Violence and Victims, Volume 18, Number 1, February, 2003

Vicarious Traumatization, Secondary

Traumatic Stress, and Burnout in
Sexual Assault and Domestic Violence
Agency Staff
Stephanie Baird
Sharon Rae Jenkins
University of North Texas, Denton

This study investigated three occupational hazards of therapy with trauma victims: vicar-
ious trauma and secondary traumatic stress (or “compassion fatigue”), which describe
therapists’ adverse reactions to clients’ traumatic material, and burnout, a stress response
experienced in many emotionally demanding “people work” jobs. Among 101 trauma
counselors, client exposure workload and being paid as a staff member (vs. volunteer)
were related to burnout sub-scales, but not as expected to overall burnout or vicarious
trauma, secondary traumatic stress, or general distress. More educated counselors and
those seeing more clients reported less vicarious trauma. Younger counselors and those
with more trauma counseling experience reported more emotional exhaustion. Findings
have implications for training, treatment, and agency support systems.

Keywords: trauma; occupational hazards; domestic violence; therapy; sexual abuse;

burnout

A
ssisting victims of violence can be an emotionally hazardous occupation for the
helpers. Two lines of research may be useful in better understanding the nature of
occupational stressors that may affect these helpers and in developing strategies
to prevent and relieve stress-related difficulties. Burnout is a well-studied response to the
interpersonal stresses of work with people who are in emotionally demanding situations,
such as most human service professions (Cherniss, 1980; Maslach, 1976, 1982, 1996;
Pines, 1993; Pranger & Brown, 1992). More recently, there is growing clinical and
research evidence that counselors who work with traumatized clients may develop reac-
tions specific to the traumatic nature of the clients’ material. Service providers, such as
sexual assault counselors or police officers with high numbers of sexually assaulted
clients, may begin to suffer post-traumatic stress disorder (PTSD) symptoms after fre-
quent exposure to sexual assault survivors (Alexander, de Chasney, Marshall, Campbell,
Johnson, & Wright, 1989; Astin, 1997; Farrenkopf, 1992; Martin, McKean & Veltkamp,
1986; McCann & Pearlman, 1990a, 1990b; Pearlman & Mac Ian, 1995; Remer &
Ferguson, 1995; Schauben & Frazier, 1995). Symptoms such as these have been called
secondary traumatic stress or compassion fatigue (Figley, 1983, 1995) and vicarious trau-
ma (McCann & Pearlman, 1990a, 1990b).

© 2003 Springer Publishing Company 71

72 Baird and Jenkins

The present study examines the presence and degree of trauma-related and burnout
symptoms in both sexual assault and domestic violence agency volunteer and paid staff in
relation to their job roles and their degree of exposure to clients. Some individuals may be
drawn to sexual assault or domestic violence work because they wish to help others
through assaults that they themselves have experienced. However, therapists and coun-
selors having a personal history of sexual assault show more symptoms related to psy-
chological trauma than do non-assaulted personnel (Cunningham, 1996; Follete, Polusny,
& Milbeck, 1994; Kassam-Adams, 1995; Pearlman & Mac Ian, 1995). Burnout, a
response to more tedious kinds of interpersonal stress in helping professions, has not been
found related to trauma symptoms (Schauben & Frazier, 1995), but it has been associated
with work overload, emotional involvement, and being younger or less educated
(Maslach, 1982).

BURNOUT

Maslach (1982) defined and measured burnout as a specific occupational stress syn-
drome occurring when human service professionals become emotionally exhausted,
begin to dehumanize their clients, and lose a sense of personal accomplishment at work
(Maslach Burnout Inventory [MBI], Maslach & Jackson, 1981). The resulting exces-
sive stress or dissatisfaction can lead to psychological withdrawal from work as a cop-
ing strategy (Cherniss, 1980). Burnout is associated with lowered morale and
psychological well-being (Cherniss, 1980), reduced self-esteem and coping, more
daily life hassles (Maslach, 1987), and damaging levels of marital conflict, greater
mental illness, and suicide (Maslach, 1976). Van deWater (1996) found that counselors
feeling adequately trained for trauma work reported less work-related burnout.
According to Schauben and Frazier (1995), working with more sexual violence sur-
vivors was not related to burnout. However, compared to other counselors, sexual
assault therapists report more emotional exhaustion (Johnson & Hunter, 1997). Only
one study investigated burnout in domestic violence shelter workers, finding that,
although they were moderately distressed, the shelter workers did not meet burnout cri-
teria on the MBI (Brown & O’Brien, 1998). Burnout among trauma workers and vol-
unteers is an understudied area.
The main risk factor for burnout appears to be employment in a “people work” job that
presents routine high levels of interpersonal demands and inadequate structural support
for meeting those demands. Work overload has been related to burnout in hospital and
community mental health counselors (Prosser, Johnson, Kuipers, Szmuckler, Bebbington,
and Thorncroft, 1997), in health care professionals working with AIDS (Bennett, Kelaher,
& Ross, 1993), and in drug and alcohol service employees (Price & Spence, 1994, p. 67).
A high caseload and longer work hours, specifically involving continuous, direct contact
with clients, have been related to greater stress and more negative attitudes (Maslach,
1976).
Positive aspects of job structure, on the other hand, may ameliorate burnout. Work
autonomy has been related to lower burnout in hospital social workers (Oktay, 1992).
Social support (i.e., supervision, communication, praise) is generally correlated with
lower burnout in counselors (Ross, Altmaier, & Russell, 1989), AIDS assistance volun-
teers (Maslanka, 1996), and residential counselors for emotionally disturbed children
(Kruger, Botman, & Goodenow, 1991).
Staff Trauma, Stress, and Burnout 73

Among personal characteristics, higher education, greater work expectations, and

younger age have all been associated with burnout (Bennett, Kelaher, & Ross, 1993;
Maslach, 1982; Maslanka, 1996; Oktay, 1992; Price & Spence, 1994). There is substan-
tial evidence that length of counseling experience is related to levels of burnout, particu-
larly among experienced counselors (Jupp & Shaul, 1991; Pines & Maslach, 1978).
However, less experienced therapists are also at risk, possibly due to the inexperience with
handling stressors and difficulties (Farber, 1990; Ross, Altmaier, & Russell, 1989).

SECONDARY TRAUMATIC STRESS/COMPASSION FATIGUE

Figley (1983) defined secondary traumatic stress, which he later called compassion
fatigue (Figley, 1995), as the experiencing of emotional duress in persons who have had
close contact with a trauma survivor, which may include family members as well as ther-
apists. Secondary traumatic stress disorder (STSD) symptoms include re-experiencing the
survivor’s traumatic event, avoidance, and/or numbing in response to reminders of this
event, and persistent arousal (Figley, 1995). These are nearly identical to the symptoms of
PTSD, except that the person suffering the trauma may develop PTSD, while the person
close to the victim/survivor, through hearing about the trauma, may develop the corollary
syndrome, STSD. One early example was Alexander and colleagues (1989) descriptions
of the parallels between the experiences of sexual assault survivors and the symptomatic
reactions of research nurses who coded file data describing the cases but who never met
the survivors.
Figley (1995) developed an instrument to measure secondary traumatic stress, the
Compassion Fatigue Self-Test for Psychotherapists (CFST), which has two sub-scales tap-
ping the dimensions of compassion fatigue and burnout. This instrument is self-adminis-
tered, self-scored, and self-interpreted; scoring instructions and interpretations are
provided in Figley, 1995. Most of the questions inquire about general feelings relating to
experiences of trauma; some request this information specifically in relation to working
with trauma clients. Other questions were designed to identify features of burnout, partic-
ularly in relation to the work environment. However, the conceptualization of burnout that
was used differs from that measured by the MBI, and the MBI and the CFST’s burnout
sub-scale are not highly correlated (Jenkins & Baird, 2002).

VICARIOUS TRAUMA

In developing their construct of vicarious trauma, McCann and Pearlman (1990a) pro-
posed that the therapist’s cognitive world can be altered by verbal exposure to the client’s
traumatic material. The main symptoms of vicarious trauma involve cognitive shifts that
may be paralleled by intrusive imagery rather than the full spectrum of PTSD symptoms.
The cognitive content areas in which losses or disruptions occur are safety, trust, esteem,
intimacy, and control regarding both self and others. These cognitive schema changes may
have a negative effect on the therapist’s feelings, relationships, and nonwork life as well
as work with clients. Painful images and emotions related to the client’s traumatic mem-
ories may become incorporated into the therapist’s imagery system of memory. This re-
experiencing or avoidance of specific aspects of their client’s traumatic memories
becomes tangible via flashbacks, dreams, painful emotions, or intrusive thoughts.
74 Baird and Jenkins

Using their constructivist self-development theory (McCann & Pearlman, 1990b),

Pearlman (1996) and her colleagues developed and refined the Traumatic Stress Institute
Belief Scale, Revision L (TSI-BSL), to measure the five need-schema content areas of
vicarious trauma. Each has two sub-scales, one regarding the self and the other regarding
other people. Lobel’s (1997) qualitative study of vicarious effects in 10 therapists treating
female survivors of adult sexual assault found that 70% experienced vicarious trauma seen
in negative, long-term changes in cognitive schema (particularly the frame of reference
schema) and 80% reported positive changes in schema (particularly the esteem schema)
and vicarious enrichment (positive effects from work).

RISK FACTORS FOR COUNSELORS’ ADVERSE REACTIONS TO

TRAUMATIZED CLIENTS

Research to date on secondary traumatic stress has used both Figley’s measure (the CFST)
and the Impact of Event Scale (IES; Horowitz, Wilner, & Alvarez, 1979; Zilberg, Weiss,
& Horowitz, 1982), a widely used measure of PTSD symptoms. Kassam-Adams (1995)
focused on psychotherapists’ work with sexually traumatized clients, whereas Lee (1996)
recruited marriage and family counselors as participants, and Good (1996) a sample of art
therapists and other mental health professionals. Several studies have used the TSI-BSL
to examine vicarious trauma in therapists working with victims of sexual violence (Brady,
Guy, Poelstra, & Brokaw, 1999; Cunningham, 1996; Schauben & Frazier, 1995; Simonds,
1997), whereas others have evaluated symptoms in relation to trauma survivors more gen-
erally (Lee, 1996; Pearlman & Mac Ian, 1995; Walton, 1997).
The major identified risk factors for secondary traumatic stress and vicarious trau-
ma in this still emerging research literature are being less experienced, having a per-
sonal trauma history, and greater exposure to traumatized individuals, usually
represented by having a heavier caseload of survivors. Less experienced therapists may
identify more with the client than with the role of a therapist; for them, coping with the
transference and countertransference dynamics of trauma therapy may present over-
whelming strain (Neuman & Gamble, 1995). (Countertransference differs from sec-
ondary traumatic stress and vicarious trauma in that countertransference focuses on the
possible consequences of the counselor’s past experiences for the client. Secondary
traumatic stress and vicarious trauma are concerned with the negative ramifications of
exposure to the client for the counselor.) Distinguishing these may be extremely chal-
lenging for the less experienced or previously traumatized therapist. Neuman and
Gamble describe countertransference reactions common in new therapists that may
make trauma counseling especially challenging. On the theory that more experienced
counselors should be less vulnerable to symptoms themselves, Good (1996) found that
therapists who had more years in practice showed lower secondary traumatic stress
symptom levels. More experienced therapists typically had lower vicarious trauma lev-
els (Simonds, 1997), especially if they had more experience treating trauma victims
(Cunningham, 1996; Pearlman & Mac Ian, 1995).
Several researchers have investigated the service provider’s own history of victimiza-
tion as a vulnerability factor for negative reactions to clients’ traumas linked to greater lev-
els of PTSD-like symptoms. Therapists who are survivors of violence may have
particularly difficult responses to traumatized clients’ disclosures. Professionals in both
Staff Trauma, Stress, and Burnout 75

law enforcement and mental health having a history of childhood abuse have reported sig-
nificantly higher levels of trauma survivor-like symptoms than their nonabused counter-
parts (Follette, Polusny, & Milbeck, 1994). Counselors have shown more symptoms of
secondary traumatic stress if they have a personal history of traumatic events (Kassam-
Adams, 1995) or PTSD symptoms related to previous events (Good, 1996). By some
reports, counselors with a history of personal victimization had more vicarious trauma
(Cunningham, 1996; Lee, 1996; Pearlman & Mac Ian, 1995), but other findings did not
substantiate this (Schauben & Frazier, 1995; Simonds, 1997).
Regarding caseload exposure, two studies have found PTSD symptoms related to
amount of work done with trauma survivors, Kassam-Adams (1995) with sexually trau-
matized clients, and Lee (1996) with a variety of traumatic events common to clients of
marriage and family counselors. Trauma client workload has been related inconsistently
to vicarious trauma; some studies found a positive association (Cunningham, 1996;
Pearlman & Mac Ian, 1995; Schauben & Frazier, 1995), but others did not (Brady, Guy,
Poelstray, & Brokaw, 1999; Lee, 1996) or were inconclusive (Simonds, 1997).

THE PRESENT STUDY

Secondary traumatic stress, vicarious trauma, and burnout are all forms of occupational
stress that may affect those working with victims of violence because that work involves
direct exposure to interpersonally demanding clients. They are similar in being more com-
mon among younger and/or less experienced individuals, and they can result in lowered
job satisfaction, increased job turnover, absenteeism, and the deterioration of client serv-
ice or care. Secondary traumatic stress and vicarious trauma refer to the same observed
phenomena: the changes in a therapist or other person who is exposed to someone recov-
ering from a traumatic event. However, they differ in that secondary traumatic stress
emphasizes clinically observed DSM-IV PTSD-based symptomatology of relatively sud-
den onset, whereas vicarious trauma, a theory-driven construct, emphasizes more gradual,
covert, and permanent changes in cognitive schema. In contrast, burnout has not been
linked specifically to victimization history or trauma exposure (Schauben & Frazier,
1995), but to workplace variables such as job structure, workload, and lack of support in
human service occupations. Secondary traumatic stress and vicarious trauma have not yet
been related to workplace conditions. The present study tested hypotheses highlighting
these similarities and differences:
Hypothesis 1. Less experienced and younger personnel will report more secondary
traumatic stress, vicarious trauma, general distress, and burnout compared to more expe-
rienced and older personnel.
Hypothesis 2. Greater workload exposure to sexual assault and/or domestic violence
survivors is correlated with higher rates of secondary traumatic stress, vicarious trauma,
burnout, and general distress.
We also explored comparisons of volunteer and paid staff, a distinction rarely made in
this literature. We assumed that the latter would have more exposure to traumatized clients
than volunteer staff and thus would be more at risk for these forms of occupational stress.
However, they may also be more highly trained and experienced, which may outweigh stress
exposure. Furthermore, compared to volunteers, paid staff may experience more structural
support from their agencies and each other that would also buffer them against stress.
76 Baird and Jenkins

METHOD

Participants
Of the 101 participants, 35% (n = 35) were sexual assault counselors, and 17% (n = 17)
were domestic violence counselors. Almost half (48%, n = 49) indicated that they worked
at a dual-purpose sexual assault and domestic violence agency. Almost half were volun-
teers (45.5%) and more than half paid (63.4%), with some people who both volunteered
and were paid at their centers (14.9%). Participants’ titles included counselor, therapist,
psychologist, intern, crisis worker, hotline worker, caseworker, case manager, supervisor,
director, and educator. There were 96 females and four males, and one who marked nei-
ther, with ages normally distributed from 21 to 65 (see Table 1). Over half (51.5%) were
married. The majority were heterosexual (93%), Anglo-American (77.2%), and/or
Protestant (65.3%). Most held a bachelor’s (36.6%) or master’s degrees (46.5%) in a men-
tal health field (62.4%) and averaged a yearly income of $20,000 or more (75.3%) or
$10,000 or under (12.9%). Additional analyses of this data set are reported in Jenkins and
Baird (2002).
Comparison of the means and standard deviations of the basic demographic data with
previously published research indicates that the sample does not differ radically from
other similar populations and norms, supporting external validity. Characteristics of the
total populations from each participating agency were obtained in order to compare the
pooled sample with the total population. The demographic characteristics of this sampled
population did not deviate notably from the total populations volunteering or working at
the centers.

Procedure
The agencies’ human resource/volunteer directors were approached via letter and follow-
up telephone call as to the center’s interest in research participation. Following confirma-
tion of participation, data-gathering visits were arranged; most coincided with scheduled
staff meetings or volunteer in-services. The researcher typically attended the meetings and
introduced the topic of the study, distributed questionnaires and consent forms, and
assured participants of confidentiality. After gathering completed questionnaires, the
researcher answered questions, provided participants with a lecture about vicarious trau-
ma, and facilitated a discussion regarding methods of prevention and intervention.
Participant response rate is difficult to calculate, as questionnaires were mostly distrib-
uted and collected on the same occasion. On approximately four (out of 11) distribution
opportunities (approximately 10 meetings and one mailing), questionnaires were distrib-
uted at one meeting and collected the following week or left with staff and volunteers to
return via mail. Response rates for these occasions ranged from approximately 50% to
100%.

Instruments
Sexual Assault/Domestic Violence Agency History and Activities (Exposure to
Clients). Subjects provided information concerning their work or counseling experience,
particularly regarding the past month. Questions regarded:
1. weeks spent working at this center;
2. time since last meaningful contact with trauma survivor;
Staff Trauma, Stress, and Burnout 77

TABLE 1. Sample’s Demographic Characteristics

Demographic Variables (N = 101) f (%)
Gender
Female 96 95.0
Male 4 4.0
Missing Data 1 1.0
Age
21-25 16 16.0
26-35 21 21.0
36-45 24 24.0
46-55 22 22.0
56-65 9 9.0
Missing Data 9 9.0
Marital Status
Single (never married, committed) 18 17.8
Married 52 51.5
Living w/Partner 8 7.9
Divorced/Separated/Widowed 22 21.8
Missing Data 1 1.0
Race/Ethnicity
Asian 1 1.0
African-American/Black 3 3.0
Anglo-American/White 78 77.2
Hispanic 9 8.9
Middle-Eastern/European 2 2.0
Missing Data 8 7.9
Education
High School/GED 14 13.9
College Graduate 37 36.6
Master’s Degree 47 46.5
Doctorate 1 1.0
Missing Data 2 2.0
Income
less than $10,000 13 12.9
$10,001 to $15,000 5 5.0
$15,001 to $20,000 2 2.0
$20,001 to $30,000 35 34.7
more than $30,000 41 40.6
Missing Data 5 5.0

3. numbers of clients counseled in an average week in the past month; and

4. average hours per week (in the last month) spent working directly with this popu-
lation in various capacities.

The questions regarding weekly counseling and hourly exposure to clients were combined
into a single index of direct counseling exposure by adding together the hours per average
week (in the past month) spent engaged in three types of direct counseling (individual
therapy, group therapy, crisis intervention).
78 Baird and Jenkins

Compassion Fatigue Self-Test for Psychotherapists. Participants completed the

Compassion Fatigue Self-Test for Psychotherapists (CFST) developed by Charles Figley to
assess the degree of secondary traumatization/compassion fatigue in professionals who work
with trauma survivors. This instrument contains two sub-scales, providing measures of com-
passion fatigue/secondary traumatic stress (CFST-CF) and burnout (CFST-BO). For the pres-
ent study, all responses have also been added together for a total scale score (CFST-SUM).
This instrument employs a five-point Likert scale (1 = rarely; 5 = very often) response sys-
tem and contains 40 questions. Reported internal consistency reliability ranges from .94 to .86
(Figley, 1995). “Therapist” was changed to “staff and volunteer” (or some variation). For the
current study, Cronbach’s alphas were .84 for the CF sub-scale, .83 for the burnout sub-scale,
and .90 for the total summed score (CFST-SUM). Validity analyses reported by Jenkins and
Baird (2002) supported the CFST-CF scale as a measure of secondary traumatic stress, but
they found that CFST-BO does not correlate highly with MBI scores.
TSI Belief Scale, Revision L. Vicarious traumatization was measured with the TSI Belief
Scale, Revision L (TSI-BSL; Pearlman, 1996). This 80-item, seven-point Likert scale (0 =
disagree strongly, 6 = agree strongly) instrument measures disruptions in five cognitive
schemas areas, each regarding self and others, yielding 10 sub-scales. A total score was cal-
culated from the sum of all responses; a higher score indicates greater disruption. Reported
internal consistency reliability overall is .98, with sub-scale reliabilities from .77 to .91. For
the current study, a Cronbach’s alpha of .95 for the total score and alphas from .62 to .83 for
each of the 10 sub-scales were calculated. Jenkins and Baird (2002) found a strong correla-
tion of TSI-BSL total scores with CFST-CF scores but not with MBI scores.
The Maslach Burnout Inventory. The Maslach Burnout Inventory (MBI) is a 22-
item, self-report inventory which produces three scales and a total burnout score. The
Emotional Exhaustion (EE) scale entails being mentally and emotionally overextended
and exhausted by one’s work. Depersonalization (DP) refers to a detached and imper-
sonal response toward one’s clients. Personal Accomplishment (PA) is related to enjoy-
ment of competence and success in a job working with people (Maslach, 1996) and is
reversed to indicate burnout. The MBI questions the respondent on the frequency (how
often) with which various feelings related to burnout occur during their work year. The
seven-point Likert scale for the MBI ranges from never (0) to every day (6). Each of the
three sub-scale scores is the sum of designated responses. The total score is the sum of
all responses with the PA item scores reversed.
Validity and reliability is supported by alpha coefficients of .86 (Kruger et al., 1991)
and .83 (Maslach & Jackson, 1981) for the total scale. Other research has shown that the
MBI has adequate test-retest reliability, and convergent and discriminant validity
(Maslach & Jackson, 1981; and reviewed in Pranger & Brown, 1992). The current study
obtained Cronbach’s alphas of .91 for the Emotional Exhaustion (MBI-EE) sub-scale, .81
for Depersonalization (MBI-DP), .92 for Personal Accomplishment (MBI-PA), and .91 for
the total score (MBI-SUM).
The Symptom Checklist-90, Revised. Symptoms of general psychological distress
were assessed utilizing the Symptom Checklist-90, Revised (SCL-90-R). This widely
used multidimensional self-report inventory contains 90 Likert scaled items rated from 0
(not at all) to 4 (extremely). The SCL-90-R examines general psychological symptom pat-
terns for the last seven days. The Global Severity Index (GSI), derived by averaging all
responses, was chosen as the most sensitive indicator of general psychological distress
(Derogatis, 1983). For the current study, a Cronbach’s alpha of .77 for the GSI was cal-
culated.
Staff Trauma, Stress, and Burnout 79

RESULTS

Descriptive Statistics
Volunteer and paid staff differed significantly on several demographic variables. More
educated personnel were more likely to be paid rather than volunteers (see Table 2), and
were more likely to be counselors and not crisis workers. Regarding occupational vari-
ables, paid staff had seen clients more recently, spent more hours per average week coun-
seling clients, had more clients per week, more often worked primarily as counselors
(46%) or managers (31%) and less often crisis workers (18%), had been with the agency
longer, and had more paid counseling experience, both in general and with trauma sur-
vivors. Volunteer workers were more often crisis workers (65%) than counselors (22%).
Paid staff and volunteer staff did not differ in personal trauma histories or self-ratings on
the CFST, TSI-BSL or SCL-90-R GSI. Paid staff had significantly higher scores on all
three MBI sub-scales but not on MBI-SUM than did people who only volunteered. More
educated participants also had lower TSI-BSL total scores, r(99) = -.33, p < .001, and
higher MBI-EE and MBI-PA scores than did those who were less educated (see Table 3).
When the above differences in educational level, counseling experience, and client expo-
sure were controlled using ANCOVA, only MBI-EE differentiated paid workers from
volunteers.

TABLE 2. T-Test Comparisons of Paid Staff and Volunteers

Measure Group n M SD t rpb
Educationa Paid 64 2.5 0.7 2.81** .27**
volunteer 39 2.0 0.8
Paid counseling Paid 65 51.0 73.5 3.62*** .30**
experienceb volunteer 38 9.3 43.4
Paid trauma counseling Paid 65 47.0 62.9 5.70*** .40***
experienceb volunteer 38 2.0 7.0
Exposure hours/week Paid 64 14.1 10.9 4.69*** .40***
volunteer 38 5.3 8.0
N of clients/week Paid 62 10.4 8.4 3.98*** .35***
volunteer 36 4.4 6.3
MBI-SUM Paid 63 27.7 15.3 1.26 .12
volunteer 37 24.3 11.8
MBI-EE Paid 63 16.7 9.7 4.95*** .41***
volunteer 37 8.6 6.6
MBI-DP Paid 63 3.7 4.4 3.00** .25**
volunteer 37 1.7 2.2
MBI-PA Paid 63 40.6 5.9 2.94** .33**
volunteer 37 34.0 12.9
Note. MBI = Maslach Burnout Inventory. SUM = Total score. EE = Emotional exhaus-
tion. DP = Depersonalization. PA = Personal accomplishment.
aCoded 1 = HS diploma or GED, 2 = college degree, 3 = master’s degree, 4 = doctorate.
bIn months
**p < .01. ***p < .001.
80 Baird and Jenkins

TABLE 3. Associations (Point Biserial Correlations) of Educational Level and Job

Role With Burnout
MBI- MBI- MBI- MBI-
SUM EE DP PA
Education .00 .26** .13 .29**
Job role (paid staff only, n = 63)
Counselor .05 -.00 .09 -.07
Crisis worker -.29* -.40*** -.16 -.03
Job role (volunteers only, n = 35)
Counselor -.15 .50** .23 .43**
Crisis worker .12 -.52*** -.20 -.41*
Note. MBI = Maslach Burnout Inventory. SUM = Total score. EE = Emotional exhaus-
tion. DP = Depersonalization. PA = Personal accomplishment.
*p < .05. **p < .01. ***p < .001 (1-tailed).

Volunteers in counseling and crisis job roles showed contrasting patterns of association
with burnout sub-scales but not overall burnout (Table 3). Volunteers (but not paid staff)
in counseling roles (compared to other roles) had more emotional exhaustion but also
more personal accomplishment. Volunteers in crisis roles displayed the opposite pattern.
Among paid staff, crisis workers were less likely than others to rate themselves as burned
out and also less likely to rate highly on the CFST-SUM (r(62) = -.25, p < .05).

Hypothesis-Testing Analyses
Hypothesis 1. Less experienced and younger personnel will report more secondary
traumatic stress, vicarious trauma, burnout, and general distress compared to more expe-
rienced and older personnel. This hypothesis was not supported for experience and par-
tially supported for age. More experienced workers, generally and with trauma
specifically, reported greater MBI-PA (r(96) = .22 and .24) respectively, both p < .05, and
more trauma-experienced workers showed more MBI-EE (r(96) = .22, p < .05). However,
months of paid counseling experience and of trauma counseling were not related to any
overall symptom scales. Compared to their older coworkers, younger participants often
had higher MBI-SUM and MBI-EE scores (r (88) = -.22 and -.23) respectively and both
p < .05, but other scores did not differ.
Hypothesis 2. Greater occupational exposure to sexual assault and/or domestic violence
survivors will be correlated with higher rates of secondary traumatic stress, vicarious trau-
ma, burnout, and general distress symptomatology. This hypothesis was not supported.
Using Pearson correlations, hours counseling clients and numbers of clients in an average
week were both significantly and positively related to the MBI-EE and MBI-PA scores but
not to overall burnout (see Table 4). An unexpected and disconfirming negative relationship
between number of clients and TSI-BSL total score was significant (r (94) = -.21, p < .05).
Neither the SCL-90-R nor any of the CFST scales was related to client exposure.
Separate correlation matrices of exposure and outcome variables were produced for
paid staff and volunteer staff, revealing stronger associations for volunteer than paid staff.
Volunteers who spent more hours counseling per average week had higher scores on the
Staff Trauma, Stress, and Burnout 81

TABLE 4. Associations of Client Exposure With Burnout

MBI- MBI- MBI- MBI-
SUM EE DP PA
Exposure (paid staff only, n = 61-62)
Hours -.10 -.00 -.03 .23
Clients .04 .14 .08 .19
Exposure (volunteers only, n = 34-35)
Hours -.26 .53*** .40* .56***
Clients -.42* .22 .10 .51**
Hours = number of hours in an average week spent counseling clients.
Clients = number of clients counseled in average week.
*p < .05. **p < .01. ***p < .001 (1-tailed).

MBI-EE, DP, and PA sub-scales, but they did not have higher MBI-SUM scores (see Table
4). Contrary to the hypothesis, volunteers (but not paid staff) who saw more clients had
higher MBI-PA scores and lower MBI-SUM scores.

Follow-Up Multiple Regression Analyses

To examine the relative importance of the various correlates of the symptom scales, each
was regressed on the set of variables having significant bivariate associations with it.
Educational level was the strongest predictor of TSI-BSL scores, and only number of
clients contributed even marginally (p < .10) significant unique variance; together, they
explained 14% of the variance in TSI-BSL scores. For CFST-SUM, no variables were sig-
nificant. For MBI-SUM, only age was significant, and not being a crisis worker added
marginally significant unique variance (p < .10), with 7% of the variance explained.
Among the MBI sub-scales, both age and being a crisis worker contributed significant
unique variance to MBI-EE, together explaining 31% of the variance. Both being a crisis
worker and hours of client exposure explained significant unique variance in MBI-PA, and
they accounted for 21% of the variance together. Counseling experience and being paid
vs. volunteer staff did not add significantly to any of these analyses.

DISCUSSION

This study was conceived to examine the presence and correlates of secondary traumatic
stress, vicarious trauma, burnout, and general distress, comparing volunteer and paid staff
working with sexual assault and/or domestic violence survivors. Hypothesis 1 found experi-
ence and age unrelated to any of the symptom total scales with the exception of younger par-
ticipants showing a little more burnout. However, more trauma-experienced workers reported
both more emotional exhaustion and correspondingly greater feelings of personal accom-
plishment. Hypothesis 2 investigated symptoms relating to client exposure and found com-
plex patterns that were more often contrary to expectations. Workers who saw more clients
actually had fewer vicarious trauma symptoms. Overall burnout was likewise unexpectedly
related to seeing fewer clients, but only for volunteers. Also only for volunteers, seeing more
clients for more hours was related to greater self-rated personal accomplishment.
82 Baird and Jenkins

For Hypothesis 2, workers with more exposure to sexual assault/domestic violence

clients showed no evidence of greater secondary traumatic stress, vicarious trauma, over-
all burnout, or general distress. Previous studies testing this hypothesis for trauma-related
symptoms usually used measures such as percentage of client caseload having traumatic
experiences, but in the present sample this was uniformly 100%. Measuring the severity
of trauma symptoms among this sample’s clients would have offered a more precise meas-
ure of exposure. Secondary traumatic stress or vicarious trauma should be more common
among counselors seeing more severely traumatized clients, but it was not possible to
measure client characteristics in this study. More interesting in these results is the pattern
of correlations between exposure variables and burnout sub-scales, which is suggestive of
a balance between stressors and satisfaction. Management and human resources practices
that support volunteers’ feelings of personal accomplishment may provide a good antidote
to the emotional exhaustion that Johnson and Hunter (1997) found to be characteristic of
this particularly stressful work. The rewarding work may provide a buffer against the
development of negative effects, consistent with the theoretical description of the
sequence of burnout processes offered by Maslach (1982).
Both in theory and previous research, burnout is related to general workplace conditions
not measured here, including chronic, nontraumatic, interpersonal demands from clients and
other people and lack of structural social support for stress management. Of the job features
included, job role (not being a crisis worker, but rather a counselor or manager) explained
the most variance in burnout for the sample as a whole. Other workplace conditions, for
example, for managers supervising stressed personnel, may explain workers’ burnout better
than clients’ trauma symptoms do. Among these stressors for volunteers, especially crisis
workers, may be idleness; “routine” caseloads were highly variable over time, and 15 indi-
viduals (who comprised 35% of volunteers and 42% of crisis workers) had not seen a client
for at least two weeks. Volunteers who saw fewer clients had higher MBI-SUM scores.
The findings suggest that burnout-producing demands may be greater for most paid
staff than for most volunteer staff, but so are the rewards of personal accomplishment that
go with heavier caseloads. Paid staff had significantly higher MBI-EE and MBI-DP scores
than volunteers did. But because they also reported greater experience of personal accom-
plishment, they did not have higher rates of overall burnout. Apparently, despite some neg-
ative effects (high EE and DP), they were still able to experience stress-mitigating
personal rewards (high PA) from the work, as were volunteer counselors and those who
saw more clients for more hours.
The surprising association for volunteers of high overall burnout with seeing fewer
clients appears tied to the related feeling of low personal accomplishment that may act as
a burnout prophylactic for busier volunteers, especially counselors. Volunteer crisis work-
ers seem more at risk than their counterparts who are paid staff, as the latter showed less
burnout than other paid staff. Although volunteers had lighter caseloads on average than
paid staff, they may also have less access to structural support mechanisms that would
reinforce their feeling of personal accomplishment, such as case conferences, supervision,
and daily contact with staff for the caseloads they do have. Thus, they may be more vul-
nerable to burnout. In addition, volunteers may be a more heterogeneous group, on aver-
age less educated, much less experienced, and perhaps including individuals who are more
easily affected by trauma-related situations than paid staff are. Volunteers with more hours
of client contact had higher rates of depersonalization, consistent with previous research
that long-term involvement in emotionally demanding situations relates to feelings of
depersonalization toward one’s clients (Pines, 1993).
Staff Trauma, Stress, and Burnout 83

Unexpectedly for vicarious trauma, those who saw more clients had fewer cognitive
disruptions/changes (TSI-BSL total score), inconsistent with previous research which
has demonstrated the opposite effect (Cunningham, 1996; McCann & Pearlman, 1990a,
1990b; Pearlman & Mac Ian, 1995; Schauben & Frazier, 1995) and with the hypothesis
that greater exposure is related to greater vicarious trauma. However, this likely reflects
the prophylactic effect of education. Counselors were more highly educated, and more
educated personnel had lower vicarious trauma scores. Other studies have used a nar-
rower range of education. This finding is not an artifact of workload differences between
volunteer and paid staff because they did not differ in reported vicarious trauma symp-
toms. This sample’s participants may possess adequate coping skills or may be involved
with agencies having good staff support mechanisms. One such adaptive coping mecha-
nism by either workers or supervisors consistent with these results is to cut back one’s
client caseload if vicarious trauma symptoms are detected. Future research using repeat-
ed measures over time is needed to trace the personal, social, and structural processes
involved in the development of symptoms in relation to changes in workload patterns.
The breadth of the sample selected is a significant strength of this study, as it appears
representative of sexual assault and domestic violence agency paid and volunteer staff.
The inclusion of volunteers and domestic violence workers is a particular strength atypi-
cal for research in this area, which has mainly focused on paid sexual assault counselors
(Cunningham, 1996; Kassam-Adams, 1995; Pearlman & Mac Ian, 1995; Schauben &
Frazier, 1995). Most of the agencies studied appear to recognize the frequent joint occur-
rence of sexual assault and domestic violence, generally offering services for survivors of
either assault type, particularly as domestic violence survivors often also experience sex-
ual assault in the course of the domestic violence.
Future research should refine these questions by including measures of client charac-
teristics such as the specific nature of the abuse, not only sexual assault/domestic violence
but also the degree (injury-threat, life-threat) and kind (verbal, emotional, physical) of vio-
lence involved, and the severity of clients’ actual trauma symptoms (PTSD). Matching
these characteristics to those of traumatized therapists would indicate whether therapists
are more vulnerable or more resistant to secondary traumatic stress or vicarious trauma
when clients report traumas more similar to the counselor’s own. Asymptomatic clients
may evoke less secondary traumatic stress or vicarious trauma.
Few individuals in this sample reached clinically significant levels of symptomatology;
this sample scored similarly to other mental health professionals on these measures of dis-
tress. It seems likely that this relatively low prevalence of serious problems accurately rep-
resents the general population of trauma workers, most of whom probably continue to
function well in their roles on a day-to-day basis. Probably only a few are severely symp-
tomatic at any one time because appropriate supervision would promptly reduce their
caseloads, move them out of counseling roles, or encourage them to seek other careers.
Despite the low symptom levels, the topic of the study was clearly compelling to the par-
ticipants, as shown by their responses to the researcher’s presentation following data collec-
tion on the symptoms, prevention, and intervention of secondary traumatic stress, vicarious
trauma, and burnout. Often during the subsequent discussion, participants expressed tremen-
dous interest in the topic, making connections between the information provided and their
own lives. Realizations of the negative and positive aspects of working with this difficult
population were often voiced, leading to greater insight for both the participants and inves-
tigator. Future research might productively examine the nature of specific stressful trauma
counseling experiences, as well as chronic stressors, reported by this population in relation
to secondary traumatic stress, vicarious trauma, and burnout symptoms.
84 Baird and Jenkins

REFERENCES

Alexander, J. G., de Chasney, M., Marshall, E., Campbell, A. R., Johnson, S., & Wright, R. (1989).
Research note: Parallel reactions in rape victims and rape researchers. Violence and Victims, 4,
57-62.
Astin, M. C. (1997). Traumatic therapy: How helping rape victims affects me as a therapist. Women
and Therapy, 20, 101-109.
Bennett, L., Kelaher, M., & Ross, M. (1993). Burnout and coping in HIV/AIDS health care profes-
sionals. In H. van Dis & E. van Dongen (Eds.), Burnout in HIV/AIDS health care and support:
Impact for professionals and volunteers (pp. 41-51). Amsterdam: University of Amsterdam Press.
Brady, J. L., Guy, J. D., Poelstra, P. L., & Brokaw, B. F. (1999). Vicarious traumatization, spiritual-
ity, and the treatment of sexual abuse survivors: A national survey of women psychotherapists.
Professional Psychology: Research and Practice, 30, 386-393.
Brown, C., & O’Brien, K. M. (1998). Understanding stress and burnout in shelter workers.
Professional Psychology: Research and Practice, 29, 4, 383-385.
Cherniss, C. (1980). Staff burnout. Beverly Hills, CA: Sage.
Cunningham, M. (1996). Vicarious traumatization: Impact of trauma work on the clinician.
Unpublished master’s thesis, Adelphi University School of Social Work, Garden City, NY.
Derogatis, L. R. (1983). SCL-90-R administration, scoring & procedures manual-II for the r(evised)
version and other instruments of the psychopathology rating scale series. Towson, MD:
Clinical Psychometric Research.
Farber, B. A. (1990). Burnout in psychotherapists: Incidence, types, and trends. Psychotherapy in
Private Practice, 8, 35-44.
Farrenkopf, T. (1992). What happens to therapists who work with sex offenders? Journal of
Offender Rehabilitation, 18, 217-223.
Figley, C. R. (1983). Catastrophes: An overview of family reactions. In C. R. Figley & H. I.
McCubbin (Eds.), Stress and the Family II: Coping with Catastrophe (pp. 3-20). New York:
Brunner/Mazel.
Figley, C. R. (1995). Compassion fatigue as secondary traumatic stress disorder: An overview. In C.
R. Figley (Ed.), Compassion fatigue: Coping with secondary traumatic stress disorder in those
who treat the traumatized (pp. 1-20). New York: Brunner/Mazel.
Follette, V. M., Polusney, M. M., & Milbeck, K. (1994). Mental health and law enforcement pro-
fessionals: Trauma history, psychological symptoms, and impact of providing services to child
sex abuse survivors. Journal of Clinical and Consulting Psychology, 25, 275-282.
Good, D. A. (1996). Secondary traumatic stress in art therapists and related mental health profes-
sionals. Unpublished dissertation, University of New Mexico, Albuquerque.
Horowitz, M. J., Wilner, N., & Alvarez, W. (1979). Impact of event scale: A measure of subjective
stress. Psychosomatic Medicine, 41, 209-218.
Jenkins, S. R., & Baird, S. (2002). Secondary traumatic stress and vicarious trauma: A validational
study. Journal of Traumatic Stress, 15, 423-432.
Johnson, C. N. E., & Hunter, M. (1997). Vicarious traumatization in counsellors working in the New
South Wales Sexual Assault Service: An exploratory study. Work and Stress, 11, 319-328.
Jupp, J., & Shaul, V. (1991). Burn-out in student counsellors. Counseling Psychology Quarterly, 4,
157-167.
Kassam-Adams, N. (1995). The risks of treating sexual trauma: Stress and secondary trauma in
psychotherapists. Unpublished doctoral dissertation, University of Virginia, VA.
Kruger, L., Botman, H., & Goodenow, C. (1991). An investigation of social support and burnout
among residential counselors. Child & Youth Care Forum, 20, 335-352.
Lee, C. S. (1996). Secondary traumatic stress in therapists who are exposed to client traumatic mate-
rial. Unpublished dissertation, Florida State University, Tallahassee.
Lobel, J. A. (1997). The vicarious effects of treating female rape survivors: The therapist’s perspec-
tive. Unpublished dissertation, University of Pennsylvania, Philadelphia.
Staff Trauma, Stress, and Burnout 85

Martin, C. A., McKean, H. E., & Veltkamp, L. J. (1986). Post-traumatic stress disorder in police and
working with victims: A pilot study. Journal of Police Science and Administration, 14, 98-101.
Maslach, C. (1976, September). Burned-out. Human Behavior, 5, 16-22.
Maslach, C. (1982). Burnout: The cost of caring. Englewood Cliffs, NJ: Prentice-Hall.
Maslach, C. (1987). Burnout research in the social services: A critique. In D. F. Gillepsie (Ed.),
Burnout among social workers (pp. 95-105). New York: Halworth Press.
Maslach, C. (1996). The Maslach Burnout Inventory (3rd ed.). Palo Alto, CA: Consulting
Psychologists Press.
Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of
Occupational Behaviour, 2, 99-113.
Maslanka, H. (1996). Burnout, social support, and AIDS volunteers. AIDS Care, 8(2), 195-206.
McCann, L., & Pearlman, L. A. (1990a). Vicarious traumatization: A framework for understanding
the psychological effects of working with victims. Journal of Traumatic Stress, 3, 131-149.
McCann, L., & Pearlman, L. A. (1990b). Psychological trauma and the adult survivor. New York:
Brunner/Mazel.
Neuman, D. A., & Gamble, S. J. (1995). Issues in the professional development of psychotherapists:
Countertransference and vicarious traumatization in the new trauma therapist. Psychotherapy,
32, 341-347.
Oktay, J. S. (1992). Burnout in hospital social workers who work with AIDS patients. Social Work,
37(5), 433-439.
Pearlman, L. A. (1996). Psychometric review of TSI Belief Scale, Revision L. In B. H. Stamm (Ed.),
Measurement of stress, trauma, and adaptation (pp. 415-417). Lutherville, MD: Sidran Press.
Pearlman, L. A., & Mac Ian, P. S. (1995). Vicarious traumatization: An empirical study of the effects
of trauma work on trauma therapists. Professional Psychology: Research and Practice, 26,
558-565.
Pines, A. M. (1993). Burnout: An existential perspective. In W. B. Schaufeli, C. Maslach, & T.
Marek (Eds.), Professional burnout: Recent developments in theory and research. Series in
applied psychology: Social issues and questions (pp. 33-51). Washington, DC: Taylor &
Francis.
Pines, A., & Maslach, C. (1978). Characteristics of staff burnout in mental health settings. Hospital
& Community Psychiatry, 29(4), 233-237.
Pranger, T., & Brown, T. (1992). Burnout: An issue for psychiatric occupational therapy personnel?
Occupational Therapy in Mental Health, 12, 77-92.
Price, L., & Spence, S. H. (1994). Burnout symptoms among drug and alcohol service employees:
Gender differences in the interaction between work and home stressors. Anxiety, Stress, and
Coping, 7, 67-84.
Prosser, D., Johnson, S., Kuipers, E., Szmukler, G., Bebbington, P., & Thornicroft, G. (1997).
Perceived sources of work stress and satisfaction among hospital and community mental health
staff, and their relation to mental health, burnout, and job satisfaction. Journal of
Psychosomatic Research, 43(1), 51-59.
Remer, R., & Ferguson, R. (1995). Becoming a secondary survivor of sexual assault. Journal of
Counseling & Development, 73, 407-413.
Ross, R., Altmaier, E., & Russell, D. (1989). Job stress, social support, and burnout among coun-
seling center staff. Journal of Counseling Psychology, 36, 464-470.
Schauben, L. J., & Frazier, P. (1995). Vicarious trauma: The effects on female counselors of work-
ing with sexual violence survivors. Psychology of Women Quarterly, 19, 49-64.
Simonds, S. L. (1997). Vicarious traumatization in therapists treating adult survivors of childhood
sexual abuse. Unpublished dissertation, The Fielding Institute, Santa Barbara, CA.
Van de Water, R. C. (1996). Vicarious traumatization of therapist: The impact of working with trau-
ma survivors. Unpublished dissertation, Boston University, MA.
Walton, D. T. (1997). Vicarious traumatization of therapists working with survivors: An investigation
of the traumatization process including therapists’ empathy style, cognitive schemas and role of
protective factors. Unpublished doctoral dissertation, Temple University, Philadelphia, PA.
86 Baird and Jenkins

Zilberg, N. J., Weiss, D. S., & Horowitz, M. J. (1982). Impact of Event Scale: A cross-validation
study and some empirical evidence supporting a conceptual model of stress response syn-
dromes. Journal of Consulting and Clinical Psychology, 50, 407-414.

Acknowledgments. We thank the participating centers and shelters: Denton County Friends of the
Family, Rape Crisis Center of Collin County, Victims’ Outreach, Women’s Haven of Fort Worth,
Women’s Center of Fort Worth, Brighter Tomorrows (Grand Prairie), Rape Crisis and Child Center
of Dallas, and New Beginnings Center (Garland).

Offprints. Requests for offprints should be directed to Sharon Rae Jenkins, PhD, Department of
Psychology, University of North Texas, P.O. Box 311280, Denton, TX 76203-1280. E-mail:
jenkinss@unt.edu