mental Sci. 194, 185–191. Yoshimasu, K., Sugahara, H., Tokunaga, S., Akamine, M., Kondo, T., Fujisawa, K., Miyashita, K., Kubo, C., 2006. Gender differences in psychiatric symptoms related to suicidal ideation in Japanese patients with depression. Psychiatry Clin. Neurosci. 60, 563–569. Yu, X.N., Lau, J.T., Zhang, J., Mak, W.W., Choi, K.C., Lui, W.W., Chan, E.Y., 2010. Posttraumatic growth and reduced suicidal ideation among adolescents at month 1 after the Sichuan earthquake. J. Affect. Disord. 123, 327–331. Zhang, J., Wu, Z., Fang, G., Li, J., Han, B.X., 2010. Development of the Chinese age norms of CES-D in urban area. Chin. Mental Health J. 24, 139–143. Zhou, X., Wu, X., Fu, F., An, Y., 2015. Core belief challenge and rumination as predictors of PTSD and PTG among adolescent survivors of the Wenchuan earthquake. Psychol. Trauma Theory Res. Pract. Policy 7, 391–397. Zhou, X., Wu, X.C., Zeng, M., Tian, Y., 2016. The relationship between emotion regulation and PTSD/PTG among adolescents after the Ya'an earthquake: the moderating role of social support. Acta Psychol. Sin. 48, 969–980.

The authors wish to thank all those who kindly volunteered to participate in the study. Funding This study was supported by National Natural Science Foundation of China (31500912). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Conflict of interest The authors have no competing interests to report. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jad.2018.04.030. Reference Agerbo, E., Qin, P., Mortensen, P.B., 2006. Psychiatric illness, socioeconomic status, and marital status in people committing suicide: a matched case sibling control study. J. Epidemiol. Community Health 60, 776–781. Allison, S., Roeger, L., Martin, G., Keeves, J., 2001. Gender differences in the relationship between depression and suicidal ideation in young adolescents. Aust. & N.Z. J. Psychiatry 35, 498–503. Ariapooran, S., Rajabi, M., Goodarzi, A., 2016. Relationship between social support, time perspective and suicide ideations in patients with multiple sclerosis. Asian Soc. Sci. 12, 192–200. Bush, N.E., Skopp, N.A., Mccann, R., Luxton, D.D., 2011. Posttraumatic growth as protection against suicidal ideation after deployment and combat exposure. Mil. Med. 176, 1215–1222. Cao, X., Zhong, B., Xiang, Y., Ungvari, G.S., Lai, K.Y.C., Chiu, H.F.K., Caine, E.D., 2015. Prevalence of suicidal ideation and suicide attempts in the general population of China: a meta-analysis. Int. J. Psychiatry Med. 49, 296–308. Carmassi, C., Stratta, P., Calderani, E., Bertelloni, C.A., Menichini, M., Massimetti, E., Rossi, A., Dell'Osso, L., 2016. Impact of mood spectrum spirituality and mysticism symptoms on suicidality in earthquake survivors with PTSD. J. Religion Health 55, 641–649. Chan, C.L., Wang, C.W., Qu, Z., Lu, B.Q., Ran, M.S., Ho, A.H., Yuan, Y., Zhang, B.Q., Wang, X., Zhang, X., 2011. Posttraumatic stress disorder symptoms among adult survivors of the 2008 Sichuan earthquake in China. J. Trauma Stress 24, 295–302. Chen, S.L., Lee, C.S., Yen, A.M., Chen, H.H., Chan, C.C., Chiu, S.Y., Fann, J.C., Chang, J.C., 2016. A 10-year follow-up study on suicidal mortality after 1999 Taiwan earthquake. J. Psychiatr. Res. 79, 42–49. Cheng, S.T., Chan, A.C., 2005. The Center for Epidemiologic Studies Depression Scale in older Chinese: thresholds for long and short forms. Int. J. Geriatr. Psychiatry 20, 465–470. Cheng, Y., Wang, F., Wen, J., Shi, Y., 2014. Risk factors of post-traumatic stress disorder (PTSD) after Wenchuan earthquake: a case control study. PLoS One 9, e96644. Cohen, S., Syme, S.L., 1985. Issues in the study and application of social support. Creamer, M., Bell, R., Failla, S., 2003. Psychometric properties of the impact of event scale—revised. Behav. Res. Therapy 41, 1489–1496. De Lange, A.W., Neeleman, J., 2004. The effect of the September 11 terrorist attacks on suicide and deliberate self-harm: a time trend study. Suicide Life Threat. Behav. 34, 439–447. Denney, J.T., Rogers, R.G., Krueger, P.M., Wadsworth, T., 2009. Adult suicide mortality in the United States: marital status, family size, socioeconomic status, and differences by sex. Soc. Sci. Q. 90, 1167–1185. Goenjian, A.K., Walling, D., Steinberg, A.M., Roussos, A., Goenjian, H.A., Pynoos, R.S., 2009. Depression and PTSD symptoms among bereaved adolescents 6(1/2) years after the 1988 Spitak earthquake. J. Affect. Disord. 112, 81–84. Guo, J., Fu, M., Xing, J., Qu, Z., Wang, X., 2017a. Coping style and posttraumatic growth among adult survivors 8 years after the 2008 Wenchuan earthquake in China. Personal. Individ. Diff. 111, 31–36. Guo, J., He, H., Fu, M., Han, Z., Qu, Z., Wang, X., Guan, L., 2017b. Suicidality associated with PTSD, depression, and disaster recovery status among adult survivors 8 years after the 2008 Wenchuan earthquake in China. Psychiatry Res. 253, 383–390. Guo, J., Wang, X., Yuan, J., Zhang, W., Tian, D., Qu, Z., 2015. The symptoms of posttraumatic stress disorder and depression among adult earthquake survivors in China. J. Nervous Mental Dis. 203, 469–472. Hirsch, J.K., Barton, A.L., 2011. Positive social support, negative social exchanges, and suicidal behavior in college students. J. Am. Coll. Health 59, 393–398. Horwitz, A.G., Hill, R.M., King, C.A., 2011. Specific coping behaviors in relation to adolescent depression and suicidal ideation. J. Adolesc. 34, 1077–1085. Jin, Y., Xu, J., Liu, D., 2014. The relationship between post traumatic stress disorder and post traumatic growth: gender differences in PTG and PTSD subgroups. Soc. Psychiatry .Psychiatr. Epidemiol. 49, 1903–1910. Juurlink, D.N., Herrmann, N., Szalai, J.P., Kopp, A., Redelmeier, D.A., 2004. Medical

95