Journal of Social and Clinical Psychology, Vol. 26, No. 3, 2007, pp.

385–403
Linley andWell–Being
Therapist Joseph

THERAPY WORK AND THERAPISTS’ POSITIVE

AND NEGATIVE WELL–BEING
P. ALEX LINLEY
Centre for Applied Positive Psychology

STEPHEN JOSEPH
University of Nottingham

A substantial literature testifies to the potential negative effects of therapy work on

therapists. However, little is known about the potential positive effects of this work.
The present study explored both positive aspects (personal growth, compassion
satisfaction) and negative aspects (compassion fatigue, burnout) of therapists’
well–being in 156 therapists. Analyses of occupational factors (personal therapy,
supervision, therapeutic training and practice orientation, length of therapy career,
current therapy workload, personal trauma history, gender) and psychological fac-
tors (sense of coherence, social support, empathy, the bond from the working alli-
ance) revealed hypothesized associations with therapist well–being. The findings
illustrate some of the factors associated with positive and negative well–being in
therapists, while the discussion considers future research directions and possible
implications for counseling and clinical practice.

Research on the well–being of therapists has been focused almost exclu-

sively on the negative costs of caring, rather than the personal growth
and satisfaction that therapists may themselves experience as they seek
to facilitate these developmental experiences in their clients. These neg-
ative costs of caring have been labelled as burnout (Maslach & Jackson,
1982) and compassion fatigue (Figley, 1999) in therapists generally, or as
vicarious traumatization (McCann & Pearlman, 1990), contact victim-
ization (Courtois, 1988), secondary posttraumatic stress reaction
(Dutton & Rubenstein, 1995; Figley, 1995), and secondary traumatic

We thank Beth Hudnall Stamm for her helpful comments on our use of the Compassion
Satisfaction–Fatigue Self Test, and Mohammedreza Hojat for his helpful comments on our
use of the Jefferson Scale of Physician Empathy.
Address correspondence to P. Alex Linley, Ph.D., Centre for Applied Positive Psychol-
ogy, Barclays Venture Centre, University of Warwick Science Park, Coventry, CV4 7E2,
United Kingdom; E–mail: alex@cappeu.org

385
386 LINLEY AND JOSEPH

stress (Stamm, 1999) in therapists specifically working with the trauma-

tized. The specific effects of vicarious exposure to traumatic events have
been distinguished from more general concepts of vicarious exposure,
such as countertransference and burnout, through the recognition that
secondary traumatic stress is a response to traumatic events disclosed by
the client, but which the therapist has not experienced directly (Danieli,
1985). Secondary traumatization in therapists is believed to include
many of the symptoms reported by trauma victims themselves, such as
intrusions, avoidance, and physiological symptoms of hyperarousal
(Chrestman, 1999; Courtois, 1988; Dutton & Rubenstein, 1995;
Kassam–Adams, 1999).
The relationships between these different constructs indicative of
some of the negative costs of caring are not clear, but given their similari-
ties, Figley (2002) suggested that vicarious traumatization and burnout
may both be viewed as latent variables that contributed to compassion
fatigue (which he suggested could be used interchangeably with the
terms secondary traumatic stress and secondary traumatic stress disor-
der; Figley, 1999). However, these constructs have not proven empiri-
cally separable (Figley & Stamm, 1996; Stamm, 2005), and the consensus
view now is that they share many features, but differ primarily in their
affective domains, which are self–inefficacy in relation to burnout, and
fear in relation to vicarious traumatization (Barbanel, Saakvitne, &
Stamm, 2003; Larsen, Stamm, & Davis, 2002).
There has been longstanding concern about these negative costs of
caring, and the need for greater attention to therapists’ self–care is a re-
current theme in the literature (Baker, 2003). The deleterious effects of
working with the suffering of others have been reported across a range
of therapists, including domestic violence counselors (Iliffe & Steed,
2000) and especially therapists working with the traumatized (e.g.,
Chrestman, 1999; Kassam–Adams, 1999; Ortlepp & Friedman, 2002;
Pearlman & Mac Ian, 1995).
However, in parallel with the growth of positive psychology (e.g.,
Seligman & Csikszentmihalyi, 2000), there is now a strongly emerging
interest in the positive aspects of human experience. In this study we
were interested in exploring the variables that may be associated with
both positive and negative aspects of therapists’ well–being. This
study builds on the extant research literature concerned with burnout
and vicarious traumatization to consider the variables that might be as-
sociated with positive well–being in therapists. While there are early
studies providing incidental or anecdotal data in support of the posi-
tive effects of therapy work, research has not generally addressed this
question directly. This is surprising, given that the ProQOL/Compas-
sion Fatigue and Satisfaction Test added a positive aspect to measure
THERAPIST WELL–BEING 387

compassion satisfaction (one of the positive effects of caring) in 1996

(Stamm, 2002), and there is also a broader awareness of the potential
positive effects of this work on therapists (Stamm, Varra, Pearlman, &
Giller, 2002).
In an early study in the area of vicarious reactions, Raphael, Singh,
Bradbury, and Lambert (1983–1984) examined disaster response work-
ers attending the Granville rail disaster in Sydney, Australia. Of their 95
respondents, 33 reported that they felt more positive about their own life
as a result of their disaster work. Similarly, Radeke and Mahoney (2000)
described how, relative to research psychologists, professional psychol-
ogists reported more emotional exhaustion, but also felt more positive
influence from their work, including feeling wiser and more experi-
enced about life (cf. Linley, 2003; Smith, Staudinger, & Baltes, 1994).
Likewise, members of the clergy consistently have to deal with the be-
reavement and suffering of others, and it has been shown that this vicari-
ous exposure can lead to posttraumatic growth (Profitt, Calhoun,
Tedeschi, & Cann, 2002). Similar phenomena have been reported in pop-
ulations of funeral directors (Linley & Joseph, 2005) and disaster
response workers (Linley & Joseph, 2006).
Specifically within the trauma therapy literature, there have been an-
ecdotal suggestions of the personal growth and positive change that
therapists may experience through their work with clients. For example,
McCann and Pearlman (1990) presented vicarious traumatization reac-
tions as “an area of potential growth for the helper” (p. 146). Schauben
and Frazier (1995) documented that through dealing with the existential
issues that arose from working with traumatized clients, female rape
counselors appreciated witnessing the resilience and growth of their cli-
ents, “noting that they [the counselors] also grow and change as a result
of their work with survivors” (p. 62). Tedeschi and Calhoun (2004) noted
that “in listening to clients with respect for their strength and ability to
change, we find ourselves changed for the better” (p. 416).
Just as vicarious traumatization has been demonstrated in a range of
therapists and others working with those in suffering, so Linley, Jo-
seph, Cooper, Harris, and Meyer (2003) hypothesized that positive
changes may also be found following vicarious exposure. They sur-
veyed a sample of 108 British citizens who had been vicariously ex-
posed to the September 11, 2001 terrorist attacks through television
viewing. They found that participants who more closely identified
with the view that the attacks were an attack on their values and beliefs
were more likely to report positive changes. They interpreted this em-
pathic response as one of the possible mechanisms through which posi-
tive changes following vicarious exposure may be mediated. This shar-
ing of the experience, even through vicarious exposure, can then
388 LINLEY AND JOSEPH

provide the trigger for both positive and negative changes, a phenome-
non that has also been documented in husbands of women with breast
cancer (Weiss, 2002).
In one of the early empirical studies that documented the potential for
positive change in therapists as a result of their clinical work, Linley, Jo-
seph, and Loumidis (2005) explored the role of the personality construct
sense of coherence. They hypothesized that, on the basis of a previous in-
tegrative theoretical review of personality constructs associated with
adversarial growth (Linley, 2003), therapists with a higher sense of co-
herence would report less negative changes and more positive changes
as a result of their trauma work. This hypothesis was supported.
The present study set out to investigate salient factors that may be as-
sociated with positive and negative aspects of therapist well–being. Pos-
itive well–being in therapists was conceptualized and measured by the
variables of personal growth, positive psychological changes, and com-
passion satisfaction. Negative well–being in therapists was conceptual-
ized and measured by the variables of negative psychological changes,
burnout, and compassion fatigue.
We considered nine occupational factors drawn from the previous lit-
erature: personal psychotherapy (previous or current); clinical supervi-
sion; personal trauma history; therapist gender; therapeutic training ori-
entation; therapeutic practice orientation; the length of time
respondents had been working as therapists; and their current workload
as therapists. We also identified four psychological factors for investiga-
tion in their relation with adversarial growth in therapists: the sense of
coherence personality construct; empathy; the bond from the working
alliance; and social support. Taken together, these occupational and psy-
chological factors begin to suggest some of the variables that may
influence positive and negative well–being in therapists.

METHOD
PROCEDURE
Four hundred questionnaire packs were distributed by mail to individ-
ual practitioners. Participants were selected from the clinical psycholo-
gist and counselling psychologist sections of The Directory of Chartered
Psychologists & The Directory of Expert Witnesses 2002/2003 (British Psy-
chological Society, 2002), and the Counselling and Psychotherapy Resources
Directory 2002 (British Association for Counselling and Psychotherapy,
2002). Two hundred individual practitioners were randomly selected
from each directory using the online Research Randomizer
(www.randomizer.org).
THERAPIST WELL–BEING 389

Participants were invited to take part in a Therapist Experiences Survey

regarding their experiences of working with their clients and how this
work may have influenced them. The questionnaires within the pack
were systematically varied across four different orderings to control for
possible order effects. Responses were anonymous, although partici-
pants were invited to include their details separately for entry into a
prize draw for a £25 (c. US $45) gift voucher of their choice.
One hundred fifty six completed questionnaires were returned, to-
gether with three questionnaires returned blank and two undeliverable,
giving a total response rate of 40%. This compares with previous studies
of therapists that have reported response rates ranging from 32%
(Pearlman & Mac Ian, 1995) to 58% (Pope & Feldman–Summers, 1992).
However, given that the 40% of therapists who responded are likely not
fully representative of all therapists, any generalizations of our findings
should be made with caution.

PARTICIPANTS
Participants were 156 therapists (122 women, 34 men), ranging in age
from 27 to 85 years (M = 53.67, SD = 10.90). Participants were predomi-
nantly white (97%), and were married (64%), divorced (17%), or single
(10%). They were qualified to diploma level (39%), Masters level (32%),
or doctoral level (14%), and worked as therapists either part time (58%)
or full time (42%). They had been working as therapists for between two
and 40 years (M = 15.10, SD = 8.71), and were currently spending an av-
erage of between one and 30 hours per week with their clients (M = 12.64,
SD = 6.60). The majority of the sample had received personal therapy in
the past (78%), but were not receiving personal therapy currently (77%),
although they did have formal supervision or support for their work as a
therapist (90%). Therapists either worked in individual practice (41%),
in more than one practice setting (42%; i.e., individual practice as well as
either group, clinic, or hospital settings), or in a group (3%), clinic (4.5%),
or hospital (4.5%) setting.

MEASURES
Crisis Support Scale (Joseph, Andrews, Williams, & Yule, 1992). A seven
item measure of social support that taps both practical (“Are people help-
ful in a practical sort of way?”) and emotional (“Are you able to talk about
your thoughts and feelings?”) support, and social support satisfaction
(“Overall, are you satisfied with the support you receive?”). Participants were
asked about the “support that you receive during your work as a thera-
pist.” Test–retest reliabilities have not been reported, but a recent review
390 LINLEY AND JOSEPH

endorsed the psychometric properties of the scale (Elklit, Pedersen, &

Jind, 2001).
Jefferson Scale of Physician Empathy (JSPE; Hojat et al., 2002). A
20–item measure of therapist empathy, scored using a 7-point Likert for-
mat scale (1 = strongly disagree; 7 = strongly agree). The JSPE was selected
in preference to the more general Interpersonal Reactivity Index (Davis,
1983) because it is more clearly anchored to the experiential world of the
therapist, and so provides a more specific assessment of empathy within
the therapeutic setting, rather than empathy as a general construct. Mi-
nor amendments were made so that the items were consistent with ther-
apy rather than medical interventions. These included replacing “pa-
tient” with “client,” and using the more general “treatment” rather than
the specific “medical or surgical treatment.” Sample items include “I try
to imagine myself in my clients’ shoes when providing care to them” and “I try
to think like my clients in order to render better care.” Test–retest reliability
over three–four months was reported at r = .65 (Hojat et al., 2002).
Working Alliance Inventory, Form T–Bond subscale (WAI–Bond; Horvath
& Greenberg, 1989). For the present study we used the 12–item
WAI–Bond, a measure of the positive personal attachments between the
client and therapist, assessing themes such as mutual trust, acceptance,
and confidence. Sample items include “I am genuinely concerned for my cli-
ents’ welfare” and “I appreciate my clients as people.” Higher scores indicate
a greater perceived bond between the therapist and their clients.
Test–retest reliabilities have not been reported.
Professional Quality of Life Scales (ProQOL; Stamm, 2002). A 3 0 – i te m
scale with three 10–item subscales assessing Burnout, Compassion satis-
faction, and Compassion fatigue. Four items are reverse–scored. Sample
items include: “I feel overwhelmed by the amount of work or the size of my
caseload I have to deal with” (Burnout); “I find it difficult to separate my per-
sonal life from my life as a helper” (Compassion fatigue) and; “I get satisfac-
tion from being able to help people” (Compassion satisfaction). Minor
amendments were made to four items so that these items were general
rather than trauma–specific. Higher scores indicate greater burnout,
compassion fatigue, and compassion satisfaction respectively.
Test–retest reliabilities have not been reported.
Sense of Coherence Scale – Short form (SOC–13; Antonovsky, 1987). A
13–item self–report measure of one’s general orientation to life, scored
using a 7–point (1 to 7) Likert format scale, giving a potential range of 13
to 91. Five questions have a negative formulation and are re-
verse–scored. The SOC–13 contains items assessing a person’s percep-
tion of the world as comprehensible (five items, e.g., “Do you have very
mixed up feelings and ideas?” scored 1 = very often, 7 = very seldom or never),
manageable (four items, e.g., “Do you have the feeling that you’re being
THERAPIST WELL–BEING 391

treated unfairly?” scored 1 = very often, 7 = very seldom or never), and mean-
ingful (four items, e.g., “How often do you have the feeling that there’s little
meaning in the things you do in your daily life?” scored 1 = very often, 7 = very
seldom or never). Higher scores indicate a greater sense of coherence (i.e.,
that the world is perceived as comprehensible, manageable, and mean-
ingful). Test–retest reliability over six months was reported at r = .77
(Antonovsky, 1993).
Posttraumatic Growth Inventory (PTGI; Tedeschi, & Calhoun, 1996). A
21–item self–report measure of personal growth, scored using a 6–point
Likert format scale (0 = “I did not experience this change as a result of my ther-
apy work”; 5 = “I experienced this change to a very great degree as a result of my
therapy work”). Sample items include “A sense of closeness with others” and
“Appreciating each day.” All 21 items are positively scored, yielding a po-
tential range of 0 to 105, where higher scores indicate greater experience
of growth. Test–retest reliability over two months was reported at r = .71
(Tedeschi & Calhoun, 1996), and at r = .53 over six months (Linley & Jo-
seph, 2006). The PTGI was conceptualised as a measure of personal
growth.
Changes in Outlook Questionnaire (CiOQ; Joseph, Williams, & Yule, 1993).
A 26–item self–report measure of positive and negative psychological
changes “following your work as a therapist,” scored using a six–point
Likert format scale (1 = strongly disagree; 6 = strongly agree). The CiOQ has
two sub–scales: positive psychological changes (11 items, e.g., “I value
my relationships much more now”; “I don’t take life for granted anymore”),
and negative psychological changes (15 items, e.g., “I have very little trust
in other people now”; “I feel very much as if I’m in limbo”). The positive
change subscale has a range of 11 to 66, and the negative change subscale
a range of 15 to 90, with higher scores indicating greater reports of posi-
tive and negative psychological changes respectively. A recent study
supported the factor structure, internal reliability, and validity of the
measure (Joseph et al., 2005). Test–retest reliabilities over six months
have been reported at r = .58 for both the positive change subscale and
the negative change subscale (Linley & Joseph, 2006).

DATA ANALYSES
Group differences according to the occupational factors clinical supervi-
sion, personal therapy (previous or current), personal trauma history
(all dichotomized, yes or no), and gender were assessed using
multivariate analysis of variance. Associations between therapeutic
training orientations, therapeutic practice orientations, length of time
working as a therapist, hours worked per week as a therapist, and the
outcome variables (personal growth, positive changes, compassion sat-
392 LINLEY AND JOSEPH

isfaction, negative changes, compassion fatigue, and burnout) were

assessed using Pearson’s correlation.
Associations between the psychological factors (sense of coherence,
empathy, therapeutic alliance, and social support) and the outcome
variables were assessed using Pearson’s correlation. The prediction of
each outcome variable from the psychological factors was estimated us-
ing multiple regression with simultaneous entry of the four
psychological factors.
Given the early stages of empirical work exploring positive changes in
therapists, and the high participant to item ratio (156 participants to 129
items), we opted to use a .05 significance level for the analyses. While
mindful of the potential inflation of a Type I error, we believe this is ap-
propriate for the generation of areas of investigation that should guide
future research at these early stages. The positive well–being variables
were intercorrelated (highest r = .62, p < .001, for positive changes and
personal growth), and the negative well–being variables were
intercorrelated (highest r = .45, p < .001, for burnout and compassion fa-
tigue). However, we opted not to combine them for two reasons. First,
these intercorrelations were not so high as to suggest multicollinearity
(where a suggested correlation of .7 is indicated; Tabachnik & Fidell,
2001). Second, we opted at this early stage of research to consider the
variables individually and so inform the development of future research
questions.

RESULTS
Table 1 provides the descriptive statistics for the study variables.

OCCUPATIONAL FACTORS AND THERAPIST CHANGES

Personal Therapy. Therapists who had either received personal ther-
apy previously, or were receiving personal therapy currently, reported
more personal growth and positive changes, and less burnout (see Table
2).
Clinical Supervision. Participants who answered yes to the question
“Do you receive formal supervision or support for your work as a therapist?”
reported greater levels of personal growth (M = 65.82, SD = 20.05) than
did therapists who answered no (M = 51.20, SD = 17.40), F(1, 139) = 7.50,
p < .01. All other group comparisons for the supervision variable were
nonsignificant, largest F(1, 139) = 3.78, p > .05.
Personal Trauma History. Therapists who answered yes to the ques-
tion “Do you have a personal trauma history?” reported greater levels of
personal growth (M = 71.61, SD = 17.38) than did therapists who an-
 TABLE 1. Descriptive Statistics for Study Variables

Variable Alpha ( ) M SD Observed Range

Sense of Coherence (SOC) .79 69.32 8.86 42–87
Empathy (JSPE) .64 119.53 9.88 92–139
Therapeutic Bond (WAI–B) .80 67.01 6.34 50–79
Social Support (CSS) .79 38.93 5.97 15–49
Posttraumatic Growth (PTGI) .94 64.42 20.08 0–105
Positive Changes (CiOQ) .78 47.61 7.38 11–63
Compassion Satisfaction .83 37.43 5.89 20–50

393
Negative Changes (CiOQ) .80 24.95 7.87 15–60
Compassion Fatigue .70 10.27 4.80 2–26
Burnout .61 18.56 4.68 8–31

Note. SOC = Sense of coherence. JSPE = Jefferson Scale of Physician Empathy. WAI–B = Working Alliance Inventory—Bond subscale. CSS = Crisis Support Scale. PTGI =
Posttraumatic Growth Inventory. CiOQ = Changes in Outlook Questionnaire.
 TABLE 2. MANOVA Showing Differences in Positive and Negative Therapist Well–Being According to Previous and Current Personal Therapy

Previous Personal Therapy Current Personal Therapy

Variable Previous M SD F(1, 140) Current M SD F(1, 140)
Positive Well–Being
PTGI Yes 67.18 19.29 8.75** Yes 72.25 16.95 6.47*
No 54.71 20.28 No 62.19 20.49
CiOQ–P Yes 48.53 7.09 9.06** Yes 50.80 6.00 7.51**
No 44.29 7.60 No 46.63 7.53
CS Yes 37.26 5.98 .15 Yes 38.80 5.68 1.85

394
No 37.68 5.37 No 37.03 5.94
Negative Well–Being
CiOQ–N Yes 24.54 7.97 .80 Yes 24.40 7.56 .10
No 26.35 7.58 No 25.17 7.99
CF Yes 10.25 4.86 .00 Yes 9.91 5.00 .28
No 10.53 4.55 No 10.39 4.77
Burnout Yes 18.19 4.41 4.46* Yes 17.06 4.21 4.66*
No 20.09 5.22 No 19.04 4.74

Note. PTGI = Posttraumatic Growth Inventory. CiOQ–P = Changes in Outlook Questionnaire – Positive changes. CS = Compassion Satisfaction. CiOQ–N = Changes in Out-
look Questionnaire – Negative changes. CF = Compassion Fatigue. *p < .05; **p < .01.
THERAPIST WELL–BEING 395

swered no (M = 61.70, SD =20.56), F(1, 136) = 5.60, p < .05. All other group
comparisons for the personal trauma history variable were
nonsignificant, largest F(1, 136) = 2.56, p > .10.
Gender. Women reported greater levels of personal growth (M = 66.97,
SD = 19.61) than did men (M = 55.21, SD = 19.30), F(1,141) = 8.41, p < .01.
This was also the case for positive changes, with women reporting more
positive changes (M = 48.29, SD = 6.68) than did men (M = 45.15, SD =
9.21), F(1, 141) = 5.62, p < .05. All other gender comparisons were
nonsignificant, largest F(1, 141) = 3.57, p > .05.
Therapeutic Training and Practice Orientations. Therapists whose train-
ing was primarily transpersonal in orientation were more likely to re-
port positive psychological changes as a result of their therapeutic work.
In contrast, therapists whose training was primarily cognitive–behav-
ioral in orientation were less likely to report positive psychological
changes as a result of their therapeutic work. Similarly, therapists from a
cognitive–behavioral training orientation were significantly more likely
to report symptoms of burnout as a result of their therapeutic work. This
same pattern of findings also held for therapists’ current therapeutic
practice orientation (see Table 3).
Lifetime Therapy Work. Therapists who reported a greater length of
time working as a therapist reported more negative psychological
changes (r = .16, p < .05) and more compassion fatigue (r = .20, p < .01). All
other correlations were nonsignificant.
Current Therapy Workload. Therapists who reported a greater number
of hours per week spent with clients in therapy reported more personal
growth (r = .23, p < .01) and more positive psychological changes (r = .19,
p < .05). All other correlations were nonsignificant.

PSYCHOLOGICAL FACTORS AND THERAPIST CHANGES

Multiple regression of the therapist well–being variables on the psycho-
logical factors, with simultaneous entry of the four psychological fac-
tors, showed that the therapeutic bond was the best predictor of positive
psychological changes and compassion satisfaction. In contrast, sense of
coherence was the best predictor of less negative psychological change
and compassion fatigue, while sense of coherence and the therapeutic
bond were the best predictors of less burnout (see Table 4).

DISCUSSION
This study provides a broad analysis of possible occupational and psy-
chological factors that influence both positive and negative aspects of
therapist well–being. The literature dealing with the effects of therapy
 TABLE 3. Therapeutic Training and Current Practice Orientation Correlates with Positive and Negative Therapist Well–Being

Positive Therapist Well–Being Negative Therapist Well–Being

PTGI CiOQ–P CS CiOQ–N CF Burnout
Therapeutic Training
Client–centered .16* .13 .05 –.10 –.01 –.03
Psychodynamic .19* .08 .02 –.12 –.08 –.06
Cognitive–behavioral –.18* –.19** –.02 .10 .07 .29***
Existential .13* .03 –.02 .12 .08 .13*
Transpersonal .34*** .28*** .19** .03 –.04 –.09
Integrative .17* .18* .05 –.08 .02 .06
Eclectic .10 .13 .18* .01 .04 .02

396
Current Practice
Client–centered .14* .17* .05 –.08 .03 .01
Psychodynamic .10 .10 –.02 –.09 –.07 –.09
Cognitive–behavioral –.16* –.14* .02 .11 .05 .20**
Existential .13 .07 –.08 .05 .11 .11
Transpersonal .24*** .19** .18* .03 .07 –.10
Integrative .10 .11 .03 –.05 .10 .09
Eclectic –.01 .04 .14* .01 .07 .08

Note. PTGI = Posttraumatic Growth Inventory. CiOQ–P = Changes in Outlook Questionnaire—Positive changes. CS = Compassion Satisfaction. CiOQ–N = Changes in
Outlook Questionnaire—Negative changes. CF = Compassion Fatigue. *p < .05; **p < .01; ***p < .001. All one–tailed tests.
THERAPIST WELL–BEING 397

TABLE 4. Multiple Regression of Positive and Negative Therapist Well–Being Variables

on Psychological Factors

B SE (B) t
Positive Well–Being
Posttraumatic Growth (PTGI)
SOC –.29 .21 –.13 –1.38
JSPE .44 .17 .22 2.53*
WAI–B .45 .30 .14 1.49
CSS .23 .29 .07 .78
Positive Changes (CiOQ)
SOC .00 .07 –.03 –.33
JSPE .00 .06 .12 1.48
WAI–B .37 .11 .31
CSS .23 .10 .18 2.22*
Compassion Satisfaction
SOC .00 .05 –.44 –5.10***
JSPE .00 .05 .16 2.20
WAI–B .43 .08 .45 5.53***
CSS .00 .07 .06 .85
Negative Well–Being
Negative Changes (CiOQ)
SOC –.41 .07 –.47 –5.96***
JSPE –.20 .06 –.26 –3.54***
WAI–B .00 .10 .04 .48
CSS –.13 .10 –.10 –1.30
Compassion Fatigue
SOC –.23 .05 –.44 –5.10***
JSPE .00 .04 .05 .62
WAI–B .00 .07 –.13 –1.49
CSS .00 .06 .04 .49
Burnout
SOC –.15 .04 –.27 –3.40***
JSPE .00 .04 .03 .42
WAI–B –.26 .06 –.34 –4.22***
CSS .00 .06 –.12 –1.58
Note. Sense of coherence, empathy, therapeutic bond, and social support were entered simultaneously to
predict each positive and negative therapist change variable. PTGI = Posttraumatic Growth Inventory.
CiOQ = Changes in Outlook Questionnaire. SOC = Sense of coherence. JSPE = Jefferson Scale of Physi-
cian Empathy. WAI–B = Working Alliance Inventory—Bond subscale. CSS = Crisis Support Scale. *p <
.05; ***p < .001.

work on the therapist has almost exclusively focused on the deleterious

effects of this work. However, recent trends suggesting a more positive
focus in psychology (e.g., positive psychology; Seligman &
Csikszentmihalyi, 2000) suggest there is also a need to address the posi-
tive well–being that therapists may experience in relation to their
therapy work.
Nine occupational factors were investigated, and found to be associ-
ated with positive and negative well–being in therapists. Therapists
who either had been, or were currently receiving personal therapy in re-
398 LINLEY AND JOSEPH

spect of their therapeutic work reported more positive psychological

changes and less burnout. This finding points to the efficacy of personal
therapy not only in protecting therapists from negative changes, but also
in facilitating their own positive changes and personal growth (cf.
Macran, Stiles, & Smith, 1999). Greater levels of personal growth were
reported by therapists receiving clinical supervision, by therapists who
had a personal trauma history, and by female therapists. These findings
point to the value of clinical supervision in facilitating personal growth
in therapists, and further suggest that a personal trauma history is not
simply a risk factor for vicarious traumatization (Pearlman & Mac Ian,
1995), but may also be a facilitator of greater personal growth. The find-
ing that female therapists reported higher levels of positive psychologi-
cal changes is consistent with previous research in the adversarial
growth field (Linley & Joseph, 2004).
Therapeutic training and practice orientations revealed consistent as-
sociations with both positive and negative therapist well–being in thera-
pists. Therapists of a humanistic and transpersonal orientation were
more likely to report personal growth and positive change as a result of
their therapy work. This finding makes intuitive sense, given that the
theme of suffering providing opportunities for learning and growth is a
central part of humanistic and transpersonal psychotherapy (Joseph,
2001). In contrast, therapists of a cognitive–behavioral orientation were
significantly less likely to report personal growth and positive change,
and significantly more likely to report symptoms of burnout. It may be
that therapists of a cognitive–behavioral orientation may more typically
work with client groups displaying more severe and chronic psycholog-
ical conditions, thus limiting the therapists’ opportunities for personal
growth and positive well–being. This finding requires replication, but
suggests a potential area of concern, given that cognitive–behavioral
orientations are prevalent amongst current therapeutic training and
practice within counseling and clinical psychology.
Therapists who had been doing therapeutic work for longer overall
time periods reported more negative psychological changes and burn-
out, suggesting that a lifetime career in therapeutic work may not be
conducive to personal satisfaction and growth. This is consistent with
the findings of Chrestman (1999) and Kassam–Adams (1999), who
showed that therapists with higher caseloads of trauma clients reported
more trauma–related symptoms. In contrast, in the present study, thera-
pists who worked a greater number of hours per week reported more
personal growth and positive changes. This inference is not causal, and
it would seem more likely that therapists who are benefiting from their
work would take on greater caseloads, rather than greater caseloads
facilitating personal growth and positive change.
THERAPIST WELL–BEING 399

Based on a review of the limited empirical evidence to date, we hy-

pothesized that positive well–being in therapists would be associated
with the sense of coherence personality construct, empathy, the thera-
peutic bond, and social support. These hypotheses were supported. Fur-
ther analyses indicated that the therapeutic bond may be the key compo-
nent in this relationship. Consistent with previous work (Linley, Joseph,
Cooper et al., 2003), the therapeutic bond may represent the therapist’s
empathic connection with his or her clients, and thus serve as the chan-
nel through which the therapist experiences positive psychological
changes in grappling vicariously with the suffering and distress of his or
her clients. Future research could explore this question more fully by
teasing out whether the therapeutic bond moderates therapists’ expo-
sure to their clients’ distressing material and positive psychological
changes in the therapist.
In contrast, the sense of coherence personality construct was found to
be the factor most protective against negative psychological changes
and compassion fatigue, while sense of coherence and the therapeutic
bond were the factors most protective against burnout (cf. Linley et al.,
2005). Future research should explore whether therapists’ sense of co-
herence moderates their exposure to their clients’ distressing material,
and the therapist’s negative psychological changes.
The primary limitation of this study is its cross–sectional design,
which precludes consideration of causal inferences. Hence, while we
may suggest, for example, that therapeutic training and practice orienta-
tions predispose therapists to greater or lesser potential for positive and
negative well–being, this is likely confounded by other factors. These
may include the therapists’ personal characteristics that prompted them
to pursue a particular theoretical orientation, and the nature of the client
groups with whom therapists of different therapeutic orientations may
work. The same issues apply to each of the relationships reported above,
and further longitudinal research is required before a more comprehen-
sive understanding of the processes that lead to positive and negative
well–being in therapists can be reached.
The strengths of the study include its consideration of a range of occu-
pational and psychological factors that may be associated with both pos-
itive and negative well–being in therapists, and it provides a firm empir-
ical foundation from which future research in this area may proceed.
The study findings point toward a range of directions for future re-
search, and also indicate areas of implication and concern for counseling
and clinical practice, as we now discuss.
Given the early stage of research in this area, these clinical implica-
tions are made tentatively. However, if these findings are replicated in
other studies, there are several matters of importance for effective coun-
400 LINLEY AND JOSEPH

seling and clinical practice that are mindful of the experiences of the
therapist. From an occupational perspective, personal therapy and su-
pervision appear to be good facilitative experiences for positive well–be-
ing in therapists, and should be encouraged as a regular part of any ther-
apist’s clinical work (cf. Macran, Stiles, & Smith, 1999). Personal therapy
also appears protective against burnout, replicating previous research
in this area. The finding that practitioners working from a cognitive–be-
havioral perspective were less likely to report positive changes and
more likely to report burnout suggests an important need for awareness
of these issues, given the prevalence of the cognitive–behavioral orienta-
tion in current clinical training and practice. Cognitive–behavioral prac-
titioners may particularly benefit from interventions to facilitate their
personal growth and relieve their sense of burnout.
The sense of coherence personality construct appears to be a good pro-
tective factor against negative psychological changes, compassion fa-
tigue, and burnout. As such, it may usefully be considered for assess-
ment and screening of therapists who, low in sense of coherence, may
find themselves at risk for the negative psychological effects of provid-
ing therapy. In contrast, the therapeutic bond appears to be the best pre-
dictor of positive therapist well–being, and may be usefully conceptual-
ized as the empathic channel through which therapists can
constructively engage with their vicarious experience of the suffering
and distress of their clients (cf. Martin, Garske, & Davis, 2000).
Overall, these findings point to a range of factors that are associated
with both positive and negative well–being in therapists. While further
research is clearly needed, there are clear implications for counseling and
clinical practice. Given that the therapist is second only to the client as a
factor predictive of therapeutic success (Wampold, 2001), it is imperative
that therapists take steps to ensure that they are functioning at their best in
the therapeutic relationship. Facilitating their own personal well–being
and avoiding burnout is clearly one way in which this can be achieved.

REFERENCES

Antonovsky, A. (1987). Unraveling the mystery of health: How people manage stress and stay
well. San Francisco, CA: Jossey Bass.
Antonovsky, A. (1993). The structure and properties of the sense of coherence scale. Social
Science and Medicine, 36, 725–733.
Baker, E. K. (Ed.). (2003). Caring for ourselves: A therapist’s guide to personal and professional
well–being. Washington, DC: American Psychological Association.
Barbanel, L., Saakvitne, K. W., & Stamm, B. H. (2003). Fostering resilience in response to terror-
ism among mental health workers. Retrieved May 23, 2006 from: http://www.apa.org/
psychologists/pdf/mentalhealthworkers.pdf
THERAPIST WELL–BEING 401

British Association for Counselling and Psychotherapy. (2002). Counselling and Psychother-
apy Resources Directory 2002 (15th ed.). Rugby, UK: Author.
British Psychological Society. (2002). The directory of chartered psychologists & The directory of
expert witnesses 2002/2003. Leicester, UK: Author.
Chrestman, K. R. (1999). Secondary exposure to trauma and self-reported distress among
therapists. In B. H. Stamm (Ed.), Secondary traumatic stress: Self–care issues for clini-
cians, researchers, and educators (pp. 29–36). Lutherville, MD: Sidran Press.
Courtois, C. A. (1988). Healing the incest wound: Adult survivors in therapy. New York:
Norton.
Danieli, Y. (1985). The treatment and prevention of long term effects and intergenerational
transmission of victimization: A lesson from Holocaust survivors and their chil-
dren. In C. R. Figley (Ed.), Trauma and its wake: The study and treatment of posttraumatic
stress disorder. New York: Brunner/Mazel.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidi-
mensional approach. Journal of Personality and Social Psychology, 44, 113–126.
Dutton, M. A., & Rubenstein, F. L. (1995). Trauma workers. In C. R. Figley (Ed.), Trauma and
its wake: Secondary traumatic stress disorder (Vol. 3). New York: Brunner/Mazel.
Elklit, A., Pedersen, S. S., & Jind, L. (2001). The Crisis Support Scale (CSS): Psychometric
qualities and further validation. Personality and Individual Differences, 31, 1291–1302.
Figley, C. R. (1995). Compassion fatigue: Coping with secondary traumatic stress disorder in those
who treat the traumatized. New York: Brunner/Mazel.
Figley, C. R. (1999). Compassion fatigue: Toward a new understanding of the costs of car-
ing. In B. H. Stamm (Ed.), Secondary traumatic stress: Self–care issues for clinicians, re-
searchers, and educators (2nd ed., pp. 3–28). Lutherville, MD: Sidran Press.
Figley, C. R. (2002). Introduction. In C. R. Figley (Ed.), Treating compassion fatigue (pp. 1–14).
New York: Brunner/Routledge.
Figley, C. R., & Stamm, B. H. (1996). Psychometric review of the Compassion Fatigue Self
Test (CFST). In B. H. Stamm (Ed.), Measurement of stress, trauma, and adaptation.
Lutherville, MD: Sidran Press.
Hojat, M., Gonnella, J. S., Nasca, T. J., Mangione, S., Vergare, M., & Magee, M. (2002). Physi-
cian empathy: Definition, components, measurement, and relationship to gender
and specialty. American Journal of Psychiatry, 159, 1563–1569.
Horvath, A. O., & Greenberg, L. S. (1989). Development and validation of the Working Al-
liance Inventory (WAI). Journal of Counseling Psychology, 36, 223–233.
Iliffe, G., & Steed, L. G. (2000). Exploring the counselor’s experience of working with per-
petrators and survivors of domestic violence. Journal of Interpersonal Violence, 15,
393–412.
Joseph, S. (2001). Psychopathology and therapeutic approaches: An introduction. Basingstoke,
UK: Palgrave.
Joseph, S., Andrews, B. Williams, R., & Yule, W. (1992). Crisis support and psychiatric
symptomatology in adult survivors of the Jupiter cruise ship disaster. British Journal
of Clinical Psychology, 31, 63–73.
Joseph, S., Linley, P. A., Andrews, L., Harris, G., Howle, B., Woodward, C., & Shevlin, M.
(2005). Assessing positive and negative changes in the aftermath of adversity:
Psychometric evaluation of the Changes in Outlook Questionnaire. Psychological
Assessment, 17, 70–80.
Joseph, S., Williams, R., & Yule, W. (1993). Changes in outlook following disaster: The pre-
liminary development of a measure to assess positive and negative responses. Jour-
nal of Traumatic Stress, 6, 271–279.
Kassam–Adams, N. (1999). The risks of treating sexual trauma: Stress and secondary
trauma in psychotherapists. In B. H. Stamm (Ed.), Secondary traumatic stress:
402 LINLEY AND JOSEPH

Self–care issues for clinicians, researchers, and educators (pp. 37–48). Lutherville, MD:
Sidran Press.
Larsen, D., Stamm, B. H., & Davis, K. (2002). Telehealth for prevention and intervention of
the negative effects of caregiving. Traumatic StressPoints, 16(4). Retrieved May 23,
2006 from: http://www.istss.org/publications/TS/Fall02/telehealth.htm
Linley, P. A. (2003). Positive adaptation to trauma: Wisdom as both process and outcome.
Journal of Traumatic Stress, 16, 601–610.
Linley, P. A., & Joseph, S. (2004). Positive change following trauma and adversity: A re-
view. Journal of Traumatic Stress, 17, 11–21.
Linley, P. A., & Joseph, S. (2005). Positive and negative changes following occupational
death exposure. Journal of Traumatic Stress, 18, 751–758.
Linley, P. A., & Joseph, S. (2006). The positive and negative effects of disaster work: A pre-
liminary investigation. Journal of Loss and Trauma, 11, 229–245.
Linley, P. A., Joseph, S., Cooper, R., Harris, S., & Meyer, C. (2003). Positive and negative
changes following vicarious exposure to the September 11 terrorist attacks. Journal
of Traumatic Stress, 16, 481–485.
Linley, P. A., Joseph, S., & Loumidis, K. (2005). Trauma work, sense of coherence, and posi-
tive and negative changes in therapists. Psychotherapy and Psychosomatics, 74,
185–188.
Macran, S., Stiles, W. B., & Smith, J. A. (1999). How does personal therapy affect therapists’
practice? Journal of Counseling Psychology, 46, 419–431.
Martin, D. J., Garske, J. P., & Davis, K. M. (2000). Relation of the therapeutic alliance with
outcome and other variables: A meta–analytic review. Journal of Consulting and Clin-
ical Psychology, 68, 438–450.
Maslach, C., & Jackson, S. E. (1982). Burnout in health professions: A social psychological
analysis. In G. Sanders & J. Suls (Eds.), Social psychology of health and illness (pp.
227–251). Hillsdale, NJ: Erlbaum.
McCann, I. L., & Pearlman, L. A. (1990). Vicarious traumatization: A framework for under-
standing the psychological effects of working with victims. Journal of Traumatic
Stress, 3, 131–149.
Ortlepp, K., & Friedman, M. (2002). Prevalence and correlates of secondary traumatic
stress in workplace lay trauma counselors. Journal of Traumatic Stress, 15, 213–222.
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.
Pope, K. S., & Feldman–Summers, S. (1992). National survey of psychologists’ sexual and
physical abuse history and their evaluation of training and competence in these ar-
eas. Professional Psychology: Research and Practice, 23, 353–361.
Profitt, D. H., Calhoun, L. G., Tedeschi, R. G., & Cann, A. (2002, August). Clergy and crisis:
Correlates of posttraumatic growth and well–being. Poster presented at the Annual Con-
vention of the American Psychological Association, Chicago, IL.
Radeke, J. T., & Mahoney, M. J. (2000). Comparing the personal lives of psychotherapists
and research psychologists. Professional Psychology: Research and Practice, 31, 82–84.
Raphael, B., Singh, B., Bradbury, L., & Lambert, F. (1983–1984). Who helps the helpers? The
effects of a disaster on the rescue workers. Omega, 14, 9–20.
Schauben, L. J., & Frazier, P. A. (1995). Vicarious trauma: The effects on female counselors
of working with sexual violence survivors. Psychology of Women Quarterly, 19, 49–64.
Seligman, M. E. P., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction.
American Psychologist, 55, 5–14.
Smith, J., Staudinger, U. M., & Baltes, P. B. (1994). Occupational settings facilitating wis-
THERAPIST WELL–BEING 403

dom–related knowledge: The sample case of clinical psychologists. Journal of Con-

sulting and Clinical Psychology, 62, 989–999.
Stamm, B. H. (Ed.). (1999). Secondary traumatic stress: Self–care issues for clinicians, research-
ers, and educators (2nd ed.). Lutherville, MD: Sidran Press.
Stamm, B. H. (2002). Measuring compassion satisfaction as well as fatigue: Developmental
history of the Compassion Fatigue and Satisfaction Test (CFST). In C. R. Figley (Ed.),
Treating compassion fatigue (pp. 107–119). New York: Brunner/Routledge.
Stamm, B. H. (2005). The ProQOL Manual: The Professional Quality of Life Scale: Compassion
satisfaction, burnout, and compassion fatigue/secondary trauma scales. Baltimore, MD:
Sidran Press.
Stamm, B. H., Larsen, D., & Davis–Griffel, K. (2002). ProQOL, the Professional Quality of Life
Scale: Compassion satisfaction and fatigue subscales: R–III. Retrieved from
http://www.isu.edu/~bhstamm/tests.htm, December 20, 2002.
Stamm, B. H., Varra, E. M., Pearlman, L. A., & Giller, E. (2002). The helper’s power to heal and
to be hurt – or helped – by trying. Washington, DC: Register Report: A Publication of
the National Register of Health Service Providers in Psychology. Retrieved May 23,
2006 from: http://www.isu.edu/~bhstamm/tests/S_VT%20One%20Page%20
Handout.pdf
Tabachnik, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Needham
Heights, MA: Allyn & Bacon.
Tedeschi, R. G., & Calhoun, L. G. (1996). The Posttraumatic Growth Inventory (PTGI):
Measuring the positive legacy of trauma. Journal of Traumatic Stress, 9, 455–471.
Tedeschi, R. G., & Calhoun, L. G. (2004). A clinical approach to posttraumatic growth. In P.
A. Linley & S. Joseph (Eds.), Positive psychology in practice (pp. 405–419). Hoboken,
NJ: Wiley.
Wampold, B. E. (2001). The great psychotherapy debate: Models, methods, and findings.
Mahwah, NJ: Erlbaum.
Weiss, T. (2002). Posttraumatic growth in women with breast cancer and their husbands:
An intersubjective validation study. Journal of Psychosocial Oncology, 20, 65–80.