Assessing the association between probable posttraumatic stress disorder symptoms and quality of life among emergency service workers using path analysis

Abstract

BACKGROUND:

Emergency service workers have highly stressful occupations; the stressors encountered can contribute to the development of mental disorders such as depression, anxiety, and posttraumatic stress disorder (PTSD).

OBJECTIVE:

The present study used a conceptual model and survey to identify variables influencing the association between probable PTSD and quality of life (QOL) in emergency service workers.

METHOD:

PTSD was assessed using the Impact of Event Scale-Revised. QOL was assessed using the Medical Outcomes Study 36-Item Short-Form Health Survey. Path analysis was used to determine whether stress-coping (Brief Coping Orientation to Problems Experienced [Brief COPE] scores), social support (Multidimensional Scale of Perceived Social Support [MSPSS] scores), and resilience (Connor–Davidson Resilience Scale scores) explain the association between probable PTSD and QOL among 220 emergency service workers in Japan.

RESULTS:

Impact of Event Scale-Revised scores were significantly positively associated with Brief COPE Active coping scores and significantly negatively associated with MSPSS scores. Brief COPE Active coping and MSPSS scores were significantly positively associated with Connor–Davidson Resilience Scale scores, which were in turn significantly positively associated with Medical Outcomes Study 36-Item Short-Form Health Survey scores.

CONCLUSION:

Active coping in response to stressful situations is important for maintaining good mental health among emergency service workers. Active coping and social support may strengthen resilience, and resilience may improve QOL. Screening for mental health and QOL should include simultaneous assessment of stress-coping, social support, and resilience. Although the findings of this cross-sectional study are important, it could not confirm a causal relationship between PTSD and QOL.

Keywords

Firefighter resilience quality of life coping strategy social support posttraumatic stress disorder

1 Introduction

Emergency service workers, such as firefighters, emergency medical technicians, and military personnel, have highly stressful occupations and often encounter fatal accidents and disasters [1 –3], as well as dangerous environments (e.g., extremely high temperatures, poisonous dust, fires, and collapsing buildings) [4 –7]. A previous study reported that emergency service workers often experience negative mental health outcomes [8]. Evidence suggests that the stressors experienced by emergency service workers can contribute to the development of mental disorders, such as depression, anxiety, and posttraumatic stress disorder (PTSD) [2 , 10]. Factors thought to affect the likelihood of PTSD symptoms in this population include age, sex, education level, length of employment in emergency services, lack/receipt of continuous psychological care, exposure/non-exposure to human remains, stress-coping, social support, and resilience [7 , 11–17]. These factors may also influence emergency service workers’ quality of life (QOL). Indeed, a study that investigated the harmful effects of PTSD among emergency service workers reported reduced QOL [18]. QOL is an important concept for understanding workers’ mental and physical health. The World Health Organization [19] defines QOL as “individuals’ perceptions of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards, and concerns.” Several studies that focused on emergency service workers highlighted the importance of assessing QOL. For example, Schnurr et al. [20] reported that war veterans with mental disorders had a relatively low QOL. Another study reported that retired non-emergency service workers had superior mental health, according to their scores on a QOL measure, compared with retired emergency service workers [18]. A similar study by Bartlett et al. [21] reported that emergency service workers experienced sleep disturbances in association with their work, and that their QOL was affected by sleep disturbances [22]. In addition, other studies reported that stress-coping, social support, and resilience mediated the QOL of firefighters, clinical populations, and disaster victims with PTSD, respectively [23 –25]. Connor et al. [26] argued that resilience is an important aspect of mental health status and several studies have reported that resilience affects the QOL of participants with mental disorders [27 –29]. Furthermore, the factors influencing stress-coping, social support, and resilience are related, and evidence suggests that resilience is associated with both stress-coping and social support [30, 31]. Therefore, resilience may be a crucial concept for stress studies.

The Brief Coping Orientation to Problems Experienced (Brief COPE) scale is a useful inventory for surveying the numerous stress-coping strategies [32]. The Brief COPE comprises subscales that assess Self-distraction, Active coping, Using emotional support, Using instrumental support, Venting, Positive reframing, Planning, Humor, Acceptance, and Self-blame. Previous studies on firefighters that used the Brief COPE showed significant associations of various stress-coping strategies with PTSD [33, 34]. Studies that categorized the coping strategies of the Brief COPE (e.g., as adaptive and maladaptive strategies) revealed relationships between these categories and the mental health (Mental Component Summary [MCS]) and physical health components (Physical Component Summary [PCS]) of QOL [35]. Given previous findings of both positive and negative associations of stress-coping strategies with measures of PTSD, resilience, and QOL, it is important to clarify the stress-coping strategies used by study participants.

In Japan, there is little mental health support for emergency service workers, and most workers are required to manage any mental distress that arises from their work on their own [36]. Therefore, it is necessary to identify ways to mitigate the low QOL associated with PTSD among emergency service workers in Japan. To the best of our knowledge, few studies have examined the relationships among stress-coping, social support, resilience, and QOL in emergency service workers responsible for rescuing victims from fires, traffic accidents, and disasters. There are likely to be complex associations between QOL and various factors in this population; an understanding of these associations is urgently needed.

2 Methods

2.1 Study aim, design, and setting

This survey study aimed to identify the effects of stress-coping, social support, and resilience (after adjusting for confounding factors) on the QOL subscale scores of emergency service workers using parallel and series models. Specifically, in an exploratory survey, we assessed whether different types of stress-coping behaviors differentially influenced the Brief COPE subscale scores of emergency service workers. On the basis of the findings of previous studies, we hypothesized that associations between probable PTSD and QOL would be modulated by stress-coping, social support, and resilience, as shown in Fig. 1 [23–25 , 31].

Fig. 1

Path model of the effect of PTSD (IES-R) on QOL (SF-36), as mediated by social support (MSPSS), Active coping (Brief COPE), and resilience (CD-RISC). Abbreviations: PTSD, posttraumatic stress disorder; QOL, quality of life; IES-R, Impact of Event Scale-Revised; MSPSS, Multidimensional Scale of Perceived Social Support; CD-RISC, Connor–Davidson Resilience Scale; SF-36, Medical Outcomes Study 36-Item Short-Form Health Survey; MCS, Mental Component Summary; PCS, Physical Component Summary; Brief COPE, Brief Coping Orientation to Problems Experienced. The Brief COPE includes 14 subscales: Self-distraction, Active coping, Denial, Substance use, Emotional support, Instrumental support, Behavioral disengagement, Venting, Positive reframing, Planning, Humor, Acceptance, Religion, and Self-blame. **p<0.01; ***p<0.001.

2.2 Study population

We conducted a multicenter cross-sectional survey involving fire stations in Japan. The participants were firefighters and emergency medical technicians aged 20–65 years. A similar study that surveyed QOL among emergency service workers with PTSD reported a small effect size [18], where Cohen suggested that 0.02 represents a small effect size for multiple regression analyses [37]. In the study design phase, we assumed that our statistical analysis would include multiple regression. G*Power software [38] was used for the power analysis, which indicated that 395 participants would be needed for this study assuming a significance level of α= 0.05, power of 80%, and effect size of 0.02. In total, 404 participants expressed interest in participating in this study. However, 31 participants withdrew before data collection began. We therefore sent the questionnaires to 373 participants, 220 of whom returned completed surveys (response rate = 58.9%). This research was conducted with the consent of the participating fire departments. Participant recruitment started in April 2015 and was completed in August 2015. Ethical approval was granted by the Clinical Research Ethics Committee of XXX University (approval number: 1731). All participants provided written informed consent by post.

2.3 Measurements

The Impact of Event Scale-Revised (IES-R) scale is the most commonly used measure to screen emergency service workers for mental health problems [39 –43]. The IES-R has a three-factor structure (eight items assessing Intrusion, eight items assessing Avoidance, and six items assessing Hyperarousal). The scale was developed by Weiss (2004) to assess probable PTSD among individuals exposed to severe stressors. A previous study reported that scores on all three subscales (Intrusion, Avoidance, and Hyperarousal) showed statistically significant differences between individuals with and without PTSD [44]. In addition, the IES-R has adequate internal consistency [42, 44] and is considered a useful measure of PTSD symptoms.

Responses to IES-R items are made on a five-point rating scale (0=“not at all,” 1=“a little bit,” 2=“moderately,” 3=“quite a bit,” 4=“extremely”). A non-clinical study of Japanese firefighters reported a cut-off total IES-R score of 24 for PTSD of clinical concern [45]. We included the scores of all three IES-R subscales in our analyses.

The Brief COPE inventory measures how often individuals use various coping strategies to handle difficult situations [32]. The Brief COPE comprises 14 factors, each with two items: Self-distraction, Active coping, Denial, Substance use, Using emotional support, Using instrumental support, Behavioral disengagement, Venting, Positive reframing, Planning, Humor, Acceptance, Religion, and Self-blame. Responses are made on a four-point rating scale (1=“I haven’t been doing this at all,” 2=“I’ve been doing this a little bit,” 3=“I’ve been doing this a moderate amount,” 4=“I’ve been doing this a lot”). Scores for the 14 subscales range from 2 to 8, with higher scores indicating greater use of a given coping method. We entered the 14 Brief COPE subscale scores into the analysis separately.

We evaluated social support with the 12-item Multidimensional Scale of Perceived Social Support (MSPSS), which measures perceptions of support from three sources: family, friends, and significant others [46]. Responses are made on a seven-point rating scale (1=“very strongly disagree,” 2=“strongly disagree,” 3=“mildly disagree,” 4=“neutral,” 5=“mildly agree,” 6=“strongly agree,” 7=“very strongly agree”). Total MSPSS scores (divided by 12) range from 1 to 7. Higher scores indicate a greater amount of support.

We used the Connor–Davidson Resilience Scale (CD-RISC) to assess resilience. This 25-item scale classifies respondents as high, intermediate, or low resilience. Responses are made on a five-point scale (0=“not true at all,” 1=“rarely true,” 2=“sometimes true,” 3=“often true,” 4=“true nearly all of the time”) [26]. Total CD-RISC scores range from 0 to 100, with higher scores indicating greater resilience.

Finally, the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) [47] is frequently used to screen participants for health-related QOL issues and assess general health status. The original SF-36 comprises eight factors (physical functioning, role limitations due to physical health, role limitations due to emotional problems, vitality, mental health, social functioning, bodily pain, and general health) and two composite scores (PCS and MCS). The summary score for each component is based on norm scores for the Japanese population, where the mean is 50.0 and the standard deviation is 10.0 [47]. We used an online scoring service to calculate the scores for the three SF-36 components. Higher SF-36 scores indicate better QOL.

2.4 Covariates

Covariates included age, period of employment, education level, and marital status. Age and period of employment were continuous variables, and education level was a binary variable (i.e., graduated high school or college/technical school/university). Because covariates can moderate associations between variables, covariates that were not moderators were treated as confounding factors.

2.5 Analyses

2.5.1 Assumption testing and sample characteristics

The normality of the data for each measure was assessed using probability–probability plots and Shapiro–Wilk tests. In addition, homogeneity of variance was assessed using Levene’s test. This study enrolled a non-clinical sample; therefore, many of the emergency service workers were likely healthy, which may skew the results.

2.5.2 Correlation analysis with adjustment for covariates

We subjected the IES-R, Brief COPE, MSPSS, CD-RISC, and SF-36 scores to partial correlation analysis with 2,000 bootstrap replications and adjustment for covariates. In addition, we checked for multicollinearity between the Brief COPE subscale scores and MSPSS scores because the two measures evaluate similar traits protective against mental disorders [48]. Multicollinearity, which can bias the results, was considered present when measures had very strong correlations [49]. In this analysis, values of 0.90–1.00 were taken to indicate a very strong correlation, while values of 0.70–0.89 indicated a strong correlation, values of 0.50–0.69 indicated a moderate correlation, values of 0.30–0.49 indicated a weak correlation, and values of 0.00–0.29 indicated a very weak or no correlation [50].

2.5.3 Multiple linear regression, path analysis, and confirmatory factor analysis

We used linear regression analysis to analyze the associations between the independent (IES-R score) and dependent (SF-36 score) variables. The IES-R and SF-36 scores were also analyzed using confirmatory factor analysis, which was conducted to evaluate whether subcategory scores were more suitable as predictor or outcome variables, and to develop a multiple path model for the above measures. We considered variables such as age, length of employment, and education level as potential moderators if the independent variable×moderator interaction was significant [51]. The total effect was calculated as the sum of the direct and total indirect effects. The effect of PTSD (indexed by three latent IES-R variables) on QOL (indexed by the PCS and MCS subdomains of the SF-36) was determined on the basis of coping (Brief COPE scores), social support (MSPSS scores), and resilience (CD-RISC).

2.5.4 Inference methods, model fit indices, and missingness

A Bayesian inference method was used for the path analysis. The Bayesian method does not depend on the sample size or normality of the data distribution [52], and the results can be analyzed using Markov chain Monte Carlo algorithms [53]. We could not apply pre-specified information in our path analysis using Bayesian inference, which exploits the prior distribution of data to improve the accuracy of inferences [54]. Model fit was evaluated using the posterior predictive p-value (PPP), comparative fit index (CFI), Tucker–Lewis index (TLI), and root mean square error of approximation (RMSEA). A PPP cut-off value of < 0.10 was used to determine whether the model should be rejected [55]. CFI and TLI values > 0.95 and an RMSEA value < 0.08 indicate an acceptable fit of the data [56, 57]. In addition, we conducted path analysis with Bonferroni correction because multiple comparisons involving the Brief COPE may cause type I errors. Missing values were handled using the full information maximum likelihood method. This approach assumes that data are missing completely at random (MCAR) or at random, and available data can be analyzed for missingness [58]. Data missing in a non-random way may lead to misinterpretation of the results. To determine whether the missing data were MCAR, we conducted Little’s MCAR test [59, 60]. A non-significant difference indicated that the missing data were MCAR [61]. A sensitivity analysis, which can reveal bias, was also conducted to further examine the data using a list-wise deletion method [58].

We conducted confirmatory factor analysis and regression, path, and moderation analyses using Mplus software (version 8.4; Muthén & Muthén, Los Angeles, CA, USA). Other analyses were conducted using Stata software (version 13; Stata Corp LP, College Station, TX, USA).

3 Results

3.1 Participant characteristics

This study used a cross-sectional design. Table 1 shows the participants’ demographic characteristics. Questionnaires were sent to 373 participants, 220 of whom responded. All participants were male. Less than 2% of the participants had missing data. Half of the participants were aged 30–39 years and most had worked at a fire station for > 10 years. Although few participants showed poor mental health, the IES-R scores indicated that 55 (25%) had probable PTSD (IES-R score of≥25; data not shown in the table). Three Brief COPE strategies were frequently used: Active coping, Planning, and Acceptance. The SF-36 MCS and PCS scores were above 50.0. Both the normality and homogeneity of variance assumptions were violated for all measures. Little’s MCAR test was not significant (χ² = 16.83, p = 0.99); therefore, missing values were likely to be MCAR.

Table 1
Demographic characteristics of Japanese emergency service workers

N = 220 % Mean SD Min Max

Age, y 99.1 39.12 7.61 23 61

18–29 8.2

30–39 48.2

40–49 33.2

>50 9.5

Period of employment, y 98.6 17.98 8.17 2 43

Education High school 51.4

Above college 47.7

Marital status Single 6.45

Married 93.55

IES–R

Intrusion 100.0 5.90 4.93 0 20

Avoidance 100.0 5.46 5.13 0 23

Hyperarousal 100.0 3.80 3.55 0 16

MSPSS 100.0 70.19 10.82 31 84

CD–RISC 99.5 58.67 17.20 0 96

Brief COPE

Self–distraction 99.5 4.21 1.39 2 8

Active coping 100.0 5.43 1.37 2 8

Denial 99.1 2.58 0.93 2 7

Substance use 99.5 3.53 1.68 2 8

Emotional support 99.1 4.26 1.35 2 8

Instrumental support 99.5 4.69 1.55 2 8

Behavioral disengagement 99.5 3.32 1.17 2 8

Venting 99.1 4.27 1.37 2 8

Positive reframing 99.5 4.87 1.32 2 8

Planning 99.1 5.53 1.45 2 8

Humor 99.5 3.96 1.37 2 8

Acceptance 99.1 5.62 1.44 2 8

Religion 99.5 2.93 1.18 2 8

Self–blame 99.5 3.95 1.45 2 8

SF–36

MCS 98.1 55.56 7.84 21.4 65.1

PCS 98.1 54.16 6.37 18.9 65.3

		N = 220 %	Mean	SD	Min	Max
Age, y		99.1	39.12	7.61	23	61
	18–29	8.2
	30–39	48.2
	40–49	33.2
	>50	9.5
Period of employment, y	98.6	17.98	8.17	2	43
Education	High school	51.4
	Above college	47.7
Marital status	Single	6.45
	Married	93.55
IES–R
	Intrusion	100.0	5.90	4.93	0	20
	Avoidance	100.0	5.46	5.13	0	23
	Hyperarousal	100.0	3.80	3.55	0	16
MSPSS		100.0	70.19	10.82	31	84
CD–RISC		99.5	58.67	17.20	0	96
Brief COPE
	Self–distraction	99.5	4.21	1.39	2	8
	Active coping	100.0	5.43	1.37	2	8
	Denial	99.1	2.58	0.93	2	7
	Substance use	99.5	3.53	1.68	2	8
	Emotional support	99.1	4.26	1.35	2	8
	Instrumental support	99.5	4.69	1.55	2	8
	Behavioral disengagement	99.5	3.32	1.17	2	8
	Venting	99.1	4.27	1.37	2	8
	Positive reframing	99.5	4.87	1.32	2	8
	Planning	99.1	5.53	1.45	2	8
	Humor	99.5	3.96	1.37	2	8
	Acceptance	99.1	5.62	1.44	2	8
	Religion	99.5	2.93	1.18	2	8
	Self–blame	99.5	3.95	1.45	2	8
SF–36
	MCS	98.1	55.56	7.84	21.4	65.1
	PCS	98.1	54.16	6.37	18.9	65.3

Abbreviations: SD, standard deviation; IES-R, Impact of Event Scale-Revised; MSPSS, Multidimensional Scale of Perceived Social Support; CD-RISC, Connor-Davidson Resilience Scale; SF-36, MOS 36-Item Short-Form Health Survey; MCS, Mental Component Summary; PCS, Physical Component Summary.

3.2 Bootstrapped partial correlations

Table 2 shows the partial correlations between measures. IES-R subscale scores were significantly negatively correlated with the SF-36 summary scores (MCS: r=–0.40 to –0.28, ps < 0.001) and MSPSS scores (r=–0.25 to –0.18, p < 0.01 to 0.001). The scores for three IES-R subscales were significantly positively correlated with all of the Brief COPE subscale scores, except for Humor (r = 0.14 to 0.47, p < 0.05 to 0.001). CD-RISC scores were significantly positively correlated with seven Brief COPE subscale scores (Active coping: r = 0.25, p < 0.001; Instrumental support: r = 0.18, p < 0.05; Venting: r = 0.18, p < 0.01; Positive reframing: r = 0.38, p < 0.001; Planning: r = 0.40, p < 0.001; Humor: r = 0.27, p < 0.001; and Acceptance: r = 0.26, p < 0.01) and MSPSS scores (r = 0.16, p < 0.05), and significantly negatively associated with two Brief COPE subscale scores (Denial: r=–0.15, p < 0.05; Behavioral disengagement: r=–0.20, p < 0.01). CD-RISC scores showed a significant but relatively weak positive correlation with SF-36 MCS scores (r = 0.31, p < 0.001).

Table 2
Bootstrapped partial correlations between IES-R, SF-36, Brief COPE, MSPSS, and CD-RISC scores

IES–R SF–36 MSPSS p CD–RISC p

Avoidance p Intrusion p Hyperarousal p MCS p PCS p

IES–R Avoidance 1

Intrusion 0.85 * 1

Hyperarousal 0.79 * 0.83 * 1

SF–36 MCS –0.28 * –0.30 * –0.40 * 1

PCS –0.04 –0.02 –0.07 0.01 1

MSPSS –0.19 –0.18 –0.25 * 0.40 * 0.07 1

CD–RISC –0.07 –0.05 –0.13 0.31 *** 0.08 0.16 * 1

Brief–COPE Self–distraction 0.43 * 0.31 * 0.30 * –0.30 * –0.08 –0.16 * 0.06

Active coping 0.17 * 0.19 0.21 * –0.14 –0.02 –0.03 0.25

Denial 0.46 * 0.30 * 0.28 * –0.12 0.02 –0.06 –0.15 *

Substance use 0.24 0.22 0.19 ** –0.15 * –0.05 –0.17 * –0.02

Emotional support 0.29 0.20 0.23 *** –0.17 * –0.06 0.09 0.14

Instrumental support 0.12 0.12 0.14 * –0.10 –0.02 0.13 0.18 *

Behavioral disengagement 0.47 * 0.30 * 0.38 * –0.28 * –0.08 –0.21 –0.20

Venting 0.13 0.15 * 0.13 –0.08 –0.10 0.10 0.18

Positive reframing 0.23 * 0.19 0.17 –0.03 0.08 0.05 0.38 ***

Planning 0.08 0.15 * 0.12 –0.03 0.04 0.07 0.40 *

Humor –0.01 –0.08 –0.05 0.09 –0.01 0.08 0.27 *

Acceptance 0.15 * 0.14 * 0.11 0.02 0.00 –0.06 0.26

Religion 0.32 * 0.33 * 0.34 * –0.14 0.03 –0.10 –0.00

Self–blame 0.41 * 0.42 * 0.46 * –0.34 * –0.07 –0.13 –0.07

		IES–R	SF–36		MSPSS	p	CD–RISC
IES–R	Avoidance	1
	Intrusion	0.85	***	1
	Hyperarousal	0.79	***	0.83	***	1
SF–36	MCS	–0.28	***	–0.30	***	–0.40	***	1
	PCS	–0.04		–0.02		–0.07		0.01		1
MSPSS		–0.19	**	–0.18	**	–0.25	***	0.40	***	0.07	1
CD–RISC		–0.07		–0.05		–0.13		0.31	***	0.08	0.16	*	1
Brief–COPE	Self–distraction	0.43	***	0.31	***	0.30	***	–0.30	***	–0.08	–0.16	*	0.06
	Active coping	0.17	*	0.19	**	0.21	***	–0.14		–0.02	–0.03		0.25	**
	Denial	0.46	***	0.30	***	0.28	***	–0.12		0.02	–0.06		–0.15	*
	Substance use	0.24	**	0.22	**	0.19	**	–0.15	*	–0.05	–0.17	*	–0.02
	Emotional support	0.29	**	0.20	**	0.23	***	–0.17	*	–0.06	0.09		0.14
	Instrumental support	0.12		0.12		0.14	*	–0.10		–0.02	0.13		0.18	*
	Behavioral disengagement	0.47	***	0.30	***	0.38	***	–0.28	***	–0.08	–0.21	**	–0.20	**
	Venting	0.13		0.15	*	0.13		–0.08		–0.10	0.10		0.18	**
	Positive reframing	0.23	***	0.19	**	0.17	**	–0.03		0.08	0.05		0.38	***
	Planning	0.08		0.15	*	0.12		–0.03		0.04	0.07		0.40	***
	Humor	–0.01		–0.08		–0.05		0.09		–0.01	0.08		0.27	***
	Acceptance	0.15	*	0.14	*	0.11		0.02		0.00	–0.06		0.26	**
	Religion	0.32	***	0.33	***	0.34	***	–0.14		0.03	–0.10		–0.00
	Self–blame	0.41	***	0.42	***	0.46	***	–0.34	***	–0.07	–0.13		–0.07

Abbreviations: IES-R, Impact of Event Scale-Revised; SF-36, MOS 36-Item Short-Form Health Survey; MCS, Mental Component Summary; PCS, Physical Component Summary; RCS, Role/social Component Summary; MSPSS, Multidimensional Scale of Perceived Social Support; CD-RISC, Connor-Davidson Resilience Scale. *p<0.05; **p<0.01; ***p<0.001. Covariates were adjusted for age, length of employment, education, and marital status. Bootstrapped partial correlation estimations obtained using 2000 samples.

3.3 Regression and path analysis with multiple variables

Regression analysis showed that the latent factor structure of the SF-36 (based on the MCS and PCS scores; β= –0.500, p < 0.001) was predicted by that of the IES-R (based on the Intrusion, Avoidance, and Hyperarousal scores; data not shown in the tables). Tables 3 4 and Fig. 1 show the results of path models including multiple variables and adjusted for confounding factors, including 3 IES subscale scores, 14 Brief COPE subscale scores, MSPSS scores, and 2 SF-36 component scores (PCS and MCS). All pathway requirements were fulfilled by the Active coping and Positive reframing Brief COPE hypothesis models. The model fit indices of the two path models indicated an acceptable model fit. The direct effects (c) between the IES-R and SF-36 latent factor structures were –0.430 to –0.427. The latent factor structure of the IES-R was significantly negatively associated with MSPSS scores (a1: β= –0.225 to –0.223, ps < 0.01), and significantly positively associated with the Brief COPE Active coping and Positive reframing scores (a2: β= 0.200 to 0.208, ps < 0.01). CD-RISC scores were significantly positively associated with MSPSS scores (b2: β= 0.151 to 0.175, ps < 0.01) and the Brief COPE Active coping and Positive reframing scores (b3: β= 0.255 to 0.364, ps < 0.001). There were significant positive associations between CD-RISC scores and the latent factor structure of the SF-36 (d1: β= 0.431 to 0.441, ps < 0.001). All direct and total indirect effect pathways with Brief COPE Active coping scores were statistically significant, but the Positive reframing scores were not.

Table 3
Coefficients of the path analysis between PTSD (IES-R) and QOL (SF-36), explained by coping (Brief COPE subscales), support (MSPSS), and resilience (CD-RISC)

Self– Active Denial Substance Emotional Instrumental Behavioral Venting Positive Planning Humor Acceptance Religion Self–

distraction coping use support support disengagement reframing blame

β p β p β p β p β p β p β p β p β p β p β p β p β p β p

IES–R ⟶ SF–36 (c) –0.432 * –0.427 * –0.440 * –0.428 * –0.430 * –0.426 * –0.430 * –0.434 * –0.430 * –0.432 * –0.421 * –0.427 * –0.421 * –0.431 *

IES–R ⟶ MSPSS (a1) –0.228 –0.225 –0.226 * –0.226 –0.223 –0.224 –0.230 –0.225 –0.223 –0.224 –0.224 –0.226 –0.225 –0.227

IES–R ⟶ Brief COPE ^a (a2) 0.339 * 0.200 0.372 * 0.241 * 0.244 * 0.134 0.424 * 0.148 0.208 0.139 –0.055 0.145 0.362 * 0.478 *

MSPSS ⟶ SF–36 (b1) 0.393 * 0.378 * 0.367 * 0.385 * 0.373 * 0.399 * 0.388 * 0.380 * 0.404 * 0.383 * 0.396 * 0.375 * 0.395 * 0.390 *

MSPSS ⟶ CD–RISC (b2) 0.177 0.175 ** 0.157 * 0.166 ** 0.156 * 0.146 * 0.129 * 0.149 * 0.151 ** 0.145 * 0.148 * 0.184 0.166 0.158 *

Brief COPE ^a⟶ CD–RISC (b3) 0.077 0.255 * –0.109 0.004 0.120 0.160 –0.184 0.166 0.364 * 0.381 * 0.242 * 0.264 * 0.008 –0.056

CD–RISC ⟶ SF–36 (d1) 0.435 * 0.431 * 0.417 * 0.430 * 0.419 * 0.435 * 0.424 * 0.426 * 0.441 * 0.427 * 0.432 * 0.423 * 0.432 * 0.429 ***

PPP 0.019 0.197 0.015 0.205 0.081 0.146 0.005 0.189 0.147 0.166 0.192 0.246 0.258 0.140

RMSEA 0.078 0.048 0.076 0.048 0.061 0.054 0.086 0.049 0.054 0.053 0.049 0.043 0.042 0.055

CFI 0.967 0.987 0.969 0.987 0.979 0.984 0.961 0.986 0.984 0.985 0.986 0.990 0.990 0.984

TLI 0.927 0.972 0.931 0.972 0.954 0.964 0.913 0.969 0.964 0.966 0.970 0.978 0.978 0.964

	Self–	Active	Denial	Substance	Emotional	Instrumental	Behavioral	Venting	Positive	Planning	Humor	Acceptance	Religion	Self–
IES–R ⟶ SF–36 (c)	–0.432	***	–0.427	***	–0.440	***	–0.428	***	–0.430	***	–0.426	***	–0.430	***	–0.434	***	–0.430	***	–0.432	***	–0.421	***	–0.427	***	–0.421	***	–0.431	***
IES–R ⟶ MSPSS (a1)	–0.228	**	–0.225	**	–0.226	***	–0.226	**	–0.223	**	–0.224	**	–0.230	**	–0.225	**	–0.223	**	–0.224	**	–0.224	**	–0.226	**	–0.225	**	–0.227	**
IES–R ⟶ Brief COPE ^a (a2)	0.339	***	0.200	**	0.372	***	0.241	***	0.244	***	0.134		0.424	***	0.148		0.208	**	0.139		–0.055		0.145		0.362	***	0.478	***
MSPSS ⟶ SF–36 (b1)	0.393	***	0.378	***	0.367	***	0.385	***	0.373	***	0.399	***	0.388	***	0.380	***	0.404	***	0.383	***	0.396	***	0.375	***	0.395	***	0.390	***
MSPSS ⟶ CD–RISC (b2)	0.177	**	0.175	**	0.157	*	0.166	**	0.156	*	0.146	*	0.129	*	0.149	*	0.151	**	0.145	*	0.148	*	0.184	**	0.166	**	0.158	*
Brief COPE ^a⟶ CD–RISC (b3)	0.077		0.255	***	–0.109		0.004		0.120		0.160		–0.184		0.166		0.364	***	0.381	***	0.242	***	0.264	***	0.008		–0.056
CD–RISC ⟶ SF–36 (d1)	0.435	***	0.431	***	0.417	***	0.430	***	0.419	***	0.435	***	0.424	***	0.426	***	0.441	***	0.427	***	0.432	***	0.423	***	0.432	***	0.429	***
PPP	0.019	0.197	0.015	0.205	0.081	0.146	0.005	0.189	0.147	0.166	0.192	0.246	0.258	0.140
RMSEA	0.078	0.048	0.076	0.048	0.061	0.054	0.086	0.049	0.054	0.053	0.049	0.043	0.042	0.055
CFI	0.967	0.987	0.969	0.987	0.979	0.984	0.961	0.986	0.984	0.985	0.986	0.990	0.990	0.984
TLI	0.927	0.972	0.931	0.972	0.954	0.964	0.913	0.969	0.964	0.966	0.970	0.978	0.978	0.964

Abbreviations: PTSD, posttraumatic stress disorder; QOL, quality of life; IES-R, Impact of Event Scale-Revised; MSPSS, Multidimensional Scale of Perceived Social Support; CD-RISC, Connor-Davidson Resilience Scale; SF-36, MOS 36-Item Short-Form Health Survey; MCS, Mental Component Summary; PPP, posterior predictive p-value; RMSEA, root mean square error of approximation; CFI, comparative fit index; TLI, Tucker-Lewis index; SD, standard deviation. *p<0.05, **p<0.01; ***p<0.001. ^aBonferroni corrected p-values (p < 0.0036). β values are standardized. Covariates were adjusted for age, length of employment, education, and marital status.

Table 4

Specific indirect effects, total indirect effect, and total effect in the path analysis

	Active coping					Positive reframing
	β	p	95% CI		Posterior	β	p	95% CI		Posterior
					SD					SD
MSPSS (a1⟶b1)	–0.095	**	–0.184	–0.032	0.039	–0.101	**	–0.193	–0.033	0.041
MSPSS ⟶ CD–RISC (a1⟶b2⟶d1)	–0.018	**	–0.045	–0.003	0.011	–0.016	**	–0.040	–0.002	0.010
Brief COPE ^a ⟶ CD–RISC (a2⟶b3⟶d1)	0.024	**	0.006	0.055	0.013	0.036	*	0.011	0.076	0.017
Total indirect effect	–0.090	*	–0.190	–0.012	0.046	–0.098		–0.186	0.006	0.034
Total effect	–0.590	***	–0.810	–0.382	0.108	–0.575	***	–0.797	–0.363	0.111

Abbreviations: MSPSS, Multidimensional Scale of Perceived Social Support; CD-RISC, Connor-Davidson Resilience Scale; PPP, posterior predictive p-value; RMSEA, root mean square error of approximation; CFI, comparative fit index; TLI, Tucker-Lewis index; CI confidence interval; SD, standard deviation. *p<0.05; **p<0.01; ***p<0.001. ^aBrief COPE subscales were Active coping and Positive reframing. β values are standardized. Covariates were adjusted for age, length of employment, education, and marital status. Model fit values of the two path models: Active coping: PPP = 0.197, RMSEA = 0.048, CFI = 0.987, TLI = 0.972; Positive reframing: PPP = 0.147, RMSEA = 0.054, CFI = 0.984, TLI = 0.964.

Table 5 shows the results of sensitivity analysis using the list-wise deletion method. The model fit indices for the Brief COPE Active coping subscale scores had similar values when using the full information maximum likelihood and list-wise deletion methods, and the sensitivity analysis revealed an acceptable model fit (PPP = 0.180, RMSEA = 0.054, CFI = 0.984, TLI = 0.965).

Table 5

Path analysis using the list-wise deletion method

	Active coping
	β	p	95% CI		Posterior
					SD
IES–R ⟶ SF–36 (c)	–0.410	***	–0.601	–0.242	0.091
IES–R ⟶ MSPSS (a1)	–0.189	**	–0.340	–0.071	0.069
IES–R ⟶ Brief COPE ^a (a2)	0.211	**	0.071	0.344	0.070
MSPSS ⟶ SF–36 (b1)	0.364	***	0.205	0.532	0.084
MSPSS ⟶ CD–RISC (b2)	0.191	**	0.057	0.319	0.067
Brief COPE ^a ⟶ CD–RISC (b3)	0.253	***	0.124	0.376	0.065
CD–RISC ⟶ SF–36 (d1)	0.406	***	0.237	0.584	0.090
MSPSS (a1⟶b1)	–0.093	**	–0.184	–0.029	0.040
MSPSS ⟶ CD–RISC (a1⟶b2⟶d1)	–0.019	**	–0.047	–0.003	0.011
Brief COPE ^a ⟶ CD–RISC (a2⟶b3⟶d1)	0.026	**	0.007	0.059	0.013
Total indirect effect	–0.086	*	–0.192	–0.004	0.048
Total effect	–0.611	***	–0.838	–0.389	0.114

Abbreviations: IES-R, Impact of Event Scale-Revised; MSPSS, Multidimensional Scale of Perceived Social Support; CD-RISC, Connor-Davidson Resilience Scale; SF-36, MOS 36-Item Short-Form Health Survey; MCS, Mental Component Summary; PPP, posterior predictive p-value; RMSEA, root mean square error of approximation; CFI, comparative fit index; TLI, Tucker-Lewis index; CI, confidence interval; SD, standard deviation. *p<0.05; **p<0.01; ***p<0.001. ^aBrief COPE subscale was Active coping. β values are standardized. Covariates were adjusted for age, length of employment, education, and marital status. Model fit in a path model: Active coping: PPP = 0.180, RMSEA = 0.054, CFI = 0.984, TLI = 0.965.

3.4 Moderation analysis

In the path analysis with moderators, all pathways between the IES-R, Brief COPE, MSPSS, CD-RISC, and SF-36 scores were statistically significant for the model for Active coping strategies. However, the moderation analysis based on this model yielded non-significant results because the potential moderators were not significant (data not shown in the table). The confounding factors controlled for in the path analysis were age, length of employment, education level, and marital status.

4 Discussion

This study aimed to determine whether stress-coping, social support, and resilience explain the association between PTSD symptoms and QOL among firefighters and emergency medical technicians working in emergency services in Japan. In addition, we sought to determine how stress-coping behavior influences components of QOL to clarify the associations among the above measures. The results of the models supported our initial hypothesis that PTSD symptoms are directly and indirectly associated with mental health status and QOL via all three variables. Furthermore, in the Active coping Brief COPE hypothesis model, there were significant associations among all measures.

4.1 Effects of independent variables

This study found that the prevalence of probable PTSD in emergency service workers was 25%. The prevalence of PTSD in the general Japanese population is approximately 1%, which is clearly lower than that of the emergency service workers in our study [62]. Therefore, emergency service workers may be more prone to developing mental disorders than the general population. Repeated exposure to traumatic scenes might lead to feelings of numbness in firefighters, which in turn may negatively impact their mental health [63]. In this study, more severe PTSD symptoms reduced mental health functioning in terms of QOL, which highlights the importance of identifying effective methods to prevent individuals from developing PTSD. A similar result was reported by a study that compared firefighters with PTSD and a non-emergency service group [18].

This study showed that greater use of certain stress-coping strategies was associated with PTSD symptoms. Previous studies have indicated that greater use of maladaptive coping strategies (e.g., self-distraction, denial, behavioral disengagement, and self-blame) is predicted by more severe mental disorders [33 , 65]. Holubova et al. [66] reported that participants with mental disorders tend to engage in maladaptive rather than adaptive coping strategies, and that these maladaptive coping strategies are associated with lower QOL. The Japanese emergency service workers in our study engaged in three maladaptive coping strategies (i.e., self-distraction, behavioral disengagement, and self-blame), which in turn may have decreased mental health-related QOL among those with PTSD symptoms.

We found that less social support was associated with more severe PTSD symptoms. Therefore, social support may help prevent the exacerbation of PTSD [24]. Previous studies on war veterans with PTSD reported that lower family support was associated with PTSD [67]. Moreover, war veterans with PTSD exhibited poorer social functioning than those without PTSD [24]. Importantly, social support may improve the QOL of rescue workers. Our findings showed that rescue workers who used more adaptive coping strategies (e.g., Active coping, Instrumental support, Venting, Positive reframing, Planning, Humor, and Acceptance), and those who used fewer maladaptive strategies (e.g., Denial and Behavioral disengagement), had greater resilience, consistent with previous studies [30, 68]. In addition, greater resilience was associated with better mental health status in terms of QOL, but not PTSD symptoms. We found no association between PTSD symptoms and resilience, which is inconsistent with a previous study involving emergency department patients that found that lower resilience predicted a greater impact of PTSD [30]. Given this discrepancy in results, we suggest that PTSD symptoms are an indirect predictor of mental health-related QOL, which can be explained by stress-coping and social functioning via resilience.

4.2 Main findings of the multiple path analysis

The path models for the active coping strategy supported our research hypothesis. The findings of the multiple path analysis revealed that PTSD symptoms were directly associated with stress-coping and social support, which were indirectly related to QOL via resilience. The p-values for the association between PTSD symptoms and resilience in the bootstrapped partial correlation and path analyses were not significant, suggesting that resilience was not directly associated with probable PTSD. In contrast, the association between PTSD symptoms and QOL in the multiple path model was significant. Resilience, as a third variable, was associated with stress-coping and social support. Previous studies have reported a positive association between stress-coping and resilience [30], and that receiving social support from a manager influences firefighters’ resilience [69]. We found a significant positive association between resilience and QOL among emergency service workers. Furthermore, via resilience, both stress-coping and social support may help prevent the development of PTSD symptoms and maintain good QOL. Palmes et al. [70] stated that resilience explains the association between coping strategies and QOL among older adults, and postulated that coping strategies may promote QOL through resilience. As a strategy to cope with difficulties, active coping was found to be a particularly important behavior to maintain mental health and QOL among emergency service workers, although it requires practice and social support. Resilience may link active coping and social support, and thus influence mental health-related QOL. An example of active coping is talking about one’s feelings of anxiety and receiving acceptance from another person as a form of social support. A previous review indicated that asking for social support is associated with stress-coping, which prevents PTSD. In addition, the review indicated that this stress-coping behavior may aid recovery from exposure to a stressor, trauma symptoms, and long-term mental distress [71].

The main finding of our study was that, via resilience, active coping and social support exerted a protective effect against PTSD and improved mental health-related QOL. Thus, it is necessary to cope with stressors before PTSD develops [9]. Encouraging both active stress-coping and social support can improve resilience [30, 31] and, as a result, QOL [23 –25]. However, in the absence of these protective factors emergency service work may lead to PTSD symptoms. Although the data analyzed in this study were 8 years old, it nevertheless makes a valuable contribution by addressing the gap in the literature regarding the roles of stress-coping, social support, and resilience in the association between PTSD and QOL. It should also be noted that a previous study found that mindfulness-based exposure therapy reduced PTSD symptoms; this therapy might help prevent mental disorders in emergency service workers [72].

Mental health screening for emergency service workers at risk of PTSD should include an evaluation of resilience. If, as indicated in our study, stress-coping is negatively associated with QOL, failure to consider resilience could lead to inaccurate assessments of mental health because resilience may improve QOL. In addition, resilience refers to the ability to recover from difficult situations and maintain mental health; thus, if resilience is reduced, an individual may experience mental health problems. Therefore, it is important that healthcare providers and medical practitioners consider the behaviors of emergency service workers. However, the present findings are only speculative; given the cross-sectional design of the study, a causal relationship between PTSD and QOL (modulated by stress-coping, social support, and resilience) could not be confirmed.

4.3 Evaluation of the theoretical model using structural equation modeling

This study evaluated the relationship between probable PTSD and QOL using structural equation modeling. Previous PTSD studies have reported significant associations among stress-coping, social support, resilience, and QOL, but did not consider the complex interrelationships of resilience with stress-coping and social support [22 , 25]. The significant direct and total indirect (via active coping and social support) pathways between probable PTSD and QOL in this study show that complex relationships between measures can be predicted in QOL studies. Furthermore, sensitivity analysis using a list-wise deletion method indicated that the hypothesis model had low sampling bias. Structural equation modeling is a useful and flexible technique to test complex associations among measures [73, 74].

4.4 Limitations

This study had several limitations. First, the sample only included male participants. The observed association between PTSD symptoms and QOL may have differed if female participants had also been included in the study given the potential role of sex. However, previous studies on PTSD and QOL also failed to examine the effect of sex, including a study that focused on male veterans [20]. Female emergency service workers have a higher risk of suicide compared with their male counterparts, or with female workers in other occupations [75]. Further studies are needed including female rescue workers, who might experience lower QOL because of more severe PTSD. Second, the long-term influence of resilience on the mental health of emergency service workers should be examined in longitudinal studies. Third, because this study used a cross-sectional design, we could not confirm a causal relationship between QOL and probable PTSD. Fourth, the present study had an insufficient sample size; the power analysis showed that 373 participants were required given the number of variables. Fifth, most previous studies on stress/trauma examined past traumatic experiences and work stressors, where repeated exposure to traumatic scenes while working might negatively impact mental health [63]. However, we did not collect such information. Sixth, the exploratory nature of the Bayesian inference analysis may have introduced inferential errors. Bayesian analysis may have low statistical power, especially for small effect sizes, although very few previous studies adjusted their analyses accordingly [76]. To determine whether the above problems affected our results, further confirmatory studies using the hypothesis model employed in the present study are needed. Seventh, we used the IES-R to screen for probable PTSD. However, this scale is inferior to other screening tools, such as the PTSD Checklist, because it does not pertain specifically to PTSD [77]. Eighth, the present study enrolled a non-clinical sample; therefore, many of the emergency service workers were likely healthy, which may have skewed the results. Additionally, the results cannot be generalized to general populations worldwide because all of our participants were Japanese emergency service workers, who may have different characteristics compared with other workers. Finally, we could not consider the impact of the coronavirus disease 2019 pandemic (COVID-19) because the data in this study are 8 years old. COVID-19 can cause severe respiratory distress and death. Rescue workers may have come into contact with infected people, and witnessed traumatic events and deaths, during the pandemic, potentially leading to mental disorders [78]. It is necessary to carefully assess the impact of the COVID-19 pandemic on the mental health of emergency service workers. A longitudinal study with a larger sample (including both males and females) that also considers the effects of repeated exposure to traumatic incidents is required to validate our results.

5 Conclusion

Japanese emergency service workers have a higher rate of PTSD and lower QOL than the general population. However, the results of this study suggest that active coping strategies protect against the development of PTSD symptoms and thus improve mental health-related QOL among emergency service workers. Active coping is important to overcome difficulties, and may include talking about feelings of anxiety and receiving acceptance from others. This behavior, and the associated social support, can enhance resilience, which in turn helps emergency service workers maintain their mental health and QOL. Additionally, the emergency service workers in this study employed three maladaptive coping strategies (i.e., self-distraction, behavioral disengagement, and self-blame), which can decrease mental health-related QOL among those with PTSD symptoms. Given that the association between coping strategies and social support was modulated by resilience, assessing resilience, the ability to cope with stress, and social support could improve the accuracy of mental health assessments for emergency service workers. It should be noted that, because the present study used a cross-sectional design, conclusions cannot be drawn regarding the causality of the relationship between PTSD symptoms and QOL. A longitudinal study is required to clarify the mechanisms by which stress-coping, social support, and resilience mediate the association between PTSD and QOL; female emergency service workers should also be included to assess the effects of gender.

Ethics statement

All procedures involving human participants were conducted in accordance with the ethical standards of the institutional and/or national research committee, and the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Ethical approval was granted by the Clinical Research Ethics Committee of Chiba University (approval number 1731).

Informed consent

All patients provided written informed consent to participate.

Competing interests

The authors declare that they have no conflicts of interest.

Footnotes

Acknowledgments

The authors would like to thank the rescue workers who participated in the study. They especially thank Dr. Noriyoshi Takei for his advice on the statistical methodology. They also thank Sarina Iwabuchi, PhD, and Michael Irvine, PhD, from Edanz (https://jp.edanz.com/ac) for editing drafts of this manuscript.

Funding

This work was supported by Pfizer Japan Inc. (no grant number assigned). The funder had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

Berninger

, Webber

, Niles

, Gustave

, Lee

, Cohen

, et al. Longitudinal study of probable post-traumatic stress disorder in firefighters exposed to the World Trade Center disaster. Am J Ind Med. 2010;53(12):1177–85.

Sakuma

, Takahashi

, Ueda

, Sato

, Katsura

, Abe

, et al. Post-traumatic stress disorder and depression prevalence andassociated risk factors among local disaster relief andreconstruction workers fourteen months after the Great East Japan Earthquake: a cross-sectional study. BMC Psychiatry. 2015;24;15:58.

Stellman

, Smith

, Katz

, Sharma

, Charney

, Herbert

, et al. Enduring mental health morbidity and social function impairment in world trade center rescue, recovery, and cleanup workers: the psychological dimension of an environmental health disaster. Environ Health Perspect. 2008;116(9):1248–53.

Slottje

, Witteveen

, Twisk

JWR

, Smidt

, Huizink

, van Mechelen

, et al. Post-disaster physical symptoms of firefighters and police officers: role of types of exposure and post-traumatic stress symptoms. Br J Health Psychol. 2008;13(Pt 2):327–42.

Brackbill

, Hadler

, DiGrande

, Ekenga

, Farfel

, Friedman

, et al. Asthma and posttraumatic stress symptoms 5 to 6 years following exposure to the World Trade Center terrorist attack. JAMA. 2009;302(5):502–16.

McFarlane

. Epidemiological evidence about the relationship between PTSD and alcohol abuse: the nature of the association. Addict Behav. 1998;23(6):813–25.

Chiu

, Niles

, Webber

, Zeig-Owens

, Gustave

, Lee

, et al. Evaluating risk factors and possible mediation effects inposttraumatic depression and posttraumatic stress disordercomorbidity. Public Health Rep. 2011;126(2):201–9.

Chang

C-M

, Connor

, Lai

T-J

, Lee

L-C

, Davidson

JRT

. Predictors of posttraumatic outcomes following the 1999 Taiwan earthquake. J Nerv Ment Dis. 2005;193(1):40–6.

Katsavouni

, Bebetsos

, Malliou

, Beneka

. The relationship between burnout, PTSD symptoms and injuries in firefighters. Occup Med. 2016;66(1):32–7.

10.

Duckworth

. Psychological problems arising from disaster work. Stress Med. 1986;2(4):315–23.

11.

Sifaki-Pistolla

, Chatzea

V-E

, Vlachaki

S-A

, Melidoniotis

, Pistolla

. Who is going to rescue the rescuers? Post-traumatic stress disorder among rescue workers operating in Greece during the European refugee crisis. Soc Psychiatry Psychiatr Epidemiol. 2016;52(1):45–54.

12.

Perrin

, DiGrande

, Wheeler

, Thorpe

, Farfel

, Brackbill

. Differences in PTSD prevalence and associated risk factors amongWorld Trade Center disaster rescue and recovery workers. Am J Psychiatry. 2007;164(9):1385–94.

13.

Pietrzak

, Feder

, Singh

, Schechter

, Bromet

, Katz

, et al. Trajectories of PTSD risk and resilience in World Trade Center responders: an 8-year prospective cohort study. Psychol Med. 2014;44(1):205–19.

14.

Wisco

, Marx

, Wolf

, Miller

, Southwick

, Pietrzak

. Posttraumatic stress disorder in the US veteran population: results from the National Health and Resilience in Veterans Study. J Clin Psychiatry. 2014;75(12):1338–46.

15.

Haynes

, Kelly

, Warner

, Quan

, Krakow

, Bootzin

. Cognitive Behavioral Social Rhythm Group Therapy for Veterans with posttraumatic stress disorder, depression, and sleep disturbance: Results from an open trial. J Affect Disord. 2016;192:234–43.

16.

Brown

, Mulhern

, Joseph

. Incident-related stressors, locus of control, coping, and psychological distress among firefighters in Northern Ireland. J Trauma Stress. 2002;15(2):161–8.

17.

Regehr

, Hill

, Knott

, Sault

. Social support, self-efficacy and trauma in new recruits and experienced firefighters. Stress Health. 2003;19(4):189–93.

18.

Bracken-Scally

, McGilloway

, Gallagher

, Mitchell

. Life after the emergency services: an exploratory study of well being and quality of life in emergency service retirees. Int J Emerg Ment Health. 2014;16(1):223–31.

19.

Development of the World Health Organization WHOQOL-BREF quality of life assessment. The WHOQOL Group. Psychol Med. 1998;28(3):551–8.

20.

Schnurr

, Lunney

, Bovin

, Marx

. Posttraumatic stress disorder and quality of life: extension of findings to veterans of the wars in Iraq and Afghanistan. Clin Psychol Rev. 2009;29(8):727–35.

21.

Bartlett

, Smith

, Tran

, Vujanovic

. Understanding mental health among military veterans in the fire service. Psychiatry Res. 2018;267:394–9.

22.

Chen

Y-S

, Chen

M-C

, Chou

FH-C

, Sun

F-C

, Chen

P-C

, Tsai

K-Y

, et al. The relationship between quality of life and posttraumatic stress disorder or major depression for firefighters in Kaohsiung, Taiwan. Qual Life Res. 2007;16(8):1289–97.

23.

Nachar

, Guay

, Beaulieu-Prévost

, Marchand

. Assessment of the Psychosocial Predictors of Health-Related Quality of Life in a PTSD Clinical Sample. Traumatology. 2013;19(1):20–7.

24.

Tsai

, Harpaz-Rotem

, Pietrzak

, Southwick

. The role of coping, resilience, and social support in mediating the relation between PTSD and social functioning in veterans returning from Iraq and Afghanistan. Psychiatry. 2012;75(2):135–49.

25.

Crestani Calegaro

, Canova Mosele

, Lorenzi Negretto

, Zatti

, Miralha da Cunha

, Machado Freitas

. The role of personality in posttraumatic stress disorder, trait resilience, and quality of life in people exposed to the Kiss nightclub fire. PLoS ONE. 2019;29;14(7):e0220472.

26.

Connor

, Davidson

JRT

. Development of a new resilience scale: the Connor-Davidson Resilience Scale (CD-RISC). Depress Anxiety. 2003;18(2):76–82.

27.

Post

, Pardeller

, Frajo-Apor

, Kemmler

, Sondermann

, Hausmann

, et al. Quality of life in stabilized outpatients with bipolar I disorder: Associations with resilience, internalized stigma, and residual symptoms. J Affect Disord. 2018;238:399–404.

28.

Mizuno

, Hofer

, Suzuki

, Frajo-Apor

, Wartelsteiner

, Kemmler

, et al. Clinical and biological correlates of resilience in patients with schizophrenia and bipolar disorder: A cross-sectional study. Schizophr Res. 2016;175(1-3):148–53.

29.

Nakazawa

, Noda

, Ichikura

, Okamoto

, Takahashi

, Yamamura

, et al. Resilience and depression/anxiety symptoms in multiplesclerosis and neuromyelitis optica spectrum disorder. Mult SclerRelat Disord. 2018;25:309–15.

30.

Thompson

, Fiorillo

, Rothbaum

, Ressler

, Michopoulos

. Coping strategies as mediators in relation to resilience and posttraumatic stress disorder. J Affect Disord. 2018;225:153–9.

31.

, Ou

. Resilience and quality of life among Wenchuan earthquake survivors: the mediating role of social support. Public Health. 2014;128(5):430–7.

32.

Carver

. You want to measure coping but your protocol’s too long: consider the brief COPE. Int J Behav Med. 1997;4(1):92–100.

33.

Chamberlin

MJA

, Green

. Stress and Coping Strategies Among Firefighters and Recruits. J Loss Trauma. 2010;15(6):548–60.

34.

Tugade

, Fredrickson

, Barrett

. Psychological Resilience and Positive Emotional Granularity: Examining the Benefits of Positive Emotions on Coping and Health. J Pers. 2004;72(6):1161–90.

35.

Yang

, Yao

, Wu

, Wang

, Liu

, Wang

, et al. Quality of Life and Its Related Factors in Chinese Unemployed People: A Population-Based Cross-Sectional Study. Int J Environ Res Public Health. 2016;13(8):E797.

36.

Noda

, Asano

, Shimizu

, Hirano

. Assessing Subgroup Differences in Posttraumatic Stress Disorder Among Rescue Workers in Japan With the Impact of Event Scale-Revised. Disaster Med Public Health Pre. 2019;13(2):309–318.

37.

Cohen

. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. New York City: Routledge; 2013.

38.

Erdfelder

, Faul

, Buchner

. GPOWER: A general power analysis program. Behav Res Methods Instrum Comput. 1996;28:1–11.

39.

Asukai

, Kato

, Kawamura

, Kim

, Yamamoto

, Kishimoto

, et al. Reliability and validity of the Japanese-language version of the impact of event scale-revised (IES-R-J): four studies of different traumatic events. J Nerv Ment Dis. 2002;190(3):175–82.

40.

Matsuoka

, Nishi

, Nakaya

, Sone

, Noguchi

, Hamazaki

, et al. Concern over radiation exposure and psychological distress among rescue workers following the Great East Japan Earthquake. BMC Public Health. 2012;12:249.

41.

Chung

I-S

, Lee

M-Y

, Jung

S-W

, Nam

C-W

. Minnesota multiphasic personality inventory as related factor for post traumatic stress disorder symptoms according to job stress level in experienced firefighters: 5-year study. Ann Occup Environ Med. 2015;27:16.

42.

John Wilson

, Teren Keane

. Assessing Psychological Trauma and PTSD. 2nd ed. New York City: The Guilford Press; 2004.

43.

Cetin

, Kose

, Ebrinc

, Yigit

, Elhai

, Basoglu

. Identification and posttraumatic stress disorder symptoms in rescue workers in the Marmara, Turkey, earthquake. J Trauma Stress. 2005;18(5):485–9.

44.

Beck

, Grant

, Read

, Clapp

, Coffey

, Miller

, et al. The impact of event scale-revised: psychometric properties in a sample of motor vehicle accident survivors. J Anxiety Disord. 2008;22(2):187–98.

45.

Fushimi

. Posttraumatic stress in professional firefighters in Japan: rescue efforts after the Great East Japan Earthquake (Higashi Nihon Dai-Shinsai). Prehospital Disaster Med. 2012;27(5):416–8.

46.

Zimet

, Dahlem

, Zimet

, Farley

. The Multidimensional Scale of Perceived Social Support. J Pers Assess. 1988;52(1):30–41.

47.

Hays

, Sherbourne

, Mazel

. The RAND 36-Item Health Survey 1. 0. Health Econ. 1993;2(3):217–27.

48.

Roohafza

, Afshar

, Keshteli

, Mohammadi

, Feizi

, Taslimi

, et al. What’s the role of perceived social support and coping styles in depression and anxiety? J Res Med Sci. 2014;19(10):944–9.

49.

Vatcheva

, Lee

, McCormick

, Rahbar

. Multicollinearity in Regression Analyses Conducted in Epidemiologic Studies. Epidemiol. 2016;6(2):227.

50.

Hinkle

, Wiersma

, Jurs

. Applied Statistics for the Behavioral Sciences, International Edition. 5th ed. Belmont, CA: Wadsworth Publishing Co Inc; 2009.

51.

Baron

, Kenny

. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51(6):1173–82.

52.

Muthen

. Bayesian Analysis In Mplus: A Brief Introduction [Internet]. Los Angeles, CA: 2018 [cited 2018 Oct 7] Available from: https://www.statmodel.com/download/IntroBayesVersion%203.pdf

53.

Muthén

, Asparouhov

. Bayesian structural equation modeling: a more flexible representation of substantive theory. Psychol Methods. 2012;17(3):313–35.

54.

Yuan

, MacKinnon

. Bayesian mediation analysis. Psychol Methods. 2009;14(4):301–22.

55.

Cain

, Zhang

. Fit for a Bayesian: An Evaluation of PPP and DIC for Structural Equation Modeling. Struct Equ Model. 2019;26(1):39–50.

56.

Tucker

, Lewis

. A reliability coefficient for maximum likelihood factor analysis. Psychometrika. 1973;38:1–10.

57.

, Bentler

. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Equ Model. 1999;6(1):1–55.

58.

Nakagami

, Sugihara

, Takei

, Fujii

, Hashimoto

, Murakami

, et al. Effect of physical state on pain mediated through emotional health in rheumatoid arthritis. Arthritis Care Res. 2019;71(9):1216–1223.

59.

Little

RJA

. A Test of Missing Completely at Random for Multivariate Data with Missing Values. J Am Stat Assoc. 1988;83:1198–202.

60.

Van Ness

, Murphy

, Araujo

KLB

, Pisani

, Allore

. The Use of Missingness Screens in Clinical Epidemiologic Research Has Implications for Regression Modeling. J Clin Epidemiol. 2007;60(12):1239–45.

61.

. Little’s Test of Missing Completely at Random. Stata J. 2013;13(4):795–809.

62.

Kawakami

, Tsuchiya

, Umeda

, Koenen

, Kessler

. Trauma and posttraumatic stress disorder in Japan: Results from the World Mental Health Japan Survey. J Psychiatr Res. 2014;53:157–65.

63.

, Ws

, Ck

, B

. Firefighting and mental health: Experiences of repeated exposure to trauma. Work. 2016(4);53:737–44.

64.

Brooks

, Dunn

, Amlôt

, Greenberg

, Rubin

. Social andoccupational factors associated with psychological distress anddisorder among disaster responders: a systematic review. BMC Psychol. 2016;4:18.

65.

Saxon

, Makhashvili

, Chikovani

, Seguin

, McKee

, Patel

, et al. Coping strategies and mental health outcomes of conflict-affected persons in the Republic of Georgia. Epidemiol Psychiatr Sci. 2017;26(3):276–86.

66.

Holubova

, Prasko

, Ociskova

, Grambal

, Slepecky

, Marackova

, et al. Quality of life and coping strategies of outpatients witha depressive disorder in maintenance therapy –a cross-sectionalstudy. Neuropsychiatr Dis Treat. 2018;14:73–82.

67.

Wright

, Kelsall

, Sim

, Clarke

, Creamer

. Support mechanisms and vulnerabilities in relation to PTSD in veterans of the Gulf War, Iraq War, and Afghanistan deployments: a systematic review. J Trauma Stress. 2013;26(3):310–8.

68.

Fino

, Mema

, Russo

. War trauma exposed refugees and posttraumatic stress disorder: The moderating role of trait resilience. J Psychosom Res. 2020;129:109905.

69.

Bernabé

, Botia

. Resilience as a mediator in emotional social support’s relationship with occupational psychology health in firefighters. J Health Psychol. 2016;21(8):1778–86.

70.

Palmes

, Trajera

, Ching

. Relationship of Coping Strategies and Quality of Life: Parallel and Serial Mediating Role of Resilience and Social Participation among Older Adults in Western Philippines. Int J Environ Res Public Health. 2021;18(19):10006.

71.

Olff

, Langeland

, Gersons

BPR

. The psychobiology of PTSD: coping with trauma. Psychoneuroendocrinology. 2005;30(10):974–82.

72.

King

, Block

, Sripada

, Rauch

, Giardino

, Favorite

, et al. Altered Default Mode Network (DMN) resting state functionalconnectivity following a mindfulness-based exposure therapy forPosttraumatic Stress Disorder (PTSD) in combat veterans ofAfghanistan and Iraq. Depress Anxiety. 2016;33(4):289–99.

73.

Preacher

, Rucker

, Hayes

. Addressing Moderated Mediation Hypotheses: Theory, Methods, and Prescriptions. Multivar Behav Res. 2007;42(1):185–227.

74.

Preacher

, Hayes

. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods. 2008;40(3):879–91.

75.

Milner

, Witt

, Maheen

, LaMontagne

. Suicide among emergencyand protective service workers: A retrospective mortality study in Australia, 2001 to 2012. Work. 2017;57(2):281–7.

76.

Kennedy

. Beware of inferential errors and low power with Bayesian analyses: Power analysis is needed for confirmatory research. J Parapsychol. 2015;79:53–64.

77.

Creamer

, Bell

, Failla

. Psychometric properties of the Impactof Event Scale –Revised. Behav Res Ther. 2003;41(12):1489–96.

78.

Easterbrook

, Brown

, Millman

, Van Blyderveen

, Lanius

, Heber

, et al. The mental health experience of treatment-seeking military members and public safety personnel: a qualitative investigation of trauma and non-trauma-related concerns. Health Promot Chronic Dis Prev Can. 2022;42(6):252–60.