A Scoping Review of Vicarious Trauma Interventions for Service Providers Working With People Who Have Experienced Traumatic Events

Health and human service providers who aid traumatized individuals frequently experience vicarious trauma (VT). Although VT plays a critical role in service providers’ mental health and well-being, as well as in the quality of their service provision, little information is available concerning the development and implementation of VT interventions for service providers. To advance the development of evidence in this area, we undertook a scoping review in which we reviewed existing interventions intended to address VT among service providers working with traumatized clients. Searches of electronic databases were conducted to identify studies published in peer-reviewed journals, with no date restrictions. Over 1,315 citations were reviewed, and a total of 27 studies were included in the final review. The findings show that VT interventions in the literature can be divided broadly into four categories: psychoeducation, mindfulness intervention, art and recreational programs, and alternative medicine therapy. The VT interventions reviewed generally showed promise in their key outcomes, including reductions in secondary trauma stress, compassion fatigue, burnout, and other mental health outcomes. However, the current body of research is lacking both in rigor and in specificity regarding the definition of VT. Furthermore, existing VT interventions are generally self-care based and tend to focus on general stress management rather than addressing the specific effects of VT. Therefore, we call for an increase in efforts to tailor VT interventions to different service settings and participant characteristics, as well as greater attention to developing primary VT interventions at the organizational level.

Health and human service providers working with individuals who have experienced trauma (e.g., abuse and assault, grief and loss, human trafficking, military combat, natural disasters, and terrorism) are at risk of developing vicarious trauma (VT) through repeated exposure to and empathic engagement with their clients' and patients' traumatic experiences (Raunick et al., 2015;Sprang et al., 2019). Even though service providers' VT is a typical response to working with traumatized people (Figley, 1995;Molnar et al., 2017), such symptoms may negatively affect service providers themselves, resulting in a broad range of cognitive, emotional, and behavioral changes (Bercier & Maynard, 2015). Worryingly, VT symptoms may also affect service quality, causing a deterioration in provider decision making and service delivery (Branson, 2019;Schauben & Frazier, 1995).
In the last few decades, researchers and practitioners have developed different interventions to prevent and decrease service providers' trauma-related symptoms (Bercier & Maynard, 2015;Molnar et al., 2017). However, little comprehensive information has been available regarding VT intervention research, including effectiveness and limitations (Bercier & Maynard, 2015;Molnar et al., 2017;Sprang et al., 2019). The scarcity of such research highlighted a critical need to examine and synthesize systematically existing VT intervention studies to help advance evidence development in this area (Molnar et al., 2017;Sprang et al., 2019).

VT, Secondary Trauma Stress, and Compassion Fatigue
Broadly, VT refers to significant, indirect experiences of distress resulting from empathic engagement with clients who experienced trauma (World Health Organization, 2013). VT is conceptually linked to and overlaps with other two constructs, namely, secondary traumatic stress (STS) and compassion fatigue (CF) (Jenkins & Baird, 2002;Newell & MacNeil, 2010). VT, STS, and CF have been often used interchangeably (Newell & MacNeil, 2010;Rauvola et al., 2019) or grouped together as a broad construct (e.g., empathy-based stress) explaining service providers' comprehensive trauma-related symptoms (Rauvola et al., 2019). Given that VT, STS, and CF share common causes and consequences for service provision, these terms are understandably used interchangeably. However, many researchers have urged the conceptual differentiation of these constructs and some researchers have identified them as symptomatically different constructs (Molnar et al., 2017;Newell & MacNeil, 2010;Rauvola et al., 2019). Pearlman and Saakvitne (1995) defined VT as a change of cognitive schemas resulting from consistent empathic engagement with traumatized individuals (Newell & MacNeil, 2010). That is, indirect exposure to trauma changes service providers' cognitive schemas and beliefs about the self, others, and the world (Cohen & Collens, 2013;Moulden & Firestone, 2007;Pearlman & Saakvitne, 1995). For example, a domestic violence shelter worker who regularly helps women who have been victimized by their male partners may start to believe that all men will become violent (Molnar et al., 2017). While VT referred to a process of cognitive change, the construct of STS placed more emphasis on the outward symptoms (Figley, 1995). The symptoms of STS were similar to the full range of post-traumatic stress disorder (PTSD) symptoms, including intrusive thoughts, insomnia, chronic irritability or angry outbursts, fatigue, difficulty concentrating, avoidance of clients, and client situations (Figley, 1995;Jenkins & Baird, 2002). That is, service providers exposed to indirect trauma might show the same PTSD symptoms experienced by individuals directly exposed to trauma. In the most recent version of the Diagnostic Statistical Manual (the DSM-5), the PTSD diagnosis now includes an expanded definition of trauma that includes those who have experienced secondary trauma in their professional role (American Psychiatric Association, 2013). Figley (1995) later changed the name of STS to CF, which applied to broader populations and goes beyond those professions where compassion is an expected service (Ludick & Figley, 2017). CF was best defined as a syndrome consisting of a combination of STS symptoms and professional burnout (Figley, 1995;Newell & MacNeil, 2010).
Put together, initially, VT-related symptoms seem to be defined as different concepts. However, the boundaries remain ambiguous and much of the literature still use these constructs interchangeably (Bercier & Maynard, 2015;Sprang et al., 2019). Thus, though we recognize their different natures, our review on interventions related to preventing or treating VT included not only VT intervention programs but also STS and CF interventions, which allowed us to achieve a comprehensive review of existing VT interventions. However, for readers' ease, we called all of the interventions addressing service providers' VT, STS, and CF "VT intervention" here.

Previous Reviews on VT Interventions and the Current Study
While diverse interventions to prevent and ameliorate service providers' VT-related symptoms have been developed and studied, there have been limited efforts to date to review this research (Rauvola et al., 2019). For example, Everly (1999) explored the effectiveness of group psychological debriefings on VT with emergency care providers through a meta-analysis. However, this review only focused on a particular type of VT interventions (i.e., debriefing) and, thus, did not review a range of relevant literature. In addition, the review included not only emergency workers but also trauma victims in four sample studies, without a specific rationale. Bercier and colleagues (2015) attempted to conduct a systematic review of studies on interventions for STS with mental health workers. However, no studies met their inclusion criteria for full-text review, given issues with study design (nonrandomized experimental or quasi-experimental design) or participant characteristics (nonmental health workers). Consequently, they concluded that the existing evidence lacked the rigor necessary for a review and called for renewed evaluation efforts with stronger research methods (Bercier & Maynard, 2015). Recently, Sprang and colleagues (2019) attempted to review evidence from STS interventions and to convene meetings of STS experts in order to discuss best practices in treatment approaches and to consider strategies for moving the field forward. They found that a variety of STS interventions have been launched, including self-help programs, structured workshops, and in-person and online trainings. However, the field still lacked evidence that can be used to inform future STS assessments and interventions. While Sprang et al.'s study provided a thoughtful discussion of existing studies and future directions, their work remained a largely descriptive review rather than a systematic collection and analysis of evidence.
Despite the prior review efforts to date, calls to synthesize and evaluate empirical evidence on VT interventions have continued (e.g., Molnar et al., 2017). Moreover, there is a pressing need to summarize and assess extant VT interventions research both to take stock of the research conducted so far and to help move the field forward. For these reasons, we aimed to conduct a comprehensive scoping review of existing VT intervention research.

Method
Considering that the existing literature regarding VT interventions was heterogeneous in nature, we adopted a scoping review approach, underpinned by Arksey and O'Malley's (2005)  literature in a field of interest. This approach addresses broad areas of evidence, exploring breadth rather than depth, and is useful when a topic is still in its early stages of development (Arksey & O'Malley, 2005). In order to strengthen rigor for this review method, Arksey and O'Malley developed a framework for conducting a scoping review, which includes five key phases: identifying the initial research questions, identifying relevant studies, study selection, charting the data, and collating, summarizing, and reporting the results (Arksey & O'Malley, 2005). Our review included these phases.

Identifying the Research Questions
The focus of our review was the exploration of key aspects of VT interventions for service providers working with people who have experienced traumatic events. Specifically, we posed the following research questions to guide the search: (1) What is the extent, range, and nature of VT intervention studies for service providers? (2) What is known from this existing literature about the effectiveness of the VT interventions? (3) What are the implications for future practices and research based on the findings? Arksey and O'Malley (2005) suggested that a wide definition of key words for search terms should be adopted to obtain "broad coverage" of the available literature. Initial key concepts and search terms were developed to capture a broad range of literature regarding VT. Then, our research team consulted with a university reference librarian to refine search key words and appropriate literature databases based on our research questions. Using the recommended databases and search terms formats, beginning in October 2019 and ending in November 2019, we conducted the literature search through five electronic databases: PsycINFO, Social Service Abstract, PubMed, Cumulative Index to Nursing and Allied Health Literature, and Web of Science. We created four search terms categories, namely, VT terms, intervention terms, service provider terms, and client terms, and used combinations of the terms in the four categories (Table 1).

Study Selection
We examined the resulting articles based on the following inclusion criteria: (1) original research in peer-reviewed journals; (2) full texts available online with no exclusions with regard to publication dates; (3) human subjects; (4) study participants/sample includes professionals working with people experiencing traumatic events such as abuse and assault, grief and loss, human trafficking, military combat, natural disasters, and terrorism; (5) intervention-based studies focusing on VT, STS, and CF; and (6) peer-reviewed articles available in English. Thus, we excluded nonempirical articles, such as program manuals, books and book chapters, concept papers, and nonpeer review/gray literature reports on empirical studies (using qualitative, quantitative, or mixed methods), including unpublished research reports or dissertations. Our search yielded 1,315 articles (see Figure 1). After the removal of duplicate articles (n ¼ 256), we reviewed each article's title and abstract based on the inclusion and exclusion criteria. The review produced a selection of 157 articles for full text review. Our team evaluated the articles in-depth by reading the full text and excluding articles (n ¼ 140) if they did not cover VT, intervention research, or human subjects, if they were in the form of narratives, expert opinions, or review articles, or if they were not in English. Following the full-text review, a total of 17 studies were selected for the study. In addition, in an effort to be as comprehensive as possible in evidence identification, we conducted a backward search of  Kim et al. 3 the sources cited in each article as a means of identifying additional literature that might have been missed in our database searches. Our snowball search efforts added 10 articles meeting inclusion criteria for this review.

Data Charting and Collating
To chart and collate our review of selected articles systematically, our research team developed a data abstraction tool. We designed this tool to obtain detailed information to the extent that such information was presented in each article, including information on VT interventions, study aims and method, study results, strengths, and limitations. Three members of our research team independently read full texts and then gathered detailed information using the abstraction tool. Before this process fully began, all team members read the same two articles and shared findings both to pilot the tool and to reach review consensus and consistency. After all abstractions were completed, one member of the research team independently reviewed each abstraction along with the article form to ensure accuracy and completeness across all reviews.

Summarizing and Reporting the Results
The final stage of Arksey and O'Malley's (2005) scoping review framework summarizes and reports findings. Table 2,  Table 3, and Table 4 provides key information for each of the reviewed studies, noting the study characteristics, the types of VT interventions evaluated, and relevant findings.
1) Canada 2) Third year medical trainees/ students (n ¼ 59) 3) Canadian undergraduate medical school 1) Burnout and CF program 2) Seven essential elements of the program evaluation process (Haji et al., 2013) 3) A self-reflection exercise to increase awareness of one's own signs and symptoms of distress and a review of evidence for strategies aimed at increasing resilience and coping skills 4) In-person group workshop: One workshop, length not specified  2) Secondary outcomes: Depression, anxiety, anger, and hope/ hopelessness 1) No significant differences were found between treatment group, placebo, and control group in any outcomes   (2018) 1. Short and feasible intervention, validated measures, and mixed methods 2. Small sample size, no control/comparison group, only two data collection points, and lack of detail on qualitative analysis Allen et al. (2017) 1. In-depth discussion program facilitated by psychiatrists, three data collection points, and validated measures 2. Small sample size, no control/comparison group, and little detail on the intervention program Berger & Gelkopf (2011) 1. Detailed curriculum, validated measures, control/comparison group (wait-list), rigorous analysis strategies, and relatively large sample size 2. Intensive program sessions may be less feasible in certain types of settings and only two data collection points Berger et al. (2016) 1. Adapted training to address the needs of diverse participants, given the ethnic/cultural backgrounds, control/ comparison group, relatively large sample size, and three data collection points 2. Intensive program sessions may be less feasible in certain types of settings and the measures of optimism and personal self-efficacy were based on single items Cieslak et al. (2016) 1. Strongly evidence-based intervention model (targeting self-efficacy as well as using cognitive therapy), control/ comparison group, relatively large sample size, control/comparison group, and rigorous analysis strategies 2. Selection bias (online participation), high attrition, large dropout rate, and only two data collection points deFouchier & Kedia (2018) 1. CBT-based feasible and cost-effective intervention, clinical psychologists administered the intervention, relatively large sample size, and mixed methods 2. No control/comparison group, no description of qualitative methods used, and only two data collection points Flarity et al. (2016) 1.

Study Characteristics
This scoping review included 27 empirical studies, which were diverse in their samples, methods, intervention programs, and findings.
Study setting and participants. The 27 articles identified were published between 2008 and 2019, with the majority published within the past 6 years (k ¼ 23, 85%). More than half of the studies were conducted in the United States (k ¼ 16, 59%), whereas the rest of the studies were conducted in other countries (k ¼ 11, 41%; e.g., Australia, South Africa, Canada, Poland, Scotland, China, Israel, and Central African Republic). The majority of the studies (k ¼ 19, 70%) investigated the efficacy of VT interventions for health care providers working in diverse settings, including palliative care, cancer care, obstetrics/gynecology, and hospital emergency and trauma centers. Meanwhile, the other eight studies (k ¼ 8, 30%) were targeted to human service professionals who respond to clients' violence and trauma issues in a variety of settings, including counselors, lawyers, social workers, and teachers. Of the documents reviewed, three studies examined VT interventions for novice professionals, such as nurses entering the field (Wylde et al., 2017), medical student interns (Tucker et al., 2017), and mental health interns (Novoa & Cain, 2014). In addition, of the total literature, three studies sampled professionals experiencing dual trauma in war, terror, and natural disaster affected areas (Berger et al., 2016;Berger & Gelkopf, 2011;Weidlich & Ugarriza, 2015). In these situations, service providers may suffer symptoms of both primary trauma (as a result of the war, terror, or disaster itself) and secondary trauma (as a result of working with traumatized clients and patients).
Research method. The majority of the empirical studies (k ¼ 18, 66%) were quantitatively focused research; only two studies (7%) used qualitative methods, namely, in-depth interviews and focus group interviews (Newman et al., 2015;Slatyer et al., 2017). Seven articles (25%) adopted mixed method designs that included collecting quantitative evaluation data and qualitative interviews. With the exception of Hatzipapas et al.'s (2017) study, which adopted a qualitatively focused mixed design, the other six studies used a quantitatively focused mixed method design. For articles using a quantitative approach, most of the studies (k ¼ 14, 77%) adopted preexperimental designs, which is a one-group pre-and posttest design  � Vicarious trauma (VT) interventions mostly targeted health care professionals and were delivered as the following types of programs, psychoeducation, mindfulness interventions, art and recreational programs, and alternative medicine therapies � VT interventions were generally self-care based and tended to focus on general stress reduction and health promotion rather than addressing the specific effects of VT � The VT interventions generally showed positive effect in their key outcomes. In particular, interventions delivered over the longer term in a group setting showed the most promise in addressing service providers' VT symptoms. However, future systematic evaluations are needed to determine the efficacy of such interventions, as well as investigate the efficacy of different intervention approaches without a control or comparison group. Only four studies (14%) adopted a randomized control trial (RCT) design, which assigned control and intervention groups through random assignment (Berger & Gelkopf, 2011;Boone & Castillo, 2008;Cieslak et al., 2016;Novoa & Cain, 2014). Of the four RCT studies, two studies used an RCT design using an online format but did not provide specific information for their RCT process (Boone & Castillo, 2008;Cieslak et al., 2016). Another two studies (7%) used quasi-experimental designs with a comparison group but not random assignment (Berger et al., 2016;Wylde et al., 2017). Overall, apart from the two qualitative studies, most studies adopted simple pretest-posttest group design with two data collection points (k ¼ 18, 66%). Only seven studies adopted longitudinal design with three or more data collection points (k ¼ 7, 25%).
Major measurements. Of the 25 quantitative and mixed-method studies, 21 studies (77%) collected data from participants concerning VT-relevant constructs, such as STS, CF, compassion satisfaction, post-traumatic stress, and secondary posttraumatic growth, as their primary outcomes. The Professional Quality of Life Scale (ProQOL) Version 5 (Stamm, 2010) was the most commonly used scale for measuring VT-relevant constructs (k ¼ 14, 51%). Using Figley's (1995) definition of CF and its constructs, Stamm (2010) developed the ProQOL-5 scale (Wells-English et al., 2019). The 30-item ProQOL-5 includes three subscales: Compassion Satisfaction, Burnout, and Secondary Trauma Stress where respondents indicate how frequently they have experienced each in the past 30 days (Stamm, 2010). Consistent with this, Wylde's (2017) study used the Compassion Fatigue Self-Test (Figley & Stamm, 1996) which is a previous version of the ProQOL-5. Compassion Fatigue Self-Test is a 66-item test measuring individuals' levels of compassion satisfaction, CF, and burnout in their role as a helper. The other two studies measured VT-relevant outcomes by the Secondary Traumatic Stress Scale (STSS) which contains three 17-item subscales (Intrusion, Avoidance, and Arousal). The STSS assessed the frequency of STS symptoms experienced by service providers in the past 7 days using a 5point, Likert-type response format, congruent with the DSM-IV-TR definition of PTSD (Bride et al., 2004). These three instruments, ProQOL-5, Compassion Fatigue Self-Test, and STSS, have demonstrated strong internal reliability in previous studies (Bride et al., 2004;Figley & Stamm, 1996;Stamm, 2010). Notably, the other four studies (14%) measured the efficacy of VT interventions through general mental health outcomes such as general PTSD, perceived stress, avoidance, anxiety, depression, and workplace engagement (Buchanan et al., 2018;de Fouchier & Kedia, 2018;Hatzipapas et al., 2017;O'Mahony et al., 2016). Aside from primary outcomes, nine studies used additional secondary measures (e.g., professional sense of self-efficacy, self-esteem, hope, a sense of mastery, perceived level of stress, self-care, resiliency, ways of coping, and attitudes toward alcohol use) to examine the effectiveness of VT interventions (Berger et al., 2016;Berger & Gelkopf, 2011;Cieslak et al, 2016;Meadors & Lamson, 2008;Wahl et al., 2018;Weidlich & Ugarriza, 2015;Williamson, 2019;Xu et al., 2016;Zajac et al., 2017). With respect to the measurement of qualitative studies, interview questions mainly consisted of brief feedback on the intervention (e.g., overall subjective thoughts about the intervention). For example, the interview asked broad questions on the strengths and weaknesses of the program (Slatyer et al., 2017) and how the program made participants feel or how it affected their work (Newman et al., 2015).

Key Findings
In the reviewed studies, the majority of the VT interventions focused on psychoeducation (k ¼ 17, 62%) and mindfulness (k ¼ 5, 18%) approaches, with overlap in their key activities. The other program types include art and recreational programs, such as a drumming group, aerobic laughter therapy, and poetry therapy (k ¼ 3, 11%) and complementary and alternative medicine, such as auricular acupuncture and energy therapy (k ¼ 2, 7%). Comprehensive details are provided regarding key program activities, delivery methods, effectiveness, and limitations.

Psychoeducation
Key activities and delivery methods. Seventeen studies implemented and evaluated psychoeducational-focused interventions. Generally, the psychoeducational approach included (1) education on the early symptoms of VT, how to be cognizant of VT, and physical and psychological signs and symptoms of VT; (2) education on self-care and stress management activities and techniques (e.g., mindfulness practices); and (3) development and application of individual self-care plans. However, the findings also showed considerable diversity across all these interventions' theoretical underpinnings and program activities. For example, some VT interventions used evidencebased psychological models such as cognitive behavior therapy (e.g., Cielak et al., 2016;de Fouchier & Kedia, 2018). Alternatively, some interventions focused on brief, professional stress management seminars with education on self-care activities. Notably and in addition, some psychoeducational interventions included additional components or methods including professional skills trainings (Berger & Gelkopf, 2011;Williamson, 2019) and peer support networks (Wahl et al., 2018). With respect to delivery methods, psychoeducationalfocused programs tended to consist of in-person group sessions, including didactic presentations and small activities. However, the length and intensity of the interventions ranged from 1 to 12 sessions and from 1 to 6 hr each session meeting. Specifically, about half of the psychoeducational programs (k ¼ 8) provided a short program through a one-time session rather than programs with multiple sessions. Notably, some interventions were implemented through unique formats, such as an intensive 3-day workshop (Berger et al., 2016), psychiatrist-led Balint-style discussion groups (Allen et al., 2017), an online format (Cieslak et al., 2016;Williamson, 16 TRAUMA, VIOLENCE, & ABUSE XX(X) 2019), and hybrid formats, which included both in-person and online meetings (Wahl et al., 2018).
Effectiveness. The majority of the psychoeducational programs (k ¼ 13, 76%) showed promise with regard to key outcomes, which include VT, STS, CF, burnout, and other negative psychological symptoms, including anxiety, as measured with standard scales, especially the ProQOL-5. However, four studies showed mixed findings across multiple outcome measures and in follow-up (Klein et al., 2017;Weidlich & Ugarriza, 2015;Williamson, 2019;Zajac et al., 2017). As an example, a resilience intervention for health care workers showed a positive change in compassion satisfaction and a small reduction in burnout, but no change in STS during pre-and post-follow-up (Klein et al., 2017). Even though some studies were conducted using strong research designs, including as examples, large sample sizes, longitudinal designs, and randomization (e.g., Berger et al., 2016;Cieslak et al, 2016), most of the psychoeducational studies (k ¼ 10, 37%) used one-group, pre-and posttest design, without control or comparison groups.

Mindfulness
Key activities and delivery methods. Five studies adopted mindfulness interventions to prevent and reduce service providers' experience of VT. Even though there were variations in major program components and delivery methods, these interventions generally covered an introduction to mindfulness and mindfulness practices such as meditation, yoga, and body movement. However, some mindfulness interventions also included psychoeducation (Dutton et al., 2017;Slatyer et al., 2017). For example, a brief mindful self-care and resiliency program provided a 1-day psychoeducational workshop on CF, resiliency, and mindfulness, followed by four weekly mindfulness skill seminars (Slatyer et al., 2017). In terms of delivery methods, mindfulness interventions generally consisted of in-person group sessions, except for one smartphone-based mindfulness training (Wylde et al., 2017). These interventions delivered content through a combination of mindfulness seminars and practices and generally seemed to have more or lengthier sessions compared to psychoeducational programs. In particular, one study implemented a 4-day retreat for violence advocates to help participants find balance and heal their minds, bodies, and spirits based on a holistic approach (Dutton et al., 2017).

Effectiveness.
A review of all the mindfulness interventions demonstrated their effects, not only on VT-related symptoms (e.g., STS, burnout, and PTSD) but also on outcomes related to mind-body wellness (e.g., somatic symptoms, insomnia, and mindfulness). However, while a retreat-based intervention showed positive effects of reduced VT, which lasted until the 3-month follow-up, the effects were no longer seen by the 6month follow-up (Dutton et al., 2017). Meanwhile, a smartphone-based mindfulness training showed similar or better effects of reduced VT compared to traditional, in-person training format (Wylde et al., 2017). Overall, mindfulness interventions study used a single group pre-and postdesign, except for one study with a quasi-experimental design (Wylde et al., 2017). In qualitative findings, participants commonly reported greater attention to self-acceptance and mindful awareness and less self-criticism through mindfulness (Dutton et al., 2017;Slatyer et al., 2017). Even though qualitative studies provided a brief explanation of the data analysis process, the authors lacked an explanation of the steps they took to ensure trustworthiness (e.g., triangulation, peer debriefing, and audit trail; Lietz & Zayas, 2010).

Art and Recreational Programs
Key activities and delivery methods. Three art and recreational programs (online poetry therapy, aerobic laughter therapy, and a drumming group program) were implemented to decrease professionals' trauma-related symptoms (Boone & Castillo, 2008;Hatzipapas et al., 2017;Newman et al., 2015). Overall, these interventions adopted novel program approaches that were empirically supported in previous studies. However, the interventions seemed to be unstructured in that the descriptions of the intervention programs lacked a detailed explanation. For instance, the poetry therapy designed for domestic violence counselors consisted of reading poetry provided by researchers and then writing emotionally expressive pieces that allowed the counselors to address and process their secondary PTSD through reflection on the poetry (Boone & Castillo, 2008). This program was implemented across three sessions, using online platforms; however, specific information on program delivery was not provided. Aerobic laughter therapy for community care workers consisted of physical warm-up activities, breathing techniques, and laughter exercises and was implemented 10-15 min a day for a month (Hatzipapas et al., 2017). The drumming group program, designed for health care professionals at a psychiatric hospital, was delivered by a psychologist who had experience in drumming (Newman et al., 2015). This program was held 3 times per week for 30 min before work for 18 months. Each session consisted of teaching drumming techniques and playing together as a means of expression.
Effectiveness. Domestic violence advocates who participated in the online poetry therapy with the online RCT study showed a decrease in secondary PTSD compared to a control group (Boone & Castillo, 2008). The study of the aerobic laughter therapy adopted a qualitative-focused, mixed method design (Hatzipapas et al., 2017). Qualitative findings showed that participants reported more positive emotions, positive coping, improved interpersonal relationships, and improvement in their work after exposure to laughter therapy. Quantitative findings showed that this laughter therapy was effective on reducing participant stress, anxiety, and depression. The drumming group for health care providers also used a qualitative research approach and showed that participants experienced a sense of belonging, relaxation, emotional expression, and an escape 1454 TRAUMA, VIOLENCE,& ABUSE 23(5) from VT (Newman et al., 2015). Even though two qualitative studies provided detailed data analysis techniques to assure research quality, they appeared to be largely based on exploratory research.

Alternative Medicine Programs
Key activities and delivery methods. The reviewed studies included two alternative medicine programs (Buchanan et al., 2018;Novoa & Cain, 2014). An auricular acupuncture program was implemented for health care providers in the cardiovascular division of a hospital (Buchanan et al., 2018). In this therapy, an acupuncturist placed needles in five specific points on the participants' external ear in order to facilitate the release of negative emotions associated with specific organs (e.g., the liver with anger, the lungs with sadness, and the kidney with fear). This therapy was implemented 5 times for 30 min, over the course of 16 weeks. Another alternative medicine program was an energy therapy called Reiki, which was delivered to different mental health service providers (Novoa & Cain, 2014). In this therapy, a Reiki practitioner started at the head and worked toward the feet, keeping their hands approximately 1-2 in. away from the participant's body. This therapy aimed to heal using a holistic approach that incorporated mind, body, and spirit. A Reiki practitioner implemented four one-on-one, in-person sessions every week for 4 weeks.
Effectiveness. Auricular acupuncture therapy with a one-group, pre-and posttest design showed a significant reduction in participant anxiety and an increase in their overall work engagement scores compared to scores at baseline (Buchanan et al., 2018). However, this study did not measure vicarious traumarelated outcomes, even though the program aimed to cover CF. The Reiki treatment for mental health professionals adopted an RCT design and showed no significant difference in compassion satisfaction, STS, or burnout between the treatment and control groups (Novoa & Cain, 2014).

Discussion
Given the critical role that VT plays in service providers' mental health and well-being, as well as in the quality of service provision, it is important to understand the existing evidence on interventions used to address service providers' VT. Through this current study, we aimed to investigate and examine the existing interventions designed to address service providers' VT-related symptoms. Specifically, we sought to determine (a) the scope of currently available published literature on addressing service providers' VT related to symptoms, (b) the effectiveness of VT interventions, and (c) the implications for future practice and research.

Scope of VT Intervention Research
As previously mentioned, we identified 27 articles, published between 2008 and 2019, which evaluated interventions created to address health and human service providers' VT-related symptoms, including STS and CF. The majority of the literature on VT interventions was conducted in the United States and adopted quantitative research or quantitatively focused mixed method designs. Generally, the studies reviewed adopted a single-group research design, without a control or comparison group with only four using RCT designs. VT-related outcomes were largely measured using a ProQOL-5 Scale, which includes CF, burnout, and secondary trauma stress. Literature on VT interventions mostly targeted health care professionals, while the rest of the literature targeted human service providers serving clients exposed to violence, abuse, and disaster. Although VT interventions that have been empirically investigated to date are heterogeneous and multidimensional in their approaches, the literature can be broadly divided into four categories: psychoeducation, mindfulness intervention, art and recreational programs, and alternative medicine therapy. We also note here that the review findings show that the key activities of the first two categories-psychoeducation and mindfulness interventions-often overlapped. Accordingly, the approaches used for VT interventions may not always be distinct in either practice or research.

Effectiveness of VT Interventions
Due to the variations of study population, outcome measures, and research designs, assessing and establishing the promise of the various VT interventions determined in this review were not possible. Accordingly, and consistent with the scoping review approach taken for this study, this discussion largely reflects summaries of intervention outcomes but also delves into the details of statistical findings when possible.
Overall, findings showed that, regardless of intervention type, the VT interventions reviewed generally showed promising findings not only for primary outcomes related to VT symptoms but also in secondary outcomes (e.g., self-efficacy, job satisfaction, and mindfulness). In particular, the qualitative studies reviewed here showed that participants commonly reported more mindful awareness, less self-criticism, greater relaxation, an increased sense of belonging, and an escape from VT. However, the review findings also showed some mixed outcomes, with VT interventions being effective in decreasing some VT-related symptoms, but not effective in other outcomes (Klein et al., 2017). Such results indicate the multidimensional nature of VT-related symptoms and suggest that VT-related symptoms may need to be addressed using a tailored approach to each aspect of VT responses. In addition, the review findings tended to show that interventions delivered over the longer term in a group setting may help address service providers' VT symptoms through peer support and comprehensive program approaches that address the complex nature of VT.
Despite the overall positive trends in the studies' outcomes, we also need to consider the studies' research design and limitations (Bercier & Maynard, 2015;Sprang et al., 2019). Generally, the studies reviewed adopted a single-group research design, without a control or comparison group, used small samples, and

18
TRAUMA, VIOLENCE, & ABUSE XX(X) limited short-term follow-up. Concerning the studies' follow-up periods, given that VT is multidimensional and presents complex symptoms, prior research has strongly recommended that VT needs to be measured through long-term follow-up (O'Mahony et al., 2016). In addition, only a few studies measured participants' baseline trauma symptoms or previous trauma history (e.g., Wylde et al., 2017). Given that service providers' personal trauma histories are critical risk factors on VT symptoms, studies that did not consider previous trauma might fail to control for confounding effects on the study outcomes. In the case of the qualitative research, studies often lacked information regarding how robust research methods were used to ensure study trustworthiness. To sum, while the investigated VT intervention research shows promise, the inconsistency and lack of rigor across this body of research means that little can be confidently said about the effectiveness of any approach to prevent VT, particularly when the rigor of these studies is paired with the heterogeneous nature of the intervention approaches.

Implication for Practice and Policy
This scoping review demonstrates that existing VT interventions are generally self-care based and focus on stress management rather than addressing the specific symptoms of VT depending on service setting and type of trauma. Therefore, we call for an increase in efforts to tailor VT interventions to different service settings and participant characteristics, as well as greater attention to developing primary VT interventions at the organizational level. We detail specific recommendations in the following paragraphs. First, we found that existing VT interventions, in particular psychoeducation and mindfulness interventions, generally focused on generic wellness and job stress reduction and used a similar format of key activities, such as increasing self-care, selfawareness, and mind-body skills. Given that VT-related symptoms are multidimensional, VT interventions should focus on trauma-related alterations in emotions, beliefs, and behaviors as well as on occupational skill development. Notably, VT interventions need to address specific effects of VT symptoms depending on service setting (e.g., health care workers are exposed to grief and loss, while those in a violence service setting are exposed to violence and abuse), type of trauma (e.g., indirect or dual, frequency, duration, and severity), and service providers' individual characteristics. Thus, practitioners need to provide programs tailored to particular service settings, taking into account likely sources, types of trauma, and participants' individual characteristics and backgrounds (Nuttman-Shwartz, 2015).
We also found that existing VT interventions seem to lack attention to potential protective and risk factors for VT. Prior studies show that VT-related symptoms and levels differ depending on individual characteristics including gender and age (Baum, 2016;Benuto et al., 2018), exposure to previous trauma (Rauvola et al., 2019) as well as work experience (Benuto et al., 2018). A recent systematic review regarding VT among child welfare and child protection professionals showed that personal trauma history was the most consistent risk factor for vicarious traumatization (Molnar et al., 2020). Thus, we recommend that future interventions should consider and attend to key factors such as participants' age, race, gender, previous trauma history, and work experience in order to provide participants with focused, meaningful interventions that are tailored to their backgrounds, circumstances, and needs.
In addition, VT intervention programs must be designed with clear program targets and goals, defining whether they seek to offer preventative intervention or ameliorative treatment. With the exception of a few studies targeting students and interns (Tucker et al., 2017;Wylde et al., 2017), it was unclear in our review whether programs aimed to prevent potential VT symptoms or decrease already existing VT symptoms. Sprang and colleagues suggested that universal and primary preventative VT psychoeducation can be considered a prevention intervention, which aims to reduce risk factors and strengthen protective factors for all staff (Sprang et al., 2019). On the other hand, VT intervention can be delivered as treatment intervention which consists of selective, secondary intervention and clinical treatment for staff and who are experiencing STS (Sprang et al., 2019). Such interventions may helpfully include motivational interviewing and cognitive therapy which focused on clinical approach of service providers' VT symptoms (Sprang et al., 2019). For all these reasons, designing frameworks for VT interventions which consider target group's characteristics should precede the development and implementation of VT interventions.
We also found that the VT interventions reviewed were mostly based on individual-based self-care approach. Even though individual self-care appears essential to decrease VTrelated symptoms, primary prevention on an organizational level may also be necessary to structurally and contextually lower the risk of potential VT among providers (Sprang et al., 2019). Theories of work stress as well as empirical studies have suggested that organizational culture, support, and resources may determine how service providers understand and respond to their VT-related symptoms (Molnar et al., 2017;Rauvola et al., 2019;Sprang et al., 2019). Accordingly, VT prevention activities should be implemented both at the organizational and individual levels. Suggestions for creating a VT-informed organizational culture might include increasing trauma-specific supervision, developing peer support networks, and providing opportunities for clinicians to receive mental health services and online support (Branson, 2019;Elwood et al., 2011). For example, VTinformed organizational readiness guides (Hallinan et al., 2019), STS-informed organizational assessments (Sprang et al., 2019), and training in STS core competencies in trauma-informed supervision (National Child Traumatic Stress Network, 2018) can be useful tools to assess and diagnose organizational capacity and readiness to address VT and can support organizations in providing VT-informed supervision to professionals at risk of developing VT (Sprang et al., 2019).
Despite the limitations of the research, the studies we reviewed suggest potentially promising approaches that can inform practitioners who are considering implementing VT interventions. In particular, the review findings suggest that Kim et al. 19 interventions delivered over the longer term in a group setting may help address service providers' VT symptoms. Behavioral and cognitive theories of PTSD treatment posit that trauma symptoms can be decreased through consistent emotional adjustment to traumatic experiences (O'Mahony et al., 2016).
Refresher courses or periodic monitoring may increase awareness and processing of work-related traumas, which in turn may lead to inhabituation and an eventual improvement in well-being (Klein et al., 2017;O'Mahony et al., 2016). In addition, peer-group-based interventions can contribute to a sense of group cohesion and support, which can contribute to the prevention and mitigation of secondary traumatization symptoms (Berger & Gelkopf, 2011;Newman et al., 2015).
Lastly, policymakers and funders should encourage the development and evaluation of evidence-based programs and protocols designed to address VT among service providers. Although this review showed that there are potentially promising interventions for addressing VT, the field continues to need the development and evaluation of tailored programs. Moreover, such programs, once developed, should be rigorously studied for efficacy. In particular, we urge funders to support sufficient resources for innovative intervention development and rigorous evaluation, which includes funding support for randomized designs, large samples, and follow-up data collection.

Research Implications
Future studies need to clearly define VT-related symptoms and assess the efficacy of interventions, using appropriate measurement for those specific VT-related symptoms. Across the articles we reviewed, there seemed to be a lack of conceptual clarity regarding trauma-related symptoms (e.g., VT, STS, and CF). Some professionals used CF as synonymous to STS, while others use the term to describe a broad range of symptoms that include STS, as well as burnout. In addition, some studies reported on VT-related outcomes, using only general mental health outcomes. As described above, conceptualizing VT-related symptoms can be challenging, considering that the symptoms are multidimensional and may overlap (Newell & MacNeil, 2010;Rauvola et al., 2019;Sprang et al., 2019). However, the use of measurements without consideration for the different underlying definitions VT-related symptoms can lead to unclear outcomes. In this context, we strongly agree with Molnar et al.'s (2017) suggestion that future research should focus on empirically disentangling and operationalizing the concepts of VT, STS, and CF, so that assessment tools can be tailored to be sensitive to these distinctions (Molnar et al., 2017).
Furthermore, no single instrument to assess fully the multidimensional domains of VT-related symptoms is currently available (Molnar et al., 2017;Sprang et al., 2019). Even though existing measurements such as the STSS and ProQOL-5 Scale have shown high validity and reliability in previous studies (Sprang et al., 2019), the scales do not seem to cover VT-related cognitive change, a core concept of VT. Thus, the development of comprehensive validated measures to assess VT-related symptoms is necessary (Branson, 2019). Such measures are essential to identify underlying mechanisms of VT that can serve as targets for VT interventions (Sprang et al., 2019).
Given the limitations of the reviewed research's study designs described above, we echo and underscore others' call for rigorous research including randomized design, long-term follow-up to investigate the effects of interventions aimed at preventing and/or ameliorating VT (Bercier & Maynard, 2015;Sprang et al., 2019). Furthermore, our review of qualitative studies showed that many did not specify how they tried to prevent threats to quality and ensure trustworthiness (Lietz & Zayas, 2010). Given that qualitative findings can provide significant contributions in the assessment of participants' cognitive and emotional changes, future qualitative research should use rigorous approaches and report on efforts to ensure study quality.
Lastly, we found that the existing interventions we reviewed were largely focused on health care settings dealing with disease and grief and loss; only a small number of interventions were developed for health and human service providers serving violence victims. Given that many studies have identified that service providers working with violence and trauma are at higher risk of VT-related symptoms (Baum, 2016;Cohen & Collens, 2013;Raunick et al., 2015), research conducted in settings that offer services to victims of violence should be a priority.

Limitations
Like all studies, the review has limitations. Given that VT is a multidimensional concept, we included a number of search terms in order to comprehensively search for related articles. However, our search terms might not have covered all of the existing literature on VT-related interventions. In addition, we only focused on VT interventions in health and human service settings, focusing especially on those interventions geared to health and mental health providers. This left out other potentially affected professions, including animal service providers and first responders. We also limited our review to peer-reviewed literature, excluding gray literature and non-peer-reviewed websites. Given the sizable number of peer-reviewed studies concerning the review topic as well as research questions that guided this review, we focused our efforts on the peer-reviewed literature. Nonetheless, the practice and gray literatures may also hold important insights concerning VT that could be explored in future efforts.

Conclusion
The scoping review shows that VT interventions have been implemented across diverse settings with promising outcomes. However, this review suggests that VT interventions should be tailored to specific service settings and study participants' diverse backgrounds. Furthermore, there is a need for methodologically robust studies to identify effective VT treatment interventions. Researchers have also suggested that VT offers potential positive aspects by providing an opportunity for increased professional growth and resilience, through-for example-vicarious post-traumatic growth (Arnold et al.,20 TRAUMA,VIOLENCE,& ABUSE XX(X) 2005), vicarious resilience (Hernández et al., 2007), and shared resilience in a traumatic reality (Nuttman-Shwartz, 2015). We acknowledge that such professional resilience and growth through exposure to trauma can be achieved when the VT is appropriately addressed through effective intervention programs. Thus, developing and implementing VT interventions, based on tailored approaches, should be a primary and essential task, in order to support service providers in pursuing the pathways from VT to professional resilience and growth.

Summary of Implications for Research, Practice, and Policy
Implications for research. Future studies are needed to clearly define VT-related symptoms and assess the efficacy of interventions, using appropriate measures for specific VT-related symptoms. The development of comprehensive validated measures to assess VT-related symptoms is necessary to identify underlying mechanisms of VT that can serve as targets for VT interventions. Rigorous research to investigate the effects of VT intervention should be conducted using randomized research designs, long-term follow-up data collection, as well as strategies to increase the robustness and trustworthiness of research, including in the case of qualitative studies. Given that studies have identified that service providers working with violence and trauma are at high risk of VTrelated symptoms, as a priority for future research, we call on researchers to conduct intervention research in settings that offer services to victims of violence.

Implications for practice and policy.
VT interventions need to address specific effects of VT symptoms depending on service setting, type of trauma, and service providers' individual backgrounds and characteristics. VT intervention programs must be designed with clear program targets and goals, including whether they seek to offer preventative interventions or ameliorative treatments. In addition to individual-level approaches, primary prevention on an organizational level is necessary to structurally and contextually lower the risk of potential VT among service providers. Funders and policymakers should encourage program development and evaluation. In addition, funders need to provide sufficient resources for the rigorous evaluation of VT interventions.

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.