Indirect Trauma in the Field Practicum: Secondary Traumatic Stress, Vicarious Trauma, and Compassion Fatigue Among Social Work Students and Their Field Instructors

Carolyn Knight

A sample of BSW students and their field instructors was assessed for the presence of indirect trauma, including secondary traumatic stress, vicarious trauma, and compassion fatigue. Results indicated that students were at greater risk of experi- encing vicarious trauma than their field instructors and research participants in previous studies. Risk factors for students and social workers included age, experi- ence, gender, and placement in a child welfare setting. Race was a risk factor only for students. The results underscore the protective effects of advance preparation, agency support, and agency supervision, but also suggest that specific education about indirect trauma is lacking. Also discussed are the links among indirect trauma, burnout, and students’ career aspirations, as well as suggestions for future research.

Keywords: indirect trauma, vicarious trauma, secondary traumatic stress, compassion fatigue, field practicum

It is increasingly recognized that helping professionals are likely to be trauma- tized as a result of their work with challenging clients, such as those with his- tories of trauma. Stamm notes, “The great controversy about helping- induced trauma is not ‘Can it happen?’ but ‘What shall we call it?’” (1997, p. 1). In this article the term indirect trauma will be used (Knight, 2009), which refers to the range of reactions that clinicians can experience as a result of their work, such as secondary traumatic stress (STS), vicarious trauma (VT), and compassion fatigue (CF).

STS refers to “symptoms nearly identical to PTSD [posttraumatic stress dis- order] except that exposure to a traumatizing event experienced by one person

Carolyn Knight, PhD, MSW, is professor at the University of Maryland, Baltimore County.

Journal of Baccalaureate Social Work32

becomes a traumatizing event for the second person” (Stamm, 1999, p. 11). VT involves changes in the clinician’s views of self and others that result from hearing clients’ stories of victimization and witnessing clients’ pain firsthand (McCann & Pearlman, 1990). Prac titioners are at risk of developing a world- view characterized by suspicion and personal vulnerability (Pearlman & Saakvitne, 1995). CF can occur with any client and refers to the practitioner’s inability to empathize (Collins & Long, 2003).

Numerous studies have documented a relationship between a client’s trau- ma history and indirect trauma (Bober & Regehr, 2006). Previous research also has documented positive connections among difficult- to- engage clients, CF, and STS (Figley, 1995). Professionals with less education and those who are newer to their jobs appear to be at greater risk of indirect trauma, as are those with the most and least experience working with trauma survivors (Baird & Jenkins, 2003; Meyers & Cornille, 2002; Steed & Bicknell, 2001). Pro – fessionals who report feeling supported in their work experience fewer symp- toms, whereas those who describe feeling unsupported appear to be at greater risk (Ortlepp & Friedman, 2002).

Women may be more likely to experience indirect trauma than men, but this may reflect their greater willingness to acknowledge these reactions (Kassam- Adams, 1999). Women may be more vulnerable because they are more likely to have experienced trauma themselves, a factor that has been found to increase the risk of indirect trauma in some studies (Baird & Kracen, 2006; Nelson- Gardell & Harris, 2003).

Indirect trauma is not the same as burnout, which is defined as “feelings of hopelessness and difficulties in dealing with work and in doing [one’s] job effectively” (Stamm, 2005, p. 12). There is evidence, however, that similar var – iables influence both burnout and indirect trauma and that a practitioner’s risk of burnout increases when she or he experiences indirect trauma (Bell, Kulkarni, & Dalton, 2003; Salston & Figley, 2003).

Study Rationale

Indirect trauma among social workers has been less widely studied, but available research has generated findings consistent with those noted previously. A recent study of licensed social workers in one state found that almost all of the respon- dents exhibited at least one symptom of STS, and a minority met the criteria for PTSD (Bride, 2007). Further, a study of child welfare workers in one state, most of whom were social workers, indicated that rates of CF were relatively high (Conrad & Kellar- Guenther, 2006). Further, indirect trauma and burnout among social workers appear to be distinct but related phenomena (Perron & Hiltz, 2006; Siebert, 2005). In one study, indirect trauma was more likely to occur among younger, less experienced social workers (Adams, Matto, &

Harrington, 2001). In this same study, there was no association between indirect trauma and the amount of time spent with traumatized clients. Cunningham, on the other hand, found that interpersonal victimization of clients, especially sexual abuse, produced more symptoms of indirect trauma (2003).

The relationship between education and risk of indirect trauma has received insufficient empirical attention. There is some evidence that prepar- ing clinicians in advance for the challenges associated with working with trau- ma survivors may have a protective effect (Bell et al., 2003; Dane, 2000). How – ever, research indicates that few professionals receive adequate preparation for working with trauma survivors (Knight, 1997; Shackelford, 2007). Finally, virtually no attention has been paid to the existence and implications of indi- rect trauma among students in any mental health discipline— including social work— in field practicum and internship experiences.

Research Questions

The exploratory study described in this article examined the existence of indi- rect trauma among a nonmatched sample of baccalaureate social work stu- dents and their field instructors. In addition to ascertaining the extent to which participants exhibited three manifestations of indirect trauma, the pur- pose of the study was threefold.

First, the author was interested in determining the role of educational and supervisory variables in mitigating or intensifying the effects of indirect trau- ma. Based on the limited research available, it was anticipated that respon- dents who reported learning about or being prepared for indirect trauma would evidence fewer signs of it. It also was anticipated that supervisory support— from the field instructor for students and from the agency supervisor or peers for field instructors— would serve as a protective effect. Second, the author sought to substantiate previous findings regarding the influence of per- sonal, client, professional, and organizational variables on the risk of indirect trauma. Third, the author examined the relationship between indirect trauma and burnout among both students and field instructors, and between indirect trauma and career aspirations among students.

Research Design

Selection of Subjects

All students from one undergraduate program who were in the field placement and their field instructors were included in the study. A total of 81 students and 72 field instructors received the surveys (some field instructors supervised more than one student).

Indirect Trauma in the Field Practicum 33

Research Instruments

All respondents were asked to complete three instruments, two of which meas- ure indirect trauma. The author developed the third instrument, which col- lected background details on respondents.

The Professional Quality- of- Life Scale (ProQOL) This 30-item self- report instrument is the most widely used measure of secondary traumatic stress, compassion satisfaction, and burnout among mental health professionals. Respondents are asked how frequently they experienced a variety of reactions within the last 30 days on a 6-point Likert scale (Stamm, 2005).

Three subscale scores are generated. The compassion satisfaction subscale measures “the pleasure [that is] derive[d] from being able to do [one’s] work well” (Stamm, 2005, p. 5). The lower the score, the greater the risk of CF. Burnout is defined as “feelings of hopelessness, and difficulties in dealing with work or in doing [one’s] job effectively” (p. 5). Higher scores indicated a high- er risk of burnout. The STS subscale is “about . . . work- related, secondary exposure to extremely stressful events [which may include] being afraid, hav- ing difficulty sleeping, having images of the upsetting event pop into [one’s] mind, or avoiding things that remind one of the event” (p. 5). Higher scores indicate a heightened risk of STS.

Good alpha reliabilities have been found for each subscale: compassion sat- isfaction, .87; secondary traumatic stress, .80; burnout: .72. The construct validity of the instrument also has been documented (Stamm, 2005). Fol – lowing the recommendation of Stamm, the author substituted “social worker” for “helper.”

Trauma and Attachment Belief Scale (TABS) The TABS also is a self- report and is the most widely used measure of VT among professionals and psycho- logical trauma among clinical and nonclinical samples (Pearlman, 2003). Respondents are asked to agree or disagree with 84 items, using a 6-point Likert scale. An overall score is generated, as well as scores on 10 different sub- scales. The overall score has been found to have good internal consistency (α =.96). The average alpha score for the 10 subscales is .79 and ranges from .67 to .87 (Pearlman, 2003). The TABS has good face and construct validity, although intercorrelations between the subscales can be fairly high, ranging from .22 to .75.

The overall TABS score “provides an index of the respondent’s overall level of disruption in areas that are important to maintaining healthy relation- ships” (Pearlman, 2003, p. 14). Pearlman (p. 16) defines each of the subscales as follows:

• Self- safety: the need to feel secure and reasonably invulnerable to harm inflicted by oneself or others

• Other- safety: the need to feel that cherished others are reasonably protect- ed from harm inflicted by oneself or others

Journal of Baccalaureate Social Work34

• Self- trust: the need to have confidence in one’s own perceptions and judgment • Other- trust: the need to depend or rely on others • Self- esteem: the need to feel valuable and worthy or respect • Other- esteem: the need to value and respect others • Self- intimacy: the need to feel connected to one’s own experiences • Other- intimacy: the need to feel connected to others • Self- control: the need to manage one’s feelings • Other- control: the need to manage interpersonal situations

Raw scores are converted to standardized T- scores. In all cases, higher scores indicate a greater risk of VT.

Background details Respondents supplied basic demographic information about themselves. They also were asked a variety of questions that previous research indicated might be associated with indirect trauma. Each question used a 3-point Likert scale. Respondents were asked the extent to which they worked with four types of clients: “mandated,” “aggressive,” “difficult to en – gage,” and “with histories of trauma.” They also were asked whether their education prepared them for “negative personal reactions to clients”—the def- inition of indirect trauma in this study— and whether their agencies assisted staff in dealing with negative personal reactions. Students were asked how much they had learned about negative reactions in social work classes and how often they discussed them in class and with peers. Field instructors were asked whether their agencies provided them with any preparation for negative reactions they might encounter in their work.

In addition, students were asked to evaluate the helpfulness of their field instructor in assisting them with negative personal reactions and how fre- quently they discussed negative reactions in supervision. Field instructors were asked how frequently they talked in supervision, with colleagues, and with family and friends about negative personal reactions they experienced. Finally, students were asked the extent to which their field experience strength- ened or lessened their desire to be a social worker and to estimate the likelihood that they would seek employment in a setting similar to their field placement.

Method

Students received the research packet in their social work methods course, taken concurrent with the field practicum. Distribution occurred in late April, approximately 3 weeks before the end of the year- long practicum experience. Students were asked to complete the instruments outside of class and to return them to the author’s campus mailbox or to their methods instructor. The author sent e- mail reminders, and methods instructors also reminded students to complete the instruments.

Field instructors received the research packet via regular mail and were asked to complete the instruments and return them in the postage- paid envelope that

Indirect Trauma in the Field Practicum 35

Journal of Baccalaureate Social Work36

was provided. The author sent an e- mail to all field instructors in advance informing them about the study. Follow- up e- mail reminders also were sent.

A cover letter accompanied the surveys and explained the purpose of the research as the author’s interest in understanding social workers’ negative personal reactions. The letter assured confidentiality, explained that all responses were anonymous and confidential, and confirmed that the research project had been approved by the author’s institutional review board.

Results

Characteristics of Participants and Field Settings

Forty- two students completed and returned the research instruments, repre- senting 51.8% of all seniors in field placement. Thirty- nine were female. Twenty- two described themselves as White; 14 indicated they were African American. The average age of the students was 32.3 years, and their ages ranged from 20 to 54. Twenty- six reported having less than 1 year or no prior experience in social work or a related discipline.

Fifty- one (70.8%) of the field instructors completed and returned the instruments. Forty- four were female. Forty- two described themselves as White; seven described themselves as African American. The average age of the supervisors was 50.4 years, and their ages ranged from 26 to 68. Each had an MSW. Twenty- five field instructors indicated they had 15 or more years of practice experience, and six had practiced less than 5 years. Thirty- two indi- cated they had 5 or more years of experience in the field agency.

Students were most likely to report they worked in an education or school setting (n=13) or in child welfare (n=12). More than one quarter reported they worked with difficult- to- engage clients “a great deal” (n=16). A number of students reported having no experience with three of the client groups (trau- ma histories: n=11; mandated: n=17; aggressive: n=10).

Field instructors also were most likely to report they worked in an educa- tional/school setting (n=12) or child welfare agency (n=11). Forty of the field instructors reported working with clients with trauma histories “a great deal.” Fewer field instructors reported working with the other client groups “a great deal”: mandated: n=15; difficult to engage: n=20; aggressive: n=17.

Manifestations of Indirect Trauma Among Participants

Virtually all students and their field instructors reported some negative reac- tions associated with their work with clients (students: n=40; field instructors: n=50). Student and field instructor average scores on the compassion satisfac- tion subscale of the ProQOL were slightly higher than those of 1,000 mental health professionals in previous studies, as reported by Stamm (2005), indicat- ing a lower risk of CF (see Table 1). Compared to the same professionals, stu-

Indirect Trauma in the Field Practicum 37

dents and field instructors had lower average scores on the STS subscale. Difference of means tests between student and field instructor scores for STS and CF were nonsignificant.

Students’ scores on the TABS overall and for each subscale were lower than a sample of outpatient mental health clients without histories of trauma as reported by Pearlman (2003), but they were higher than or essentially equal to those of a sample of respondents who identified themselves as trauma ther- apists for all but one subscale, other safety (see Table 2). Field instructors’ aver- age scores generally were lower than both research samples.

Students’ VT scores were consistently higher than those of field instructors (see Table 2). Difference- of- means tests revealed significant differences between the two groups for the overall TABS score (t=–3.247, df=90, p≤.002) and 6 subscales (self- safety: t=–3.037, df=90, p≤.003; self- trust: t=–1.999,

Table 1 Mean scores for Professional Quality of Life (ProQOL) subscales: BSW students, field instructors, and research participants

BSW BSW field Research students instructors participantsa

Subscale N=42 N=51 N=1000

Compassion satisfaction 41.3 42.7 37.0 Secondary traumatic stress 11.2 10.3 13.0 Burnout 18.5 18.3 22.0

aStamm, 2005

Table 2 Trauma and Attachment Belief Scale (TABS) mean T-scores for composite scale and 10 subscales: BSW students, field instructors, and two research samples

BSW Field Outpatient Trauma students instructors sample therapistsa

N=42 N=51 N=22 N=266

Composite TABS 48.3 41.8 54.6 44.0 Self-safety 46.4 38.4 53.9 45.4 Other safety 44.8 40.1 53.9 45.4 Self-trust 48.0 43.0 54.2 46.1 Other trust 44.4 36.0 52.1 44.9 Self-esteem 46.5 44.8 55.7 46.4 Other esteem 51.4 44.8 51.4 45.2 Self-intimacy 52.6 46.2 51.9 43.1 Other intimacy 48.7 43.1 56.6 47.7 Self-control 48.7 45.0 56.2 45.8 Other control 44.5 40.1 52.3 45.8

aPearlman, 2003

df=87, p≤.049; other trust: t=–3.669, df=86, p≤.000; other esteem: t=–3.285, df=89, p≤.001; self- intimacy: t=–2.948, df=89, p≤ .004; other intimacy: t=–2.658, df=89, p≤.009).

Manifestations of Burnout Among Participants

Student and field instructor average scores on the burnout subscale of the ProQOL were essentially equal (see Table 1). When compared to research par- ticipants in previous studies reported by Stamm (2005), both students and field instructors had lower scores, indicating a lower risk of burnout.

Agency Support, Supervision, and Education

Seventeen of the 42 students and 33 of the 51 field instructors reported they were “not at all” or “somewhat” prepared by their social work education for negative personal reactions. Ten students and 6 field instructors reported that the agency did not provide assistance with their negative reactions “at all,” whereas 17 students and 22 field instructors stated their agencies provided “some” assistance. Twenty- two students indicated they learned “some” about negative reactions in their social work classes. Sixteen instructors reported that their agencies did not prepare them “at all” for negative reactions they might experience in the setting, and 20 stated they received “some” preparation.

Four of the 40 students who reported experiencing negative reactions stat- ed they were unable to talk about these reactions in class, but 18 said they were able to talk “a great deal.” Twenty- four of these same students reported they talked with fellow students “a great deal.” Twenty- five indicated that they were able to talk with their field instructors “a great deal,” and 12 reported they talked “some.” Twenty- one indicated their instructors helped “a great deal”; 15 stated they received “some” help.

Among the 50 field instructors who reported experiencing negative reac- tions, 36 reported they talked with colleagues “a great deal.” Forty- six indicat- ed that they talked with family and friends “a great deal” or “some” about their reactions. Among the field instructors who experienced negative reactions and were in supervision (n=22), 4 indicated they “never” talked with their super- visors; 14 said they talked “some.” Among the 10 instructors who experienced negative reactions and were in peer supervision, 8 reported talking about their reactions “a great deal.”

Correlates of Indirect Trauma Among Students

Correlation analysis using Kendall’s tau- b as a measure of association for ordi- nal and rank- order variables examined the relationship between determinant and dependent measures. There were 13 dependent measures: the compassion

Journal of Baccalaureate Social Work38

Indirect Trauma in the Field Practicum 39

satisfaction and traumatic stress subscales of the ProQOL and the overall and 10 subscale scores on the TABS. There were a number of statistically signifi- cant associations, but these tended to be weak. Three of the VT subscales (self- safety, other safety, other trust) were not associated with any variables.

Intercorrelations among the CF, STS, and VT measures also were examined. To simplify this analysis, the composite TABS score was used as the sole meas- ure of vicarious trauma. CF was associated with VT (τ= .241, p≤ .020, n=42).

Personal characteristics Women scored higher on one VT subscale, self- trust (see Table 3). Students who described themselves as White scored lower on compassion satisfaction, indicating a greater risk of CF and also exhibited more signs of VT overall and on two subscales (self- trust and other esteem). Younger students and those with less experience also were at greater risk of CF and scored higher on one VT subscale (other control).

Client characteristics Students who reported more frequent contact with mandated clients had higher scores on 4 VT subscales (self- trust, self- intimacy, other intimacy, and self- control). Students who more frequently worked with aggressive clients had higher STS scores.

Education variables Students who reported they were not prepared by their education were at greater risk for CF, as were those who reported they did not learn about negative personal reactions in class and were not able to talk in class or with fellow students about their negative reactions. Not talking with fellow students also was associated with higher scores on two VT subscales (self- trust and other intimacy). Learning about negative reactions was associ- ated with a decreased likelihood of VT (overall, other intimacy, and self- control), as was the ability to talk in class (other intimacy).

Field instruction and agency variables Students who were unable to talk with their field instructors as well as those who believed their field instructors did not assist them with their personal reactions were at greater risk of CF. Field instructor assistance was positively associated with one measure of VT (self- esteem). Students who reported their field agencies did not provide assistance were at greater risk of CF and scored higher on one VT subscale (self- trust).

Agency type, categorized into child welfare settings and other, was associat- ed with three measures. Students who worked in child welfare settings had higher secondary traumatic stress scores and higher scores on the self- esteem and self- control subscales.

Association among indirect trauma, burnout, and career aspirations CF was associated with a diminished desire to practice social work, a decreased likeli- hood that the student would seek employment in a setting similar to the field practicum, and a higher risk of burnout. Those students with higher scores on the self- esteem subscale were less likely to report they would seek employment in a setting similar to their field site. Burnout itself was not associated with either career aspiration variable.

Journal of Baccalaureate Social Work40 Ta

bl e

3 C

Stop Stressing. Your Academic Success is Guaranteed.