REGULAR ARTICLES

Supportive Therapy as the

Treatment Model of Choice

DAVID J. HELLERSTEIN, M.D.

HENRY PINSKER, M.D.

RICHARD N. ROSENTHAL, M.D.

STEVEN KLEE, PH.D.

psychotherapy, according to

Supportive and expressive techniques in psy- S upportive

chotherapy can be located on a continuum. most definitions, has the objectives of im-
mediately reducing anxiety and maintaining
Traditionally, psychotherapy has been ori-ented
a positive patient-therapist relationship with
toward the expressive end of the contin-
minimal focus on transference. Several defi-
uum, applying the model of psychoanalytic or nitions of supportive therapy can be found in
expressive therapy to all therapy. The the recent literature. Narrow definitions-
authors propose that for most patients, the for instance that of the American Psychiatric
model for individual dynamic psychotherapy Association’s Commission on Psychiatric
shonld be based on concepts from the support-ive Therapies’-focus on a supportive relation-
end of the continuum. ship between an impaired patient and an
(The Journal of Psychotherapy Practice authoritative therapist who provides advice,
and Research 1994; 3:300-306) reassurance, and reality testing. In the spec-
trum of treatment, this approach is at the
opposite end from the expressive approach.
Broader defmitions 24 describe supportive
therapy as having many of the characteristics
of exploratory or expressive psychotherapy.
Buckle? summarizes Kernberg’s view of sup-
portive therapy as a treatment that “does not
use interpretation, partially uses clarification
and abreaction, and primarily uses sugges-
tion and environmental intervention” (p.
515). Novalis et al.6 proposed a definition that
emphasizes what supportive therapy does
Received November 16, 1993; revised March 29, 1994;

acceptedApril3, 1994. From the Short-Term Psychother-

apy Research Program, Beth Israel Medical Center, and
the Department of Psychiatry, Mount Sinai School of
Medicine, New York. Address correspondence to Dr.
Hellerstein, Psychiatric Outpatient Services, Beth Israel
Medical Center, 1st Avenue and 16th Street, New York,
NY 10003.
Copyright © 1994 American Psychiatric Press, Inc.

VOLUME S #{149}NUMBER 4 #{149}FALL 1994

 HELLERSTEIN ETAL. 301

rather than what it is. He defines these actions chotherapy outcome studies have thus far not
as 1) reducing behavioral dysfunctions, 2) demonstrated that one therapy is better than
reducing subjective mental distress, 3) sup- another for most patients,9 although it has
porting and enhancing the patient’s been consistently observed that therapy is better than
strengths, coping skills, and capacity to use no therapy.’#{176}
environmental supports, 4) maximizing
treatment autonomy, and 5) facilitating max- MODELOF SUPPORTIVE
imum possible independence from psychiat- T H E R A P Y

ric illness. Pinsker7 has suggested that

supportive therapy should not be thought of For the most part, supportive therapy has
as a type of therapy, but rather as a “shell” of been defined by contrasting it to expressive
techniques used by therapists of diverse the- or exploratory treatment, producing a defini-
a
oretical orientations; their specific rationales tion that is essentially set of subtractions
for these techniques may include cognitive as from a standard. Although supportive ther-
apy is not based on a specified theory of
well as dynamic explanations for what is

below the surface. personality development or a theory of men-

Supportive therapy has been described as tal illness,7 it is possible to extract from the
the appropriate treatment for lower function- various definitions and descriptions of sup-
ing patients and for patients who are too portive therapy a statement of its essential
fragile or too unmotivated to participate in characteristics and underlying premises. This
more demanding expressive therapy, which body of premises and assumptions, and the
might have more chance of leading to per- techniques based on them, we characterize as
sonality change. The treatment of choice for the model of supportive therapy.
the patient who is analyzable is supposed to Elements of individual dynamic support-
be psychoanalysis or psychoanalytically ori- ive psychotherapy include 1) strengthening

ented therapy. In everyday practice, most psy- the therapeutic alliance; 2) using direct mea-
chotherapy involves a blend of approaches, sures to relieve symptoms and to minimize
aptly characterized by Luborsky8 as being on development of anxiety within the therapy;
a continuum from expressive to supportive, 3) focusing on seW-esteem, adaptive skills,
with midpositions of expressive-supportive and psychological (or ego) functions (mea-
and supportive-expressive. To meet the sures may include reassurance, encourage-
patient’s needs, elements of pure expressive ment, praise, advice, reframing, clarification,
treatment are withheld or supportive ele- confrontation, education); 4) attention to
ments are added. However, the conceptual negative aspects of the patient-therapist rela-
framework of individual dynamic therapy is tionship when present, but not to positive
intertwined with a model of treatment, such transference; and 5) minimal interpretation
113
as the expressive model, that often has lasting of unconscious
impact in that the psychoanalytic techniques Table 1 compares the two models. Be-
of abstinence and nongratification are cause the model of supportive psychotherapy
adopted as the habitual stance of the thera- does not specify that analysis of transference
pist. If a body of evidence had demonstrated must be the basis for change, the therapist is
that exploratory therapy is more effective “real” to the patient. When psychotherapy is
than supportive therapy for most patients, based on the assumptions, premises, and
this would be justifiable. If evidence demon- techniques of expressive therapy (the expres-
strated that exploratory therapy is consis- sive model), the therapist gives up only as
tently best for a specific group of patients, much neutrality as necessary.’4 With the sup-
that would be reason to recommend it, at portive therapy model, the therapist is no
least for that group. However, controlled psy- more neutral than necessary. It is important

JOURNAL OF PSYCHOTHERAPY PRACTICE AND RESEARCH

302 SUPPORTIVE THERAPY MODEL

to note that this 180-degree shift does not can appropriately guide psychotherapy for a
imply that supportive therapy is undisci- wide range of clinical problems. The differ-
plined or that the relationship should ever be ential application of strategies developed for
anything but professional. Transferential is- treatment of low-functioning patients thus
sues are examined only if they threaten to defines a psychotherapeutic model also
disrupt the treatment. Because the model broad enough to serve the needs of higher
does not specify that the unconscious must be functioning patients.
made conscious, techniques developed to en-
courage the free flow of unconscious material P R o P o sA L

are not used. Because the model does not

postulate that continuing anxiety is desirable We propose that psychotherapy for most pa-
to motivate the patient to work at therapy, tients, both long-term and brief, should be
anxiety-reducing measures are employed based on the supportive psychotherapy
readily. Maladaptive patterns are identified model. To use an analogy drawn from the
from the patient’s accounts of relationships computer world, the supportive model of
with and feelings about figures in his or her therapy would be the “default” model. When
current life or past. Although the patterns we propose that a supportive therapy model
may be replicated in the patient-therapist re- be the basis of treatment for the majority of
lationship, no effort is made to encourage patients, we are not talking about the sup-
this. When this replication occurs, the thera- portive therapy of authoritarian advice, sug-
pist will not interpret it to the patient unless gestion, and environmental manipulation.
it threatens to interfere with the treatment. What we are talking about is a therapy
It is important that the reader under- grounded in some theoretically based under-
stand that what we are describing is not a new standing of mental functioning, a treatment
therapy. Supportive therapy has usually been approach with a well-defined body of thera-
presented as a set of techniques, not as a peutic techniques. These techniques are not
therapy based on a model, but it is our thesis a series of expedient departures from a
that widely used approaches and techniques monolithic model of psychoanalytic psycho-
of supportive therapy and the assumptions therapy. They are coherent expressions of the
that underlie them constitute a model that model of supportive therapy.’2”3
TABLE!. Comparison of critical aspects of supportive and expressive therapies

Supportive Expressive

Therapeutic alliance is of utmost importance. Therapeutic alliance is of utmost importance.

is
Conversational style: patient is encouraged Flow of material important: patient is encouraged

to be goal-directed. to say whatever comes to mind.

Conversational style: therapist is responsive, a real Therapist maintains abstinence to the extent

person in a disciplined relationship. possible.

Emphasis on understanding. Emphasis on affect.
Direct measures are used to enhance self-esteem Self-esteem improves as a by-product of improved

(praise, encouragement). function and freedom from symptoms.

To the extent possible, therapy-related anxiety is avoided. Therapy-related anxiety is accepted if not disruptive.
Defenses are supported unless distinctly maladaptive. Defenses are often challenged.

Clarification and confrontation are employed. Clarification, confrontation, and interpretation are

employed.
Negative transference is examined; positive Analysis of transference is a major focus.
transference is not discussed.

VOLUMES #{149}NUMBER 4 #{149}FALL 1994

HELLERSTEIN ETAL. 303

There are a number of potential advan- knowledge that therapeutic activity is coher-ent
tages to the use of the supportive psychother- and purposeful.
apy model. It may be more palatable for
many patients than an abstinent psychoana- MoDELs AND DEPARTURES

lytic approach. The style of supportive psy- F R 0 M M o o t.: I.

chotherapy is conversational and aims to A therapist applying the expressive model
reduce, not increase, anxiety and to improve
self-esteem. In contrast, the abstaining traditionally
uses supportive techniques as
stance of exploratory therapy is socially un- needed, but this is always seen as a departure
usual and can be anxiety provoking. If this from the basic psychoanalytically oriented
approach cannot be proved to be more effi- model.’6 However, if the default model of
treatment is supportive therapy, whenever a
cacious than supportive therapy, why subject
patients to such discomfort? Moreover, such technique from analytic therapy is employed
an approach does not provide the patient a (such as silent listening, parrying questions,
or focus on transferential material), it must
pattern of normal human interaction, and
many patients are quickly repelled by it. The- be recognized as a departure from the model.
oretical considerations may determine the The therapist should have a rationale for its
therapist’s approach, but patients often use use. Just as support has been seen as a possi-
their therapists as models of correct thinking ble contaminant of analytic therapy, analytic
and speaking. Because the goals of treatment techniques must be seen as potential contam-
in supportive therapy are formulated explic- inants of supportive therapy. When the sup-
itly, seeking to avoid implications that the portive model isused, honest praise or other
therapist is the possessor of a body of secret esteem-building comments are volunteered
knowledge, the therapy provides a realistic when the opportunity presents itself, and the
model of collaborative effort. Although cog- therapist is always alert to avoid ambiguities
nitive-behavioral approaches have not been that might foster anxiety. If the therapist asks
found to be less effective than supportive a question, for example, the patient should
therapy, the exacting requirements of moni- always know what the therapist has in mind.
toring thoughts, keeping logs, and doing In expressive therapy, answering the patient’s
homework do not meet the needs of the question directly, without exploring its mean-
large number of individuals who prefer treat- ing, is a departure from the model. In sup-
ment that involves an interactive relation- portive psychotherapy, it is the nonresponse
ship. It is interesting to note that, according to a question that is a departure from the
to a study of therapists’ preferences, practi- model, because therapy is conducted as a
tioners of behavior therapy prefer an interac- conversation, not a monologue or an interro-
tive relationship when they seek treatment gation. In contrast to expressive therapy,
for themselves.’5 where transference is a major focus, in sup-
It is probable that far more patients are portive therapy, transferential aspects of the
involved in treatment that resembles support- patient-therapist relationship are a focus only
ive therapy than in primarily expressive treat- when negative interaction threatens to dis-
ment. Because supportive therapy for the rupt the treatment. Fantasy and free associa-
most part has not been formally taught, most tion are not encouraged. Character defenses
therapists have figured out for themselves and neurotic defenses are challenged only if
how to administer it. The fact that it has been maladaptive. Therapy is considered success-
conceptualized primarily as a series of de- ful if the patient achieves his or her goals. It
partures from something else deprives is not essential that affective change occur or
therapists, until they have considerable expe- that the patient acquire insight about the
rience, of the satisfaction associated with the origins of the disturbance.

JOURNAL OF PSYCHOTHERAPY PRACTICE AND RESEARCH

304 SUPPORTIVE THERAPY M0IEI.

IMPLICATIONS OF jects treated at the Menninger Clinic with as

RESEARCH FINDINGS much as 30 years’ follow-up. In a naturalistic
a
follow-up study with complex design that
A growing psychotherapy-outcome literature involved 1) review of initial treatment, termi-
a
suggests that 1) positive outcome in psycho- nation, and follow-up assessment and 2) a
therapy is frequently related to a positive pa- synthesis of treatment course and outcome,
tient-therapist relationship’7”8 and 2) there is he compared 22 psychoanalyses and 20 ex-
little evidence for differential efficacy for spe- pressive or supportive therapies in patients
cific psychotherapeutic approaches.’9’2#{176} whose diagnoses were severe neuroses, char-
Smith et al.,2#{176}for instance, in their often-cited acter neuroses, and impulse neuroses.
1977 paper “Meta-Analysis of Psychotherapy Wallerstein concluded that the results from
a
Outcome,” review of 400 outcome studies, psychoanalysis were more limited than ex-
found convincing evidence for the efficacy of pected, and that supportive interventions
psychotherapy; however, “few important dif- were surprisingly important in achieving sig-
ferences in effectiveness” could be estab- nificant and lasting change: “Changes
lished between very different types of achieved through supportive therapies...
psychotherapy. Furthermore, they found that provided often enough just as much struc-
the greatest overall effects of all forms of tural change and proved just as stable and
therapy were on 1) reduced fear and anxiety enduring as the changes achieved through
and 2) increased seLf-esteem. Both are areas expressive/analytic therapies” (p. 203).
of specific focus for supportive These preliminary research findings also
Further, a small but growing literature point to the importance of distinguishing
suggests that supportive therapy approaches theory from technique. For instance, psycho-
may have significant efficacy themselves, analytic theory of mental function may be
comparable with, if not at times better than, informative and helpful, but employing a psy-
exploratory or behavioral approaches. Conte choanalytic technique such as abstinence
and Plutchik2’ reviewed the evidence for the may be a less effective way to achieve change
efficacy of supportive psychotherapy. We than a more flexible, conversational support-
briefly describe some of these studies below. ive technique. It is worth emphasizing that
In a study of behavioral versus supportive the aforementioned authors did not employ
therapy in phobic patients, Zitrin et al.22’23 any consistent definition of supportive ther-
found no difference in effectiveness between apy; nevertheless, their findings suggest that
behavioral therapy and supportive therapy, more attention should be paid to the import-
both of which produced moderate to marked ance of supportive techniques in psychother-
benefit. In a population of schizophrenic pa- apy treatment and research.
tients, Stanton et al.24 and Gunderson et al.25
studied two forms of psychotherapy: an ex- SUPPORTIVE THERAPY

ploratory, insight-oriented therapy com- AS THE DEFAULT”

pared with a reality-adaptive, supportive TREATMENT

therapy. They found minimal outcome differ-

ence between the two treatments, but the Therapists who were trained in the 1950s,
supportive therapy subjects had less recidi- ‘60s, and ‘70s learned that
vism and increased independent function-
ing, in addition to less time spent in the The therapist’s task is to help the patient
hospital. Thus, supportive therapy was felt to understand himself by bringing uncon-
be the treatment of choice. scious ideas and memories into his con-
In another, more provocative, article, sciousness through verbalization of
. .
Wallerstein26 reviewed outcome for 42 sub- them.. The patient must be left as

VOLUME S #{149}NUMBER 4 #{149}FALL 1994

HELLERSTEIN ETAL. 305

freely as possible to develop spontane- development of transference responses, and

ously those reactions determined by his only when specifically indicated should trans-
childhood experience... the less . the ference be interpreted.
patient really knows about you the There is ample precedent in medicine
greater the chance to make transfer- for holding the more invasive and more ex-
ences, which are precious material for pensive treatments in reserve for use only
the therapeutic process.27 (pp. 23-24) when the gentler treatment has been found
ineffective (for example, use the cheapest
The proposition that supportive psycho- broad-spectrum antibiotic with the fewest
therapy could be the default treatment must side effects unless the pathogen is resistant or
appear quite foreign to those who attempt to unusually virulent). Analogously, supportive
give the “gold” of psychoanalysis whenever psychotherapy provides the broadest suffi-
possible, reverting to what may be seen as a cient and effective coverage for most prob-
diluted form of treatment if the patient’s lems presented for psychotherapy. Specific
motivation, ego structure, or constitution are types of treatment problems demonstrated to
not up to the challenge. According to Rock- be unresponsive to supportive therapy might
land,3 some experienced therapists have then suggest different modes of treatment,
stated that they do not know how to do sup- such as expressive or cognitive-behavioral
portive therapy. therapy.
In sum, we are proposing that the ap- It is clear that psychoanalytic therapy is
proach to most individual psychotherapy be no longer seen as the acme of therapies.
based on the model of supportive therapy. Cognitive and interpersonal therapies, which
Contemporary psychotherapy attempts to emphasize the patient’s learning, have be-
provide each patient with the optimum mix come widely accepted. Treatment based on
of supportive, supportive-expressive, and ex- Self-Theory requires of the therapist a much
pressive elements. The traditional approach, greater degree of interactivity and focus on
using expressive therapy as the model, calls the patient-therapist relationship. Rock-
for rationalization of any departure from this land” has noted that in the opinion of depart-
basic model. If supportive therapy is the ment chairmen and training directors, the
model, it is departure from that model that importance of psychoanalytically oriented
must be rationalized. We propose that not psychotherapy has declined in relation to the
only supportive therapy, but also supportive- importance of supportive therapy. Neverthe-
expressive and expressive-supportive therapy less, there have been no articles on education
should be based on the model of supportive in supportive herapy, O27 and supportive
therapy. We are proposing that unless there therapy is rarely taught to psychiatric train-
are specific indications to the contrary, all ees. Although supportive-expressive and ex-
therapy should be based on a real, but disci- pressive-supportive therapies are employed
plined, relationship between patient and far more often than psychoanalysis, teaching
therapist, should involve direct measures to dynamic psychotherapy presents the student
enhance the patient’s self-esteem, and should simultaneously with a body of theory about
avoid measures that diminish self-esteem personality formation and a model of treat-
(such as nonresponse, subtle argumentative ment based on it. When psychodynamic psy-
style, or questions that may be perceived as chotherapy is taught, the assumptions of
attacking).’2 Therapy should attempt to allay expressive therapy are omnipresent. If sup-
anxiety and should avoid measures that are portive therapy is to provide the model on
associated with continuation of anxiety. Only which most therapy is based, supportive as
when specifically indicated should therapy be well as analytic techniques must be taught,
conducted in a manner designed to facilitate supervised, and mastered during training.

JOURNAL OF PSYCHOTHERAPY PRACTICE AND RESEARCH

306 SUPPORTIVE ThERAPY MODEL

C 0 N C L U S I 0 N S lytic model of therapy. This treatment should

be guided by the rationales of supportive

In the past, conventional views have limited therapy and should use the techniques of

the use of supportive therapy to patients with supportive therapy. It is not the therapy by

whom expressive techniques are not ex-pected to unskilled counselors that was often recom-mended
obtain good results. We suggest that in the 1960s and ‘70s, but instead
these conventional views may deprive many should be a supportive therapy based on thor-
patients of the benefits of an effective form ough knowledge of personality development
of psychotherapy that is more comfortable, and psychopathology, carried out by practi-
more palatable, and, for many patients, more tioners who have had specific training and
effective than treatment in which the supervision in this modality.
therapist’s style is based on the psychoana-

F F F RE NC F S
R
I. Karasu TB et al (eds): The Psychiatric Therapies.
1988; 42:53-66
Washington, DC, American Psychiatric Association, 16. Eissler KR: Notes on problems of technique in the
1984 psychoanalytic treatment of adolescents. Psychoanal
2. Rockland LH: SupportiveTherapy: A Psychodynamic Study Child 1958; 13:223-254
Approach. New York, Basic Books, 1989 17. Hartley DE, StrUpp HH: The therapeutic alliance: its
3. Winston A, Pinsker H, McCullough L: A review of relationship to outcome in brief psychotherapy, in
supportive psychotherapy. Hosp Community Psychi- Empirical Studies of Psychoanalytical Theories, vol 1,
atry 1986; 37:1105-1114 edited by MaslingJ. Hillsdale, NJ, Analytic Press, 1983
4. Kernberg 0: Supportive psychotherapy: a re-evalua- 18. HorvathA0, SymondsBD: Relation between working
tion. Psychiatry 1982; 5:480-487 alliance and outcome in psychotherapy: a meta-anal-
5. Buckley P: Supportive therapy: a neglected treat- ysis.J Consult Clin Psychol 1991; 38:139-149
ment. Psychiatric Annals 1986; 16:515-521 19. Smith ML, Glass GV: Meta-analysis of psychotherapy
6. Novalis PN, Rojcewicz SJ, Peele R Clinical Manual of
outcomestudies. Am Psychol1977; 132:752-760
Supportive Psychotherapy. Washington, DC, Ameri-
20. Smith ML, Glass GV, MillerTI: The Benefits of Psy-
can Psychiatric Press, 1993 chotherapy. Baltimore, Johns Hopkins University Press, 1980
7. PinskerH: The role of theory in teaching supportive
therapy.AmJ Psychother(in press) 21. Conte HR, Plutchik R: Controlled research in sup-
8. LuborskyL: Principlesof Psychoanalytic Psychother- portive psychotherapy. Psychiatric Annals 1986; 16:530-533
apy: A Manual for Supportive-Expressive (SE) Meth-
ods. New York, Basic Books, 1984
22. Zitrin CM, Klein DF, Woerner MG: Behavior therapy,
9. Luborsky L, Singer B, Luborsky L: Comparative stud-
supportive psychotherapy, imipramine, and phobias.
ies of psychotherapies: is it true that “everyone has
Arch Gen Psychiatry 1978; 35:307-316
won and all must have prizes”? Arch Gen Psychiatry
23. Klein DF, ZitrinCM, WoernerMG, et al: Treatment
1975; 32:995-108
of phobias:behaviortherapyand supportive psycho-
10. Stiles WB, Shapiro DA, Elliott R: Are all psychothera-
therapy. Are there any specific ingredients? Arch Gen
pies equivalent? Am Psychol 1986; 41:165-180 Psychiatry 1983; 40:139-145
11. Rockland LH: A review of supportive therapy, 1986- 24. Stanton AH, GundersonJC, Knapp PH, et al: Effects
1992. Hosp Community Psychiatry 1993; 44:1053.- of psychotherapy in schizophrenia, I: design and
1060 implementation of a controlled study. Schizophr Bull
12. Pinsker H, Rosenthal RN, McCullough L: Supportive 1984; 10:520-563
dynamic psychotherapy, in Handbook of Short-Term
25. GundersonJG, Frank AF, Katz HM, et al: Effects of
Dynamic Therapy, edited by Crits.Christoph P. New
psychotherapyin schizophrenia,II: comparative out-
York, Basic Books, 1991, pp 220-247
come of two forms of treatment. Schizophr Bull 1984;
13. Pinsker H, Rosenthal R: Beth Israel Medical Center 10:564-598
Supportive Psychotherapy Manual (abstract). Social 26. Wallerstein RS: The psychotherapy research project
and Behavior Sciences Documents 1988; 18:2886 of the Menninger Foundation: an overview.J Consult
14. Bellak L: Emergency Psychiatryand Brief Psychother- Clin Psychol 1989; 57:195-205
apy. New York, Grune and Stratton, 1978 27. Colby KM: A Primer for Psychotherapists. New York,
15. NorcrossJ: The therapist’s therapist. AmJ Psychother
Ronald Press, 1951

VOLUME S #{149}NUMBER 4 #{149}FALL 1994