researchc
researchc
researchc
HEIFETZ
The present study focuses on the phenomenon of therapist burnout—a problem of rapidly
increasing public and professional concern. Two-hour semistructured interviews were
conducted with a heterogeneous group of psychotherapists (N = 60) in order to investigate
their experiences of therapeutic practice. According to therapists, professional satisfaction
derives from the ability to promote a helpful therapeutic relationship; dissatisfaction stems
primarily from lack of therapeutic success; and burnout is primarily a consequence of the
nonreciprocated attentiveness, giving, and responsibility demanded by the therapeutic
relationship. The data suggest that although therapists expect their work to be difficult and
even stressful, they also expect their efforts to "pay off." Burnout is not only psychologically
debilitating to therapists, it also critically impairs the delivery of mental health services.
work (Freudenberger & Robbins, 1979; Greenson, 1966, Kubie, 1971; Schlict, 1968;
Wheelis, 1958,1963); (c) reflections on the specific difficulties encountered by beginning
psychotherapists (Adams, 1974; Book, 1973; Chessick, 1971; Halleck & Woods, 1962;
Merklin & Little, 1967; Roback, Webersinn, & Guion, 1971; Ungerleider, 1965);
and (d) observations, unsupported by research evidence, regarding the disillusioned
state of the psychotherapeutic community (Frank, 1963; Kernberg, 1968; Rogow,
1970). In addition there is a small job-satisfaction literature that, with one exception
(Farber & Heifetz, 1981), has focused exclusively on psychiatrists (Daniels, 1974;
Maciver & Redlich, 1959; Rogow, 1970). In short, despite the growing popularity
of psychotherapy—as both a career choice for students and as a response alternative
for troubled individuals—no previous research has investigated the nature of burnout
in a heterogenous group of psychotherapists.
Method
SUBJECTS
Subjects were drawn from a northeastern metropolitan community of approximately
350,000. The list of potential subjects was compiled from the rosters of three major
treatment facilities in the area as well as from a composite list, derived from several
sources, of privately practicing psychotherapists in the area. All psychiatrists, psy-
chologists, and social workers from these lists (N = 215) were considered eligible for
the study. Ninety-five randomly selected therapists were contacted initially by letter
and then by phone, and 60 (63.2%) agreed to participate in the study. No significant
differences were found in comparing the acceptance rate of male and female therapists;
of private and institutionally based therapists; or of psychiatrists, psychologists, and
social workers.
The final sample consisted of 36 men and 24 women, including 21 psychiatrists,
24 psychologists, and 15 social workers. Chi-square analysis revealed that males in
the present sample were disproportionately represented among the psychiatrists,
whereas females were disproportionately represented among social workers, X 2 (2)
= p < .001. Among these professionals, 41 considered their practices primarily in-
stitutional, 17 as primarily private, and 2 as evenly split. Historically, psychotherapists
have tended to be disproportionately Jewish (Henry, Sims, & Spray, 1971)—in the
BARRY A. FARBER received his PhD from Yale University in 1978. He is currently an assistant
professor in the clinical psychology program at Teachers College, Columbia University. His
current research focuses on the effects of psychotherapy on psychotherapists and on burnout among
human service professionals.
LOUIS J. HEIFETZ received his PhD from Harvard University in 1974. He is currently an
associate professor at the Syracuse University School of Education. He has coauthored Steps
to Independence, a skills-training series for parents of developmentatly disabled children, and
has demonstrated in his research the potential of parents as teachers and therapists. Current
research concerns the sociaj/'psychology of evolving consumer-provider relationships in human
services.
THIS ARTICLE was adapted from a paper delivered at the annual,meeting of the American
Psychological Association, Montreal, September 1980.
RESEARCH SUPPORT FOR THIS PROJECT was provided by a grant-in-aidfrom Sigma Xi.
REQUESTS FOR REPRINTS should be sent to Barry A. Farber, Department of Psychology, Box
57, Teachers College, Columbia University, New York, New York 10027. -
PROCEDURE
RELIABILITY
Interview questions generated both nominal and ordinal data. The nominal data were
of two types: data resulting from questions permitting only single responses (exclusive
response categories) and data resulting from questions permitting multiple responses
(multiple response categories).
Interrater reliability for the multiple response category nominal data was measured
by overall percentage of agreement between two independent raters. Computed on
this basis, reliability was 71.2%. Interrater reliability for both the exclusive response
category nominal data and the ordinal data was measured by kappa (Cohen, 1960),
utilizing a computer program developed by Cicchetti, Aivano, and Vitale (1977). The
range of observed agreement on those items relevant to the present study was between
.75 and 1.00. Reliability of all relevant items was statistically significant at the .05
level or better.
Results
only the sex of the therapist significantly affected the proportion of those citing this
viewpoint, with female therapists tending to ascribe to this view more frequently than
male therapists. Other factors cited in accounting for burnout included overwork
(22.2%), the general difficulty of dealing with patient problems (20.4%), discourage-
ment as a function of the slow and erratic pace of therapeutic work (18.5%), the ten-
dency of therapeutic work to raise personal issues in psychotherapists (13.0%), the
Sex
Male (n = 36)
n 18 25 27 8 8 12 27 9
% 50.0 69.4 75.0 24.2 24.2 36.4 75.0 25.0
Female (n = 24)
n 19 17 18 5 14 2 21 3
% 79.2 70.8 75.0 22.7 63.6 9.1 87.5 12.5
X 2 (D 4.02* .03 .09 — — 10.35* .73
Profession
Psychiatrist (n = 21)
n 9 14 14 2 6 9 16 5
% 42.9 66.7 66.7 10.0 30 45.0 76.2 23.8
Psychologist (n = 24)
n 17 17 •20 8 8 3 19 5
% 70.8 70.8 83.3 36.4 36.4 13.6 79.2 20.8
Social worker (n = 15)
n 11 11 11 3 8 2 13 2
% 73.3 73.3 73.3 23.1 61.5 15.4 86.7 13.3
2
X (2) 4.86 .20 1.69 — 11.77 .62
Setting
—
Institution
n 26 30 29 11 16 8 37 4
% 63.4 73.2 70.7 28.9 42.1 21.1 90.2 9.8
Private
n 9 10 14 2 6 5 9 8
% 52.9 58.8 82.4 13.3 40.0 33.3 52.9 47.1
X 2 U) .20 .58 .35 2.08 8.04**
Caseload
—
Light
n 12 11 11 7 5 1 11 4
% 80.0 73.3 73.3 50.0 35.7 7.1 73.3 26.7
Moderate
n 17 21 25 6 11 7 25 4
% 58.6 72.4 86.2 22.2 40.7 25.9 86.2 13.8
Heavy
n 8 10 9 0 6 6 12 4
% 50.0 62.5 56.3 — 42.9 42.9 75.0 25.0
2
X (2) 3.17 .59 4.97 — 11.39 — 1.37
Experience level
—
Inexp. '
n 12 14 11 4 10 1 12 3
% 80.0 93.3 73.3 26.7 66.7 6.7 80.0 20.0
Early career
n 14 15 18 6 6 6 19 5
% 58.3 62.5 75.0 27.3 27.3 27.3 79.2 20.8
Experienced
n 11 13 11 3 6 7 17 4
% 52.4 61.9 76.2 16.7 33.3 38.9 81.0 19.0
X 2 (2) 3.01 5.19 .04 10.40 .02
general passivity of therapeutic work (13.0%), and the isolation involved in therapeutic
Therapists felt that they were especially prone to transient feelings of burnout when
stresses at home lowered their threshold for coping with daily therapeutic frustration
and impaired their ability to attend effectively to the needs of their patients. Without
the impingement of these external stresses, most (63.6%) felt that they could see a
maximum 4-6 patients a day before becoming depleted. A smaller minority felt that
they could see 7-8 patients (18.2%) or even 9-10 patients per day (7.3%) before feeling
depleted. The balance of the sample declined to cite a specific figure, contending that
their maximum number of patient hours was variable, although primarily a function
of the type, frequency, and spacing of patients. As Table 1 indicates, only sex sig-
nificantly affected therapists' views regarding a maximum number of patient hours,
with male therapists generally allowing that they could see a greater number of patients
before becoming depleted. Many therapists (40.8%) felt that they were particularly
prone to burnout during winter months; a smaller proportion felt that they were most
vulnerable to burnout in the spring (16.3%) or summer (14.3%). It is also of note that
75% of those therapists surveyed felt that professional duties or concerns at least oc-
casionally "spill over" into leisure hours. This tendency was reported significantly
more often by psychologists and social workers than by psychiatrists, x2 (2) = 8.14,
p < .05.
Most therapists (73.7%) cited "lack of therapeutic success" as the single most stressful
aspect of therapeutic work. Chi-square analyses revealed no significant subgroup
differences in the proportion of therapists citing this response (see Table 1 ). In a related
vein, 25% of the therapists in the sample admitted to occasionally feeling disillusioned
with the therapeutic enterprise. Moreover, an additional 55%, although denying
current feelings of disillusionment, felt that they had need to reassess the goals and
limitations of psychotherapy in order to guard against such feelings. Twenty percent
of the sample reported no feelings of disillusionment; many therapists in this category
stated in one form or another that they were not disillusioned because they felt that
therapy, as they practiced it, "works." For the purposes of statistical analyses, the
1
Note that percentages may add to more than 100% with questions permitting multiple responses.
first two response categories, that is, "overt disillusionment" and "defending against
feelings of disillusionment," were combined. Subsequent chi-square analyses indicated
that neither profession, sex, case load, nor experience level significantly affected the
proportion of the sample that tended toward disillusionment in one form or another.
Whether a therapist had been in personal psychotherapy also did not affect the tendency
toward disillusionment, X 2 0) = 2.16, ns. As Table 1 indicates, only clinical setting
significantly affected this disposition, with institutionally based therapists more fre-
quently admitting to either overt feelings of disillusionment or the necessity of defending
against such feelings. ,
Most therapists found the role of support systems essential. All those who could
utilized supervisory relationships to help them through Hifficult moments; of those
who were not currently being supervised, 51.1% relied on the informal support of
colleagues. In addition to a social support system that served to attenuate anxieties
and restore faith, virtually all therapists expressed the need for an activity outlet, such
as hobbies or sports, that could provide relief for stored-up tensions.
Therapeutic work, of course, provides- satisfaction as well as stresses. The data
indicate that therapists' sources of satisfaction stemmed primarily from helping patients
change (75%), from an increased understanding of human behavior (50%), and from
a sense of intimate involvement with patients (40%). In comparison to their more
experienced counterparts, inexperienced therapists were particularly likely to derive
satisfaction from this sense of intimate involvement, X 2 (2) = 6.03, p < .05.
Discussion
The primary source of stress for therapists is lack of therapeutic success, that is, the
inability to promote positive change in patients. And the primary factor underlying
burnout, according to therapists, is the nonreciprocated attentiveness and giving that
are inherent within the therapeutic relationship. Taken together, these findings suggest
that therapists expect their work to be difficult and even stressful, but they also expect
their efforts to "pay off." Constant giving without the compensation of success ap-
parently produces burnout.
The data suggest that there are certain inherent difficulties in all'work with patients,
primarily difficulties relating to the nature of the therapeutic role (e.g., the requirements
of attentiveness, responsibility, detached concern) and difficulties relating to the nature
of the therapeutic process (e.g., the slow, often erratic pace of therapeutic progress).
In addition working conditions (e.g., excessive work load, organizational politics) can
create additional sources of stress, particularly for institutionally based therapists.
However, these stresses are for the most part accepted as inevitable and even necessary
components of the job; they can, in normal circumstances, be-dealt with moderately
well, with only minimal erosion of one's faith in psychotherapy. It is when psycho-
therapeutic work is particularly frustrating and only minimally successful—and this
may often be the case when one is overworked or dealing with suicidal, homicidal,
depressed, 'or especially resistant patients—that disillusionment and burnout occur.
Frank (1963) stated that those who seek psychotherapeutic care have, as a common
characteristic, feelings of demoralization: "They feel powerless to change the situation
or themselves" (p. 314). The data of the present study suggest that an analogous
process may occur with mental health workers, namely, that those who become burned
out have, as a common denominator, perceptions that their efforts are inconsequen-
tial.
and research that takes place outside the confines of the therapist's office (Grinberg,
1963; Marmor, 1953; Rose, 1974). Professional activities other than therapeutic work
might continue to affirm the therapist's sense of expertise but in less emotionally
draining contexts.
Dissatisfaction and burnout among therapists may potentiate "radical career
changes" among therapists as well as increase demands for alternative sources of sat-
isfaction (Sarason, 1977). Moreover, if it has not done so already, therapist burnout
will surely affect substantially the delivery of mental health services, particularly to
that growing segment of the population that can ill afford to bear the costs of private
professional help.
REFERENCES
Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.
Alternative Proxies: