Content uploaded by Bruce Wampold
Author content
All content in this area was uploaded by Bruce Wampold on Jun 11, 2016
Content may be subject to copyright.
Placebo Effects in Psychotherapy: A Flawed Concept
and a Contorted History
Bruce E. Wampold
University of Wisconsin–Madison and Modum Bad
Psychiatric Center, Vikersund, Norway
Nickolas D. Frost and Noah E. Yulish
University of Wisconsin–Madison
In this article, we examine the history and controversy of the placebo in psychotherapy.
In medicine, beginning in the 1950s, the use of the placebo in double-blind randomized
trials has allowed the identification of specific treatments. It seemed logical to research-
ers that this strategy could be fruitfully adapted for studying psychotherapy. However,
Rosenthal and Frank (1956) noted that the analogy would likely fail because there is no
such thing as an “inert psychotherapy.” Notwithstanding the problems with psycho-
therapy placebos, psychotherapy research moved forward with efforts to make infer-
ences about specificity using placebo types control groups. The discussion of specificity
and the therapeutic value of the common factors raise the issue of whether psycho-
therapy is in and of itself a placebo.
Keywords: common factors, placebo, psychotherapy, specificity
The pharmacopoeias of European medicine of
the 17th century contained such substances as
Vigo’s plaster (viper’s flesh with live worms and
frogs), fox lungs, moss from the skulls of victims
of violent death, Gascoyne’s powder (Bezoar, am-
ber, pearls, crab’s eyes and claws, and coral),
human urine, various sexual organs, excreta (from
many different sources), human placenta, saliva
from fasting individuals, and wood lice (Shapiro
& Shapiro, 1997b). However, as science and the
scientific method were evolving in Europe in the
Cartesian context, it became apparent that most of
the substances in the pharmacopoeias were not
effective. Indeed, only a very few seemed to be
effective for particular diseases (e.g., foxglove for
congestive heart conditions and cinchona bark for
malaria) (Shapiro & Shapiro, 1997b). We now
call the inert substances given to please the patient
placebos. But the concept of the placebo, and its
related concept the placebo effect, are complex
and have a rather contorted history.
In 1785 the term placebo entered the medical
lexicon and was applied to treatments that were
known to be ineffective physiochemically but sat-
isfied the patient’s desire to be treated (Shapiro &
Shapiro, 1997b). The term, according to Walach
(2003), originated from the Latin psalm verse,
“Placebo Domino in regione vivorum” (“I shall
please the Lord in the land of the living”), which
was sung in the Middle Ages as a prayer at the
deathbed. Because others were often paid to do
the singing, the term placebo became associated
with a “nearly fraudulent replacement of the real”
(Walach, p. 178). A placebo, since its origin, has
carried a tainted connotation—those who admin-
istered a substance simply to please the patient
became repugnant and those who claimed that a
“placebo” was curative would risk being labeled a
charlatan. Nevertheless, the effects of placebos are
real and now constitute a legitimate area of inquiry
(see, e.g., Benedetti, 2011,2014;Price, Finniss, &
Benedetti, 2008).
(Very) Brief History of Placebo Effects and
Placebo Control Groups in Medicine
The use of randomized designs in medical
and psychological research are so ingrained in
Bruce E. Wampold, Department of Counseling Psychol-
ogy, University of Wisconsin–Madison, and Modum Bad
Psychiatric Center, Vikersund, Norway; Nickolas D. Frost
and Noah E. Yulish, Department of Counseling Psychology,
University of Wisconsin–Madison.
Correspondence concerning this article should be ad-
dressed to Bruce E. Wampold, Department of Counseling
Psychology, 335 Education Building, 1000 Bascom Mall,
University of Wisconsin, Madison, WI 53705. E-mail:
bwampold@wisc.edu
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Psychology of Consciousness: Theory, Research, and Practice © 2016 American Psychological Association
2016, Vol. 3, No. 2, 108–120 2326-5523/16/$12.00 http://dx.doi.org/10.1037/cns0000045
108
our minds that many of us forget that they are
relatively recent inventions. Early examples of
rudimentary designs using blinding, and pla-
cebo type groups date back at least to efforts to
discredit Mesmer around 1784 (see Kaptchuk,
1998 for a thorough review of the history of
these designs). In the early twentieth century
significant methodological and statistical ad-
vances were made to the randomized design to
increase the validity of the inferences of these
designs and these advances were utilized in
three areas. The first application involved ex-
amining procedures in education, where in the
early 1920s the treatment group design was
“being sold to American school superintendents
as the ‘control experiment’ and touted as a key
element in comparing the ‘efficiency’ of various
administrative measures” by the American Psy-
chological Association (Danziger, 1990,p.
114). The scientific bases of these designs were
developed and presented in McCall’s (1923)
volume How to Experiment in Education. Dur-
ing the same period, Sir Ronald Fisher, who was
developing methods to study agricultural prac-
tices, developed further the notion of random-
ization and derived various statistical methods
for making inferences from the data produced
by such designs, the best known of which is the
analysis of variance (Danziger, 1990;Gehan &
Lemak, 1994;Shapiro & Shapiro, 1997b). Fish-
er’s volumes Statistical Methods for Research
Workers (Fisher, 1925) and The Experimental
Design (Fisher, 1935) formed, with little mod-
ification, the bases for all randomized designs in
medicine and the social sciences.
Medicine needed to take randomized design
one step further. The goal of modern medicine
was to establish that the benefits of any medical
treatment were attributable to the physiochemi-
cal properties of the interventions and not to the
patient’s expectations, hopes, or other psycho-
logical processes, thereby establishing the speci-
ficity of the purported active ingredients of the
procedures. To rule out threats to validity attrib-
utable to these psychological factors, researchers
in the late 1930s began to use double-blind pla-
cebo studies in the United States and the United
Kingdom, but the method did not take root, ap-
parently because placebo carried a negative con-
notation (Gehan & Lemak, 1994;Shapiro & Sha-
piro, 1997a,1997b). Gradually however, the
acceptance of the randomized double-blind pla-
cebo design spread. Harry Gold, a pharmacologist
and one of those given credit for the development
of the placebo control group design in medicine
the United States, participated in several confer-
ences at Cornell University on the subject in the
late 1940s and early 1950s and he became the first
professor of Clinical Pharmacology, a new disci-
pline (Kaptchuk, 1998;Shapiro & Shapiro,
1997b). As noted by Shapiro and Shapiro (1997b):
Gold advocated a comparison between “an allegedly
potent agent and a blank of such physical properties as
to render a distinction between the two impossible
except through some pharmacologic potency which
may exist...[the recommended] double-blind proce-
dure which calls for an investigation in which neither
the patient nor the doctor is aware of the identify of the
two agents until the results are in and analyzed. This is
imperative to avoid the influence of subconscious bias”
(Gold, 1954, p. 724). The statement by Gold culmi-
nated 20 years of pioneering study of methods with
which to reliably and validly evaluate the effectiveness
of new drugs (p. 148).
The work of Gold as well as others became
the basis for the procedures used by the United
States Food and Drug Administration (FDA) to
approve drugs for marketing the drug for a
specific disorder (Shapiro & Shapiro, 1997b).
The logic of the design is understood by exam-
ining Figure 1, where patients hypothetically
were randomly assigned to three conditions: no
treatment, a pill placebo, and active medication.
Let’s assume the study is conducted well in that
blinding is accomplished to the extent possible,
the measures are psychometrically sound, and
the analysis is correctly conducted. The course
of the disorder for those assigned to no treat-
ment (ideally) represents the natural history of
the disorder as well as other possible effects
(e.g., regression toward the mean). Indeed, pa-
tients in no treatment groups seem to improve
over time (e.g., depression; Minami, Wampold,
Serlin, Kircher, & Brown, 2007). Taking a pill
placebo has an effect for many disorders—the
separation between the outcome for patients
taking the placebo versus the outcome for pa-
tients taking no medication or receiving no in-
tervention is what is commonly called the pla-
cebo effect, the size of which is an issue which
will be discussed shortly. The separation be-
tween the active medication and the placebo can
be labeled as the specific treatment effect—this
is the effect that medicine cherishes. Indeed, as
part of the approval process, the FDA requires
109PLACEBO EFFECTS IN PSYCHOTHERAPY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
at least two trials that show the drug is superior
to a pill placebo (Spielmans & Kirsch, 2014).
1
In medical clinical trials, inclusion of a no
treatment control group is uncommon, as the
placebo’s therapeutic effect is of no interest. It
is the effects of the drug that are attributable to
the active ingredients that are of interest in
medicine—the goal is to rule out effects attrib-
utable to hope, expectation, the ritual of taking
a medication, conditioning, and so forth. A pre-
ponderance of the substances in the pharmaco-
poeias of European medicine of the 17th cen-
tury mentioned earlier would fail under the
scrutiny of the FDA requirements, and with
good reason.
Despite the lack of interest in, or even the
disdain of placebos, there have been attempts in
medicine to estimate the size of the placebo
effect. In an oft cited study, Beecher (1955)
reviewed many studies and concluded that the
placebo “had a high degree of therapeutic ef-
fectiveness in treating subjective responses” in
about a third of patients (p. 1606), justifying the
title of the article “The Powerful Placebo.”
However, in a meta-analysis of studies that con-
tained a placebo condition and a no-treatment
condition, with the title “Is the placebo power-
less?” which was used to foreshadow the re-
sults, Hróbjartsson and Gøtzsche (2001)
claimed, “We found little evidence in general
that placebos had powerful clinical effects” (p.
1594; see also Hróbjartsson & Gøtzsche, 2004).
However, these results were challenged in a
reanalysis of the same trials based on concep-
tual and statistical grounds, with the conclusion,
again in the title, “The Placebo is Powerful”
(Wampold, Minami, Tierney, Baskin, & Bhati,
2005), creating a bit of a contentious inter-
change between the various authors (see Hrób-
jartsson & Gøtzsche, 2006;Hróbjartsson &
Gøtzsche, 2007;Wampold, Imel, & Minami,
2007a,2007b). As well, Vase, Riley, and Price
(2002) noted that clinical studies are not de-
signed to detect placebo effects and con-
firmed this conjecture in a meta-analysis that
showed that placebo experiments (i.e., studies
designed to investigate the placebo effect)
produced larger effects than clinical studies.
Moreover, the experimental studies have pro-
duced evidence of neurological and physio-
logical effects as well as subjective reports of
placebos (Benedetti, 2014;Price et al., 2008).
Finally, Kirsch and colleagues have provided
convincing evidence that to a great extent the
effects of antidepressants are placebo effects
(Kirsch, 2000,2002,2009;Kirsch, 2010;
Kirsch, Moore, Scoboria, & Nicholls, 2002;
Kirsch & Sapirstein, 1998).
Having a general notion of placebos and pla-
cebo effects as well as the designs using placebo
controls in medicine, we turn our attention to
these topics in the context of psychotherapy.
Placebos and Psychotherapy—The
Beginning
In the 1950s when Gold was discussing pla-
cebos in medicine, the idea of the placebo, as a
means to understand how psychotherapy works
and as a control condition to establish specific-
ity, came to the attention of psychotherapy re-
searchers. In 1956, two prominent psychother-
1
Actually, superiority to pill placebo is not required, but
typically most such comparisons form the basis of the
evidence submitted to the FDA by the manufacturer.
Figure 1. Conceptualization of natural history, placebo effects, and specific treatment
effects.
110 WAMPOLD, FROST, AND YULISH
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
apy researchers, David Rosenthal, of the
National Institute of Mental Health, and Jerome
Frank, of the School of Medicine at Johns Hop-
kins, authored a seminal article entitled “Psy-
chotherapy and the Placebo Effect.” This article
foreshadowed the issues that were to bedevil
psychotherapy research for the coming de-
cades—all too often we repeat mistakes because
we neglect literature that we might classify as
“ancient.”
Rosenthal and Frank (1956) understood the
nature of placebos and the emerging placebo
control group in medicine: “The relief of any
particular complaint by a given medication is
not sufficient evidence for the specific effect of
the medicine on this complaint unless it can be
shown that the relief is not obtained as a placebo
effect” (p. 296). Analogously to medicine, “psy-
chotherapists have theories of personality and
psychotherapy and therapeutic actions in the
belief that these are the active agents which
produced the desired results” (p. 296). Conse-
quently, Rosenthal and Frank indicated that pla-
cebo controls could be used to establish speci-
ficity in psychotherapy:
To show that a specific form of [psychotherapy] pro-
duces more than a nonspecific placebo effect it must be
shown that its effects are stronger, last longer, or are
qualitatively different from those produced by the ad-
ministration of placebos, or that it affects different
types of patients (p. 297). If we do not control for
nonspecific factors like the placebo effect, we cannot
know whether effects predicted from a theory lead to
or result from improvement based on the nonspecific
effect (p. 299).
They went in a more prescriptive fashion to
explain what was needed:
1. A theory of personality and psychological
distress (neurosis, maladjustment, etc.).
2. Predictions of effects in the patient or cli-
ent consequent to psychotherapy, in ac-
cord with the theory.
3. Demonstration of a relationship between
the predicted effects and some criterion of
improvement.
4. Demonstration that the predicted effects
and their relationship to the improvement
criterion are not primarily attributable to
the patient’s conviction that therapy will
help him. This will permit greater confi-
dence that the relationship found is spe-
cific to the therapeutic technique derived
from the theory (p. 299).
On the face of it, it would seem that
Rosenthal and Frank (1956) were suggesting
that psychotherapy research follow the medical
model in that the placebo control group could be
used to establish the specificity of various psy-
chotherapeutic approaches. But this was not the
case. They realized that the effectiveness of
psychotherapy may indeed be related to the
ways in which the placebos work: “It may well
be that the efficacy of any particular set of
therapeutic operations lies in their analogy to a
placebo in that they enhance the therapist’s and
patient’s conviction that something useful is
being done” (p. 296). They went on to say
emphatically that these conditions could not be
met, especially the fourth condition:
The fourth condition has not been met in any research
of which we are aware. It is not possible to set up an
experiment precisely analogous to comparison of a
medication with a placebo because there is no such
thing as inert psychotherapy in the sense that placebos
are pharmacologically inert.
Since 1956 when Rosenthal and Frank laid
out the dilemma about using placebo control
groups in psychotherapy, there have been em-
pirical and logical analyses of the problems with
using psychotherapy placebos (Bowers &
Clum, 1988;Colloca & Benedetti, 2009;Critelli
& Neumann, 1984;Grünbaum, 1981;Horvath,
1988;Kirsch, 1997;Shapiro & Morris, 1978;
Wampold & Imel, 2015;Wampold et al., 2005;
Wilkins, 1983,1984). The problems with pla-
cebo controls are many, including the lack of
blinding, the fact that the active treatment and
the psychological placebo are distinguishable,
the logical impossibility of creating a placebo
with inert ingredients, the fact that the active
ingredients depend on relationship factors for
their delivery, among others (see Kirsch,
Wampold, & Kelley, 2016). Despite these prob-
lems, researchers in psychotherapy continue to
put their faith in establishing specificity with
placebo type control groups, a topic to which
we know turn.
Placebo Type Controls and a History of
Improper Inference
Shortly after Rosenthal and Frank’s (1956)
article on the use of placebos in psychotherapy
111PLACEBO EFFECTS IN PSYCHOTHERAPY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
research, psychotherapy researchers began to
design control groups to establish the efficacy
and specificity of treatments. These placebo
treatments, labeled in various ways, were used
first, it seems, by behavior therapists, primarily
in the area of systematic desensitization. A brief
review of a few studies from this era is infor-
mative historically.
Systematic desensitization (SD) was devel-
oped by Joseph Wolpe (Wolpe, 1958) as a treat-
ment for phobias and is based theoretically on
the mechanism of reciprocal inhibition. In this
procedure the patient is imaginally exposed to
the feared stimulus when in a state that is in-
compatible to fear, such as when relaxed. The
initial step of the procedure involves the cre-
ation of a hierarchy of feared situations, from
least to most threatening. Next the patient is
assisted to acquire relaxation skills (i.e., the
incompatible response). Finally the patient pro-
gresses through the developed hierarchy by
imagining their feared situation while relaxed
and only ascending the hierarchy when the cur-
rent situation is tolerated without anxiety. There
is some debate about whether an anxiety com-
peting response to the stimulus is needed or
whether the absence of an aversive consequence
to the stimulus is sufficient (i.e., an extinction
explanation). A number of studies in the early
1960s demonstrated that SD was an effective
and efficient treatment for phobias. However,
there were disagreements about what the ev-
idence demonstrated in terms of how SD
worked, and these disagreements rested
largely on the nature of placebo-type control
groups.
Gordon Paul investigated the mechanisms of
SD using performance anxiety created by public
speaking. In a series of studies, Paul and col-
leagues (see Paul & Shannon, 1966) compared
SD with an alternative treatment (insight ori-
ented psychotherapy) and an “attention-placebo
treatment,” the latter of which was included to
“determine the degree of improvement which
might be attributed to nonspecific treatment ef-
fects such as: expectation of relief; attention,
warmth, and interest of the therapist (therapeu-
tic relationship; suggestion, and ‘faith’” (p.
127). Participants in the placebo condition were
told the treatment was a method of training for
working under stress and were given a pill pla-
cebo described as having a tranquilizing effect
at the beginning of every session. Participants
were then given what was described as a stress-
ful task (e.g., identifying particular sounds in an
audio tape) and the therapists, who did not
interact with the participants, purportedly re-
corded the patient’s responses. The SD par-
ticipants improved to a greater extent than the
participants in the insight oriented and the
attention-placebo control group, supporting
not only the efficacy of SD, but the mecha-
nisms of change, as purportedly the attention-
placebo treatment controlled for nonspecific
factors.
About the same time as Paul’s investigation,
Lang, Lazovik, and Reynolds (1965) compared
SD with relaxation and hypnotic states as the
counterconditions to anxiety, to “pseu-
dotherapy,” and no treatment controls. Psue-
dotherapy was “an effort...toinvolve the
subject in a treatment procedure that was ther-
apeutically neutral except for the therapist-
client relationship” (p. 396). Effort was made
for the pseudotherapy to be similar in several
respects to the SD, including providing a cogent
explanation for the procedure, creation of the
hierarchy, and involvement of relaxation and
hypnotic states. However, the therapist in this
condition prevented the patient from phobic re-
actions in the session by steering the conversa-
tion away from material related to the phobia.
The sessions consisted mainly of using the
items in the hierarchy to begin a conversation
about the patient’s life that did not elicit anxi-
ety. The results showed that SD was superior
to no-treatment and pseudotherapy; moreover
pseudotherapy was not more effective than no
treatment. The authors concluded, “These re-
sults imply that the reduction in fear follow-
ing desensitization does not stem from a sug-
gestion to change implicit in being ‘in
therapy’” (p. 400).
The results of the two experiments on SD
described here, as well as other studies at the
time, seemed to support not only the efficacy of
SD but the conditioning mechanisms hypothe-
sized to be responsible for the benefits. How-
ever, the results of these studies generally ig-
nored the credibility of placebos as well as the
expectations created by the treatments, two
processes that were thought to be critical to
success of psychotherapy by some theorists
(e.g., Frank, 1961). These processes were of
interest to Kirsch, who set out to examine
credibility and expectations as alternative ex-
112 WAMPOLD, FROST, AND YULISH
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
planation for the efficacy of SD (Kirsch &
Henry, 1977;Kirsch, Tennen, Wickless, Sac-
cone, & Cody, 1983).
In one study Kirsch and Henry (1977) explic-
itly tested whether the benefits of SD were
attributable to extinction, a hypothesized mech-
anism of SD, or attributable to the credibility of
the treatment. The effort to test the alternative
hypothesis (viz., that credibility of the treatment
was responsible for the benefits of SD) provided
guidance for the design of the control groups for
SD. In this study of treatments for speech anx-
iety, two control groups were designed to be as
credible as SD and credibility was assessed as a
mediating variable. The first control group, la-
beled “operant desensitization,” was designed
to rule out extinction as an explanation by fol-
lowing presentation of the stimulus with elec-
trical shocks, which theoretically should in-
crease anxiety. However, the subjects were told
that the shocks served to “punish” the anxiety.
The second control group, labeled “systematic
ventilation,” involved having the participant de-
scribe neutral situation in the participants’ his-
tory (e.g., first pet) by free associating, and was
described to the participants as a “combination
of ‘directed attention’ and ‘free association’ to a
series of childhood incidents that ‘recent re-
search has established as being related to stage
freight” (p. 1054). Credibility ratings did not
differ among the three groups, ruling out a
threat due to differential credibility. Moreover,
no differences on the outcomes variables were
found, indicating that the two control groups
were as effective as SD, despite one group (viz.,
operant desensitization) that should have failed
if extinction was explanatory. Finally, and im-
portantly, credibility scores for the participants
predicted reduction of fear in all three groups.
Clearly, to be a viable comparison, placebo-
type controls need to be credible (see Borkovec
& Nau, 1972).
In another study, Kirsch and colleagues
(Kirsch et al., 1983) examined the role of ex-
pectations in producing the benefits of SD. In
this study of snake phobias, the authors differ-
entiated expectancies for improvement, self-
efficacy a` la Bandura (Bandura, 1977), and re-
sponse expectancies (i.e., expectations that one
will experience fear). The goal was to design a
treatment which would produce comparable re-
sponse expectancies but contain no procedures
that would produce benefits by any known psy-
chological theory. SD was compared to with
condition designed to induce positive response
expectancies using a variation of the “system-
atic ventilation” used in the previous study, and
to a waiting list control group. Participants in
the expectancy modification group benefited as
much as SD participants, and both groups were
superior to the waiting list control group. More-
over, within each treatment, expectancies were
correlated with improvement, and to a greater
extent than credibility ratings. It appeared, from
these results, that response expectancies ex-
plained the results better than either condition-
ing or self-efficacy.
Over the years, SD has fallen away as treat-
ment for anxiety, but not because it was not
effective (i.e., it was never shown not to be
superior to no treatment). Although difficult to
know, perhaps the demonstration that the con-
ditioning mechanism was faulty led to this de-
cline. Nevertheless, this brief review of SD lit-
erature shows how important the design of the
placebo-type controls in psychotherapy studies
are for making proper conclusions. Indeed,
when poorly designed, SD was superior to pla-
cebo type comparisons, but when well de-
signed, with an alternative explanation in mind,
SD was not superior to the credible and expec-
tation producing comparisons. Kirsch’s efforts
to design placebo groups that test alternative
hypotheses are important historically, but ap-
parently did not serve to be a model for the
design of placebo type controls in psychother-
apy research.
Beginning in the 1980s, a concerted effort
was made to establish various psychotherapies
as scientific by the use of randomized, clinical
trials (RCTs) (Goldfried & Wolfe, 1998), as
exemplified by the National Institute of Mental
Health Treatment of Depression Collaborative
Research Program (Elkin, Parloff, Hadley, &
Autry, 1985;Elkin et al., 1989). Notwithstand-
ing the cautions raised by Rosenthal and Frank
(1956), and the example set by Kirsch and col-
leagues, a number of researchers attempted to
design psychotherapy treatments that were de-
signed to have no purported active ingredients
but failed to adequately control for alternative
hypotheses. These conditions were given vari-
ous names, including psychological placebos,
attention controls, alternate treatments, or sup-
113PLACEBO EFFECTS IN PSYCHOTHERAPY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
portive counseling.
2
A few examples from the
PTSD literature will illustrate the difficulties
inherent in designing these psychological pla-
cebos.
Foa, Rothbaum, Riggs, and Murdock (1991)
investigated the efficacy of two cognitive–
behavioral treatments (CBT), prolonged expo-
sure and stress inoculation training for the treat-
ment of PTSD in rape victims, by comparing
them to a treatment called supportive counsel-
ing and a no treatment control group. Support-
ive counseling “was included to control for non-
specific therapy effects” (p. 716). The challenge
was to design supportive counseling so that it
contained no cognitive– behavioral procedures,
including exposure and cognitive restructuring,
as well as any other purportedly active ingredi-
ent of any other known treatment for PTSD.
Accordingly, supportive counseling contained
the following procedures:
Patients were taught a general problem-solving tech-
nique. Therapists played an indirect and uncondition-
ally supportive role. Homework consisted of the pa-
tient’s keeping a diary of daily problems and her
attempts at problem solving. Patients were immedi-
ately redirected to focus on current daily problems if
discussions of the assault occurred. No instructions for
exposure or anxiety management were included.
Patients were not allowed to talk about being
raped, a prohibition some would think unethi-
cal, because that would involve imaginal expo-
sure (imagining the rape while in a relaxed
state). Although patients were taught a possible
coping strategy, it was not individualized to the
patient, no feedback was given to the patient,
homework was not integrated into the treat-
ment—that is, as intended the patients were not
given any strategies or asked to conduct any
actions that might be useful. Consequently, the
treatment lacked many of the components that
are common to all treatments, such as viable
and cogent explanation for the patient’s disor-
der, agreement between patient and therapist
about the goals and tasks of treatment (i.e., the
components of the working alliance), a plan for
overcoming one’s difficulties, and so forth
(Wampold & Imel, 2015)—most likely support-
ive counseling would not be credible to a patient
seeking to work on her problems, expectations
that the treatment would be efficacious would
be low, and the attribution that the patients
efforts in therapy were responsible for improve-
ment would be nonexistent. Moreover, these
problems were exacerbated by the fact that the
therapists were aware of the treatments being
delivered and were knowledgeable about
whether they were giving the active treatments
or the placebos. Clearly, supportive counseling
as a placebo control for the two active treat-
ments was not analogous to comparing a drug
with a pill placebo. Nevertheless, this study
found that the two active treatments were supe-
rior to supportive counseling for PTSD symp-
toms, although not for “nonspecific” distress,
and it was concluded that the treatments acted
specifically.
But the use of placebo controls in psycho-
therapy research becomes even more complex,
as illustrated again by the PTSD literature. A
treatment called present centered therapy (PCT)
has been used in PTSD trials to “provide a
credible therapeutic alternative to control for
nonspecific therapeutic factors so that observed
effects of prolonged exposure could be attrib-
uted to its specific effects beyond the benefits of
good therapy” (Schnurr et al., 2007, p. 823).
PCT, similar to supportive counseling used by
Foa et al. (1991), intentionally omits prolonged
exposure and cognitive restructuring, but PCT
was more deliberate by intentionally addressing
maladaptive relationship patterns, providing
psychoeducation about PTSD, and teaching the
use of problem solving therapy. PCT was
guided by a manual and involved therapists who
were trained to deliver the treatment faithfully
(Frost, Laska, & Wampold, 2014). Even though
PCT did not contain what are thought to be the
necessary ingredients of beneficial PTSD treat-
ments (viz., exposure and cognitive restructur-
ing), it has been found to be as effective as CBT
containing those ingredients (Frost et al., 2014).
That is to say, when the treatment had structure,
contained a proper rationale that could be ex-
plained cogently to patients, had ingredients
that the patients could believe were remedial,
and were given by therapists who believed that
2
Supportive counseling or supportive therapy are ambig-
uous terms, as sometimes they are used to apply to placebo
type controls, and as such have no coherent treatment ra-
tionale and no therapeutic actions, other than empathic
responding. However, the term is also used to refer to a
psychodynamic treatment. Attention is needed to know
whether supportive counseling is purportedly inert (i.e., a
control) or a purportedly active treatment (Budge, Baards-
eth, Wampold, & Flu˝ ckiger, 2010).
114 WAMPOLD, FROST, AND YULISH
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
the treatment was remedial, the treatment was as
effective as the canonical evidence-based treat-
ments, a finding that has been established meta-
analytically (Baskin, Tierney, Minami, &
Wampold, 2003). Indeed there are other accepted
evidence-based treatments that originated as treat-
ments designed to be control conditions without
hypothesized specific ingredients (e.g., behavioral
activation and interpersonal therapy for depres-
sion).
One of the problems for psychotherapy re-
searchers is that any psychotherapy contains
many ingredients. For example, examination of
PTSD treatments yields a multitude of possibly
beneficial ingredients (Wampold et al., 2010),
including the following:
• Cogent psychological rationale that is ac-
ceptable to patient
• Systematic set of treatment actions consis-
tent with the rationale
• Development and monitoring of a safe, re-
spectful, and trusting therapeutic relation-
ship
• Collaborative agreement about tasks and
goals of therapy
• Nurturing hope and creating a sense of self
efficacy
• Psychoeducation about PTSD
• Opportunity to talk about trauma (i.e., tell
stories)
• Ensuring the patient’s safety, especially if
the patient has been victimized as in the
case of domestic violence, neighborhood
violence, or abuse
• Helping patients learn how to avoid revic-
timization
• Identifying patient resources, strengths,
survival skills, and intra- and interpersonal
resources and building resilience
• Teaching coping skills
• Examination of behavioral chain of events
• Exposure (covert in session and in vivo
outside of session)
• Making sense of traumatic event and pa-
tient’s reaction to event
• Patient attribution of change to his or her
own efforts
• Encouragement to generate and use social
supports
• Relapse prevention
Of course some treatments would emphasize
some of these ingredients more than others, but
it is impossible to design a comparison treat-
ment without any of these ingredients, in the
manner that a placebo pill in medicine contains
no ingredients that could possibly be remedial
for the disorder being studied.
Interestingly, there is a perspicuous example
of a control condition in the panic treatment
literature that harkens back to Kirsch’s studies
of SD. Often panic patients are taught to stop or
reduce hyperventilation in panic-inducing situ-
ations. In a study of breathing and panic, Kim,
Wollburg, and Roth (2012) manipulated breath-
ing. One group was taught to raise partial pres-
sure of carbon dioxide to counteract hyperven-
tilation using a feedback device; a second group
used the same device to lower the partial pres-
sure of carbon dioxide. These two conditions
were compared with a no-treatment control
group. Both breathing training protocols re-
duced panic severity compared to the no-
treatment group (i.e., no differences between
the two breathing interventions). The authors
concluded, “Clinical improvement must have
depended on elements common to both breath-
ing therapies...[including ] changed beliefs
and expectancies, exposure to ominous bodily
sensations, and attention to regular and slow
breathing” (p. 931).
Clearly placebo type controls have obvious
problems—and for that reason they have been
called pseudoplacebos (Wampold & Imel,
2015), a term we use from now on. Despite
these problems, researchers have over the years
insisted that pseudoplacebos are legitimate con-
trols to establish specificity, in the same manner
as in medicine. Various meta-analyses have
shown that legitimate treatments, a concept we
discuss below, are superior to pseudoplacebos,
which in turn are superior to no treatments
(Lambert & Bergin, 1994;Stevens, Hynan, &
Allen, 2000). However, as illustrated above,
when the pseudoplacebo begins to resemble a
legitimate treatment, it becomes as effective as
the “active” treatment (Baskin et al., 2003;Frost
et al., 2014). Moreover, removing the active
ingredient of treatments in dismantling designs
does not attenuate the benefits of the treatments,
a conclusion reached in two meta-analyses (Ahn
& Wampold, 2001;Bell, Marcus, & Goodlad,
2013).
The problems continue, to an absurd extent.
Recently, Honyashiki et al. (2014) conducted a
meta-analysis of CBT versus what they called
psychological placebos and concluded, despite
115PLACEBO EFFECTS IN PSYCHOTHERAPY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
very small and for the most nonsignificant dif-
ferences, that “the present study strongly sug-
gested a non-null specific component of CBT”
(p. 249). One of the psychological placebos in
the meta-analysis was a “talking control,”
which was described as follows:
Dysfunctional beliefs were not challenged; however,
the therapists were asked to show interest and warmth,
encouraging participants to discuss neutral topics such
as hobbies, sports, and current affairs. No advice or
problem solving was given, and there was little focus
on emotional issues. No suggestions for behavioral
tasks were offered (p. 256).
As innocuous as that sounds, the talking con-
trol was indeed a miserable form of therapy:
consider that the therapist was instructed to
respond, for example, to the statement “My
daughter does not like me as she never comes to
visit me” with “How many children do you
have?” (p. 256), a response no reasonably
trained psychotherapist would make. The fact
that CBT is superior to the “talking control”
does not, in our view, support a specificity con-
clusion. Moreover, such controls deviate greatly
from the cleverly designed treatments that
Kirsch used to test alternative explanations for
the benefits of SD.
There has been another problem interwoven
into this fabric that has historically led to some
specious conclusions. As we have seen, pseu-
doplacebos are not appropriate logically or em-
pirically for establishing specificity in psycho-
therapy trials. However, another issue in
psychotherapy research is whether one type of
psychotherapy is superior to another. Unfortu-
nately, the pseudoplacebo has played a nefari-
ous role here as well. Claims of superiority
often are based on comparative trials, in which
two (or more) treatments are compared in a
head-to-head manner. These comparative trials
have been summarized in meta-analyses to
make conclusions about classes of therapies.
For example, Gloaguen, Cottraux, Cucherat,
and Blackburn (1998) found that CBT was su-
perior to “other therapies.” However, the cate-
gory “other therapies” contained legitimate
therapies, such as psychodynamic therapies, but
also several others which were intended to be
controls (i.e., were pseudoplacebos). This prob-
lem was identified early on by Lambert and
Bergin (1994):
There is a strong trend toward no difference between
techniques in amount of change produced, which is
counterbalanced by indications that, under some cir-
cumstances, cognitive and behavioral methods are su-
perior even though they do not generally differ in
efficacy between themselves. An examination of se-
lected exemplary studies allows us to further explore
this matter. Research carried out with the intent of
contrasting two or more bona fide treatments show
surprisingly small differences between the outcomes
for patients who undergo a treatment that is fully in-
tended to be therapeutic (p. 158; emphasis added).
To address this problem, Wampold et al.
(1997) proposed a set of criteria for differenti-
ating treatments intended to be therapeutic (la-
beled as bona-fide) and those that were pseudo-
placebos. When these criteria were applied to
the “other” therapies in Gloaguen et al. (1998),
it was found that CBT was not superior to
treatments intended to be therapeutic but were
superior to those that were not bona fide thera-
pies (Wampold, Minami, Baskin, & Tierney,
2002). This distinction is critical to understand-
ing whether some treatments for PTSD are more
effective than others (see Benish, Imel, &
Wampold, 2008;Ehlers et al., 2010;Wampold
et al., 2010 for an interchange about this issue).
Is Psychotherapy a Placebo?
From the beginnings of psychotherapy,
claims have been made that the factors common
to all psychotherapies are what produces the
benefits (Rosenzweig, 1936), a conjecture that
has had many advocates over the years (Frank
& Frank, 1991;Garfield, 1992;Wampold,
2001;Wampold & Imel, 2015). The lists of
what might constitute common factors vary, but
a short list would contain the relationship with
the psychotherapist, expectations, instillation of
hope, the provision of a reasonable and accept-
able explanation for one’s difficulties, a thera-
peutic set of actions that the patient believes
will be helpful, and the therapeutic alliance.
Indeed, there is good evidence that the common
factors, rather than the specific ingredients of
psychotherapy, are what make therapy work
(Laska, Gurman, & Wampold, 2014;Wampold
& Imel, 2015). Not coincidentally, these factors
are endemic to the effects of placebo, raising the
issue of whether psychotherapy is a placebo, a
disturbing conjecture to some, as noted by
Kirsch (2005):
116 WAMPOLD, FROST, AND YULISH
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
There is a problem with identifying psychotherapy
with the placebo effect. A placebo is something that is
sham, fake, false, inert, and empty. Psychotherapy is
none of these. In this sense, it is different from medical
placebos, and it does not deserve the pejorative con-
notations associated with the term. (p. 797)
Kirsch goes on to address this issue:
But what does it mean to say that a particular psycho-
therapy is a placebo? It cannot mean that the treatment
is ineffective, because placebos can be effective. If this
were not the case, there would be no need for placebo
controls in any kind of outcome research, whether for
pills, surgery, or psychotherapy. The general assump-
tion seems to be that psychotherapy is a placebo if its
effects are nonspecific. This is generally stated as if the
meaning of the term nonspecific were clear. But it is
not clear at all. Specifically, it is not clear how the
effects of placebos are any less specific than any other
psychologically produced effects. (p. 797)
Despite Kirsch’s compelling argument, few
have been willing to explicitly discuss psycho-
therapy as a placebo. Indeed, a perusal of the
major sources that discuss the common factors
in psychotherapy will reveal that they either
avoid mentioning the placebo concept or dis-
cuss, as Kirsch did, that the application of the
term to psychotherapy is fraught with concep-
tual difficulties. However, decades ago Jeffer-
son Fish in a volume titled Placebo Therapy
argued that psychotherapy could be guided by
what is known about placebos and understand-
ing the mechanisms of placebos could guide the
delivery of psychotherapy (Fish, 1973).
Conclusions
The adoption of the RCTs in psychotherapy
outcome research, while providing a means to
test the efficacy of various psychotherapies, has
been plagued by logical problems. These prob-
lems were identified at the origins of the RCT in
psychotherapy, but have been often ignored,
leading to flawed conclusions about specificity.
A perusal of the psychotherapy outcome lit-
erature reveals that when placebo psychothera-
pies are compared with evidence-based psycho-
therapies (treatments with active/specific
ingredients) the evidence-based psychothera-
pies tend to demonstrate superiority. However,
as we have identified, when the placebo thera-
pies are engineered to be more structurally
equivalent these differences in outcome dissi-
pate (Frost et al., 2014). That is, when the
integrity of the placebo is not a factor, a placebo
psychotherapy that is allegedly “inert” can be
equally effective at remediating symptoms and
alleviating psychological distress. This may
seem perplexing to some, but it clearly implies
that there are shared mechanisms by which
these treatments are achieving their outcome. It
has been outlined in detail elsewhere (Wampold
& Imel, 2015) that elements common to all
psychotherapies may be the mechanisms by
which the placebo and psychotherapies achieve
their therapeutic effects. In this regard, a pla-
cebo psychotherapy is not “inert.”
References
Ahn, H., & Wampold, B. E. (2001). Where oh where
are the specific ingredients? A meta-analysis of
component studies in counseling and psychother-
apy. Journal of Counseling Psychology, 48, 251–
257. http://dx.doi.org/10.1037/0022-0167.48.3
.251
Bandura, A. (1977). Self-efficacy: Toward a unifying
theory of behavioral change. Psychological Re-
view, 84, 191–215. http://dx.doi.org/10.1037/
0033-295X.84.2.191
Baskin, T. W., Tierney, S. C., Minami, T., &
Wampold, B. E. (2003). Establishing specificity in
psychotherapy: A meta-analysis of structural
equivalence of placebo controls. Journal of Con-
sulting and Clinical Psychology, 71, 973–979.
http://dx.doi.org/10.1037/0022-006X.71.6.973
Beecher, H. K. (1955). The powerful placebo. Jour-
nal of the American Medical Association, 159,
1602–1606. http://dx.doi.org/10.1001/jama.1955
.02960340022006
Bell, E. C., Marcus, D. K., & Goodlad, J. K. (2013).
Are the parts as good as the whole? A meta-
analysis of component treatment studies. Journal
of Consulting and Clinical Psychology, 81, 722–
736. http://dx.doi.org/10.1037/a0033004
Benedetti, F. (2011). The patient’s brain: The neu-
roscience behind the doctor-patient relationship.
New York, NY: Oxford University Press.
Benedetti, F. (2014). Placebo effects: Understanding
the mechanisms in health and disease (2nd ed.).
New York, NY: Oxford University Press. http://dx
.doi.org/10.1093/acprof:oso/9780198705086.001
.0001
Benish, S. G., Imel, Z. E., & Wampold, B. E. (2008).
The relative efficacy of bona fide psychotherapies
for treating post-traumatic stress disorder: A meta-
analysis of direct comparisons. Clinical Psychol-
ogy Review, 28, 746 –758. http://dx.doi.org/10.1016/
j.cpr.2007.10.005
Borkovec, T. D., & Nau, S. D. (1972). Credibility of
analogue therapy rationales. Journal of Behavior
117PLACEBO EFFECTS IN PSYCHOTHERAPY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Therapy and Experimental Psychiatry, 3, 257–
260. http://dx.doi.org/10.1016/0005-7916(72)
90045-6
Bowers, T. G., & Clum, G. A. (1988). Relative
contribution of specific and nonspecific treatment
effects: Meta-analysis of placebo-controlled be-
havior therapy research. Psychological Bulletin,
103, 315–323. http://dx.doi.org/10.1037/0033-
2909.103.3.315
Budge, S. L., Baardseth, T. P., Wampold, B. E., &
Flu˝ckiger, C. (2010). Researcher allegiance and
supportive therapy: Pernicious affects on results of
randomized clinical trials. European Journal of
Counselling Psychology, 12, 23–39. http://dx.doi
.org/10.1080/13642531003637742
Colloca, L., & Benedetti, F. (2009). Placebo analge-
sia induced by social observational learning. Pain,
144, 28 –34. http://dx.doi.org/10.1016/j.pain.2009
.01.033
Critelli, J. W., & Neumann, K. F. (1984). The pla-
cebo. Conceptual analysis of a construct in transi-
tion. American Psychologist, 39, 32–39. http://dx
.doi.org/10.1037/0003-066X.39.1.32
Danziger, K. (1990). Constructing the subject: His-
torical origins of psychological research. Cam-
bridge, UK: Cambridge University Press. http://dx
.doi.org/10.1017/CBO9780511524059
Ehlers, A., Bisson, J., Clark, D. M., Creamer, M.,
Pilling, S., Richards, D.,...Yule, W. (2010). Do
all psychological treatments really work the same
in posttraumatic stress disorder? Clinical Psychol-
ogy Review, 30, 269 –276. http://dx.doi.org/10.1016/
j.cpr.2009.12.001
Elkin, I., Parloff, M. B., Hadley, S. W., & Autry,
J. H. (1985). NIMH Treatment of Depression Col-
laborative Research Program. Background and re-
search plan. Archives of General Psychiatry, 42,
305–316. http://dx.doi.org/10.1001/archpsyc.1985
.01790260103013
Elkin, I., Shea, M. T., Watkins, J. T., Imber, S. D.,
Sotsky, S. M., Collins, J. F.,...Parloff, M. B.
(1989). National Institute of Mental Health treat-
ment of depression collaborative research pro-
gram: General effectiveness of treatments. Ar-
chives of General Psychiatry, 46, 971–982. http://
dx.doi.org/10.1001/archpsyc.1989.018101100
13002
Fish, J. M. (1973). Placebo therapy. San Francisco,
CA: Jossey-Bass.
Fisher, R. A. (1925). Statistical methods for research
workers. London, UK: Oliver & Boyd.
Fisher, R. A. (1935). The design of experiments.
Edinburgh, UK: Oliver and Boyd.
Foa, E. B., Rothbaum, B. O., Riggs, D. S., & Mur-
dock, T. B. (1991). Treatment of posttraumatic
stress disorder in rape victims: A comparison be-
tween cognitive-behavioral procedures and coun-
seling. Journal of Consulting and Clinical Psy-
chology, 59, 715–723. http://dx.doi.org/10.1037/
0022-006X.59.5.715
Frank, J. D. (1961). Persuasion and healing: A com-
parative study of psychotherapy. Baltimore, MD:
Johns Hopkins University Press.
Frank, J. D., & Frank, J. B. (1991). Persuasion and
healing: A comparative study of psychotherapy
(3rd ed.). Baltimore, MD: Johns Hopkins Univer-
sity Press.
Frost, N. D., Laska, K. M., & Wampold, B. E.
(2014). The evidence for present-centered therapy
as a treatment for posttraumatic stress disorder.
Journal of Traumatic Stress, 27, 1– 8. http://dx.doi
.org/10.1002/jts.21881
Garfield, S. L. (1992). Eclectic pyschotherapy: A
common factors approach. In J. C. Norcross &
M. R. Goldfried (Eds.), Handbook of psychother-
apy integration (pp. 169 –201). New York, NY:
Basic Books.
Gehan, E., & Lemak, N. A. (1994). Statistics in
medical research: Developments in clinical trials.
New York, NY: Plenum Press Medical Book.
http://dx.doi.org/10.1007/978-1-4615-2518-9
Gloaguen, V., Cottraux, J., Cucherat, M., & Black-
burn, I. M. (1998). A meta-analysis of the effects
of cognitive therapy in depressed patients. Journal
of Affective Disorders, 49, 59 –72. http://dx.doi
.org/10.1016/S0165-0327(97)00199-7
Goldfried, M. R., & Wolfe, B. E. (1998). Toward a
more clinically valid approach to therapy research.
Journal of Consulting and Clinical Psychology,
66, 143–150. http://dx.doi.org/10.1037/0022-006X
.66.1.143
Grünbaum, A. (1981). The placebo concept. Behav-
iour Research and Therapy, 19, 157–167. http://
dx.doi.org/10.1016/0005-7967(81)90040-1
Honyashiki, M., Furukawa, T. A., Noma, H., Tanaka,
S., Chen, P., Ichikawa, K.,...Caldwell, D. M.
(2014). Specificity of CBT for depression: A con-
tribution from multiple treatments meta-analyses.
Cognitive Therapy and Research, 38, 249 –260.
http://dx.doi.org/10.1007/s10608-014-9599-7
Horvath, P. (1988). Placebos and common factors in
two decades of psychotherapy research. Psycho-
logical Bulletin, 104, 214 –225. http://dx.doi.org/
10.1037/0033-2909.104.2.214
Hróbjartsson, A., & Gøtzsche, P. C. (2001). Is the
placebo powerless? An analysis of clinical trials
comparing placebo with no treatment. The New Eng-
land Journal of Medicine, 344, 1594 –1602. http://dx
.doi.org/10.1056/NEJM200105243442106
Hróbjartsson, A., & Gøtzsche, P. C. (2004). Is the
placebo powerless? Update of a systematic review
with 52 new randomized trials comparing placebo
with no treatment. Journal of Internal Medicine,
256, 91–100. http://dx.doi.org/10.1111/j.1365-
2796.2004.01355.x
118 WAMPOLD, FROST, AND YULISH
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Hróbjartsson, A., & Gøtzsche, P. C. (2006). Unsub-
stantiated claims of large effects of placebo on
pain: Serious errors in meta-analysis of placebo
analgesia mechanism studies. Journal of Clinical
Epidemiology, 59, 336 –338. http://dx.doi.org/
10.1016/j.jclinepi.2005.05.011
Hróbjartsson, A., & Gøtzsche, P. C. (2007). Powerful
spin in the conclusion of Wampold et al.’s re-
analysis of placebo versus no-treatment trials de-
spite similar results as in original review. Journal
of Clinical Psychology, 63, 373–377. http://dx.doi
.org/10.1002/jclp.20357
Kaptchuk, T. J. (1998). Intentional ignorance: A his-
tory of blind assessment and placebo controls in
medicine. Bulletin of the History of Medicine, 72,
389 – 433. http://dx.doi.org/10.1353/bhm.1998
.0159
Kim, S., Wollburg, E., & Roth, W. T. (2012). Op-
posing breathing therapies for panic disorder: A
randomized controlled trial of lowering vs raising
end-tidal P(CO
2
). Journal of Clinical Psychiatry,
73, 931–939. http://dx.doi.org/10.4088/JCP
.11m07068
Kirsch, I. (1997). Specifying nonspecifics: Psycho-
logical mechanisms of placebo effects. In A. Har-
rington (Ed.), The placebo effect: An interdisci-
plinary exploration (pp. 166 –186). Cambridge,
MA: Harvard University Press.
Kirsch, I. (2000). Are drug and placebo effects in depres-
sion additive? Biological Psychiatry, 47, 733–735.
http://dx.doi.org/10.1016/S0006-3223(00)00832-5
Kirsch, I. (2002). Yes, there is a placebo effect, but is
there a powerful antidepressant drug effect? Pre-
vention & Treatment, 5, 22.
Kirsch, I. (2005). Placebo psychotherapy: Synonym
or oxymoron? Journal of Clinical Psychology, 61,
791– 803. http://dx.doi.org/10.1002/jclp.20126
Kirsch, I. (2009). Antidepressants and the placebo
response. Epidemiology and Psychiatric Sci-
ences, 18, 318 –322. http://dx.doi.org/10.1017/
S1121189X00000282
Kirsch, I. (2010). The emperor’s new drugs: Explod-
ing the antidepressant myth. New York, NY: Basic
Books.
Kirsch, I., & Henry, D. (1977). Extinction versus
credibility in the desensitization of speech anxiety.
Journal of Consulting and Clinical Psychology,
45, 1052–1059. http://dx.doi.org/10.1037/0022-
006X.45.6.1052
Kirsch, I., Moore, T. J., Scoboria, A., & Nicholls,
S. S. (2002). The emperor’s new drugs: An anal-
ysis of antidepressant medication data submitted to
the U.S. Food and Drug Administration. Preven-
tion & Treatment, 5, 23.
Kirsch, I., & Sapirstein, G. (1998). Listening to Pro-
zac but hearing placebo: A meta-analysis of anti-
depressant medication. Prevention & Treatment, 1,
2A.
Kirsch, I., Tennen, H., Wickless, C., Saccone, A. J.,
& Cody, S. (1983). The role of expectancy in fear
reduction. Behavior Therapy, 14, 520 –533. http://
dx.doi.org/10.1016/S0005-7894(83)80075-6
Kirsch, I., Wampold, B., & Kelley, J. M. (2016).
Controlling for the placebo effect in psychother-
apy: Noble quest or tilting at windmills? Psychol-
ogy of Consciousness: Theory, Research, and
Practice, 3, 121–131. http://dx.doi.org/10.1037/
cns0000065
Lambert, M. J., & Bergin, A. E. (1994). The effec-
tiveness of psychotherapy. In A. E. Bergin & S. L.
Garfield (Eds.), Handbook of psychotherapy and
behavior change (4th ed., pp. 143–189). New
York, NY: Wiley.
Lang, P. J., Lazovik, A. D., & Reynolds, D. J. (1965).
Desensitization, suggestibility, and pseudotherapy.
Journal of Abnormal Psychology, 70, 395– 402.
http://dx.doi.org/10.1037/h0022763
Laska, K. M., Gurman, A. S., & Wampold, B. E.
(2014). Expanding the lens of evidence-based
practice in psychotherapy: A common factors per-
spective. Psychotherapy, 51, 467– 481. http://dx
.doi.org/10.1037/a0034332
McCall, W. A. (1923). How to experiment in educa-
tion. New York, NY: Macmillan. http://dx.doi.org/
10.1037/13551-000
Minami, T., Wampold, B. E., Serlin, R. C., Kircher,
J. C., & Brown, G. S. J. (2007). Benchmarks for
psychotherapy efficacy in adult major depression.
Journal of Consulting and Clinical Psychology,
75, 232–243. http://dx.doi.org/10.1037/0022-006X
.75.2.232
Paul, G. L., & Shannon, D. T. (1966). Treatment of
anxiety through systematic desensitization in ther-
apy groups. Journal of Abnormal Psychology, 71,
124 –135. http://dx.doi.org/10.1037/h0023172
Price, D. D., Finniss, D. G., & Benedetti, F. (2008).
A comprehensive review of the placebo effect:
Recent advances and current thought. Annual Re-
view of Psychology, 59, 565–590. http://dx.doi
.org/10.1146/annurev.psych.59.113006.095941
Rosenthal, D., & Frank, J. D. (1956). Psychotherapy
and the placebo effect. Psychological Bulletin, 53,
294 –302.
Rosenzweig, S. (1936). Some implicit common fac-
tors in diverse methods of psychotherapy. Ameri-
can Journal of Orthopsychiatry, 6, 412– 415.
http://dx.doi.org/10.1111/j.1939-0025.1936
.tb05248.x
Schnurr, P. P., Friedman, M. J., Engel, C. C., Foa,
E. B., Shea, M. T., Chow, B. K.,...Bernardy, N.
(2007). Cognitive behavioral therapy for posttrau-
matic stress disorder in women: A randomized
controlled trial. JAMA: Journal of the American
Medical Association, 297, 820 – 830. http://dx.doi
.org/10.1001/jama.297.8.820
119PLACEBO EFFECTS IN PSYCHOTHERAPY
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Shapiro, A. K., & Morris, L. A. (1978). The placebo
effect in medical and psychological therapies. In
S. L. Garfield & A. E. Bergin (Eds.), Handbook of
psychotherapy and behavior change (2nd ed., pp.
369 – 410). New York, NY: Wiley.
Shapiro, A. K., & Shapiro, E. S. (1997a). The pla-
cebo: Is it much ado about nothing? In A. Har-
rington (Ed.), The placebo effect: An interdisci-
plinary exploration. Cambridge, MA: Harvard
University Press.
Shapiro, A. K., & Shapiro, E. S. (1997b). The pow-
erful placebo: From ancient priest to modern med-
icine. Baltimore, MD: Johns Hopkins University
Press.
Spielmans, G. I., & Kirsch, I. (2014). Drug approval
and drug effectiveness. Annual Review of Clinical
Psychology, 10, 741–766. http://dx.doi.org/10.1146/
annurev-clinpsy-050212-185533
Stevens, S. E., Hynan, M. T., & Allen, M. (2000). A
meta-analysis of common factor and specific treat-
ment effects across domains of the phase model of
psychotherapy. Clinical Psychology: Science and
Practice, 7, 273–290. http://dx.doi.org/10.1093/
clipsy.7.3.273
Vase, L., Riley, J. L., III, & Price, D. D. (2002). A
comparison of placebo effects in clinical anal-
gesic trials versus studies of placebo analgesia.
Pain, 99, 443– 452. http://dx.doi.org/10.1016/
S0304-3959(02)00205-1
Walach, H. (2003). Placebo and placebo effects - a
concise review. Focus on Alternative and Comple-
mentary Therapies, 8, 178 –187.
Wampold, B. E. (2001). The great psychotherapy
debate: Model, methods, and findings. Mahwah,
NJ: Erlbaum.
Wampold, B. E., & Imel, Z. E. (2015). The great
psychotherapy debate: The research evidence for
what works in psychotherapy (2nd ed.). New York,
NY: Routledge.
Wampold, B. E., Imel, Z. E., Laska, K. M., Benish,
S., Miller, S. D., Flückiger, C.,...Budge, S.
(2010). Determining what works in the treatment
of PTSD. Clinical Psychology Review, 30, 923–
933. http://dx.doi.org/10.1016/j.cpr.2010.06.005
Wampold, B. E., Imel, Z. E., & Minami, T. (2007a).
The placebo effect: “relatively large” and “robust”
enough to survive another assault. Journal of Clin-
ical Psychology, 63, 401– 403. http://dx.doi.org/
10.1002/jclp.20350
Wampold, B. E., Imel, Z. E., & Minami, T. (2007b).
The story of placebo effects in medicine: Evidence
in context. Journal of Clinical Psychology, 63,
379 –390. http://dx.doi.org/10.1002/jclp.20354
Wampold, B. E., Minami, T., Baskin, T. W., &
Callen Tierney, S. (2002). A meta-(re)analysis of
the effects of cognitive therapy versus ‘other ther-
apies’ for depression. Journal of Affective Disor-
ders, 68, 159 –165. http://dx.doi.org/10.1016/
S0165-0327(00)00287-1
Wampold, B. E., Minami, T., Tierney, S. C., Baskin,
T. W., & Bhati, K. S. (2005). The placebo is
powerful: Estimating placebo effects in medicine
and psychotherapy from randomized clinical trials.
Journal of Clinical Psychology, 61, 835– 854.
http://dx.doi.org/10.1002/jclp.20129
Wampold, B. E., Mondin, G. W., Moody, M., Stich,
F., Benson, K., & Ahn, H. (1997). A meta-analysis
of outcome studies comparing bona fide psycho-
therapies: Empirically, “All must have prizes.”
Psychological Bulletin, 122, 203–215. http://dx
.doi.org/10.1037/0033-2909.122.3.203
Wilkins, W. (1983). Failure of placebo groups to
control for nonspecific events in therapy outcome
research. Psychotherapy, Theory, Research, &
Practice, 20, 31–37. http://dx.doi.org/10.1037/
h0088475
Wilkins, W. (1984). Psychotherapy: The powerful
placebo. Journal of Consulting and Clinical Psy-
chology, 52, 570 –573. http://dx.doi.org/10.1037/
0022-006X.52.4.570
Wolpe, J. (1958). Psychotherapy by reciprocal inhi-
bition. Palo Alto, CA: Stanford University.
Received October 20, 2014
Revision received January 6, 2015
Accepted January 22, 2015 䡲
120 WAMPOLD, FROST, AND YULISH
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
A preview of this full-text is provided by American Psychological Association.
Content available from Psychology of Consciousness: Theory, Research, and Practice
This content is subject to copyright. Terms and conditions apply.