Discussion II

Crimes Against Caring: Exploring the Risk of Secondary Traumatic Stress, Burnout, and Compassion Satisfaction Among Child Exploitation Investigators

Patrick Q. Brady1

Published online: 9 December 2016 # Society for Police and Criminal Psychology 2016

Abstract Secondary traumatic stress (STS) and burnout are debilitating occupational hazards that inhibit helping profes- sional’s overall well-being. Much of the extant scholarship on this topic has focused on mental health and child welfare workers and not law enforcement officials who investigate Internet child exploitation. This study used data from 433 Internet Crimes Against Children (ICAC) Task Force person- nel to explore the impact of individual and work-related fac- tors associated with the risk of STS, burnout, and compassion satisfaction. Findings indicated that nearly one in four ICAC personnel exhibited low compassion satisfaction and high levels of STS and burnout. Individual-level protective factors for increasing compassion satisfaction and mitigating symp- toms of STS and burnout included having a strong social support system outside of work and the frequent use of posi- tive coping mechanisms. Work-related risk factors such as frequent indirect exposure to disturbing materials, low orga- nizational support, and frequently feeling overwhelmed at work were all associated with higher STS and burnout and lower levels of compassion satisfaction. Policy implications and future avenues of research are discussed.

Keywords Secondary traumatic stress . Burnout .

Compassion satisfaction . Child exploitation . Criminal investigation

Introduction

Investigating crimes against children is an emotionally haz- ardous occupation that can provoke negative outcomes for professionals and their organizations (Burns et al. 2008; Sprang et al. 2011). Every day, investigators are regularly exposed, both directly and indirectly, to graphic materials in- cluding stories, images, and audio/video files depicting crimes against children. For example, among a national sample of 1713 offenders arrested for child pornography, 40% of of- fenders had images of children between the ages of 3 and 6, 80% had images portraying sexual penetration of a child, and one in five (21%) possessed images and videos graphically depicting sexual violence against a child through bondage, rape, and torture (Wolak et al. 2005). As a result of their exposure to graphic materials involving children, studies have found that child maltreatment investigators (CMIs) are at an increased risk for experiencing secondary traumatic stress (STS) and burnout (Bourke and Craun 2014; Bride et al. 2007; Burns et al. 2008; Conrad and Kellar-Guenther 2006; Perez et al. 2010; Perron and Hiltz 2006; Sprang et al. 2011). This is particularly problematic as STS and burnout have been linked to a host of negative health and occupational outcomes including general anxiety, depression, high rates of absentee- ism, and turnover (Bourke and Craun 2014; Pryce et al. 2007; Travis et al. 2015).

Secondary traumatic stress describes the problematic emo- tional and behavioral reactions that manifest from indirect exposures to traumatic experiences (Figley 1995). The fifth edition of the Diagnostic and Statistical Manual of Mental

Patrick Q. Brady is currently a doctoral candidate in the Department of Criminal Justice and Criminology at Sam Houston State University, and will be an Assistant Professor in the Department of Criminology at the University of West Georgia in August 2017. His research interests focus on campus crime, burnout, and decision-making practices among inves- tigators of interpersonal violence. His work has appeared in the Journal of Interpersonal Violence, the Journal of Contemporary Criminal Justice, the Journal of School Violence, and the Journal of Criminal Justice Education.

* Patrick Q. Brady [email protected]

1 Department of Criminology, University of West Georgia, Pafford Social Science Building, Carrollton, GA 30118, USA

J Police Crim Psych (2017) 32:305–318 DOI 10.1007/s11896-016-9223-8

Disorders (DSM-5) now identifies STS as a form of PTSD that manifests through indirect exposures to traumatized pop- ulations as the result of work-related experiences (American Psychiatric Association (APA) 2013). Moreover, helping pro- fessionals are at an increased risk of developing symptoms of STS, including intrusive thoughts, avoidance, and persistent arousal that could negatively impact one’s capacity to effec- tively accomplish their responsibilities (Bride et al. 2004; Figley 1995).

The extant scholarship on stressors among law enforce- ment has primarily focused on front-line officers (Gershon et al. 2009), with a relative dearth of information identifying manifestations of investigator stress, particularly among in- vestigators in specialized units (Holt and Blevins 2011). More recently, scholars have focused their attention on explor- ing stressors among Internet Crimes Against Children Task Force (ICAC) personnel (Burns et al. 2008; Bourke and Craun 2014; Craun and Bourke 2014; Perez et al. 2010; Powell et al. 2013, 2014, 2015). This specialized unit of in- vestigators investigates, prevents, and helps prosecute of- fenders who use the Internet to exploit children. To date, how- ever, the extant scholarship among ICAC personnel has fo- cused primarily on the factors that increase the risk of STS (Bourke and Craun 2014; Perez et al. 2010), which encom- passes only one adverse outcome of the profession (Stamm 2010). Additionally, less is known about the positive aspects of child maltreatment investigations that help investigators derive a sense of pleasure and satisfaction from their work. Collectively, little is known about the individual and work- related factors associated with both the negative (e.g., burn- out) and positive facets (e.g., compassion satisfaction) of ICAC investigations.

Thus, the next logical step is to identify the key factors associated with an increased risk for adverse outcomes, while isolating important facets of the profession that can help orga- nizations promote resiliency and retention. The current study expands the literature by exploring the individual, operational, and organizational factors influencing the professional quality of life among ICAC Task Force personnel. The study begins with an overview of the conceptual framework, followed by a review of the existing literature on risk and protective factors for STS, burnout, and compassion satisfaction. Finally, find- ings, policy implications, and future avenues of research are discussed.

Conceptual Framework

The present analysis is guided by the Professional Quality of Life (ProQOL) perspective, which suggests that the perceived quality one feels as a helping professional is largely shaped by the positive (e.g., compassion satisfaction) and negative as- pects (e.g., STS and burnout) of one’s professional

responsibilities (Stamm2010).Moreover, the ProQOL is com- prised of three related constructs—compassion satisfaction, STS, and burnout—which manifest through the juxtaposition of helping professional’s individual, operational (e.g., expo- sure to trauma), and organizational characteristics. This sec- tion discusses conceptualizations, prevalence estimates, and consequences for each construct.

Secondary Traumatic Stress

Figley (1995, p. 1) coined the term secondary traumatic stress to describe the Bcost of caring^ for others in emotional dis- tress. Secondary traumatic stress, also known as compassion fatigue, encompasses the emotional consequences of working with traumatized populations and is characterized as Ba state of tension and preoccupation with traumatized patients by re- experiencing the traumatic events, avoidance/numbing of re- minders and persistent arousal associated with the patient^ (Figley 2002, p. 1435). While much of the research on STS has been conducted with mental health professionals (Baum et al. 2014), symptoms of STS have been found among inter- personal violence/sexual assault advocates (Baird and Jenkins 2003), social workers (Bride et al. 2007), nurses (Dominguez- Gomez and Rutledge 2009), forensic interviewers (Perron and Hiltz 2006), and Internet Crimes Against Children Task Force personnel (Bourke and Craun 2014). Prevalence estimates for STS range from 15.2% to 50% among CMIs (Bride et al. 2007; Bride et al. 2004, 2007; Conrad and Kellar-Guenther 2006), with previous studies identifying CMIs as having the greatest risk for STS compared to other helping professions (Sprang et al. 2011).

Using a sample of 600 US ICAC Task Force personnel (ICAC personnel), Bourke and Craun (2014) found that while most investigators exhibited low to moderate levels of STS, 25% experienced high to severe STS. Moreover, previous studies have linked STS to negative personal and professional outcomes including family/marital problems (Craun et al. 2015), lower job satisfaction, turnover intentions, overprotec- tiveness of loved ones, having a general distrust of the world (Bourke and Craun 2014; Perez et al. 2010), and burnout (Perez et al. 2010; Stamm 2010).

Burnout

Burnout is a chronic form of psychological strain that is in- creasingly common among professionals who have high levels of emotional interactions with the public (Maslach and Jackson 1981; Maslach et al. 2001). Burnout encom- passes three principle components: emotional exhaustion, de- personalization, and reduced personal accomplishment (Maslach and Jackson 1981). Emotional exhaustion is the core component of burnout and is characteristic of a depletion of emotional energy and empathy that results from stress and

306 J Police Crim Psych (2017) 32:305–318

inundation of personal and work-related demands (Maslach and Jackson 1981). Depersonalization refers to feeling de- tached, having a decreased interest in one’s work, and/or exhibiting negative or cynical attitudes towards the work or others (Maslach and Jackson 1981). Reduced personal accom- plishment occurs when individuals feel like they are not mak- ing a difference or that their work-related efforts are ineffec- tive (Maslach and Jackson 1981).

Burnout is commonplace among a wide-range of helping professionals who experience prolonged exposures to stress- ful work demands (Leiter and Schaufeli 1996; Maslach et al. 2001). Research has found prevalence estimates for burnout ranging from 39% to 75% among social workers (Siebert 2006), between 40% and 62% among CMIs (Anderson 2000; Conrad and Kellar-Guenther 2006), and 43% to 54% among ICAC personnel (Perez et al. 2010). After nearly 40 years of empirical findings, scholarship has continued to identify burnout as an occupational hazard associated with numerous negative individual and occupational consequences (Freudenberger 1977; Maslach et al. 2001).

Individuals experiencing symptoms of burnout may suffer from lowered moral and psychological well-being (Cherniss 1980), physical and mental illnesses (Maslach et al. 2001; Salston and Figley 2003), insomnia (Armon et al. 2008), work-family conflicts (Jackson and Maslach 1982), substance abuse issues (Ahola et al. 2006), maladaptive coping mecha- nisms (Padyab et al. 2013), and suicidal ideation (Maslach 1976). Moreover, burnout has been associated with organiza- tional consequences among human service organizations in- cluding absenteeism (Travis et al. 2015), increased aggres- siveness (Queirós et al. 2013), impaired interpersonal and pro- fessional relationships with coworkers and clients (Smoktunowicz et al. 2015), reduced job performance (Maslach et al. 2001; Smoktunowicz et al. 2015), and turnover (US General Accounting Office (GAO) 2003).

Compassion Satisfaction

Compassion satisfaction (CS) describes the enjoyment and pleasure one receives from being able to accomplish their professional roles and responsibilities effectively (Stamm 2002). Research suggests that it is possible for CS to coexist among CMIs who exhibit symptoms of STS and burnout (Stamm 2010). For example, Conrad and Kellar-Guenther (2006) found that while nearly half of the 363 CMIs expressed high levels of STS, 75% of staff expressed a Bhigh^ or Bgood^ potential for CS. In a randomized national sample of 532 self- identified trauma specialists, 46% exhibited high levels of CS (Craig and Sprang 2010). Compassion satisfaction may be a key component for the retention of ICAC professionals con- sidering previous studies have found that high CS can mitigate symptoms of STS and burnout (Conrad and Kellar-Guenther 2006; Perez et al. 2010). Additionally, CS has been associated

with increase collegial cohesion and organizational commit- ment (Li et al. 2014).

Individual Risk and Protective Factors Associated with the Professional Quality of Life

While the extant scholarship has identified several important personal risk and protective factors, few studies have explored these factors among ICAC personnel. Indeed, much of the literature has focused primarily on mental health professionals (Hensel et al. 2015; Lim et al. 2010), nurses (Beck 2011), and front-line officers (Andersen and Papazoglou 2015). This sec- tion provides an overview of the key personal risk and protec- tive factors linked to STS, burnout, and CS.

Secondary Traumatic Stress

In a recent meta-analysis of 38 studies of STS among mental health professionals working with trauma victims, Hensel et al. (2015) found that having a personal trauma history and lower social support were the only two individual risk factors significantly associated with STS. Moreover, demographic characteristics such as age, sex, emotional involvement, eth- nicity, and experience did not significantly impact STS. A recent study of 600 ICAC personnel found that being female, increased use of tobacco or alcohol in the past year, and using avoidant coping mechanisms were significantly associated with higher STS scores (Bourke and Craun 2014). Additionally, having a strong social support system outside of work was the only personal protective factor associated with lower STS.

Burnout

A review of the burnout literature suggests that risk and pro- tective factors vary according to each dimension of burnout. A meta-analysis of 15 studies conducted over a 20-year period found that age was the most significant factor associated with all three burnout dimensions, indicating that mental health professionals who were younger were at a higher risk for emotional exhaustion, depersonalization, and reduced person- al accomplishment (Lim et al. 2010). Additionally, having a higher level of education was significantly associated with higher levels of emotional exhaustion and personal accom- plishment. Some scholars have speculated that females, over- all, are at a higher risk for burnout (Maslach et al. 2001), yet meta-analytic studies have found that males are at a higher risk for depersonalization (Lim et al. 2010; Purvanova and Muros 2010), while females are at a higher risk for emotional exhaus- tion (Purvanova and Muros 2010).

J Police Crim Psych (2017) 32:305–318 307

Compassion Satisfaction

There exists a dearth of information on individual-level risk and protective factors for CS. Indeed, studies have shown that individual-level correlates of CS are generally subsumed by organizational factors in multivariate analyses (Craig and Sprang 2010; Kulkarni et al. 2013). According to Perez et al. (2010), however, qualitative responses from ICAC personnel suggested that most investigators felt their efforts were mak- ing a positive difference. While Bourke and Craun (2014) did not measure CS specifically, they did find that the use of positive reframing coping mechanisms were significantly re- lated to attributes of CS, including job satisfaction and taking pride on one’s work. Furthermore, additional research on the personal factors associated with CS is warranted.

Organizational Risk and Protective Factors

Child maltreatment investigators are expected to manage work-related hazards while facing additional obstacles in the form of organizational stressors, including high workload de- mands with time-sensitive deadlines (Anderson 2000), inade- quate resources and staffing, and drastic variations in inter- agency policies and procedures (Powell et al. 2013, 2014; Wright et al. 2006). While research has established a link between organizational factors and burnout (Alarcon 2011; Swider and Zimmerman 2010), relatively little attention has been given to organizational issues associated with STS and CS (Bell et al. 2003). This section describes the organizational risk and protective factors associated with the ProQOL.

Secondary Traumatic Stress

According to Hensel et al.’s (2015) meta-analysis of risk factors for STS, organizational risk factors included low work support, heavy caseloads, more frequent encounters with trauma vic- tims, and high caseload ratio. Findings from studies exploring the operational sources of STS among ICAC personnel, how- ever, tend to differ between qualitative and quantitative studies (Burns et al. 2008; Bourke and Craun 2014; Craun and Bourke 2014; Perez et al. 2010; Powell et al. 2013, 2014, 2015; Wright et al. 2006). Qualitative studies of ICAC personnel have sug- gested that adverse reactions to ICE materials are generally short term considering that most investigators tend to adopt viewing strategies and copingmechanisms that help them avoid long-term psychological harm (Burns et al. 2008; Perez et al. 2010; Powell et al. 2014, 2015).1 In fact, recent studies note that

the extent to which investigators experience aversive reactions is contingent on the media format (e.g., live Web feeds with sound vs. still photos) and content of the material (e.g., toddlers vs. adolescent victims) (Krause 2009; Powell et al. 2015). Some qualitative studies have found that ICE investigators perceive organizational stressors (heavy caseloads and lack of resources and staffing) to be much more stressful in ICE investigations than viewing graphic materials (Powell et al. 2014, 2015).

Conversely, quantitative studies of ICAC personnel have found that having a heavy caseload volume was not a signifi- cant risk factor for STS (Perez et al. 2010). Factors such as time spent working with disturbing materials (DMs), time since first exposure to DMs, self-reported difficulty with viewing DMs, and frequency of viewing DMs were significantly associated with higher STS scores (Bourke and Craun 2014; Craun and Bourke 2014; Perez et al. 2010). Additionally, coworker and supervisor support has also been identified as an important or- ganizational protective factor against STS (Bourke and Craun 2014; Craun and Bourke 2014).

Burnout

It is well established that burnout tends to manifest from or- ganizational factors rather than exposure to critical incidents (Gershon et al. 2009; Salston and Figley 2003). A meta- analysis of 231 studies of helping professionals found that organizational risk factors such as job resources (e.g., auton- omy and participation in the decision-making process), de- mands (e.g., caseload, role conflict, and role ambiguity), and organizational perceptions (e.g., turnover intentions, job satis- faction, and organizational commitment) were significantly related to all three burnout dimensions (Alarcon 2011). Findings from other meta-analytic studies and systematic re- views not only corroborate these findings, but suggest other important organizational factors increase the risk for burnout, including perceptions of unfairness, shift work, communica- tion problems, and a disconnect between personal values and organizational goals and missions (Barak et al. 2001; Maslach et al. 2001; Paris and Hoge 2010).

In one of the only studies of burnout among ICAC person- nel, Perez et al. (2010) found that greater exposure to DMs, time since first exposure to DMs, turnover intentions, in- creased protectiveness of others, and a general distrust of the public correlated with higher emotional exhaustion and deper- sonalization scores at the bivariate level. Similar to STS, hav- ing a strong support system outside of work was significantly related to lower emotional exhaustion and increased profes- sional efficacy.

Compassion Satisfaction

Few studies have explored the organizational factors associat- ed with CS, particularly among law enforcement officers and

1 It is also important to note that much of the literature on workplace stressors among ICAC Task Force personnel has stemmed from qualitative interviews with CMIs in Australia and Canada about coping mechanisms and stress reactions as a whole rather than STS specifically (Burns et al. 2008; Powell et al. 2013, 2014, 2015; Wright et al. 2006).

308 J Police Crim Psych (2017) 32:305–318

investigators. Among a national sample of trauma therapists, Craig and Sprang (2010) found that years of experience and use of evidence-based practices were the only significant fac- tors associated with CS. Among ICAC personnel, Bourke and Craun (2014) found that lower STS scores, higher frequency of exposure to DMs, supervisor and collegial support were related to attributes of CS, such as job satisfaction, turnover intentions, and taking pride in one’s work. Furthermore, addi- tional research is needed to examine the unique risk and pro- tective factors among ICAC personnel that increase CS and ameliorate symptoms of burnout and STS.

Current Study

The current study makes several contributions to the growing scholarship on STS, burnout, and CS. First, much of the liter- ature on individual and work-related stressors among ICAC personnel stems primarily from qualitative studies with inves- tigators outside of the US (Burns et al. 2008; Powell et al. 2013, 2014, 2015; Wright et al. 2006). The current study uses data from over 400 ICAC personnel across the United States to identify and isolate the key factors that increase the risk for STS and burnout and the potential for compassion satisfac- tion. Second, previous studies of ICAC personnel have had a narrow focus regarding the negative aspects of the profession (Craun and Bourke 2014; Craun et al. 2015; Bourke and Craun 2014). As a result, little is known about burnout and CS among this population and how individual and organiza- tional factors mitigate or ameliorate risk.

Data and Methods

Data were collected via a Web-based survey over a 3-month period in the Spring of 2012. Each of the 61 ICAC Task Force Commanders were sent an e-mail that contained a recruitment letter inviting them and their task force members to participate in the study.2 The solicitation described the purpose and goals of the project and contained a link to the online survey. Task Force personnel were asked to complete the survey and for- ward the survey link to other ICACTask Force personnel. One chief commander of an ICAC Task Force in a western state agreed to serve as a contact liaison for the project by posting a link to the survey on a closed listerv used exclusively by ICAC task force personnel. In order to maximize response rates (Dillman et al. 2009), three follow-up e-mails were sent at 6-week intervals to each of the 61 taskforce commanders. Additionally, follow-up e-mails were sent to the contact

liaison asking them to post the survey link to the respective listservs. Overall, 463 respondents indicating that they were affiliated with an ICAC Task Force. Of the 463 respondents, 30 did not complete they survey, yielding a final sample of 443 ICAC personnel.3

Consistent with previous studies of ICAC Task Force per- sonnel (Bourke and Craun 2014; Craun and Bourke 2014), response rates were unable to be calculated for several rea- sons. First, given the nature of the Web-based survey and snowball sampling procedures, it was not possible to obtain accurate estimates of the number of professionals having ac- cess to the closed listserv, nor how many task force com- manders forwarded the recruitment e-mail. Second, according to a US Department of Justice report to Congress (US Department of Justice (DOJ) 2010), affiliations with ICAC Task Forces are often fluid, in that number of personnel fluc- tuates based on available funding and the volume of ICAC- related complaints. As a result, the DOJ does not regularly monitor the total number of trained personnel with each task force and instead, documents the number of full-time equiva- lents (FTEs) working ICAC-related cases in fiscal year 2009 (~1660; US Department of Justice (DOJ) 2010). Furthermore, the number of FTEs would be inappropriate to calculate an accurate response rate, as multiple investigators could make up one FTE (Craun and Bourke 2014). While this sampling method limited the ability to calculate precise response esti- mates, respondents were asked to indicate the state in they were currently employed. Moreover, all but five US States were reported,4 suggesting that the data represents at least one ICAC affiliate from 56 out of the 61 organized task forces.

Dependent Variables

The present analysis focused on three primary outcomes: STS, burnout, and compassion satisfaction. The dependent vari- ables were measured using the fifth and most recent version of The Professional Quality of Life Scale (ProQOL-V) (Stamm 2010). The ProQOL-V is a 30-item self-report mea- sure consisting of three, ten-item subscales that assesses indi- vidual outcomes for the positive (e.g., compassion satisfaction) and negative aspects (e.g., STS and burnout) of working with traumatized populations. Using a 5-point Likert-type scale, respondents were asked to indicate how often (1 = never to 5 = very often) they experienced each item in the past 30 days. According to scoring procedures outlined in Stamm (2010), subscales are summed and the ProQOL-V scores are reported in a standardized format, in which the raw scores are trans- formed into a t-score. Standardization of the scores allows for

2 Contact information for ICAC task force commanders was obtained via https://www.icactaskforce.org/ Pages/TaskForceContactInfo.aspx

3 Bivariate analyses indicated no significant differences for any of the vari- ables between respondents who completed the survey and those who partially completed surveys. 4 One state from the Midwest region and four states located in the Northeast region.

J Police Crim Psych (2017) 32:305–318 309

all three subscales to be interpreted similarly with a mean of 50 and a standard deviation of 10 for each subscale. Additionally, Stamm (2010) provides cut scores that are set at the 25th and 75th percentiles, which allow for standardizes scores to be categorized into three relative risk categories (low, moderate, and high).

Secondary Traumatic Stress

Secondary traumatic stress captures the negative outcomes associated with work-related, secondary exposures to the trau- matic experiences of others (e.g., BAs a result of my position working with crimes against children, I have intrusive, fright- ening thoughts^; Stamm 2010). Using principle components factor analysis (PCA), all ten items were forced onto a single- factor structure with nine of the ten items loading onto a single construct (Cronbach’sα = 0.853). The item BI am preoccupied with more than one person I help^, was excluded from the subscale as it failed to meet the minimum loading cutoff cri- terion of 0.4 and above (λ = 4.37; loadings range, 0.41–0.79; Field 2013). According to Stamm (2010), higher scores on this subscale indicate a greater sense of fear that is driven by work-related trauma and feelings of being overwhelmed (M = 50.0, SD = 10.0).

Burnout

Burnout refers to the negative feelings associated with emo- tional exhaustion, inadequacy, and/or ineffectiveness as a helping professional (e.g., BI feel bogged down by the system^). Higher scores on this subscale are reflective of a greater risk for burnout (M = 50.0, SD = 10.0). After forcing items onto a single-factor structure, results from the factor analysis indicated that eight out of the ten items loaded appro- priately. Two items loaded below 0.4 and were subsequently excluded from the final subscale (λ = 3.87; loadings range, 0.52–0.77; Cronbach’s α = 0.695).

Compassion Satisfaction

Compassion satisfaction encompasses the positive aspects of the helping profession and refers to the fulfillment and satis- faction one receives from being able to accomplishing their roles and responsibilities effectively (e.g., BI believe I can make a difference through my work^; Stamm 2010). Higher scores on this subscale represent a greater fulfillment and sat- isfaction with one’s ability to investigate cases of child exploi- tation effectively (M = 50.0, SD = 10.0). Using PCA, items were forced onto a single-factor solution with all ten items having a primary factor loading of 0.4 and above (λ = 5.21; loadings range, 0.59–0.82). Additionally, the results of a reli- ability analysis indicated acceptable internal consistency (Cronbach’s α = 0.894).

Independent Variables

Individual Attributes

Twelve individual-level variables explored the independent effects of investigator characteristics on one’s ProQOL. Respondents were asked to indicate their parental status (0 = no children; 1 = has children) and the total number of children they had (continuous). Respondents were also asked about their perceptions of being supported in their lives out- side of work, particularly among family and friends.Home life supportwas treated as a continuousmeasure that was captured using a single-item measured on a 5-point Likert-type scale (1 = unsupported to 5 = very supported).

A slightly modified version of Eastwood and Ecklunds’s (Eastwood and Ecklund 2008) 28-item Self-Care Practices Questionnaire was used to measure how often investigators engaged in a variety of coping mechanisms to maintain a healthy mental and physical well-being. Responses were cap- tured on a 6-point Likert-type scale ranging from never (coded 1) to everyday (coded 6). Principle components analysis with varimax rotation was conducted and revealed a three-item factor structure (λ = 1.97; loadings range, 0.41–0.79). Positive coping was comprised of 12-items that resembled the frequency in which the participants engaged in positive social and work-related coping mechanisms (e.g., getting suf- ficient sleep; taking frequent breaks at work; Cronbach’s α = 0.713). Glutinous coping is a two-item measure that de- picts the frequency in which the participants engage in un- healthy eating habits (e.g., eat ‘fast food’meals and/or eat junk or snack foods; Cronbach’s α = 0.652). Spiritual coping is a five-item measure captures the frequency in which ICAC per- sonnel engage in spiritual and/or altruistic activities (e.g., at- tend church, volunteer, meditate, and journaling; Cronbach’s α = 0.527).5

Several demographic variables were included as controls, including sex (0 = female; 1 = male), age (1 = 18–29; 2 = 30– 39; 3 = 40–49; 4 = 50 and older), race (0 = non-White; 1 = White), marital status (0 = not married; 1 = married), and highest level of education obtained (1 = high school di- ploma, 2 = associated degree, 3 = bachelor’s degree, and 4 = master’s degree or higher).6 Finally, participants were asked about their trauma histories including lifetime

5 As evidenced by the Chronbach’s α, the author acknowledges that the reli- ability for the spiritual coping construct is relatively low, which may be a function of the limited number of items in the scale (Carmines and Zeller 1979). Additionally, it is important to note that the Cronbach’s α is a rather conservative estimate of the scales reliability (Carmines and Zeller 1979). 6 Due to the lack of variation necessary to include in multivariate models, race (White (92.1%), Black or African-American (0.9%), American Indian or Alaskan Native (1.2%), Asian (0.7%), Native Hawaiian or other Pacific Islander (0.7%), and Mixed (2.8)) and marital status (single (9.5%), divorced (8.8), separated (1.6%), and widowed (0.7)) were recorded as dichotomous dummy variables.

310 J Police Crim Psych (2017) 32:305–318

experiences of interpersonal violence (IPV), sexual assault (SA), physical abuse (PA), neglect, or some other self- reported traumatic experiences.7 Trauma history was dichoto- mized where respondents who indicated experiencing one or more traumatic experience at some point in their lifetime were identified as a trauma victim (0 = no trauma; 1 = trauma).

Work-Related Factors

In addition to exploring the impact of the individual attributes of investigators on one’s ProQOL, the current analysis also included nine work-related factors. Respondents were asked to indicate their overall tenure investigating crimes against children (1 = 0–5 years; 2 = 6–10 years; 3 = 11–15 years; 4 = 16 or more years), the total number of weekly hours spent investigating crimes against children (1 = 0–10 hours per week [hrs]; 2 = 11–20 hrs; 3 = 21–30 hrs; 4 = 31 or more hours), whether they were a supervisor of other task force personnel (0 = no; 1 = yes), whether they engaged in undercover duties (e.g., posing as a child online in proactive investigations) (0 = no; 1 = yes), and the average age of the child victims investigators encounter in their work (1 = 0– 4 years old, 2 = 5–7 years, 3 = 8–12 years, 4 = 13–18). Using a single 5-point Likert-type measure, organizational support was treated as a continuous variable inquiring about how well task force personnel felt supported within their or- ganization (1 = unsupported; 5 = very supported). Respondents were also asked to report how often they felt overwhelmed from their job within the past 30 days. Feeling overwhelmed was captured using a single 6-point Likert scale ranging from never (coded 1) to very frequently (coded 6).

Direct exposure referred to physically witnessing abusive behaviors against children (e.g., viewing child pornography), while indirect exposure encompassed hearing or reading about crimes against children (e.g., conducting forensic interviews, reading case files, hearing graphic testimonies in court, etc.). Using two 5-point Likert scales (1 = never to 5 = very often), task force personnel were asked to indicate how often they were directly exposed to crimes against children, and how often they were indirectly exposed to crimes against children in their current position.

Missing Data

Several steps were taken to handle missing data. First, Hertel (1976) recommends that variables should have no more than

15% missing data. The majority of the independent variables included in the analysis were missing data on less than 1% of cases in a non-monotone pattern (Allison 2002). Most of the self-care variables had missing data values ranging from 3.1% to 6.7%. Approximately 6.5% cases had missing values for IPV, 7.9% for SA, 7.2% for PA, and 8.5% for neglect, while working undercover had 7.2%.Moreover, the amount of miss- ing data in the current study falls well below Hertel’s 15% threshold.

Findings from the Little’s Missing Completely at Random (MCAR) test revealed that there were no significant differences between the missing and non-missing values for any of the included variables (χ2(4978) = 4395.62, p = 1.00; Little 1988). Considering that the missing data would limit the size of the sample substantially after listwise deletion—which could potentially bias the findings (Allison 2002)—missing data on the dependent and independent variable were imputed in SPSS using the Markov Chain Monte Carlo method (Enders 2010). Multiple imputation (MI) is a powerful, iterative process that uses existing values of other variables to estimate multiple pre- dicted values that are subsequently substituted for the missing values (Allison 2002; Enders 2010; Rubin 1996). The MI pro- cess generates five copies of completed datasets, each with a different imputation estimate for the missing values (Rubin 1996). Regression models were estimated using each imputed dataset and were interpreted using Rubin’s (1996) recommen- dations to combine the parameter estimates and standard errors into a single pooled estimate.

Analytic Strategy

Analyses for the current study proceeded in three stages. First, univariate descriptives assessed the overall distribution of the sample. Second, bivariate analyses were carried out to identify significant relationships between individual attributes and work-related factors associated with the three dependent var- iables. Prior to the estimation of the multivariate models, mod- el diagnostics were examined to ensure that there were no issues with skewness and kurtosis, multicollinearity, normali- ty, heteroskedasticity, and outliers.8 Finally, given the contin- uous nature of the dependent variables, three ordinary least squares (OLS) regression models were estimated to identify and isolate the key factors responsible for influencing the pro- fessional quality of life among ICAC personnel. To avoid

7 BOther trauma^ was an opened-ended question that was subsequently trans- formed into a dummy variable where responses were was coded 1 if the respondent self-reported an event that they considered to be traumatic. Responses ranged from killing someone in the line of duty, being involved in a major car accident, to witnessing domestic violence between parents as a child.

8 Model diagnostics revealed no issues with skewness or kurtosis, as estimates for each dependent variable fell within the appropriate range and did not exceed thresholds of ±2.0 and 7.0, respectively (Curran et al. 1996). A review of the bivariate correlations and collinearity diagnostics indicated that multicollinearity was not an issue. The variance inflation factor values (VIF) were all well below 10 and tolerance statistics were all above 0.2 for all three dependent variables (Field 2013). Finally, model bias for normality, influential outliers, and heteroskedasticity was assessed through residual statistics and a review of residual plots (Field 2013).

J Police Crim Psych (2017) 32:305–318 311

oversaturating the models, multivariate analyses only includ- ed independent variables that were significant at the bivariate level.

Results

Sample Demographics

Descriptive statistics are presented in Table 1. Sample demo- graphics revealed that the respondents were predominantly white (92.4%), male (72.0%), and between the ages of 40– 49 (40.6%). The majority of the participants held a bachelor’s degree or higher (60.1%) and were married (79.4%), with 84.3% of respondents indicating that they were a parent with an average number of two children (M = 2.0, SD = 1.3). In relation to the working environment, most of the respondents were within their first 5 years of investigating cases involving crimes against children (47.4%). More than half of the respon- dents worked more than 31 hrs a week with a task force (52.4%), with nearly a third indicating that they were a super- visor of other task force personnel (28.6%).

Secondary Traumatic Stress

Table 2 presents the results of the three OLS regression models. Model 1 presents the impact of individual and work-related factors on the risk for STS. Findings indicated that being male and feeling supported outside of work were associated with lower levels of STS. Additionally, having a history of trauma and frequently engaging in spiritual/ altruistic coping mechanisms were significantly associated with higher levels of STS. Work-related factors, such as fre- quently feeling overwhelmed, low organizational support, working cases where the average age of the child victims is relatively young, and frequent indirect exposure to crimes against children (CAC) were associated with higher STS scores. While engaging in positive coping was marginally significant with STS, other factors such as the number of children an investigator has, working undercover, the number of weekly hours spent working ICE cases, and the frequency of direct exposure to CAC were non-significant.

Burnout

Model 2 in Table 2 presents the impact of individual and work-related factors on the risk for burnout. Findings indicat- ed that having a strong support system outside of work and frequent use of positive coping mechanisms significantly mit- igated the risk of burnout. Turning to the work-related factors, low organizational support, frequently feeling overwhelmed, and frequent indirect exposure to CAC were associated with higher levels of burnout. Respondent sex, number of children,

trauma history, spiritual coping, age of the child victim, work- ing undercover, the number of weekly hours spent working ICE cases, and frequency of direct exposure to CAC were non-significant.

Compassion Satisfaction

Model 3 presented in Table 2 assesses the impact of individual and work-related factors for compassion satisfaction. Home support and frequent use of positive and spiritual coping mechanisms were significantly associated with higher levels of compassion satisfaction. Turning to work-related factors, working undercover, feeling overwhelmed less often, feeling supported within the organization, and frequent indirect expo- sure to CAC significantly increases the potential for compas- sion satisfaction. Respondents’ sex, total number of children, trauma history, age of child victim, the number of weekly hours spent working ICE cases, and the frequency of direct exposure to CAC were non-significant.

Discussion

Scholarship has established that professionals who investigate ICAC are at a heightened risk for STS and burnout (Bourke and Craun 2014; Perez et al. 2010), which can ultimately result in ICAC Task Force programs experiencing high rates of absenteeism and turnover (Bourke and Craun 2014; Travis et al. 2015;). To date, however, much of the limited research on stress among ICAC personnel is comprised of qualitative (Burns et al. 2008; Powell et al. 2013, 2014, 2015; Wright et al. 2006), and quantitative studies that have focused primar- ily on prevalence estimates and predictors of only one nega- tive aspect of the profession—STS (Bourke and Craun 2014; Perez et al. 2010). Moreover, relatively little is known about how individual and work-related factors impact the risk for burnout and low compassion satisfaction. The current study addressed these research limitations by investigating the rela- tionship between individual and work-related factors on ICAC personnel’s ProQOL.

Overall, there are several findings from the current analysis that warrant further discussion. First, consistent with previous studies (Bourke and Craun 2014; Perez et al. 2010), the ma- jority of respondents were at a low- to moderate risk for STS (75%) and burnout (75%), with 76% reporting moderate to high levels of compassion satisfaction. Overall, this finding aligns with previous studies that have found that most ICAC personnel have positive feelings about their contributions and investigative efforts while adequately managing the routine stressors that are inherent in the nature of the profession (Powell et al. 2015; Wolak and Mitchell 2009). Despite vari- ations in STS instruments, prevalence estimates for high STS in the current study are similar to estimates established in

312 J Police Crim Psych (2017) 32:305–318

Table 1 Descriptives statistics for individual and work-related factors among ICAC personnel (N = 433)

Variables and attributes M or % (freq.)a SD Min/max α value Correlation with DV

STS BO CS

Dependent variables

Secondary traumatic stress 50.0 10.0 31.8–95.0 0.853 – 0.67* −0.26* Percent at low risk 23.7% (102.6) – – – – – –

Percent at moderate risk 51.5% (222.8) – – – – – –

Percent at high risk 24.8% (107.6) – – – – – –

Burnout 50.0 10.0 24.8–81.8 0.699 0.67* – −0.41* Percent at low risk 26.0% (112.6) – – – – – –

Percent at moderate risk 49.8% (215.4) – – – – – –

Percent at high risk 24.2% (105) – – – – – –

Compassion satisfaction 50.0 10.0 11.7–69.3 0.894 −0.26* −0.41* –

Percent at low risk 24.0% (103.8) – – – – – –

Percent at moderate risk 52.6% (227.6) – – – – – –

Percent at high risk 23.4% (101.6) – – – – – –

Individual attributes

Sex – – 0–1 – −0.23* −0.17* 0.00

Female 28.0% (121.6) – – – – – –

Male 72.0% (311.4) – – – – – –

Race – – 0–1 – −0.04 −0.05 −0.01 Non-White 7.6% (33) – – – – – –

White 92.4% (400) – – – – – –

Age – – 1–4 – −0.07 −0.09 0.03

18–29 4.4% (19) – – – – – –

30–39 38.1% (165) – – – – – –

40–49 40.6% (176) – – – – – –

50 or older 16.9% (73) – – – – – –

Education – – 1–4 – 0.10* 0.02 0.09

High School diploma 22.5% (97.4) – – – – – –

Associates degree 17.4% (75.4) – – – – – –

Bachelor’s degree 41.9% (181.2) – – – – – –

Master’s degree or higher 18.2% (79) – – – – – –

Trauma history – – 0–1 – 0.14* 0.06 −0.01 Trauma victim 24.1% (104.2) – – – – – –

Not a trauma victim 75.9% (328.8) – – – – – –

Marital status – – 0–1 – −0.09 −0.09 −0.00 Not married 20.6% (89) – – – – – –

Married 79.4% (344) – – – – – –

Parental status – – 0–1 – −0.03 −0.09 −0.02 Not a parent 15.7% (68) – – – – – –

Parent 84.3% (365) – – – – – –

Total number of children 2.0 1.3 0–6 – −0.02 −0.11* −0.01 Home life support 4.0 1.1 1–5 – −0.29* −0.33* 0.31*

Positive coping 54.9 7.5 26–70 0.713 −0.30* −0.38* 0.35*

Glutinous coping 9.1 1.9 2–12 0.652 0.04 0.00 −0.03 Spiritual coping 9.4 3.6 5–25 0.527 0.07 −0.09 0.17*

Work-related factors

Supervisor – – 0–1 – −0.01 −0.01 0.07

No 71.4% (309) – – – – – –

J Police Crim Psych (2017) 32:305–318 313

previous studies (25%; Bourke and Craun 2014; 18%; Perez et al. 2010). Indeed, nearly one in four investigators reported low levels of compassion satisfaction (24%) and high levels of STS (25%) and burnout (24%). Nevertheless, this is particu- larly problematic considering that investigators with these pa- rameters are experiencing grave social, emotional, and behav- ioral consequences from their work, such as feeling overwhelmed, unsupported, and ineffective in their investiga- tive and coping abilities (Figley 1995; Maslach et al. 2001; Stamm 2010). Furthermore, in order to reduce the risk of STS and burnout, it is important to identify factors that enhance investigators’ ProQOL and overall well-being.

Second, across all three models, having a strong support system outside of work and frequently engaging in positive coping strategies were significantly related to higher levels of CS and lower STS and burnout. This finding is congruent with previous studies of STS and burnout among ICAC personnel

(Bourke and Craun 2014; Perez et al. 2010), along with existing meta-analytic and systematic reviews of the literature on risk and protective factors for STS, burnout, and CS (Hensel et al. 2015; Lim et al. 2010; Paris and Hoge 2010). Social support and positive coping strategies can help inves- tigators establish and maintain a healthy work-life balance.

Third, consistent with the prior literature (Bourke and Craun 2014; Craun and Bourke 2014; Craun et al. 2015; Hensel et al. 2015), frequently feeling overwhelmed at work and experiencing low organizational support were significant- ly associated with lower CS and higher STS and burnout. The nature of the data did not allow for a deeper investigation as to the sources of stress or types of organizational support needed. However, qualitative studies have consistently found that high workload demands, inadequate technological resources, and lack of collegial and supervisory support (Burns et al. 2008; Krause 2009; Powell et al. 2013, 2014; Wright et al. 2006)

Table 1 (continued)

Variables and attributes M or % (freq.)a SD Min/max α value Correlation with DV

STS BO CS

Yes 28.6% (124) – – – – – –

Working undercover – – 0–1 – −0.07 0.00 0.21*

No 51.5% (223.2) – – – – – –

Yes 48.5% (209.8) – – – – – –

Feeling overwhelmed 3.7 1.3 1–6 – 0.51* 0.63* −0.31* Organizational support 3.3 1.2 1–5 – −0.35* −0.48* 0.35*

Frequency of direct exposure to CAC 3.2 1.4 1–5 – 0.08 0.09 0.13*

Frequency of indirect exposure to CAC 4.3 0.8 1–5 – 0.23* 0.21* 0.06

Tenure investigating (CAC) – – 1–4 – 0.03 0.02 0.06

0–5 years 47.4% (205.2) – – – – – –

6–10 years 32.7% (141.4) – – – – – –

11–15 years 12.7% (55) – – – – – –

16+ years 7.2% (31.4) – – – – – –

Weekly hours investigating CAC – – 1–4 – 0.11* 0.20* 0.10*

0–10 hrs 18.8% (81.2) – – – – – –

11–20 hrs 16.3% (70.8) – – – – – –

21–30 hrs 12.5% (54) – – – – – –

31+ hrs 52.4% (227) – – – – – –

Average age of child victim 1–4 – −0.18 −0.08 −0.01 0–4 years old 2.2% (9.6) – – – – – –

5–7 years old 24.3% (105.2) – – – – – –

8–12 years old 59.5% (257.4) – – – – – –

13–18 years old 14.0 (60.8) – – – – – –

Note. Categorization of cut scores for STS (low risk = scores <42; moderate risk = scores ranging from 42.01 to 55.99; high risk = scores >56); burnout (low risk ≤43; moderate risk = 43.01–55.99; high risk ≥56); and compassion satisfaction (low risk = scores <44; moderate risk = 44.01–56.99; high risk ≥57) (Stamm 2010)

CAC crimes against children

*p < 0.05 a Considering that the data were imputed multiple times, raw data are presented as pooled estimates and may contain decimals (Rubin 1996)

314 J Police Crim Psych (2017) 32:305–318

collectively contribute to higher levels of stress and lower professional efficacy.

Fourth, other significant work-related factors included the average age of the child victim, the frequency of indirect ex- posure to crimes against children, and working undercover. Consistent with previous studies (Burns et al. 2008; Powell et al. 2015), the younger the average age of the victims ICAC personnel encounter, the higher the risk for developing STS. This finding further supports the results from qualitative stud- ies that have suggested that the risk and degree of the negative impact from viewing ICEmaterial is contingent on the content of the material (e.g., toddlers vs. adolescent victims) (Krause 2009; Powell et al. 2015). Furthermore, while prior studies have found that frequent exposure to CAC increases the risk for STS (Bourke and Craun 2014; Craun and Bourke 2014; Perez et al. 2010), little is known as to whether the risk for STS is dependent on the type of exposure. Findings from the current study suggest that frequent indirect exposure plays a more important role for the risk of STS, burnout, and CS. The frequency of direct exposure may not be as relevant due to the varying viewing techniques investigators develop to decrease their risk of psychological harm via exposure to CAC (Powell et al. 2015). Qualitative studies have noted that investigators cope with viewing child pornography by taking frequent breaks, dissociating, and/or maintaining an emotional distance

by focusing on the evidence and remaining analytical (Burns et al. 2008; Powell et al. 2015).

The frequency of indirect exposure was positively associ- ated with CS, as well as STS and burnout. This may be be- cause investigators understand that indirect exposures to CAC is an expected occupational hazard and that exposure to this information might actually increase motivation to protect chil- dren while ensuring they can be resilient during the process (Powell et al. 2013). Powell et al. (2015) found that some investigators reported that exposure to CAC had a positive impact on their professional efficacy. Perhaps each new case is an additional opportunity to learn more about these crimes and to develop a better understanding of offenders. Nevertheless, future studies should continue to explore the impact of indirect vs. direct exposures on both positive and negative outcomes of the profession.

Finally, working undercover was positively associated with CS, indicating that ICAC personnel working proactive inves- tigations to capture offenders trying to contact minors via the Internet exhibit high levels of satisfaction from being able to do their job well. Perhaps proactive investigations provide ICAC personnel with a greater sense of purpose and success from interacting with online offenders knowing that they have a stronger probability of preventing a child from harm. This is consistent with Mitchell et al.’s (2005) notion that the mere

Table 2 OLS regression analysis: the impact of individual and work-related factors on investigator’s professional quality of life (N = 433)

Model 1: STS Model 2: burnout Model 3: compassion satisfaction

b SE β b SE β b SE β

Individual attributes

Sex −2.80** 0.94 −1.26 −0.72 0.84 −0.32 −0.94 1.01 −0.42 Total number of children 0.34 0.33 0.44 −0.20 0.29 −0.26 −0.43 0.35 −0.56 Trauma history 1.96* 0.99 0.84 −0.01 0.85 0.00 0.25 1.02 0.11

Home life support −1.25** 0.41 −1.31 −0.81* 0.37 −0.85 1.08** 0.44 1.14

Positive coping −0.11† 0.06 −0.79 −0.13* 0.05 −0.98 0.16** 0.07 1.17

Spiritual coping 0.24* 0.11 0.87 −0.09 0.10 −0.31 0.35** 0.12 1.26

Work-related factors

Age of child victim −1.77** 0.60 −1.19 −0.20 0.52 −0.14 −0.22 0.63 −0.15 Undercover −0.72 0.82 −0.36 0.11 0.72 0.06 3.75*** 0.88 1.88

Hours investigating CAC −0.20 0.36 −0.24 0.53 0.32 0.63 0.55 0.39 0.65

Feeling overwhelmed 2.81*** 0.34 3.57 3.65*** 0.30 4.64 −1.69*** 0.37 −2.14 Organizational support −1.11** 0.37 −1.34 −1.99*** 0.33 −2.41 1.74*** 0.40 2.12

Direct exposure to CAC 0.27 0.31 0.37 0.10 0.28 0.14 0.37 0.34 0.51

Indirect exposure to CAC 1.68** 0.55 1.29 1.34** 0.48 1.03 1.21* 0.58 0.92

Constant 50.36*** 4.75 – 47.77*** 4.22 – 26.63*** 4.65 –

F 19.60*** 33.88*** 12.99***

Adjusted R2 0.370 0.510 0.274

CAC crimes against children, STS secondary traumatic stress † p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001

J Police Crim Psych (2017) 32:305–318 315

presence of ICAC personnel posing as young children may deter online offenders trying to contact minors. Nevertheless, future studies should compare and contrast the unique stressors and successes of ICAC personnel who do and do not engage in proactive investigations.

Overall, this study is not without limitations. First, data were derived from a cross-sectional online survey, which limits the ability draw causal conclusions. Future studies should use longitudinal designs and control groups to better understand the onset, stability, and temporal order of STS and burnout overtime, while exploring the factors that increase compassion satisfaction and decrease absenteeism and turn- over. Second, some of the variables in the study were single- item, self-reported measures about investigators’ personal and work-related experiences. Future work should employ valid standardized measures that can more accurately capture the key facets of ICAC investigations that increase risk for both positive and negative outcomes of the profession. Finally, ac- cording to Cieslak et al. (2014), due to the high correlations between STS and burnout, future studies should employ sep- arate measures for each construct, such as the Secondary Traumatic Stress Scale (Bride et al. 2004) and the Maslach Burnout Inventory (Maslach et al. 1997).

Conclusions and Implications

Findings from the current study highlight important individual and work-related factors that may prevent ICAC personnel from suffering from negative emotional outcomes. Addressing factors that increase CS and reduce STS and burn- out, however, rest on both ICAC investigators and their orga- nizations. Investigators should strive to maintain a healthy work-life balance that is reliant on sustaining a strong social support system outside of work and frequent engagement in positive coping strategies. Furthermore, investigators should adopt positive coping strategies that are enjoyable and condu- cive to their personal interests and lives. Organizations bear partial responsibility for establishing a healthy and supportive work environment. Supervisors and administrators of ICAC Task Forces should work collectively with front-line investi- gators to develop policies and procedures that center around self-care flexibility and ongoing prevention trainings for STS, burnout, and CS. These trainings should be targeted towards ICAC personnel, their supervisors and administrators, and the family members and friends of ICAC personnel.

Research has suggested that individuals may be experienc- ing symptoms of STS and burnout without personally being cognizant of the fact (Lederer 2007). Thus, extending train- ings to the family members and loved ones of ICAC personnel can help educate them on how to recognize symptoms and support investigators in sustaining positive coping mecha- nisms prior to the onset of STS and burnout. Tangentially,

trainings may also increase socialization opportunities among family members and friends, which can lead to support sys- tems among those closest to the investigators. Other important implications for supervisors and administrators are to engage in regular conversations and check-ins among investigators about their overall well-being, offer ample professional devel- opment opportunities, and strongly encourage/mandate the use of employee assistance programs. Finally, it is imperative that ICAC Task Force programs continue to develop and sus- tain researcher-practitioner partnerships to explore, evaluate, and establish evidence-based policies and procedures that can help improve case outcomes, job satisfaction, and employee turnover.

All in all, it is impossible for ICAC personnel to help others unless they help themselves, first. Findings from the current study suggest that CS is possible and that STS and burnout can be prevented through establishing and maintaining strong systems of support both inside and outside of work. While the nature of the work is grueling, investigators should not have to suffer the costs of caring and be afforded the opportunity to endure a healthy and positive ProQOL while effectively preventing children from online exploitation. Caring about others should never be criminal, yet ICAC Task Force pro- grams that fail to prevent STS and burnout through prevention programming, organizational support, and self-care initiatives are only further perpetuating crimes against caring.

References

Ahola K, Honkonen T, Pirkola S, Isometsä E, Kalimo R, Nykyri E, Aromaa A, Lönnqvist J (2006) Alcohol dependence in relation to burnout among the Finnish working population. Addiction 101(10): 1438–1443

Alarcon GM (2011) A meta-analysis of burnout with job demands, re- sources, and attitudes. J Vocat Behav 79(2):549–562

Allison PD (2002) Missing data: quantitative applications in the social sciences. Br J Math Stat Psychol 55(1):193–196

American Psychiatric Association (2013) Diagnostic and statistical man- ual of mental disorders, 5th edn. Arlington, VA

Anderson DG (2000) Coping strategies and burnout among veteran child protection workers. Child Abuse Negl 24(6):839–848

Andersen J P, & Papazoglou K (2015) Compassion fatigue and compas- sion satisfaction among police officers: An understudied topic. International Journal of Emergency Mental Health and Human Resilience

Armon G, Shirom A, Shapira I, Melamed S (2008) On the nature of burnout–insomnia relationships: a prospective study of employed adults. J Psychosom Res 65(1):5–12

Baird S, Jenkins SR (2003) Vicarious traumatization, secondary traumatic stress, and burnout in sexual assault and domestic violence agency staff. Violence Vict 18(1):71–86

Barak MEM, Nissly JA, Levin A (2001) Antecedents to retention and turnover among child welfare, social work, and other human service employees: what can we learn from past research? A review and metanalysis. Soc Serv Rev 75(4):625–661

316 J Police Crim Psych (2017) 32:305–318

Baum N, Rahav G, Sharon M (2014) Heightened susceptibility to sec- ondary traumatization: a meta-analysis of gender differences. Am J Orthopsychiatry 84(2):111

Beck CT (2011) Secondary traumatic stress in nurses: a systematic re- view. Arch Psychiatr Nurs 25(1):1–10

Bell H, Kulkarni S, Dalton L (2003) Organizational prevention of vicar- ious trauma. Families in Society: The Journal of Contemporary Social Services 84(4):463–470

BourkeML, Craun SW (2014) Secondary traumatic stress among internet crimes against children task force personnel: impact, risk factors, and coping strategies. Sexual abuse: a Journal of Research and Treatment 26(6):586–609

Bride BE, Jones JL, MacMaster SA (2007) Correlates of secondary trau- matic stress in child protective services workers. Journal of Evidence-Based Social Work 4(3–4):69–80

Bride BE, RobinsonMM, Yegidis B, Figley CR (2004) Development and validation of the secondary traumatic stress scale. Res Soc Work Pract 14(1):27–35

Burns CM, Morley J, Bradshaw R, Domene J (2008) The emotional impact on and coping strategies employed by police teams investi- gating internet child exploitation. Traumatology 14(2):20

Carmines E G, & Zeller R A (1979) Reliability and validity assessment (Vol. 17). Sage publications

Cherniss C (1980) Staff burnout: job stress in the human services. Sage Publications, Beverly Hills, CA

Cieslak R, Shoji K, Douglas A, Melville E, Luszczynska A, Benight CC (2014) A meta-analysis of the relationship between job burnout and secondary traumatic stress among workers with indirect exposure to trauma. Psychol Serv 11(1):75

Conrad D, Kellar-Guenther Y (2006) Compassion fatigue, burnout, and compassion satisfaction among Colorado child protection workers. Child Abuse Negl 30(10):1071–1080

Craig CD, Sprang G (2010) Compassion satisfaction, compassion fa- tigue, and burnout in a national sample of trauma treatment thera- pists. Anxiety Stress Coping 23(3):319–339

Craun SW, Bourke ML (2014) The use of humor to cope with secondary traumatic stress. Journal of Child Sexual Abuse 23(7):840–852

Craun SW, BourkeML, Coulson FN (2015) The impact of internet crimes against children work on relationships with families and friends: an exploratory study. J Fam Violence 30(3):393–402

Curran PJ, West SG, Finch JF (1996) The robustness of test statistics to nonnormality and specification error in confirmatory factor analysis. Psychol Methods 1(1):16

Dillman DA, Smyth JD, Christian LM (2009) Internet, mail, and mixed- mode surveys. The tailored design method, 3rd edn. Wiley, Hoboken, New Jersey

Dominguez-Gomez E, Rutledge DN (2009) Prevalence of secondary traumatic stress among emergency nurses. J Emerg Nurs 35(3): 199–204

Eastwood CD, Ecklund K (2008) Compassion fatigue risk and self-care practices among residential treatment center childcare workers. Resid Treat Child Youth 25(2):103–122

Enders CK (2010) Applied missing data analysis. Guilford Press Field A (2013) Discovering statistics using IBM SPSS statistics. Sage Figley CR (1995) Compassion fatigue: toward a new understanding of

the costs of caring. In: Stamm BH (ed) Secondary traumatic stress: self-care issues for clinicians, researchers, & educators. Sidran Press, Lutherville, MD, p 328

Figley CR (2002) Compassion fatigue: psychotherapists’ chronic lack of self-care. J Clin Psychol 58:1433–1441

Freudenberger HJ (1977) Burn-out: occupational hazard of the child care worker. Child and Youth Care Forum 6(2):90–99

Gershon RRM, Xianbin L, Barocas B, Canton AN, Vlahov D (2009) Mental, physical, and behavioral outcomes associated with per- ceived work stress in police officers. Criminal Justice and Behavior 36(3):275–289

Hensel JM, Ruiz C, Finney C, Dewa CS (2015) Meta-analysis of risk factors for secondary traumatic stress in therapeutic work with trau- ma victims. J Trauma Stress 28(2):83–91

Hertel BR (1976) Minimizing error variance introduced by missing data routines in survey analysis. Sociol Methods Res 4(4):459–474

Holt TJ, Blevins KR (2011) Examining job stress and satisfaction among digital forensic examiners. Journal of Contemporary Criminal Justice 27(2):230–250

Jackson SE, Maslach C (1982) After-effects of job-related stress: families as victims. J Organ Behav 3(1):63–77

Krause M (2009) Identifying and managing stress in child pornography and child exploitation investigators. Journal of Police and Criminal Psychology 24(1):22–29

Kulkarni S, Bell H, Hartman JL, Herman-Smith RL (2013) Exploring individual and organizational factors contributing to compassion satisfaction, secondary traumatic stress, and burnout in domestic violence service providers. Journal of the Society for Social Work & Research 4(2)

Li A, Early SF, Mahrer NE, Klaristenfeld JL, Gold JI (2014) Group cohesion and organizational commitment: protective factors for nurse residents’ job satisfaction, compassion fatigue, compassion satisfaction, and burnout. J Prof Nurs 30(1):89–99

Lim N, Kim EK, Kim H, Yang E, Lee SM (2010) Individual and work- related factors influencing burnout of mental health professionals: a meta-analysis. J Employ Couns 47(2):86–96

Little RJ (1988) A test of missing completely at random for multivariate data with missing values. J Am Stat Assoc 83(404):1198–1202

Lederer, SJ (2007)Nurturing the nurturer: The importance of self-care as burnout prevention for clinical therapists. (Doctoral dissertation). Retrieved from Dissertations & Theses database. (UMI No. 3264332)

Leiter MP, Schaufeli WB (1996) Consistency of the burnout construct across occupations. Anxiety Stress Coping 9(3):229–243

Maslach C (1976) Burned-out. Human Behavior 5:16–22 Maslach C, Jackson SE (1981) Themeasurement of experienced burnout.

J Organ Behav 2(2):99–113 Maslach C, Jackson SE, Leiter MP (1997) Maslach burnout inventory.

Evaluating Stress: A Book of Resources 3:191–218 Maslach C, Schaufeli WB, Leiter MP (2001) Job burnout. Annu Rev

Psychol 52:397–422 McCann IL, Pearlman LA (1990) Vicarious traumatization: a framework

for understanding the psychological effects of working with victims. J Trauma Stress 3:131–149

Mitchell KJ, Wolak J, Finkelhor D (2005) Police posing as juveniles online to catch sex offenders: is it working? Sexual Abuse: a Journal of Research and Treatment 17(3):241–267

Padyab M, Backteman-Erlanson S, Brulin C (2013) Burnout, coping, stress of conscience and psychosocial work environment among patrolling police officers. Journal of Police and Criminal Psychology:1–9

Paris M Jr, Hoge MA (2010) Burnout in the mental health workforce: a review. The Journal of Behavioral Health Services & Research 37(4):519–528

Perez LM, Jones J, Englert DR, Sachau D (2010) Secondary traumatic stress and burnout among law enforcement investigators exposed to disturbing media images. Journal of Police and Criminal Psychology 25(2):113–124

Perron BE, Hiltz BS (2006) Burnout and secondary trauma among foren- sic interviewers of abused children. Child Adolesc Soc Work J 23(2):216–234

Powell MB, Guadagno BL, Cassematis P (2013) Workplace stressors for investigative interviewers of child-abuse victims. Policing: an International Journal of Police Strategies & Management 36(3): 512–525

Powell BM, Cassematis P, Benson S, Smallbone M, Wortley R (2014) Police officers’ perceptions of the challenges involved in internet

J Police Crim Psych (2017) 32:305–318 317

child exploitation investigation. Policing: an International Journal of Police Strategies & Management 37(3):543–557

Powell M, Cassematis P, Benson M, Smallbone S, Wortley R (2015) Police officers’ perceptions of their reactions to viewing internet child exploitation material. Journal of Police and Criminal Psychology 30(2):103–111

Purvanova RK, Muros JP (2010) Gender differences in burnout: a meta- analysis. J Vocat Behav 77(2):168–185

Pryce JG, Shackelford KK, Pryce DH (2007) Secondary traumatic stress and the child welfare professional. Lyceum Books, Inc., Chicago, IL

Queirós C, Kaiseler M, da Silva AL (2013) Burnout as predictor of aggressivity among police officers. Journal of Police Studies/ Cahiers Politiestudies 1(2)

Rubin DB (1996) Multiple imputation after 18+ years. J Am Stat Assoc 91(434):473–489

Salston M, Figley CR (2003) Secondary traumatic stress effects of work- ing with survivors of criminal victimization. J Trauma Stress 16(2): 167–174

Shoji K, Lesnierowska M, Smoktunowicz E, Bock J, Luszczynska A, Benight CC, Cieslak R (2015) What comes first, job burnout or secondary traumatic stress? Findings from two longitudinal studies from the US and Poland. PLoS One 10(8):e0136730

Siebert DC (2006) Personal and occupational factors in burnout among practicing social workers: implications for researchers, practitioners, and managers. J Soc Serv Res 32(2):25–44

Smoktunowicz E, Baka L, Cieslak R, Nichols CF, Benight CC, Luszczynska A (2015) Explaining counterproductive work behav- iors among police officers: the indirect effects of job demands are mediated by job burnout and moderated by job control and social support. Hum Perform 28(4):332–350

SprangG, Craig C, Clark J (2011) Secondary traumatic stress and burnout in child welfare workers: a comparative analysis of occupational distress across professional groups. Child welfare 90(6):149

Stamm BM (2002) Measuring compassion satisfaction as well as fatigue: developmental history of the compassion satisfaction and fatigue test. In: Frigley CR (ed) Treating compassion fatigue. Brunner- Routledge, New York, NY, pp 107–119

Stamm B H (2010) The concise ProQOL manual, 2nd Ed. Pocatello, ID. Retrieved from http://www.proqol.org/uploads/ProQOL_Concise_2 ndEd_12-2010.pdf.

Swider BW, Zimmerman RD (2010) Born to burnout: a meta-analytic path model of personality, job burnout, and work outcomes. J Vocat Behav 76(3):487–506

Travis D J, Lizano E L, & Barak M E M (2015) ‘I’m so stressed!’: a longitudinal model of stress, burnout and engagement among social workers in child welfare settings. British Journal of Social Work, bct205

US Department of Justice (DOJ) (2010) The national strategy for child exploitation prevention and interdiction: a report to Congress. Retrieved from http://www.justice.gov/psc/docs/natstrategyreport. pdf

US General Accounting Office (GAO) (2003) HHS could play a greater role in helping child welfare agencies recruit and retain staff. (GAO- 03-357) Washington, D.C: US General Accounting Office

Wolak J, & Mitchell K J (2009) Work exposure to child pornography in ICAC task forces and affiliates. Crimes Against Children Research Center. Retrieved from http://www.unh.edu/ccrc/pdf/Law%20 Enforcement%20Work%20Exposure%20to%20CP.pdf.

Wolak J, Finkelhor D, & Mitchell K J (2005) Child-pornography posses- sors arrested in Internet-related crimes: Findings From the National Juvenile Online Victimization Study. Retrieved from http://scholars. unh.edu/cgi/viewcontent.cgi?article=1032&context=ccrc

Wright R, Powell MB, Ridge D (2006) Child abuse investigation: an in- depth analysis of how police officers perceive and cope with daily work challenges. Policing: an International Journal of Police Strategies & Management 29(3):498–451

318 J Police Crim Psych (2017) 32:305–318

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.