Revisiting the Compassion Fatigue,Burnout,Compassion Satisfaction,and Resilience Connection Among CISM Responders

Abstract

Previous research by Burnett and Wahl found that resilience moderately mediated the association between compassion fatigue (CF) and burnout (BO) among trauma responders. No studies have attempted to replicate and verify the research results of Burnett and Wahl. This study examined whether similar patterns of resilience, CF, BO, and compassion satisfaction (CS) would be found among a convenience sample of disaster behavioral health (DBH) responders (N = 70) attending an annual Critical Incident Stress Management training conference in Michigan. Measures included the 30-item Professional Quality of Life Scale, the 14-Item Resilience Scale, the Response to Stressful Events Scale, and a demographic questionnaire. Similar to the Burnett and Wahl’s study, resilience was found to have a significant negative correlation between CF and BO. Mediation analysis similarly found that resilience moderately mediated the relationship between CF and BO among both resilience measures.

Keywords

compassion fatigue burnout compassion satisfaction resilience mediation analysis replication

Traumatic events can have a rippling effect on individuals, emergency and disaster responders, and the communities exposed to them (whether it is directly or indirectly; Raphael, 1986), thereby prompting the need to organize some form of disaster mental health response. In fact, North and Pfefferbaum’s (2013) literature review on mental health services and interventions associated with community disasters noted that mental health should be integrated as part of a comprehensive emergency management response strategy with a focus on identifying mental health needs, triaging and referral to appropriate services, and providing appropriate mental health interventions. Embedded within this response strategy is the need to also address the impact that traumatic events will have on the disaster mental health responders who are deployed. The empathetic interactive nature of the profession with trauma survivors can result in indirect negative stress reactions such as compassion fatigue and burnout (Cieslak et al., 2014; Figley, 1995; Figley & Kleber, 1995; Palm, Polusny, & Follette, 2004; Pearlman & Saakvitne, 1995). However, recent research has tentatively shown that resilience may have an important mitigating role in reducing the devastating effects of compassion fatigue and burnout among disaster mental health providers (Burnett & Wahl, 2015).

A common construct within the trauma literature, often linked with indirect exposure to trauma events, is the phenomenon known as compassion fatigue (CF). Often, CF is synonymous with vicarious traumatization, secondary traumatic stress, or secondary traumatization (e.g., Bercier & Maynard, 2015; Figley, 1995; Ludick & Figley, 2016; Motta, 2008). CF can be broadly defined as a reduced capacity for empathy as manifested through emotional, behavioral, physical, spiritual, interpersonal, and cognitive reactions experienced by a disaster mental health service provider or any individual helping a traumatized person (McCann & Pearlman, 1990; Yassen, 1995). Moreover, Figley (1995) appropriately coined CF as the “cost of caring” due to the trauma worker’s use of empathy with survivors (see also Pearlman & Saakvitne, 1995). Research has also shown the potential consequences of developing CF, which can include emotional exhaustion associated with depersonalization or lack of accomplishment (Cieslak et al., 2014), alteration of one’s core beliefs and cognitive schemas (McCann & Pearlman, 1990; Motta, 2008; Pearlman & Mac Ian, 1995), alcohol use problems similar to survivors directly exposed to the event (Palm et al., 2004), and the increasing severity of intrusive thoughts and imagery, avoidance and emotional numbing behaviors, and persistent physiological arousal symptoms (Bercier & Maynard, 2015; Cieslak et al., 2013; Cieslak et al., 2014; Figley, 1995; Stamm, 2010).

The trauma literature is mixed on the prevalence of CF, with differences across occupational groups. For instance, Cieslak et al. (2013) found a prevalence of 19.2% among mental health providers working with military patients. Bercier and Maynard’s (2015) systematic literature review cited several studies that would suggest CF affects 5% to 15% of therapists in the clinical realm, whereas Cieslak et al. (2013) noted in their literature review multiple studies that meet the diagnostic criteria for secondary traumatic stress (i.e., 15.2% of social workers, 16.3% of oncology personnel, 32.8% of emergency nurses, and 34% of child protective service workers). Burnett and Wahl (2015) found 72% of a cross section of disaster behavioral health (DBH) and emergency preparedness personnel had scores indicative of CF on the Professional Quality of Life Scale (ProQOL). With regard to the prevalence of CF among DBH responders uniquely trained with a trauma response model, Wee and Myers (2003) observed CF among 40% of Critical Incident Stress Management (CISM) providers as a result of employing empathy among CISM recipients.

A more chronic and debilitating form of CF that is prevalent in the helping professions is burnout (Burnett & Wahl, 2015; Cieslak et al., 2013; Craig & Sprang, 2010; Maslach & Jackson, 1984; Newell & MacNeil, 2011). Oftentimes, the burnout process is linked to work-related stress (Brill, 1984). Figley (1995) posited that burnout is a gradual process that is a consequence of emotional exhaustion. Moreover, Pines and Aronson (1988) described burnout as a result of being involved in situations that have been emotionally demanding for an extended period of time. Maslach, Schaufeli, and Leiter (2001) proposed a more multifaceted definition that encompasses three parallel dimensions: emotional exhaustion (i.e., feeling of depletion and fatigue), depersonalization (i.e., cynicism), and inefficacy (i.e., reduced sense of personal accomplishment, negative self-evaluation, and job effectiveness). Studies have demonstrated a strong relationship between burnout and CF (Burnett & Wahl, 2015; Cieslak et al., 2014; Craig & Sprang, 2010).

With regard to prevalence, Morse, Salyers, Rollins, Monroe-DeVita, and Pfahler (2012) noted that 21% to 67% of mental health providers may be experiencing burnout. Among practitioners providing trauma services, Craig and Sprang (2010) found that only 5% had scores indicative of burnout on the ProQOL. A more recent study that involved a cross section of disaster mental health and emergency preparedness responders found scores indicative of burnout on the ProQOL at approximately 19% (Burnett & Wahl, 2015). For DBH CISM-trained providers, Wee and Myers (2003) found 13% were at risk of burnout.

A variable that is often associated with a form of resiliency is compassion satisfaction (CS), which refers to the contentment that one draws from his or her work, coupled with the act of helping (Craig & Sprang, 2010; Phelps, Lloyd, Creamer, & Forbes, 2009; Stamm, 2002, 2010). Higher levels of CS has been associated with lower levels of CF and burnout (Burnett & Wahl, 2015; Craig & Sprang, 2010; Ray, Wong, White, & Heaslip, 2013). For instance, Craig and Sprang (2010) observed that higher levels of CS were strongly associated with individuals who had special training in treating trauma than those without the special trauma training. Wee and Myers (2003) found approximately 89% of their respondents had a good to extremely high potential for CS.

Common within the trauma literature is the construct of resilience (e.g., Agaibi & Wilson, 2005; Bonanno, 2004, 2005; Ludick & Figley, 2016; Windle, 2011). Although several similar definitions have been touted for resilience in the literature (i.e., Ahern, Kiehl, Sole, & Byers, 2006; Bonanno, 2004; Everly, Welzant, & Jacobson, 2008; Luthar, Cicchetti, & Becker, 2000; Wagnild & Young, 1993), it can more broadly be defined, especially in this study, as the ability to adopt to and rebound from change (whether it is from stress or adversity) in a healthy, positive and growth-oriented manner. Although studies have argued that resilience is more multifaceted (Bonanno, 2005; Everly, 2012; Everly & Lating, 2013; Mancini & Bonanno, 2009), Kaminsky, McCabe, Langlieb, and Everly (2007) proposed a more interactive process between proactive resilience (resistance and immunity) and reactive resilience (ability to rebound from adversity) factors. Within this resiliency model, proactive resilience is comprised of three mechanisms (i.e., realistic expectations, fostering active optimism and self-efficacy, and enhancing neurophysiology), whereas reactive resilience consists of several structures (i.e., establishing supportive interpersonal relationships, cultivating a positive self-fulfilling expectations, having access to formal crisis intervention services, and advancing physical health), both of which foster overall psychological resilience (Everly, 2017).

The prevalence of resilience rates in the literature are varied depending on the operational definition used and the sample demographics. Windle’s (2011) review of resilience research generally noted sample population proportions found to be resilient varied from 25% to 84%. More specifically, Vanderbilt-Adriance and Shaw (2008) indicated that studies with single risk factors and studies with White, middle-class majority samples found resilient rates between 30% and 90%, with the majority grouping around 40% to 60%. In terms of disaster responders and resilience, Bonanno, Galea, Bucciarelli, and Vlahov (2006) found that 51.2% of their participants who were involved in the 9/11 World Trade Center rescue efforts were resilient. For disaster mental health responders, Burnett and Wahl (2015) found approximately 95% of sample were resilient.

Overall, research has demonstrated a relationship between CF, burnout, compassion satisfaction, and resilience (i.e., Burnett & Wahl, 2015; Cieslak et al., 2014; Cooke, Doust, & Steele, 2013; Ray et al., 2013; Slocum-Gori, Hemsworth, Chan, Carson, & Kazanjian, 2013; Sprang, Clark, & Whitt-Woosley, 2007; Thomas, 2012). Stamm (2010) has also suggested that a high level of CS joined with moderate to low CF and burnout encompasses a more positive outcome. However, a recent study utilizing mediation analysis found that resilience more moderately mediated the relationship between CF and burnout, compared to CS (Burnett & Wahl, 2015). Hence, the literature remains unclear whether resilience or CS is a stronger mediator between CF and burnout. However, CS may more likely serve as one of the multiple pathways of resiliency (Ludick & Figley, 2016).

To mitigate the potential psychological distress (i.e., CF and burnout) among DBH responders who provide crisis intervention services to those who have been exposed to traumatic events, research has suggested that receiving core and continual education training may help to foster proactive resilience (Everly et al., 2008; Myers & Wee, 2005; Schiraldi, Jackson, Brown, & Jordan, 2010). For instance, Atkins and Burnett (2016) found that disaster mental health responders trained in large and small group crisis interventions and individual and peer crisis interventions were strongly correlated with higher resilience and lower levels of burnout. One model that has been used for more than 25 years to train a cross section of DBH responders (e.g., police officers, firefighters, emergency medical personnel, social workers, psychologists, nurses, and chaplains) is the CISM program. CISM is a “comprehensive, integrated, multi-component crisis intervention system” (Everly & Mitchell, 1999; Flannery, 1998) that is “specifically designed to mitigate and, if possible, prevent the development of dysfunctional and potentially disabling posttraumatic syndromes and stress disorders” (Mitchell & Everly, 2001, p. 2). This also includes facilitating trauma survivors’ access to continued care as needed (Myers & Wee, 2005). Studies have been mixed regarding the effectiveness of CISM, specifically the crisis intervention component of critical incident stress debriefings (McNally, Bryant, & Ehlers, 2003; Myers & Wee, 2005; Pack, 2012; Roberts, 2002; Roberts & Everly, 2006; van Emmerik, Kamphuis, Hulsbosch, & Emmelkamp, 2002). However, trained providers of CISM have reported an array of symptoms of psychological stress after providing CISM services (Wee & Myers, 2003).

Present Study

Burnett and Wahl’s (2015) study initially examined compassion fatigue, burnout, compassion satisfaction, and their relationship with resilience among a cross section of DBH response providers. Using a group of DBH responders who have been specifically trained in CISM crisis intervention strategies, the purpose of this study was to examine patterns of prevalence and association among the variables of compassion fatigue, burnout, compassion satisfaction, and resilience. It was hypothesized that prevalence rates, correlations, and mediation analysis results would replicate the results of the original study.

Method

Participants

The data for this study were collected from a convenience sample of volunteer participants who were either novice or experienced CISM responders receiving training in various CISM crisis intervention methods while attending the Michigan Crisis Response Association’s (MCRA’s) 28th annual training conference held at the Kettunen 4-H Conference Center in Tustin, Michigan, from September 18 to 20, 2016. Of the 121 attendees who registered for a course at the conference, 70 useful surveys were obtained after they were checked for completeness (57.8% participation rate).

Measures

The measures used in this study included a demographic questionnaire, the ProQOL, the 14-Item Resilience Scale (RS-14), and the Response to Stressful Experiences Scale (RSES).

The demographic questionnaire asked participants to provide information about their age, gender, ethnic origin, marital status, religious affiliation, highest level of education, profession, type of agency they work for, and years of experience in their current profession. Participants were also asked to indicate which types of specialized trainings in disaster behavior health response they have completed and whether they were active members of a crisis response team. Finally, participants were asked to indicate how many DBH responses they participated in within the last year.

ProQOL

Stamm (2010) developed the ProQOL as a self-report measure to assess compassion fatigue, compassion satisfaction, and burnout based on how frequently a person has experienced certain antecedents (e.g., “I am happy” or “I feel trapped by my job as a helper”) within the past 30 days. The ProQOL is comprised of 30 items (10 items for each subscale) that are reflective of the three subscales content. Items are rated on a five-point Likert-type scale, ranging from 1 (never) to 5 (very often). Several items on the Burnout Subscale are reverse scored. Items for each subscale are summed and the average score on each subscale is 50 (SD = 10), with higher scores being indicative of the subscale’s content. Stamm (2010) reported that the ProQOL has demonstrated good construct validity and interscale correlation, with Cronbach’s alpha of .88 (Compassion Satisfaction Subscale), .81 (Compassion Fatigue Subscale), and .75 (Burnout Subscale). Cronbach’s alpha for the present study were .85 (Compassion Satisfaction Subscale), .77 (Compassion Fatigue Subscale), and .71 (Burnout Subscale).

The RS-14

The RS-14 (Wagnild, 2009) is the shorter version of the original Resilience Scale. The scale contains 14 self-report statements (e.g., “I usually manage one way or another” or “I usually take things in stride) that measure five characteristics of resilience (meaning and purpose of life, perseverance, equanimity, self-reliance, and existential aloneness). Items are rated on a seven-point Likert-type scale, ranging from 1 (strongly disagree) to 7 (strongly agree). Scores on each item are summed to yield a total score ranging from 14 to 98, with higher scores being indicative of greater perceived resilience. The RS-14 has demonstrated good construct, convergent, divergent, and discriminant validity with Cronbach’s alphas of .90 and greater (Wagnild, 2009). Cronbach’s alpha for the present study was .91.

Response to Stressful Events Scale (RSES)

The RSES (Johnson et al., 2011) is a 22-item self-report measure of individual differences in adaptive capabilities that contribute to the resilience process in response to stressful life events. Items (e.g., “During and after life’s most stressful events, I tend to . . . take action to fix things” or “. . . find opportunity for growth.”) are rated on a five-point Likert-type scale, ranging from 0 (not at all like me) to 4 (exactly like me). Item scores are summed to provide a total score ranging from 0 to 88, with higher scores depicting greater adaptive capabilities to stressful life events. The RSES has demonstrated convergent, discriminant, and concurrent validity, as well as good test–retest reliability, with Cronbach’s alpha ranging from .91 to .93 (Johnson et al., 2011). Cronbach’s alpha for the present study was .87.

Procedure

Prior to the conference, written permission to use the training conference to conduct the study was granted by MCRA. A research packet containing the informed consent document, demographic questionnaire, ProQOL, RS-14, and RSES was provided to all attendees at the beginning of their specific training course. At that time, attendees were informed about the study by the researcher and instructions were provided on how they could participate. Participants were allotted time throughout the 3-day training conference to complete and return the documents to a clearly marked collection box designated for the study at the registration table. Participation in the study was voluntary and anonymous, with informed consent being obtained. The Andrews University’s Institutional Review Board granted permission to conduct the study (IRB Protocol #16-116).

The frequency distribution cut scores as recommended by the instrument developers were used to assess the prevalence of CF, burnout (BO), CS, and resilience. The magnitude and direction of the relationship between participant scores on all measures was examined utilizing Pearson’s r. The mediation function in MBESS package version 3.3.3. (Kelley & Lai, 2012) in R 3.1.1 was used to perform a mediation analysis. For the present study, the κ² statistic was used to measure the mediation effect size by applying the benchmarks of .01, .09, and .25 (small, medium, and large effect sizes, respectively) for qualitative interpretation of mediation (Preacher & Kelley, 2011).

Results

Participant Demographics and Descriptive Statistics

Of the 70 participants, 30 were female. The age of participants ranged between 25 and 69 years, with a M_age = 45 years (SD = 11.03 years). With regards to ethnicity, 73% were White (non-Hispanic), 14% were African American, 3% were Asian, and 9% listed “Other.” Approximately 69% were married, 16% were single, 7% were divorced, and 3% were widowed. Forty-three percent indicated their religious affiliation was Protestant, 26% indicated “none,” 17% were Catholic, and 13% listed “other.” A little above 39% had obtained a master’s degree, 33% reported having a bachelor’s degree, 16% listed “other,” 7% had an associate’s degree, and 6% achieved a doctorate’s degree. For years of experience in their profession, 26% were between 20 and 29 years, 21% were between 5 and 9 years, 21% were between 10 and 19 years, 17% had less than five years, whereas 14% had 30 years or more. The four most frequently identified professions were social work (n = 18), fire services (n = 10), counseling (n = 6), and law enforcement (n = 5). Thirty-four percent were employed with a government agency, whereas 17% worked for a community mental health organization.

Overall, 69% had received specialized training in DBH response: large and small group crisis interventions (n = 40), individual and peer crisis interventions (n = 36), suicide intervention, prevention and postvention (n = 30), psychological first aid (n = 21), strategic planning for DBH response (n = 21), spiritual crisis intervention (n = 10), and family support (n = 6). Fifty-nine percent were members of an active local or state-level crisis response team. Of those participants who are members of an active crisis response team, 72% have participated in one to five crisis intervention responses within the past year.

For the ProQOL subscales, CF subscale scores ranged from 11 to 37 (M = 21.6, SD = 5.19), BO subscale scores ranged from 11 to 36 (M = 21.9, SD = 5.04), and CS subscale scores ranged from 25 to 50 (M = 39.6, SD = 5.35). RS-14 scores ranged from 54 to 98 (M = 80.7, SD = 10.4). RSES scores ranged from 39 to 83 (M = 63.4, SD = 10.7).

Prevalence

The present study examined the prevalence of compassion fatigue among novice and experienced DBH responders in Michigan who attended the CISM conference (see Table 1). Stamm (2010) utilized a standard score to indicate relative risk or protective factors on each of the three subscales with cut scores at the 25th and 75th percentiles. The standardized mean score is 50 (SD = 10) for each subscale. Therefore, scores below 43 are indicative of the risk factors for low CS and protective factors for CF and BO, whereas scores of 57 or higher are indicative of the risk factors for CF and BO and protective factors for CS. Approximately 100% of the participants scored below 43 on the CF subscale. This is in sharp contrast to Burnett and Wahl (2015), which found about 72% of their participants scoring in the high range for CF. For the BO subscale, all participants again scored below a standardized mean score of 43. Again, this is contrary to the Burnett and Wahl’s (2015) study that found at least 19% of their participants having scores indicative of above average levels of BO. One hundred percent of the participants in the present study also scored below the standardized mean score for CS, with 73.9% scoring below 43, whereas 26.1% scored in the low-average range (44-50). This finding is not consistent with Burnett and Wahl (2015), which reported 40% of their participants having scores indicative of high CS (57 and above).

Table 1.

Comparison of Patterns of Prevalence Rates Among Disaster Behavioral Health Responders.

Measure	Present studyN = 70			Burnett and Wahl (2015) N = 139
Measure	Low	Moderate	High	Low	Moderate	High
ProQOL
CF subscale	100%	0	0	0	27.8%	72.2%
BO subscale	100%	0	0	14.4%	66.7%	18.9%
CS subscale	73.9%	26.1%	0	3.7%	55.9%	40.4%
RS-14	5.8%	76.8%	17.4%	4%	73%	22%
RSES	8.7%	62.8%	28.6%	—	—	—

Note. ProQOL = Professional Quality of Life Scale; CF = compassion fatigue; BO = burnout; CS = compassion satisfaction; RS-14 = 14-Item Resilience Scale; RSES = Response to Stressful Experiences Scale.

In terms of resilience, Wagnild (2009) suggested that scores below 64 on the RS-14 are low. Hence, 5.8% of participants scored in the low resilient range. Approximately 17.4% of participants in the current study had scores indicative of high resilience, whereas 76.8% scored within the moderately low to moderately high resilient range. These results are comparable with Burnett and Wahl’s (2015), who found 4% of their participants had scored in the low resilience range, 22% scored within the high range, whereas 73% scored within the moderately low to moderately high range.

Johnson et al. (2011) suggested three acuity ranges (low, moderate, and high) to assess one’s ability to adapt to stressful life events. For participants in the present study, 28.6% fell within the high resilience range, whereas 8.7% fell within the low range. The majority of participants (62.8%) were within the moderate resilience range.

Pearson’s r Analysis

The present study examined the magnitude and direction of the relationship between CF, BO, CS, and resilience (see Table 2). Results indicated a significant large positive correlation between CF and BO (r = .56, p = .00, r² = .32), a significant medium negative correlation between CF and resilience (r = –.38, p = .00, r² = .15), and no significant correlation between CF and CS (r = .12, p = .33, r² = .01). CF accounted for 32% of the variance in BO and 15% of the variance in resilience. These results are similar to Burnett and Wahl’s (2015), who found that CF accounted for 32% of the variance in BO and 10% in resilience. Thus, the results suggest a similar pattern of higher levels of CF being associated with higher levels of BO, whereas higher levels of CF being associated with lower levels of resilience among DBH responders.

Table 2.

Comparison of Pearson r Correlation Coefficients Among ProQOL Subscales, RS-14, and RSES Scores.

	Present study	Burnett and Wahl (2015)
CF → BO	.563**	.566**
CF → CS	−.119	.056
CF → RS-14	−.382**	−.308**
CF → RSES	−.288*	—
BO → CS	−.565**	−.183*
BO → RS-14	−.616**	−.617**
BO → RSES	−.634**	—
CS → RS-14	.472**	.420**
CS → RSES	.598**	—
RS-14 → RSES	.657**	—

Note. ProQOL = Professional Quality of Life Scale; RS-14 = 14-Item Resilience Scale; RSES = Response to Stressful Experiences Scale; CF = compassion fatigue; BO = burnout; CS = compassion satisfaction.

p < .05. **p < .01.

Analysis also revealed a strong negative correlation between BO and resilience (r = –.62, p = .00, r² = .38), as well as between BO and CS (r = –.57, p = .00, r² = .32). Like the Burnett and Wahl’s (2015) study, the present study also found 38% of the variance in resilience was accounted for by BO. However, the strong significant inverse association found between BO and CS in the current study, was contrary to Burnett and Wahl (2015) who reported the same direction but a weak magnitude.

With regard to resilience as measured by the RSES (see Table 2), the results indicated a small significant negative correlation between CF and resilience (r = –.29, p = .02, r² = .08), a large significant negative correlation between BO and resilience (r = –.63, p = .00, r² = .40), and a large significant positive correlation between CS and resilience (r = .60, p = .00, r² = .36). These results would suggest that the scores on the RS-14 and RSES are comparable in measuring resilience behaviors among these variables. Scores on RSES also demonstrated good convergent validity by correlating strongly with high scores on the RS-14 (r = .66, p = .00, r² = .44).

Mediation Analysis

The relationship between CF and BO was mediated by resilience (see Table 3). Figure 1 generally depicts the mediation analysis for both, the RS-14 and RSES. For the RS-14, CF (r = –.38, p < .01) and BO (r = –.62, p < .01) both correlated with resilience, but only moderately mediated the relationship between CF and BO. Basically, 32% of the total effect of CF on BO goes through resilience. Resilience accounted for 51% of the variance in BO (R² = .512, 95% CI = [0.314, 0.659]); however, the mediation effect size (κ² = .191; bootstrapping 95% CI = [0.083, 0.344]) was medium to large. These results were similar to Burnett and Wahl’s (2015), who reported 33% of the total effect of CF on BO goes through resilience, with resilience accounting for 52% of the variance in BO and a medium to large effect size (see Table 3). Resilience as measure by the RSES in the current study also indicated similar results as the RS-14 (see Table 4). For CF (r = –.29, p < .05) and BO (r = –.63, p < .01), both correlated with resilience, but only moderately mediated the relationship between CF and BO. Approximately 27% of the total effect of CF on BO goes through resilience, with resilience accounting for 56% of the variance in BO (R² = .560, 95% CI = [0.368, 0.695]) and a medium to large effect size (ĸ² = .208; bootstrapping 95% CI = [0.033, 0.340]). Hence, the results of the current study appear to support Burnett and Wahl (2015), who showed that DBH responder participants (CISM responders in the present study) who experience a high level of compassion fatigue are more likely to report greater burnout. By detailing the involvement of resilience can help to explain this relationship. CISM and DBH responders who reported lower levels of compassion fatigue reported greater levels of resilience and in turn, resilient responders reported lower burnout levels.

Table 3.

Mediation Comparisons of Compassion Fatigue and Burnout by Resilience (RS-14).

RS-14	Present study			Burnett and Wahl (2015)
RS-14	B	t	p	B	t	p
a	−0.436	−3.913	<.001	−0.792	−3.393	<.001
b	−0.405	−8.540	<.001	−0.231	−4.885	<.001
c	0.533	6.990	<.001	0.565	5.526	<.001
c’	0.356	5.477	<.001	0.382	4.011	<.001
	Statistic	CI		Statistic	CI
R ²	.522	[0.393, 0.626]		.512	[0.314, 0.659]
κ²	.191	[0.092, 0.274]		.208	[0.083, 0.344]

Note. Preacher and Kelley (2011) defines benchmark κ² small, medium, and large effect sizes as .01, .09, and .25, respectively. RS-14 = 14-Item Resilience Scale; CI = confidence interval.

Figure 1.

Mediation analysis model for the relationship between compassion fatigue and burnout as mediated by resilience in Tables 3 and 4.

Table 4.

Mediation of Compassion Fatigue and Burnout by Resilience (RSES).

RSES	B	t	p
a	−0.607	2.460	<.01
b	−0.243	−5.898	<.001
c	0.547	5.446	<.001
c’	0.400	4.701	<.001
	Statistic	CI
R ²	.560	[0.368, 0.695]
κ²	.208	[0.033, 0.340]

Note. Preacher and Kelley (2011) defines benchmark κ² small, medium, and large effect sizes as .01, .09, and .25, respectively. RSES = Response to Stressful Experiences Scale; CI = confidence interval.

Discussion

The present study reexamined the relationship between CF, burnout, CS, and resilience but among a sample of CISM-trained responders. In essence, the results of this study had similar findings as the previous research of Burnett and Wahl’s (2015), strengthening the idea that resilience may serve as a robust mediating variable between CF and burnout.

Prevalence of CF, burnout, CS, and resilience were examined among a sample of trained CISM responders. Participant mean scores on both the CF and burnout subscales were below the ProQOL 25th percentile cut mark, which is indicative of low risk. In order words, the majority of CISM providers in this study were at minimal risk of CF and burnout. This is contrary to the original study, which found at least 19% of their participants had scores indicative of a high risk of burnout, along with all their participants scoring in the moderate to high range for CF (Burnett & Wahl, 2015). However, in an earlier study, Wee and Myers (2003) also examined incidence risk of CF and burnout among CISM providers and found 87% of their sample were at low risk of burnout, whereas 40% were in the moderate to high risk span for CF. In general, the literature is mixed on the prevalence rates of CF and burnout among health care professions that work with trauma survivors (Beck, 2011; Craig & Sprang, 2010; Newell & MacNeil, 2011; Sprang et al., 2007; Wee & Myers, 2002). Based on the results of the present study, this appears to be true also among a cross section of CISM-trained responders. Further research among a cross section of CISM-trained responders is more than likely to confirm the variability of CF and burnout rates (Cieslak et al., 2014).

Contrary to the previous study (Burnett & Wahl, 2015), the majority of participants (74%) in the current study had scores indicative of low compassion satisfaction. In other words, these participants may have problems with their job or may derive satisfaction from engaging in activities other than their job. Other studies (i.e., Craig & Sprang, 2010; Ray et al., 2013; Sprang et al., 2007) have also reported a range of CS scores among a cross section of health care providers compared with the skewed results of this study. In fact, Wee and Myers (2003) reported 89% of their CISM service providers had scores indicative of moderate to extremely high CS. One might be alarmed by this finding; however, upon examining both resilience measures, the majority of CISM participants were within the moderate to high span of scores. It is also important to note that the majority of CISM participants in this study had low CF and BO scores. One plausible explanation regarding this finding is based on the ProQOL instruction statement: “consider each of the following questions about you and your current work situation.” Participants were asked to respond to the self-report statements based on their current job-specific work situation, not how it relates to working exclusively in the trauma field requiring CISM services. Thus, the low CS scores may suggest dissatisfaction or problems at their job or a sense of fulfillment in pursuits other than their job. An equally more plausible explanation is that CS prevalence is also variable such as CF and burnout. Therefore, further research among CISM-trained service providers more likely will mimic what is observed in the general CS literature.

As in the Burnett and Wahl (2015) investigation, the current study found similar significant correlation coefficient magnitudes and directions between CF and burnout, CF and resilience, and burnout and resilience. For instance, as in the original study, 32% of the variance in CF was accounted for in the burnout scores. This result suggests that those proving CISM services to people exposed to trauma incidents are highly susceptible to developing CF symptoms, as well as burnout indicators. This finding is consistent with previous studies regarding the strong positive association between CF and burnout (Cieslak et al., 2014; Slocum-Gori et al., 2013).

As mentioned above, the present study found no significant relationship between CF and CS, but a significant inverse association between CS and burnout. This finding mimics the original study. The one exception is that the current study observed a large significant negative correlation between CS and burnout, with CS accounting 32% of the variance in burnout scores compared to 3% found by Burnett and Wahl (2015). A reasonable explanation for the disparity between the two studies may once again center on the variability of the cross section of participants.

Another important result of the present study was finding significant negative relationships between CF and resilience, as well as between burnout and resilience that were similar to Burnett and Wahl (2015). Again, results indicated that resilience was strongly correlated with lower levels of CF and burnout, with burnout exhibiting a stronger relationship for both RS-14 and RSES resilience measures (38% and 40% of the variance vs. 15% and 8% of the variance, respectively). Previous research by Cooke et al. (2013) also observed similar results that associated higher resilience with lower burnout. Therefore, the present study’s results would suggest with a high degree of confidence that one’s ability to adapt to adverse experiences is connected with the level of burnout and CF symptoms, especially as it pertains to trained CISM responders.

As in the original Burnett and Wahl (2015) study, mediation analysis was used to examine the degree of overlapping variance among, resilience, CF, and burnout to determine a causal link. Similar to the original study, the results of this study’s analysis indicated that resilience moderately mediated the relationship of CF to burnout. More specifically, resilience explained a significant amount of the relationship between these two variables. This was true for both resilience measures (RS-14 and RSES), which seem to incorporate elements of both, reactive and proactive resilience factors (Everly, 2017). In other words, resilience seems to act as a “resistance/immunity and ability to rebound from adversity buffer” between CF and burnout among trained CISM responders. One critical argument that can be made based on these findings is the need for further resilience research that empirically identifies key effective resilience pathways that trauma responders can practice to help mitigate and reduce CF, burnout, and other negative consequences associated with indirect exposure from caring for trauma survivors. For instance, Ludick and Figley (2016) suggests a salutogenic “compassion fatigue resilience sector” model that is dependent on developing the positive pathways of self-care, detachment from the suffering of those receiving services, a sense of satisfaction from helping trauma survivors, and social support.

There were several limitations with the current study. First, the participants in the study were representative of a cross section of professional disciplines (i.e., law enforcement, fire services, emergency medical services, social work, psychology, nursing, and chaplaincy) who were all trained and respond as CISM service providers. However, the increase in disasters, terrorist incidents, and other trauma events affecting those exposed to them is placing a surge in the demand for mental health services that cannot be met with traditionally trained mental health providers (Raphael, 1986). In other words, the need for non–mental health professionals to receive training in effective DBH crisis response interventions (i.e., CISM) is great. Therefore, the majority of DBH response teams will more likely be comprised of a cross section of disciplines beyond just mental health workers. Another limitation is that this study was conducted by the same principle investigator, even though a specific population was utilized. Unfortunately, this may limit the power of the results obtained. A final limitation was that this study involved a convenience sample from attendees attending a CISM training conference in Michigan and utilized self-report measures that may have had an effect on the reliability of their responses. Therefore, the results of this study may not necessarily generalize to all CISM-trained service providers.

Implications

A primary goal of this study was to explore whether similar patterns of CF, burnout, resilience, and CS would be observed among DBH responders specifically trained to provide CISM services based on the findings of Burnett and Wahl (2015). In comparison with Burnett and Wahl, the results of the present study replicated support for resilience as a crucial salutogenic factor supporting the well-being of responders providing DBH services. Therefore, the present study has implications for advancing disaster mental health research.

One major area of advancing disaster mental health research is replicability. Replicability in psychology and behavioral science research is an important aspect of any empirical science (Asendorpf et al., 2013; Nosek, Spies, & Motyl, 2012; Schmidt, 2009) and when replications are well conducted, their results can help to increase confidence in the reliability of the predicted outcome (Brandt et al., 2014). Makel, Plucker, and Hegarty’s (2012) overview of published replications in psychological research since 1990 found a replication rate of 1.07%, with the majority reporting similar findings to their original studies. Brandt et al. (2014) provided researchers with a “recipe” for conducting a close replication study, which included defining the effects and methods they intend to replicate, following the exact methodology of the original study, having sufficient statistical power, making available complete details about the replication process, and evaluating and comparing replication results critically with the results of the original study. Furthermore, replications tend to be more convincing if they are conducted by researchers independent from the original study (Brandt et al., 2014; Makel, Plucker, & Hegarty, 2012).

Unfortunately, disaster research in general is motivated by a sense of urgency, which is different from research conducted in other fields of study. Oftentimes, this type of research is not so kind in the areas of replicability requirements, experimental design, or scientific rigor, as well as how to address legitimate ethical concerns that arise (Norris, Galea, Friedman, & Watson, 2006). However, according to Norris and her colleagues, disaster mental health research is important for two reasons: (a) exploring the effects of disaster events on mental health and related constructs and (b) exploring the effectiveness of interventions implemented that help to prevent or reduce disaster-related mental health problems. For these reasons, research such as Burnett and Wahl’s (2015) and the present study is vital to the disaster mental health literature and practice.

Based on the results of this study, it is recommended that an independent researcher(s) replicate and expand on this study (especially as it relates to specific resilience practice variables) using another sample of CISM-trained and experienced service providers. This may greatly substantiate the validity of the current and original study findings, as well as offer practical approaches that can be incorporated to build resiliency among this population. It is also recommended that replication of this study occur among an international sample of CISM-trained service providers as well. Conducting further research in these areas will also provide stronger empirical evidence and literature that is more representative of the multidisciplinary composition of the DBH responders who are exposed indirectly to trauma events, thereby helping to broaden the scope of the disaster mental health literature.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biography

Harvey J. Burnett is an associate professor of Psychology and Chair of the Behavioral Sciences Department at Andrews University. He is also a police sergeant with the Buchanan Police Department and has over 20 years of law enforcement and emergency preparedness experience. Burnett is a fully licensed psychologist in the state of Michigan and serves as the Clinical Director for the Berrien County Critical Inciden Stress Management Team.

References

Agaibi

C. E.

Wilson

J. P.

(2005). Trauma, PTSD, and resilience. Trauma, Violence, & Abuse, 6, 195-216.

Ahern

R. N.

Kiehl

M. E.

Sole

L. M.

Byers

(2006). A review of instruments measuring resilience. Issues in Comprehensive Pediatric Nursing, 29, 103-125.

Asendorpf

J. B.

Conner

De Fruyt

De Houwer

Denissen

Fiedler

. . .Wicherts

J. M.

(2013). Recommendations for increasing replicability in psychology. European Journal of Personality, 27, 108-119.

Atkins

C. D.

Burnett

H. J.

(2016). Specialized disaster behavioral health training: Its connection with response, practice, trauma health, and resilience. Disaster Health, 3, 57-65.

Beck

C. T.

(2011). Secondary traumatic stress in nurses: A systematic review. Archives of Psychiatric Nursing, 25, 1-10.

Bercier

M. L.

Maynard

B. R.

(2015). Interventions for secondary traumatic stress with mental health workers: A systematic review. Research on Social Work Practice, 25, 81-89.

Bonanno

G. A.

(2004). Loss, trauma, and human resilience: Have we underestimated the human capacity to thrive after extremely aversive events? American Psychologist, 59, 20-28.

Bonanno

G. A.

(2005). Resilience in the face of potential trauma. Current Directions in Psychological Science, 14, 135-138.

Bonanno

G. A.

Galea

Bucciarelli

Vlahov

(2006). Psychological resilience after disaster: New York City in the aftermath of the September 11th terrorist attack. Psychological Science, 17, 181-186.

10.

Brandt

M. J.

Ijzerman

Dijksterhuis

Farah

F. J.

Geller

Giner-Sorlla

. . .Veer

(2014). The replication recipe: What makes for a convincing replication? Journal of Experimental Social Psychology, 50, 217-224.

11.

Brill

P. L.

(1984). The need for an operational definition of burnout. Family and Community Health, 6(4), 12-24.

12.

Burnett

H. J.

Wahl

(2015). The compassion fatigue and resilience connection: A survey of resilience, compassion fatigue, burnout, and compassion satisfaction among trauma responders. International Journal of Emergency Mental Health and Human Resilience, 17, 318-326.

13.

Cieslak

Anderson

Bock

Moore

B. A.

Peterson

A. L.

Benight

C. C.

(2013). Secondary traumatic stress among mental health providers working with the military. The Journal of Nervous and Mental Disease, 201, 917-925.

14.

Cieslak

Shoji

Douglas

Melville

Luszczynska

Benight

C. C.

(2014). A meta-analysis of the relationship between job burnout and secondary traumatic stress among workers with indirect exposure to trauma. Psychological Services, 11, 75-86.

15.

Cooke

Doust

J. A.

Steele

M. C.

(2013). A survey of resilience, burnout, and tolerance of uncertainty in Australian general practice registrars. BMC Medical Education, 13, Article 2, 1-6.

16.

Craig

C. D.

Sprang

(2010). Compassion satisfaction, compassion fatigue, and burnout in a national sample of trauma treatment therapists. Anxiety, Stress, & Coping, 23, 319-339.

17.

Everly

G. S.

(2012). Fostering human resilience: A primer on resilient leadership, psychological first aid, psychological body armor and critical incident stress management (2nd ed.). Ellicott City, MD: Chevron.

18.

Everly

G. S.

(2017). Psychological body armor: Lessons learned from neuroscience that can save your career, your marriage, and your life. Ellicot City, MD: Crisis Intervention & Critical Incident Stress Management Resources.

19.

Everly

G. S.

Lating

J. M.

(2013). Resilience: The final frontier. In Everly

G. S.

Lating

J. M.

(Eds.), A clinical guide to the treatment of the human stress response (3rd ed., pp. 143-154). New York, NY: Springer.

20.

Everly

G. S.

Mitchell

J. T.

(1999). Critical Incident Stress Management (CISM): A new era and standard of care in crisis intervention. Ellicott City, MD: Chevron.

21.

Everly

G. S.

Welzant

Jacobson

J. M.

(2008). Resistance and resilience: The final frontier in traumatic stress management. International Journal of Emergency Mental Health, 10, 261-270.

22.

Figley

C. R.

(1995). Compassion fatigue as secondary traumatic stress disorder: An overview. In Figley

(Ed.), Compassion fatigue: Coping with secondary traumatic stress disorder in those who treat the traumatized (pp. 1-20). New York, NY: Routledge.

23.

Figley

C. R.

Kleber

R. J.

(1995). Beyond the “Victim”: Secondary traumatic stress. In Kleber

Figley

Gersons

(Eds.), Beyond trauma: Cultural and societal dynamics (pp. 75-98). New York, NY: Plenum Press.

24.

Flannery

R. B.

(1998). The assaulted staff action program. Ellicott City, MD: Chevron.

25.

Johnson

D. C.

Polusny

M. A.

Erbes

C. R.

King

Litz

B. T.

. . .Southwick

S. M.

(2011). Development and initial validation of the Response to Stressful Experiences Scale. Military Medicine, 176, 161-169.

26.

Kaminsky

M. J.

McCabe

O. L.

Langlieb

Everly

G. S.

(2007). An evidence-informed model of human resistance, resilience, & recovery: The Johns Hopkins’ outcomes-driven paradigm for disaster mental health services. Brief Therapy and Crisis Intervention, 7, 1-11.

27.

Kelley

Lai

(2012). MBESS (Version 3.3.3 or higher) [Computer software and manual]. Available from http://www.cran.r-project.org/

28.

Ludick

Figley

C. M.

(2016). Toward a mechanism for secondary trauma induction and reduction: Reimagining a theory of secondary traumatic stress. Traumatology, 23(1), 1-12.

29.

Luthar

S. S.

Cicchetti

Becker

(2000). The construct of resilience: A critical evaluation and guidelines for future work. Child Development, 71, 543-562.

30.

Makel

M. C.

Plucker

J. A.

Hegarty

(2012). Replications in psychology research: How often do they really occur? Perspectives on Psychological Science, 7, 537-542.

31.

Mancini

A. D.

Bonanno

G. A.

(2009). Predictors and parameters of resilience to loss: Toward an individual differences model. Journal of Personality, 77, 1805-1831.

32.

Maslach

Jackson

S. E.

(1984). Burnout in organizational settings. In Oskamp

(Ed.), Applied social psychology annual (Vol. 5, pp. 133-153). Beverly Hills, CA: Sage.

33.

Maslach

Schaufeli

W. B.

Leiter

M. P.

(2001). Job burnout. Annual Review of Psychology, 52, 397-422.

34.

McCann

I. L.

Pearlman

L. A.

(1990). Vicarious traumatization: A contextual model for understanding the effects of trauma on helpers. Journal of Traumatic Stress, 3, 131-149.

35.

McNally

R. J.

Bryant

R. A.

Ehlers

(2003). Does early psychological intervention promote recovery from posttraumatic stress? Psychological Science in the Public Interest, 4, 45-79.

36.

Mitchell

J. T.

Everly

G. S.

(2001). Critical incident stress debriefing: An operations manual for CISD, defusing and other group crisis intervention services (3rd ed.). Ellicott City, MD: Chevron.

37.

Morse

Salyers

M. P.

Rollins

A. L.

Monroe-DeVita

Pfahler

(2012). Burnout in mental health services: A review of the problem and its remediation. Administration and Policy in Mental Health, 39, 341-352.

38.

Motta

R. W.

(2008). Secondary trauma. International Journal of Emergency Mental Health, 10, 291-298.

39.

Myers

Wee

D. F.

(2005). Disaster mental health services. New York, NY: Routledge.

40.

Newell

J. M.

MacNeil

G. A.

(2011). A comparative analysis of burnout and professional quality of life in clinical mental health providers and health care administrators. Journal of Workplace Behavioral Health, 26, 25-43.

41.

Norris

F. H.

Galea

Friedman

M. J.

Watson

P. J.

(Eds.). (2006). Methods for disaster mental health research. New York, NY: Guilford Press.

42.

North

C. S.

Pfefferbaum

(2013). Mental health response to community disasters: A systematic review. JAMA: Journal of the American Medical Association, 310, 507-518.

43.

Nosek

B. A.

Spies

J. R.

Motyl

(2012). Scientific Utopia: II. Restructuring incentives and practices to promote truth over publishability. Perspectives on Psychological Science, 7, 615-631.

44.

Pack

M. J.

(2012). Critical incident stress management: A review of the literature with implications for social work. International Social Work, 56, 608-627.

45.

Palm

K. M.

Polusny

M. A.

Follette

V. M.

(2004). Vicarious traumatization: Potential hazards and interventions for disaster and trauma workers. Prehospital and Disaster Medicine, 19(1), 73-78.

46.

Pearlman

L. A.

Mac Ian

P. S.

(1995). Vicarious traumatization: An empirical study on the effects of trauma work on trauma therapists. Professional Psychology: Research and Practice, 26, 558-565.

47.

Pearlman

L. A.

Saakvitne

K. W.

(1995). Treating therapist with vicarious traumatization and secondary traumatic disorders. In Figley

C. R.

(Ed.), Compassion fatigue: Coping with secondary traumatic stress disorder in those who treat the traumatized (pp. 150-177). New York, NY: Routledge.

48.

Phelps

Lloyd

Creamer

Forbes

(2009). Caring for carers in the aftermath of trauma. Journal of Aggression, Maltreatment & Trauma, 18, 313-330.

49.

Pines

A. M.

Aronson

(1988). Career burnout: Causes and cures. New York, NY: Free Press.

50.

Preacher

K. J.

Kelley

(2011). Effect size measures for mediation models: Quantitative strategies for communicating indirect effects. Psychological Methods, 16(2), 93-115.

51.

Raphael

(1986). When disaster strikes: How individuals and communities cope with catastrophe. New York, NY: Basic Books.

52.

Ray

S. L.

Wong

White

Heaslip

(2013). Compassion satisfaction, compassion fatigue, work life conditions, and burnout among frontline mental health care professionals. Traumatology, 19, 255-267.

53.

Roberts

A. R.

(2002). Assessment, crisis intervention, and trauma treatment: The integrative ACT intervention model. Brief Treatment and Crisis Intervention, 2, 1-21.

54.

Roberts

A. R.

Everly

G. S.

(2006). A meta-analysis of 36 crisis intervention studies. Brief Treatment and Crisis Intervention, 6, 10-21.

55.

Schiraldi

G. R.

Jackson

T. K.

Brown

S. L.

Jordan

J. B.

(2010). Resilience training for functioning adults: Program description and preliminary findings from a pilot investigation. International Journal of Emergency Mental Health, 12, 117-129.

56.

Schmidt

(2009). Shall we really do it again? The powerful concept of replication is neglected in the social sciences. Review of General Psychology, 13, 90-100.

57.

Slocum-Gori

Hemsworth

Chan

W. W.

Carson

Kazanjian

(2013). Understanding compassion satisfaction, compassion fatigue and burnout: A survey of the hospice palliative care workforce. Palliative Medicine, 27, 172-178.

58.

Sprang

Clark

J. J.

Whitt-Woosley

(2007). Compassion fatigue, compassion satisfaction, and burnout: Factors impacting a professional’s quality of life. Journal of Loss and Trauma, 12, 259-280.

59.

Stamm

B. H.

(2002). Measuring compassion satisfaction as well as fatigue: Developmental history of the compassion satisfaction and fatigue test. In Figley

C. R.

(Ed.), Treating compassion fatigue. New York, NY: Brunner-Routledge.

60.

Stamm

B. H.

(2010). The concise ProQOL manual (2nd ed.). Pocatello, ID: ProQOL.org.

61.

Thomas

J. T.

(2012). Does personal distress mediate the effects of mindfulness on professional quality of life? Advances in Social Work, 13, 561-585.

62.

Vanderbilt-Adriance

Shaw

D. S.

(2008). Conceptualization and re-evaluating resilience across levels of risk, time, and domains of competence. Clinical Child and Family Psychology Review, 11, 30-58.

63.

van Emmerik

A. A.

Kamphuis

J. H.

Hulsbosch

A. M.

Emmelkamp

P. M

. (2002). Single session debriefing after psychological trauma: A meta-analysis. The Lancet, 360, 766-771.

64.

Wagnild

G. M.

Young

(1993). Development and psychometric evaluation of the resilience scale. Journal of Nursing Measurement, 1(2), 165-178.

65.

Wagnild

G. M.

(2009). The Resilience Scale user’s guide. Worden, MT: Resilience Center.

66.

Wee

D. F.

Myers

(2002). Stress response of mental health workers following disaster: The Oklahoma City bombing. In Figley

C. F.

(Ed.), Treating compassion fatigue. Psychosocial stress series, no. 24 (pp. 57-83).

67.

Wee

Myers

(2003). Compassion satisfaction, compassion fatigue, and critical incident stress management. International Journal of Emergency Mental Health, 5(1), 33-37.

68.

Windle

(2011). What is resilience? A review and concept analysis. Reviews in Clinical Gerontology, 21, 152-169.

69.

Yassen

(1995). Preventing secondary traumatic stress disorder. In Figley

C. R.

(Ed.), Compassion fatigue: Coping with secondary traumatic stress disorder in those who treat the traumatized (pp. 178-208). New York, NY: Routledge.