See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/20964440

Psychodynamic treatment of schizophrenia: Is there a future?

Article  in  Psychological Medicine · June 1990

DOI: 10.1017/S003329170001758X · Source: PubMed

CITATIONS READS
104 966

2 authors:

Kim Mueser Howard Berenbaum

Boston University University of Illinois, Urbana-Champaign
635 PUBLICATIONS   38,673 CITATIONS    169 PUBLICATIONS   5,331 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Illness Management and Recovery View project

Community Effectiveness of Virtual Reality Job Interview Training for Adults with Severe Mental Illness View project

All content following this page was uploaded by Howard Berenbaum on 06 May 2015.

The user has requested enhancement of the downloaded file.

Psychological Medicine
http://journals.cambridge.org/PSM

Additional services for Psychological Medicine:

Email alerts: Click here

Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here

Psychodynamic treatment of schizophrenia: is there a future?

Kim T. Mueser and Howard Berenbaum

Psychological Medicine / Volume 20 / Issue 02 / May 1990, pp 253 - 262

DOI: 10.1017/S003329170001758X, Published online: 09 July 2009

Link to this article: http://journals.cambridge.org/abstract_S003329170001758X

How to cite this article:

Kim T. Mueser and Howard Berenbaum (1990). Psychodynamic treatment of schizophrenia: is there a
future?. Psychological Medicine, 20, pp 253-262 doi:10.1017/S003329170001758X

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/PSM, IP address: 203.114.109.66 on 11 Dec 2013

Psychological Medicine, 1990, 20, 253-262
Printed in Great Britain

EDITORIAL

Psychodynamic treatment of schizophrenia: is there a future?1

The psychoanalytic treatment of schizophrenia has had a long and controversial history. Freud and
his more orthodox followers felt that schizophrenics were not suitable for psychoanalysis. However,
a long list of dissenters have enthusiastically advocated this treatment (e.g. Rosen, 1947; Fromm-
Reichmann, 1950; Will, 1958; Searles, 1965; Pao, 1979). For decades the dominant treatment
model, the efficacy of psychoanalytic and psychodynamic interventions for schizophrenia has
recently been questioned (May, 1968, 1984). Despite some decline in popularity, psychodynamic
therapy continues to be an important individual treatment modality for schizophrenics.
Psychodynamic treatment is usually provided in combination with pharmacotherapy (Feinsilver,
1983; Normand & Bluestone, 1986) for schizophrenics in both in-patient and out-patient settings
(e.g. Karon & Vandenbos, 1982; Stone et al. 1983; Frosch, J983; Auerhahn & Moskowitz, 1984;
Spotnitz, 1985; Silverman & Lachmann, 1985; Feinsilver, 1986; Kernberg, 1986; Lassers, 1986;
Mann, 1986; Benedetti, 1987; Munich, 1987; Strean, 1988).
To examine whether there have been changes in the general scholarly and clinical interest in the
psychodynamic treatment of schizophrenia, we conducted a review of the literature published in
journals over the past twenty-one years. Using the Medline (National Library of Medicine, 1987)
and PsychlNFO (American Psychological Association, 1988) data bases, we identified articles using
the index terms or title words 'schizophrenia', 'psychodynamic', or 'psychoanalysis'. A total of 478
articles were located, of which 159 (33 %) described actual schizophrenic patients who had been
treated according to this approach. (Duplicate articles describing the treatment of the same patients
were counted only once.) Fig. 1 contains a plot of the number of psychodynamically-orientated
treatment articles on schizophrenia and the total number of articles on psychodynamic theory and
schizophrenia from 1966 through 1987. Inspection of the figure suggests a recent decline in the

1966 1970 1975 1980 1985

Year
FIG. 1. Plot of the number of journal articles on psychodynamic treatment of schizophrenia and total number of articles on
psychodynamic theory and schizophrenia published between 1966 and 1987 (source: Medline and PsychlNFO data bases).

1
Address for correspondence: Dr Kim T. Mueser, Medical College of Pennsylvania at Eastern Pennsylvania Psychiatric Institute,
3200 Henry Avenue, Philadelphia, Pennsylvania 19129, USA.

253
254 Editorial: Psychodynamic treatment of schizophrenia

number of treatment and theoretical articles on psychodynamics and schizophrenia, to a level

similar to that following a decline which occurred between 1975 and 1980. Although the yearly rate
of published articles on psychodynamic theory and treatment of schizophrenia have fluctuated over
the past twenty years, the number of publications per year has never dropped below ten.
Considering that the number of journal articles published is only a crude index of the current role
of psychodynamic therapy in schizophrenia since only a small fraction of practitioners publish their
treatment cases, the literature review and our own clinical experience indicate a continuing
influence, albeit not predominant, in the treatment of schizophrenia. This influence, in light of
recent negative outcome studies (Gunderson et al. 1984), makes a reassessment of the value of this
treatment timely. In this article we will briefly review the data concerning the effectiveness of
psychodynamic treatment for schizophrenia, and evaluate the implications of these results for future
treatment and research priorities.

TREATMENT OUTCOME RESEARCH

Before reviewing the controlled clinical trials, we will briefly discuss the outcome criteria that we feel
should be used to judge the effectiveness of interventions with schizophrenic patients. Schizophrenia
can affect many different aspects of functioning, and different indices of functioning tend to have
only moderate or low intercorrelations: thus a multidimensional approach to the measurement of
treatment outcome is essential (May & Tuma, 1964; Strauss & Carpenter, 1972; Avison &
Speechley, 1987). The criteria for evaluating outcome can be classified in a number of different ways,
but the following four dimensions are usually included in any comprehensive assessment: (1)
symptoms; (2) rehospitalization; (3) social adjustment; and (4) vocational adjustment.
Since the present review concerns the effects of psychodynamic treatment on the outcome of
schizophrenia, measures of theoretical constructs such as ego strength and insight which are
hypothesized to have an impact on outcome will not be considered here. Examining the effects of
psychodynamic treatment on such constructs is important for validating the theories upon which
these interventions are based. However, the clinical efficacy of treatments must be evaluated solely
on the basis of their impact on widely accepted outcome criteria, such as those described above.

CONTROLLED CLINICAL OUTCOME STUDIES

Four studies have compared psychodynamic treatment with other treatments for schizophrenia,
with the primary research workers including May (1968), Karon and VandenBos (1972, 1975),
Grinspoon (Grinspoon et al. 1972), and Gunderson (Stanton et al. 1984; Gunderson et al. 1984).
Table 1 provides an overview of the four studies. Included are: random assignment to treatment
groups, patient chronicity (number of hospitalizations and time in hospital), treatment groups,
length and duration of treatment, methodological weaknesses, and outcome results for four
dimensions (symptomatology, rehospitalization, social adjustment, and vocational adjustment).
While these studies have been reviewed before (Stanton et al. 1984; Gomes-Schwartz, 1984), the
present review goes beyond these by examining the results according to different dimensions of
outcome, and evaluating the implications of the results for future treatment and research in view
of advances in alternative treatment approaches.
Three of the four studies described in Table 1 found that psychodynamic treatment failed to exert
any beneficial effect on outcome, either alone or in combination with antipsychotic medication.
Grinspoon's study (Grinspoon et al. 1972) was unique in its treatment of chronic, institutionalized
schizophrenics. While they failed to find any beneficial effects of psychodynamic treatment, serious
methodological flaws in their design limit the conclusions that can be drawn. The May (1968) study
also failed to find any beneficial effects of psychodynamic treatment. This study did not have any
major methodological weaknesses from an experimental design perspective, but it has been
criticized for its use of inexperienced psychotherapists (Karon & VandenBos, 1970; but cf. Tuma
& May, 1974).
 Editorial: Psychodynamic treatment of schizophrenia 255

Table 1. Characteristics and results of outcome studies

Stanton et al.
Karon & VandenBos Grinspoon et al. Gunderson et al.
May (1968) (1972, 1975) (1972) (1984)

Random assignment Yes Yes1 No* Yes

Number of prior 0 0-3 2-7 0-9
hospitalization
Months of prior 0 0-2 780 0-9
hospitalization
Treatment 3 groups (H) PD(46) PD (12) 4
PD (10) PD-D (43)
PD-D (44) PD-D (12) PD-D (10) Reality
D(48) D(12) D(21) Adaptive
Milieu (43) Support therapy
ECT (47) RAS-D (52)
Length duration of RX. 2 hrs/wk PD: 5 hrs/wk then 2 hrs/wk PD-D 2 hrs/wk
for 1 yr 1 hr/wk for 20 months for 2 yrs for 2 yrs
PD-D: 3 hrs/wk then RAS-D 0-6 hrs/wk
1 hr/wk for 20 months for 2 yrs
Symptoms D = PD-D> 5 « PDSs' D = PD-D>8 RAS-D = PD-D*
ECT> PD-D> PD
PD = M D
Rehospitalization D = PD-D = ECT > For experienced — RAS-D >»
PD = M therapists: PD-D
PD = PD-D > D
For inexperienced
therapists:
PD > PD-D = D
Social adjustment — — D = PD-D > 1 0 RAS-D £ • • "
PD PD-D
Vocational adjustment — — — RAS-D > '
PD-D
Method weakness Inexperienced Only 2 experienced Non-random assignment High dropout
therapists therapists to treatment groups rate (42%)
Non-random assignment Drug only treatment Absence of
of patients to given in different drug only
experienced or environment control
inexperienced Patients very chronic group
therapists
Drug only group
transferred to
different hospital
I
Patients were randomly assigned to PD, PD-D, and D groups, but were not randomly assigned to experienced or inexperienced therapists.
* Patients in the D group were those who elected to remain at the state hospital and were compared with the PD and PD-D groups, which
were composed of those patients who consented to be transferred to a clinical research centre for treatment.
3
PD = Psychodynamic; D = Drugs; PD-D = Psychodynamic plus drugs.
4
The PD group was treated with 'active psychoanalytic psychotherapy,' whereas PD-D was treated with 'ego-analytic psychoanalytic
psychotherapy'.
s
' > ' denotes statistically significant (P < 005) difference between treatment groups for most or all relevant measures. ' > ' denotes a
significant difference between treatment groups for at least one measure, but an equal number or more non-significant differences on other
relevant measures.
' Based on Symptom Rating Sheet, Ann Arbor, and Clyde Mood Scale (May, 1968, pp. 168, 170, 173, 175).
' Based on Feldman-Drasgow Visual Verbal Test for thought disorder (Feldman & Drasgow, 1951) and clinical status interviews (Karon
& Vandenbos, 1972, p. 116). The Visual Verbal Test is not a direct measure of symptomatology, but rather is a test of abstract conceptual
thinking that has been found to discriminate schizophrenics from non-schizophrenics.
8
PD and PD-D differences based on Behavior Disturbance Index (Grinspoon et al. 1972, pp. 149-151).
• Based on Gunderson et al. (1984, pp. 592, 596).
10
PD and PD-D differences based on Hospital Adjustment Scale (Grinspoon et al. 1972, pp. 149-151).
II
Household responsibilities: RAS-D > PD-D; Significant relationships: RAS-D = PD-D (Gunderson et al. 1984, pp. 592, 596).

Gunderson et a/.'s collaborative study (Gunderson et al. 1984; Stanton et al. 1984) did not suffer
from any of the shortcomings noted in the two studies previously mentioned, and employed highly
experienced psychotherapists. This study also had by far the most comprehensive assessment of
treatment outcome of all the studies described in this paper, including multiple assessments for each
outcome dimension, and stands as an example of a well-conceived and executed treatment study.
These investigators compared reality-adaptive supportive (RAS) therapy with exploratory insight-
256 Editorial: Psychodynamic treatment of schizophrenia

orientated (EIO) psychodynamic therapy. The following descriptions illustrate the differences
between the insight orientated, psychodynamic treatment and the RAS approach of attempting to
enhance practical coping skills. RAS therapy ' focused on problems in the current living situation
of the patient. In contrast to the EIO therapy, there was little attempt to explore the past and to
seek correlates between past experience and the present. Rather, the exploration of the present was
intended to identify problems that could be solved or that could be expected to recur in the future
so that more effective coping strategies could be mapped out. Another major feature of the RAS
therapy was its focus on the patient's behavior itself rather than the covert meanings behind that
behavior' (Stanton et al. 1984, p. 535). The EIO treatment was 'a form of analytic psychotherapy
adapted from that described by Frieda Fromm-Reichmann (1950) "...it was by no means
psychoanalysis... its aim is to increase insight"' (Stanton et al. 1984, p. 536).
Gunderson et al. found that psychodynamic therapy (EIO) was clearly inferior to reality-adaptive
supportive (RAS) therapy in three out of four outcome criteria: rehospitalization, vocational
adjustment, and to a lesser extent social adjustment. The two treatments did not differ in their
impact on symptoms.
Since our summary of the Gunderson et al. data is at some variance with their own summary, a
brief discussion of this discrepancy is warranted. Gunderson et al. (1984) wrote 'when one looks at
our complete array of results, the overriding impression one gets is that in most respects the EIO
(psychodynamic) and RAS patients performed similarly' (p. 582). Nevertheless, RAS-treated
patients spent significantly less time in the hospital and more time employed than their
psychodynamically treated counterparts. Furthermore, RAS patients assumed more major
household responsibilities than psychodynamically treated patients, indicating that at least one
aspect of social adjustment was superior for the RAS than EIO group. Other dimensions of social
adjustment, such as social dysfunction and social relationships, did not differ between the two
treatments, nor did any measures of symptoms. The only advantages Gunderson et al. attribute to
the psychodynamic treatment over the RAS treatment is in the area of 'ego functioning', and even
here the treatments did not differ at a statistically significant level (May, 1984) on any of the
variables measuring this construct (adaptive regression, ego weakness, and subjective experience).
Thus, our interpretation of the data differs from Gunderson et a/.'s conclusion that the two
treatments were essentially equivalent.
Karon & VandenBos (1972, 1975) reported that psychodynamic treatment was superior to
antipsychotic drug treatment. Serious methodological shortcomings limit generalizability from this
study. Although patients were randomly assigned to one of the three treatment groups, there was
a confound between therapist and treatment modality. Half of the patients in the psychodynamic-
treatment-alone group were treated by one' experienced' therapist and half of the patients receiving
psychodynamic treatment plus antipsychotics were treated by a second 'experienced' therapist. The
remaining patients receiving therapy were treated by 'inexperienced' therapists. Since only one
experienced therapist treated patients in one group, and a different experienced therapist treated
patients in the second group, the obtained interaction between therapist experience and concomitant
pharmacotherapy may simply reflect differences between the therapists. In addition, patients in the
drug-only group were transferred to a different hospital for treatment. Thus, it is impossible to
determine whether the effects noted by Karon & VandenBos resulted from different types of
treatment, different therapists, or different treatment facilities.

NATURALISTIC STUDIES
Two reports have recently been published on lengthy followups of large samples of schizophrenic
patients treated with psychodynamic therapy. Stone (1986) followed up 72 schizophrenic patients
who received an average of 12-3 months of intensive psychodynamic therapy at the New York State
Psychiatric Institute. Ten to twenty years later, more than half of the patients were substantially
dysfunctional.
McGlashan (1984a,&) reported on the outcomes for schizophrenic patients treated at Chestnut
 Editorial: Psychodynamic treatment of schizophrenia 257

Lodge, a long-term private residential facility that specializes in intensive, psychoanalytic treatment.
Of 163 schizophrenic patients treated at the Lodge for an average of over three years, two-thirds
were functioning marginally or worse 15 years later. Thus, this large sample of schizophrenics
appears to have gained little, if anything, from their intensive psychoanalytic treatment. McGlashan
concurs with this conclusion in his summary of the effects of intensive psychodynamic treatment for
schizophrenia: 'Unfortunately, we still have not improved much on Kraepelin's work. In effect,
with chronic schizophrenia, we are still just beginning to fight' (McGlashan 19836, p. 600).

DELETERIOUS EFFECTS OF PSYCHODYNAMIC THERAPY

Strupp et al. (1977) have pointed out that psychotherapy can sometimes lead to deleterious rather
than beneficial effects. There are several reasons why a search for harmful effects of psychodynamic
treatment of schizophrenic patients is especially warranted. First, as noted earlier, Freud himself felt
that schizophrenics were not suitable candidates for psychoanalysis. Second, most schizophrenics
would match the following description of the type of patient who was found to respond poorly to
psychoanalysis in the Menninger Foundation Psychotherapy Research Project: 'patients with low
initial quality of interpersonal relationships, low initial anxiety tolerance, and low initial motivation'
(Kernberg, 1973, p. 66). Finally, Stone (1986) has suggested that psychodynamic treatment can be
harmful to schizophrenic patients because it can expose them to memories and insights that they are
unable to deal with emotionally.
Some of the data from the aforementioned outcome studies suggest that psychodynamic
treatment may have deleterious rather than beneficial effects for schizophrenic patients. Gunderson
et al. (1984) presented data on the relation between outcome and the number of months patients
were treated with psychodynamic or reality-adaptive therapy. One would expect that the longer
patients remain in an effective treatment, the better their outcomes would be. However, these
expected correlations between outcome and time in treatment were not found for the psychodynamic
treatment. In fact, when the patient sample was restricted to McLean Hospital, many of the
correlations were significantly negative. Table 2 contains correlations between the number of
months patients spent in psychodynamic or reality-adaptive therapy at McLean Hospital and six
different outcome measures. One can see from this Table that time spent in psychodynamic
treatment was consistently correlated with a poor outcome. These correlations were statistically
significant at the P < 0-05 level on two out of six outcome measures and approached significance
(/> < 010) on two additional measures. On the other hand, positive correlations were found between
outcome and time in reality adaptive therapy on five out of six measures, three of which were
statistically significant.
One might posit that the reason for the negative correlations between months in psychodynamic
treatment and outcome is that time in therapy was largely determined by the severity of the patient's

Table 2. Correlations between months in reality-adaptive supportive (RAS) or psychodynamic (PD)

therapy and amount of improvement for patients treated at McLean Hospital. Positive correlations
denote greater improvement with longer periods of treatment (adapted from Gunderson et al. 1984,
p. 597)
RAS PD

Time in hospital 0-52"* -0-30 +

Rehospitalizations -013 -015
Days employed full time 0-44' -0-75*»*»
Occupational level reached 0-26 -0-32 +
Household responsibilities 0-26 -0-39'
Significant relationships 0 39* -0-28

Note: + / > < 0 1 0 ; *P<0-05; ** P < 0 0 1 ; *** P < 0005; '*** P < 0001.
258 Editorial: Psychodynamic treatment of schizophrenia

Table 3. Correlations between months in reality-adaptive supportive (RAS) or psychodynamic (PD)

therapy and amount of improvement for patients treated at Boston University and VA Hospitals.
Positive correlations denote greater improvement with longer periods of treatment (adapted from
Gunderson et al. 1984, p. 597)
RAS PD

Time in hospital -0-24 -0-49

Rehospitalizations 003 0-68
Days employed full time 0-30 -008
Occupational level reached 019 0-30
Household responsibilities 0-23 -009
Significant relationships 002 -017

Note: " P < 0 0 1 .

illness. This explanation would not explain why the correlations were negative for the
psychodynamic treatment but positive for the reality adaptive treatment. In addition, patients who
dropped out of either therapy within the first six months of treatment did not differ from patients
who remained in treatment in pre-morbid adjustment, chronicity, or two-year outcome (Stanton
et al. 1984; Katz et al. 1984), suggesting that less severely ill patients did not tend to terminate
treatment prematurely.
The hypothesis that psychodynamic therapy had deleterious effects on schizophrenics, as
evidenced by the negative correlations between months in therapy and outcome at McLean
Hospital, must be tempered by the fact that such correlations were not obtained for patients treated
at the other two sites of the study: Boston University and the Bedford Veterans Administration
Hospital. Table 3 contains the correlations between months in RAS or psychodynamic treatment
and clinical outcome for patients treated at Boston University and the VA hospital. No consistent
pattern of correlations separates the two treatments. The significant differences between months of
RAS or psychodynamic treatment and outcome for patients treated at McLean (Table 2),
contrasted with the absence of such differences at the other two hospitals (Table 3), may reflect the
fact that the two treatments were more distinguished from each other at McLean than the other two
hospitals. For example, the McLean RAS and psychodynamic therapists were more experienced
and divergent in their attitudes and practices. In addition, there was more institutional support for
the practice of psychodynamic therapy at McLean Hospital, leading the investigators to anticipate
that' the specific effects of the EIO (psychodynamic) treatment might emerge more clearly in the
McLean setting than in the other settings' (Stanton et al. 1984, p. 569). Another reason why the
correlations at the Boston University and VA hospitals may have differed from the correlations at
McLean is that the sample size of the former group was small, approximately half the size of the
McLean group. While these data are open to different interpretations, the possibility that
psychodynamic treatment provided at McLean Hospital had negative effects cannot be dismissed.
In the follow-up study reported by Stone (1986), 20 % of the patients who received psychodynamic
therapy committed suicide, approximately double the suicide rate commonly reported for
schizophrenics (Winokur & Tsuang, 1975; Drake & Cotton, 1986; Roy, 1986; Roy et al. 1986). This
high suicide rate may be further evidence for the potentially negative effects of psychodynamic
treatment for this population.
An alternative explanation for the high suicide rate in the Stone study is that many of the
individuals diagnosed as schizophrenic may have actually been psychotic depressives or bipolar
disorder patients. Although this is a possibility, diagnoses in this study were made using DSM-III
criteria. The question of diagnosis is probably more pertinent for those studies conducted prior to
DSM-III, such as Karon & VandenBos (1972, 1975). The possibility of misdiagnosis does not
change the meaning of the failure of the studies reviewed to find a beneficial effect for
psychodynamic treatment, however. The inclusion of depressives would probably have increased
the likelihood of obtaining positive results rather than decreasing it, since there is empirical support
 Editorial: Psychodynamic treatment of schizophrenia 259

for the efficacy of at least some forms of psychodynamic treatment for depression (Hersen et al.
1984).
The thesis that treatments which are too intensive may have adverse effects on schizophrenic
patients is not new (Drake & Sedere, 1986). Psychoanalytic treatment of schizophrenia is usually
provided with greater intensity than other psychotherapeutic approaches. For example, the patients
in the Gunderson study (Gunderson et al. 1984) who received psychodynamic therapy spent more
than three times as much time in therapy as patients treated with reality adaptive therapy (20 v.
06 hrs/wk, respectively). However, intensive treatments do not invariably worsen the outcome of
schizophrenia. Social skills training, a highly structured and directive treatment, can produce
positive outcomes even when multiple sessions are held daily (Liberman et al. 1986).
Thus, it is clear that having frequent sessions will not necessarily lead to worse outcome. It is
possible, however, that therapy that is too emotionally intense may be harmful for at least some
schizophrenics. If psychodynamic treatment is harmful for schizophrenics, it is probably the
emotional intensity of the treatment rather than the frequency of sessions that is responsible.

IMPLICATIONS FOR FUTURE TREATMENT AND RESEARCH

The evidence that psychodynamic treatment worsens the outcome of schizophrenia is indirect and
debatable, largely due to the fact that only one controlled study compared schizophrenics receiving
psychodynamic therapy plus antipsychotics with patients receiving only antipsychotics (May, 1968).
On the other hand, the data supporting the efficacy of this treatment are even less convincing. These
findings contrast with the results of recent outcome studies on the efficacy of social skills training
(Bellack et al. 1984; Liberman et al. 1986; Hogarty et al. 1986) and certain forms of family therapy
for schizophrenia (Leff et al. 1982, 1985; Falloon & Pederson, 1985; Hogarty et al. 1986).
For example, Liberman et al. (1986) found that during the two years following hospital discharge,
a group of schizophrenics who received nine weeks of intensive social skills training had significantly
better social functioning, less severe symptomatology, and spent less time in the hospital than a
group of schizophrenics who received a holistic health treatment. Both groups of schizophrenics
received antipsychotic medication in addition to their other treatments while they were in the
hospital.
In a study of schizophrenics who were receiving maintenance antipsychotic treatment and who
were living in high Expressed Emotion (EE) households, Hogarty et al. (1986) found that none of
the schizophrenics receiving both psychoeducational family therapy and social skills training
relapsed over one year, compared to 20% of the schizophrenics receiving only skills training, 19%
receiving only family treatment, and 41 % receiving only supportive therapy. Two-year relapse rates
were higher for all groups (25, 42, 32 and 66% for the four treatment groups, respectively), but
nevertheless supported the efficacy of the psychosocial interventions (Hogarty et al. 1987).
Falloon and his colleagues (Falloon & Pederson, 1985; Falloon et al. 1985, 1987) reported that
17% of schizophrenic patients living in high EE homes treated for two years with behavioural
family therapy relapsed, compared to 83 % of patients treated with individual supportive therapy.
These differences for patients receiving the family treatment were paralleled by improvements in
social and vocational performance, as well as fewer rehospitalizations, less time spent in the
hospital, and lower doses of antipsychotics prescribed by psychiatrists who were blind to treatment
assignment. Last, Tarrier et al. (1988, 1989) reported that a nine-month behavioural intervention
with families significantly reduced the relapse rates of schizophrenic patients returning home after
in-patient treatment for an acute exacerbation. Patients with high EE relatives who received the
behavioural intervention had two-year relapse rates of 33 %, comparable to patients living with low
EE relatives, and significantly lower than the 59 % relapse rate of patients living with high EE
relatives who received no treatment.
We recognize that the number of studies demonstrating the efficacy of treatments such as social
skills training and family therapy may not be large and some may have methodological
260 Editorial: Psychodynamic treatment of schizophrenia

shortcomings (e.g. Leff et al. 1982; Liberman et al. 1986). Clearly, additional research validating
such treatments is warranted, especially concerning the effects of interventions on outcome criteria
other than relapse rate (e.g. social and vocational adjustment). Nonetheless, the pattern of results
has been consistently positive. Despite the failure of empirical investigations to demonstrate that
psychodynamic treatment is effective for schizophrenics, and the development of other interventions
that controlled studies suggest improve outcome, psychodynamic therapy continues to be offered
as a treatment for schizophrenia. Stanton et al. (1984) noted, 'Of the many institutions approached
about possible collaboration, most of those that had the strongest identification and tradition with
the practice of intensive psychotherapy still felt they could not randomly withhold this treatment
from patients admitted to their institutions' (p. 524). Thus, psychodynamic therapy appears to
remain an influential choice in the treatment of schizophrenia.
Resources for treating schizophrenia are limited. A probable consequence of providing
schizophrenic patients with psychodynamic treatment is that they will be deprived of other
treatments which have been demonstrated to be effective, such as social skills training or some forms
of family therapy. Since psychodynamic treatment has not been demonstrated to be effective, and
more effective treatments are available, we propose a moratorium on the use of psychodynamic
treatments for schizophrenia. Indeed, if a drug had the 'efficacy profile' of psychoanalysis it would
surely not be prescribed, and no one would have the slightest qualm about relegating it to the 'dust
bin of history'. A further need for a moratorium is that clinicians who are taught to treat
schizophrenic patients with psychodynamic techniques may otherwise refrain from taking the
necessary steps to obtain training in empirically validated clinical interventions for schizophrenia.
It is possible that some schizophrenics may benefit from psychodynamic treatment, as they might
from any of the myriad of unproven pharmacological strategies such as orthomolecular treatment.
However, at present there is no data that would permit a clinician to predict which patients would
benefit. To offer such a treatment before attempting an intervention with greater empirical support
raises ethical questions. In order to provide schizophrenics with psychodynamic treatment, the
clinician must assume that he or she either practices a superior variation of therapy than has been
previously tested or is able to identify a responsive subset of patients. To our knowledge, neither
of these assumptions has been evaluated. Perhaps patients who do not appear to respond to social
skills training or certain forms of family therapy could be treated psychodynamically? This is
possible, but our preference would be to try Gunderson et al.'s 'reality-adaptive supportive'
therapy, which outperformed the more costly psychodynamic treatment.
Just as resources for patient care are limited, so are resources for research. It would be economical
to study how the theoretical constructs presumed to be modified by psychodynamic treatment are
related to outcome among schizophrenic patients before conducting further expensive outcome
studies. Gunderson et al. (1984) reported that there was a trend for schizophrenics who received
psychoanalytic treatment to have higher scores on measures of ego functioning than schizophrenics
who received reality-adaptive therapy. Despite having higher ego functioning scores, the
psychodynamically treated patients had poorer social and vocational adjustment and were
rehospitalized more frequently than patients receiving reality-adaptive treatment. If future research
found that constructs such as ego functioning or insight were predictive of poor social adjustment
and more frequent rehospitalization, the desirability of goals to improve these constructs would
need to be questioned. Determining appropriate goals for treatment will be clearer once we have a
more thorough understanding of the variables that contribute to social and vocational adjustment,
the absence of overt psychotic symptoms, and the ability to remain out of the hospital. Once such
information has been obtained it will be possible to develop specific therapeutic interventions to
address each of the factors necessary for a successful outcome.
We have proposed a moratorium on the use of psychodynamic treatments for schizophrenia. Our
intention is not to alienate dedicated practitioners, but rather encourage them to consider treatment
alternatives. The empirical evidence indicates that in order to help schizophrenics best, clinicians
need to forego their psychodynamic formulations and instead focus on building individual and
family social competence, while decreasing ambient stress. Practitioners who have been treating
 Editorial: Psychodynamic treatment of schizophrenia 261

schizophrenics with insight-orientated approaches need not abandon their attempts to help such
patients. Their experience can be harnessed by focusing on patients' current problems in living and
their ability to cope with daily stresses. Descriptions of recent treatment approaches are readily
available to the interested clinician (e.g. Leff et al. 1982; Curran & Monti, 1982; Kelly, 1982;
Falloon et al. 1984; Anderson et al. 1986; Liberman et al. 1989) as are training opportunities.
Recent advances in the psychosocial treatment of schizophrenia, fuelled by a surge in funding for
schizophrenia research, make this an exciting and hopeful area in which to work. Although
clinicians tend to be slow to abandon their old methods of treating patients (Barlow, 1981; Backer
et al. 1986), we hope they will take advantage of these recent advances in order to improve the
outcome of their patients with schizophrenia.
KIM T. MUESER AND HOWARD BERENBAUM

Appreciation is extended to Terry Wiggins for assistance in conducting the literature search and to the
following persons for their comments on an earlier draft of this paper: Henry Beck, Alan S. Bellack, Anita
DeLongis, Daniel Klein, Arnold A. Lazarus, Rachel Lehr, Douglas Levinson, Robert P. Liberman, Gregory
Miller, Nina R. Schooler, George M. Simpson, Shimon Waldfogel, and William Wilson.
This study was supported by National Institute of Mental Health grants MH 3998 and MH 3836.

REFERENCES Family management in the prevention of morbidity of schizo-

phrenia: social outcome of a two-year longitudinal study.
American Psychological Association (1988). Thesaurus of Psycho- Psychological Medicine 17, 59-66.
logical Index Terms (Fifth Edition). American Psychological Feinsilver, D. B. (1983). Application of Pao's theories to a case study
Association: Washington, DC. of the use and misuse of medication. Psychoanalytic Inquiry 3,
Anderson, C. M., Hogarty, G. E. & Reiss, D. J. (1986). Schizophrenia 125-144.
and the Family. Guilford Press: New York. Feinsilver, D. B. (1986) (ed.) Towards a Comprehensive Model for
Auerhahn, N. C. & Moskowitz, M. B. (1984). Merger fantasies in Schizophrenic Disorders: Psychoanalytic Essays in Memory of
individual inpatient therapy with schizophrenic patients. Psycho- Ping-Nie Pao, M.D. The Analytic Press: Hillsdale, NJ.
analytic Psychology 1, 131-148. Feldman, M. J. & Drasgow, J. A. (1951). Visual-verbal test for
Avison, W. R. & Speechley, K. N. (1987). The discharged psychiatric schizophrenia. Psychiatric Quarterly (Supplement) 25, 55-64.
patient: a review of social, social-psychological, and psychiatric Fromm-Reichmann, F. (1950). Principles of Intensive Psychotherapy.
correlates of outcome. American Journal of Psychiatry 144, 10-18. University of Chicago Press: Chicago.
Backer, T. E., Liberman, R. P. & Kuehnel, T. G. (1986). Dis- Frosch, J. (1983). The Psychotic Process. International University:
semination and adoption of innovative psychosocial interventions. New York.
Journal of Consulting and Clinical Psychology 54, 111-118. Gomes-Schwartz, B. (1984). Individual psychotherapy of schizo-
Barlow, D. H. (1981). On the relation of clinical research to clinical phrenia. In Schizophrenia: Treatment, Management, and Re-
practice: current issues, future directions. Journal of Consulting habilitation (ed. A. S. Bellack), pp. 307-335. Grune and Stratton:
and Clinical Psychology 49, 147-155. New York.
Bellack, A. S., Turner, S. M., Hersen, M. & Luber, R. F. (1984). An Grinspoon, L., Ewalt, J. R. & Shader, R. I. (1972). Schizophrenia:
examination of the efficacy of social skills training for chronic Pharmacotherapy and Psychotherapy. Williams & Wilkins
schizophrenic patients. Hospital and Community Psychiatry 35, Company: Baltimore.
1023-1028. Gunderson, J. G., Frank, A., Katz, H. M., Vannicelli, M. L., Frosch,
Benedetti, G. (1987). Psychotherapy of Schizophrenia. New York J. P., Knapp, P. H. (1984). Effects of psychotherapy in schizo-
University Press: New York. phrenia: II. Comparative outcome of two forms of treatment.
Curran, J. P. & Monti, P. M. (1982) (eds.) Social Skills Training: A Schizophrenia Bulletin 10, 564-598.
Practical Handbook for Assessment and Treatment. Guilford Press: Hersen, M., Bellack, A. S., Himmelhoch, J. M. & Thase, M. E.
New York. (1984). Effects of social skill training, amitriptyline, and psycho-
Drake, R. E. & Cotton, P. G. (1986). Depression, hopelessness and therapy in unipolar depressed women. Behavior Therapy 15,21-40.
suicide in chronic schizophrenia. British Journal of Psychiatry 148, Hogarty, G. E., Anderson, C. M., Reiss, D. J., Kornblith, S. J.,
554-559. Greenwald, D. P., Javna, C. D. & Madonia, M. J. (1986). Family
Drake, R. E. & Sedere, L. I. (1986). The adverse effects of intensive psychoeducation, social skills training and maintenance chemo-
treatment of chronic schizophrenia. Comprehensive Psychiatry 27, therapy : I. One year effects of a controlled study on relapse and
313-326. expressed emotion. Archives of General Psychiatry 43, 633-642.
Falloon, I. R. H. & Pederson, J. (1985). Family management in the Hogarty, G. E., Anderson, C. M. & Reiss, D. J. (1987). Family
prevention of morbidity of schizophrenia. The adjustment of the psychoeducation, social skills training, and medication in schizo-
family. British Journal of Psychiatry 147, 156-163. phrenia: The long and short of it. Psychopharmacology Bulletin 23,
Falloon, I. R. H., Boyd, J. L. & McGill, C. W. (1984). Family Care 12-13.
of Schizophrenia. Guilford Press: New York. Karon, B. P. & VandenBos, G. R. (1970). Experience, medication
Falloon, I. R. H., Boyd, J. L., McGill, C. W., Williamson, M., and the effectiveness of psychotherapy with schizophrenics: A note
Razani, J., Moss, H. B., Gilderman, A. M. & Simpson, G. M. on Drs. May and Tuma's conclusions. British Journal of Psychiatry
(1985). Family management in the prevention of morbidity of 116, 427-428.
schizophrenia. Clinical outcome of a two year longitudinal study. Karon, B. P. & VandenBos, G. R. (1972). The consequences of
Archives of General Psychiatry 42, 887-896. psychotherapy for schizophrenic patients. Psychotherapy: Theory,
Falloon, I. R. H., McGill, C. W., Boyd, J. & Pedersen, J. (1987). Research and Practice 9, 111-119.
262 Editorial: Psychodynamic treatment of schizophrenia

Karon, B. P. & VandenBos, G. R. (1975). Issues in current research Pao, P.-N. (1979). Schizophrenic Disorders: Theory and Treatment
on psychotherapy vs. medication in treatment of schizophrenics. from a Psychodynamic Point of View. International Universities
Psychotherapy: Theory, Research and Practice 12, 143-148. Press: New York.
Karon, B. P. & VandenBos, G. R. (1982). Psychotherapy of Schizo- Rosen, J. (1947). The treatment of schizophrenic psychosis by direct
phrenia: The Treatment of Choice. Jason Aronson: New York. analytic therapy. Psychiatric Quarterly 21, 3-37.
Katz, H. M., Frank, A., Gunderson, J. G. & Hamm, D. (1984). Roy, A. (1986). Suicide in schizophrenia. In Suicide (ed. A. Roy), pp.
Psychotherapy of schizophrenia: what happens to treatment 97-112. Williams and Wilkins: Baltimore.
dropouts. Journal of Nervous and Mental Disease 172, 326-331. Roy, A., Schreiber, J., Mazonson, A. & Pickar, D. (1986). Suicidal
Kelly, J. A. (1982). Social Skills Training: A Practical Guide for behavior in chronic schizophrenic patients: A follow-up study.
Interventions. Springer: New York. Canadian Journal of Psychiatry 31, 737-740.
Keraberg, O. F. (1973). Summary and conclusions of'Psychotherapy Searles, H. F. (1965). Collected Papers on Schizophrenia and Related
and Psychoanalysis. Final report of the Menninger Foundation's Subjects. International Universities Press: New York.
Psychotherapy Research Project'. International Journal of Silverman, L. H. & Lachmann, F. M. (1985). The therapeutic
Psychiatry 11, 62-77. properties of unconscious oneness fantasies: Evidence and treat-
Kernberg, O. F. (1986). Identification and its vicissitudes as observed ment implications. Contemporary Psychoanalysis 21, 91-115.
in psychosis. International Journal of Psycho- Analysis 67, 147-159. Spotnitz, H. (1985). Modern Psychoanalysis of a Schizophrenic
Lassers, E. (1986). A psychotic recovers: from symbiosis to Patient: Theory of the Technique (second edition). Human Sciences
adulthood. Journal of Psychoanalysis 46, 350-359. Press: New York.
Leff, J., Kuipers, L., Berkowitz, R. & Sturgeon, D. (1982). A Stanton, A. H., Gunderson, J. G., Knapp, P. H., Frank, A. F.,
controlled study of social intervention in families of schizophrenic Vannicelli, M. L., Schnitzer, R. & Rosenthal, R. (1984). Effects of
patients. British Journal of Psychiatry 141, 121-134. psychotherapy in schizophrenia: I. Design and implementation of
Leff, J., Kuipers, L., Berkowitz, R. & Sturgeon, D. (1985). A a controlled study. Schizophrenia Bulletin 10, 520-563.
controlled trial of social intervention in the families of schizo- Stone, M. H. (1986). Exploratory psychotherapy in schizophrenia-
phrenic patients: two year follow-up. British Journal of Psychiatry spectrum patients. Bulletin of the Menninger Clinic 50, 287-306.
146, 594-600. Stone, M. H., Albert, H. D., Forrest, D. V. & Arieti, S. (1983).
Liberman, R. P., Mueser, K. T. & Wallace, C. J. (1986). Social skills Treating Schizophrenic Patients: A Clinico-Analytic Approach.
training for schizophrenic individuals at risk for relapse. American McGraw-Hill: New York.
Journal of Psychiatry 143, 523-526. Strauss, J. S. & Carpenter, W. T. jr. (1972). The prediction of
Liberman, R. P., DeRisi, W. J. & Mueser, K. T. (1989). Social Skills outcome in schizophrenia. I. Characteristics of outcome. Archives
Training for Psychiatric Patients. Pergamon Press: New York. of General Psychiatry 27, 739-746.
McGlashan, T. H. (1984a). The Chestnut follow-up study I. Follow- Strean, H. S. (1988). Behind the Couch. Wiley Press: New York.
up methodology and study sample. Archives of General Psychiatry Strupp, H. H., Hadley, S. W. & Gomes-Schwartz, B. (1977).
41, 475-585. Psychotherapy for Better or Worse. Jason Aronson: New York.
McGlashan, T. H. (19846). The Chestnut Lodge follow-up study II. Tarrier, N., Barrowclough, C , Vaughn, C , Bamrah, J. S., Porceddu,
Long-term outcome of schizophrenia and the affective disorders. K., Watts, S. & Freeman, H. (1988). The community management
Archives of General Psychiatry 41, 586-601. of schizophrenia: a controlled trial of a behavioural intervention
Mann, D. W. (1986). Six months in the treatment of two young with families to reduce relapse. British Journal of Psychiatry 153,
chronic schizophrenics. Psychiatry 49, 231-240. 532-542.
May, P. R. A. (1968). Treatment of Schizophrenia: A Comparative Tarrier, N., Barrowclough, C , Vaughn, C , Bamrah, J. S., Porceddu,
Study of Five Treatment Methods. Science House: New York. K., Watts, S. & Freeman, H. (1989). Community management of
May, P. R. A. (1984). A step forward in research on psychotherapy schizophrenia: a two-year follow-up of a behavioural intervention
of schizophrenia. Schizophrenia Bulletin 10, 604-607. with families. British Journal of Psychiatry 154, 625-628.
May, P. R. A. & Tuma, A. H. (1964). Choice of criteria for the Tuma, A. H. & May, P. R. A. (1974). Psychotherapy, drugs and
assessment of treatment outcome. Journal of Psychiatric Research therapist experience in the treatment of schizophrenia: a critique
2, 199-209. of the Michigan State project. Psychotherapy: Theory, Research
Munich, R. L. (1987). Conceptual trends and issues in the psycho- and Practice 12, 138-142.
therapy of schizophrenia. American Journal of Psychotherapy 41, Will, O. A. (1958). Psychotherapeutics and the schizophrenic
23-37. reaction. Journal of Nervous and Mental Disease 126, 109-140.
National Library of Medicine (1987). Medical Subject Headings - Winokur, G. & Tsuang, M. (1975). The Iowa 500: suicide in mania,
Annotated Alphabetic List, 1988. National Technical Information depression and schizophrenia. American Journal of Psychiatry 132,
Service: Bethesda, Maryland. 650-651.
Normand, W. C. & Bluestone, H. (1986). The use of pharmaco-
therapy in psychoanalytic treatment. Contemporary Psychiatry 22,
218-234.

View publication stats