Twenty years after ‘Listening to Prozac but hearing placebo’. Do we hear placebo even louder?

Abstract

An active placebo is a substance that produces side effects similar to an active ingredient while not producing the same intended therapeutic effect. The aim of this study is to review the literature on the hypothesis of the active placebo response as a mechanism of action of antidepressants.

It was found that persons who expect the occurrence of side effects of a pure placebo taken under the guise of an antidepressant present a higher degree of depressive symptoms than persons who do not expect the occurrence of side effects.

There are reasons to believe that the entirety or part of the difference in the effectiveness of antidepressants and placebo is due to the fact that participants of the clinical trials correctly guess which study group they have been assigned to.

Full text

health psychology report · volume 7(1), 9
review article
An active placebo is asubstance that produces side eects
similar to an active ingredient while not producing the
same intended therapeutic eect. The aim of this study is
to review the literature onthe hypothesis of the active pla-
cebo response as amechanism of action of antidepressants.
It was found that persons who expect the occurrence of
side eects of a pure placebo taken under the guise of
an antidepressant present ahigher degree of depressive
symptoms than persons who do not expect the occurrence
of side eects.
There are reasons to believe that the entirety or part of
the dierence in the eectiveness of antidepressants and
placebo is due to the fact that participants of the clinical
trials correctly guess which study group they have been
assigned to.
key words
antidepressants; expectations; placebo response
Wojciech Oronowicz-Jaśkowiak id
1
Przemysław Bąbel
2
Twenty years aer ‘Listening to Prozac but
hearing placebo’. Do we hear placebo even louder?
 – 1: III Department of Psychiatry, Institute of Psychiatry and Neurology, Warsaw, Poland · 2: Institute
of Psychology, Jagiellonian University, Cracow, Poland
’  – A: Study design · B: Data collection · C: Statistical analysis · D: Data interpretation ·
E: Manuscript preparation · F: Literature search · G: Funds collection
  – Wojciech Oronowicz-Jaśkowiak, Institute of Psychiatry and Neurology, 9 Sobieskiego Str.,
02-957 Warsaw, Poland, e-mail: wojciechoronowicz@gmail.com
    – Oronowicz-Jaśkowiak, W., &Bąbel, P. (2019). Twenty years aer ‘Listening to Prozac but hearing
placebo’. Do we hear placebo even louder? Health Psychology Report, 7(1), 1–8. hps://doi.org/10.5114/hpr.2019.83383
 11.09.2018 ·  04.11.2018 ·  06.12.2018 ·  07.03.2019

Wojciech
Oronowicz-
Jaśkowiak,
Przemysław Bąbel
2  
background
Depressive disorder is one of the most common
mental health disorders. According to the diagnostic
criteria of the American Psychiatric Association, the
diagnosis of major depressive disorder requires the
observation of several signs and symptoms, such as
low mood, excessive sleepiness or insomnia, weight
loss, feeling of worthlessness, and unjustied guilt
(APA, 2013). It is estimated that in the United States,
depression aects 7.00% of the population annually,
and persons aged between 18 and 29 years and over
60 years are particularly vulnerable (see APA, 2013).
According to the dominant hypothesis derived
from the biological theory of depression, the condi-
tion is associated with abnormal neurotransmission
in the central nervous system (Beck &Alford, 2009).
It is suggested that there may be apathologically in-
creased sensitivity of neurotransmier receptors in
the course of depressive disorder. As neurophysiolo-
gists have proven, if the level of aneurotransmier
is adequately high, the organism downregulates the
number of receptors, e.g. serotonin receptors, in or-
der to maintain a stable neural signal (Stahl, 1994).
However, if there is adeciency of aneurotransmit-
ter, it upregulates and increases the number of recep-
tors, e.g. serotonin receptors. It is believed that in de-
pression, the number of serotonin receptors is higher
due to low serotonin levels. erefore, if a person
suering from depression and having ahigh num-
ber of receptors is given asubstance that inhibits the
reuptake of the neurotransmier, anew balance of
neurotransmiers and receptors should be achieved.
Treatment methods based on the biological theory
of depression include pharmacotherapy with antide-
pressants. Antidepressants are aheterogeneous group
of medications and are believed to reduce signs and
symptoms of depressive disorder (see Rybakowski,
2011). Antidepressant groups include monoamine
oxidase inhibitors (MAOIs), tricyclic antidepres-
sants (TCAs), selective serotonin reuptake inhibitors
(SSRIs), and serotonin-norepinephrine reuptake in-
hibitors (SNRIs). ese medications are frequently
used in clinical practice. In the United States, sales
growth of pharmacological agents, particularly anti-
depressants, is observed (Gu, Dillon, &Burt, 2011).
e aim of this study is to review the hypothesis
of the active placebo response as amechanism of ac-
tion of antidepressants.
active placebo response
hypothesis
e biological theory is criticized, as the main evi-
dence supporting this theory is the claimed eec-
tiveness of antidepressants in the treatment of de-
pressive disorders, as well as ambiguous results of
experimental studies on the eect of the medications
on the receptor activity in the central nervous sys-
tem (Cowen, 2008). It has been suggested that em-
pirical evidence conrming the chemical imbalance
theory is insucient, however, it seems not to have
aected its popularity (Lacasse &Leo, 2015). In one
of the most interesting studies, mice that were unable
to produce serotonin were bred in order to investi-
gate whether they would display depressive symp-
toms (Angoa-Pérez et al., 2014). e mice displayed
compulsive and aggressive behaviors, but no signs of
depression were observed. Moreover, when the mice
were subjected to stress, there was no dierence in
behavior of normal mice and mice without serotonin
receptors. e researchers concluded that serotonin
cannot be the main factor contributing to the etiol-
ogy or development of depressive disorder.
It should be noted that an alternative mechanism
of action of antidepressants has been presented, and
it does not refer to the chemical imbalance theory –
the active placebo response hypothesis (Kirsch, 2009;
see Siwak, Oronowicz-Jaśkowiak, & Oronowicz-
Jaśkowiak, 2017). Selected publications suggest that
the eect of antidepressant treatment is indeed the
same as the active placebo response, which takes
place when the patient experiences side eects of
taken medications (Kirsch, 2014). It is underlined that
antidepressants cause anumber of easily noticeable
side eects, such as dry mouth (xerostomia), drowsi-
ness, anxiety, and low libido (e.g., Hughes, Lacasse,
Fuller, &Paulding-Givens, 2017). It is suggested that
if apatient discovers one of these side eects, they
conclude that they are in the study group receiving
the active substance, and not the placebo (Rabkin
etal., 1986), which aects their expectations of the
eectiveness of the treatment.
When dening the term active placebo, it can be
assumed that it is an agent that does not contain any
active ingredient (in this case, an antidepressant), but
produces similar side eects as the active treatment
(active ingredient of antidepressants) (Bąbel, 2006;
Dolińska, 2011).
It is also noted that in clinical trials, the awareness
of the fact that the participant may receive aplacebo
instead of areal medication is related to the percent-
age of patients that respond to treatment. e per-
centage of patients responding to treatment when
(1) the subjects are aware of the fact that there is no
possibility that they are in agroup receiving a pure
placebo instead of amedication, and (2) the subjects
are aware of the fact that there is apossibility that
they are receiving either amedication or apure place-
bo was 60.00% and 46.00%, respectively (Sneed et al.,
2008). Recent experimental studies have shown that
the therapeutic eect of selective serotonin reuptake
inhibitors (escitalopram was used in the study) was
two to three times higher (in terms of the reduction
of signs and symptoms of depressive disorder or so-

Twenty years
aer ‘Listening to
Prozac but hearing
placebo’
3
 7(1), 9
cial anxiety) when the patient was sure that they had
received an active substance than when the patient
was not sure whether they had been administered an
antidepressant or aplacebo (Faria et al., 2017). In ad-
dition, the study revealed that these two groups of
patients were characterized by adierent neural re-
sponse of brain areas responsible for the interaction
of cognitive and emotional processes (Faria et al.,
2017). Other studies have shown that patients’ expec-
tations towards therapy have a signicant inuence
on their depressive signs and symptoms (Rutherford
et al., 2016).
e above described observations aect the out-
comes of pharmacotherapy. It is reported that the
higher eectiveness of active medications when com-
pared with placebo is not caused exclusively by the
suggested chemical mechanism of action of amedi-
cation. is leads to overestimating the eectiveness
of pharmacotherapy (Lund, Vase, Petersen, Jensen,
&Finnerup, 2014; Kube &Rief, 2017). e interfering
factors include, among others, the conditioning pro-
cess and expectations towards eectiveness of treat-
ment (Kube &Rief, 2017).
direct evi dence supporting
the activ e placebo response
hypothesis as amechanism
of action of an tidepressants
e rst scientic report that showed aclear rela-
tionship of the placebo response with the ecacy
of antidepressant therapy was a meta-analysis of
19clinical trials: Listening to Prozac but hearing pla-
cebo (Kirsch &Sapirstein, 1998). is meta-analysis
revealed a0.90 correlation between the therapeutic
eect of antidepressant treatment and the placebo
response in the control group. e meta-analysis
reported that antidepressants and psychotherapy
all reduced depressive signs and symptoms, but the
therapeutic eect of uoxetine was stronger than the
placebo response.
It must be noted that conclusions of the study by
Kirsch and Sapirstein (1998) were criticized, as the
method for analyzing the results that was used by
the authors could result in artifacts related to the re-
gression to the average due to comparing two groups
with ahigh intensity of the examined feature (Dawes,
1998). Further criticism was also presented by Klein
(1998), but at least some of Klein’s objections regard-
ing Kirsch’s “irrelevant speculations” may not re-
main true today. Subsequent studies have extended
the research perspective by using active placebo and
dierent groups of patients.
A meta-analysis of 131 clinical trials that com-
pared the therapeutic eect of selective serotonin
reuptake inhibitors with pure placebo indicated that
antidepressants show higher eectiveness than pure
placebo; however, the dierence was established to
be too small to be clinically relevant ( Jakobsen et al.,
2017). e mean improvement dierence between
the groups was 1.94 on the Hamilton Depression
Rating Scale, while the dierence that is expected for
a therapeutic method to be recognized as eective
should be at least 3 points (Jakobsen et al., 2017).
A meta-analysis of 9 clinical trials in which active
placebo was used indicated that antidepressant treat-
ment is slightly more eective than the administra-
tion of active placebo (d=0.39) or that the dierence
between the groups is not signicant, depending on
the data analysis approach – liberal or conservative
(Moncrie, Wessely, & Hardy, 2004). On the other
hand, another meta-analysis suggests that the lack of
dierence between active placebo and antidepressants
may depend on aparticular substance (Hieronymus,
Lisinski, Nilsson, &Eriksson, 2017). It has been ob-
served that patients taking paroxetine and experienc-
ing side eects had the biggest clinical improvement
(when compared with patients taking pure placebo
and patients taking paroxetine and not experiencing
side eects); however, in the case of citalopram, the
outcomes were similar for taking citalopram and ex-
periencing side eects and taking citalopram and not
experiencing side eects, and the dierence between
groups was not signicant (p=.140).
Greenberg et al. (1994) observed asignicant cor-
relation between symptom reduction and side eects
of selective serotonin reuptake inhibitors. Impor-
tantly, the correlation applied both to signs observed
by clinicians (r=.85) and to symptoms reported by
patients (r=.96).
e above described studies may support the hy-
pothesis of the active placebo response as a mech-
anism of action of antidepressants. However, it
should be kept in mind that there are signicantly
fewer studies that use an active placebo than studies
that use apure placebo ( Jensen, Bielefeldt, & Hrób-
jartsson, 2017). Nevertheless, there is also indirect
evidence supporting the hypothesis of the active
placebo response as amechanism of action of antide-
pressants, as will be shown in the next section.
indirect evidence supporting
the activ e placebo response
hypothesis as amechanism
of action of an tidepressants
It is believed that many pharmacological agents help
reduce the severity of depression. ese medications
include, among others, psilocybin (Dydak, Sliwin-
ska-Mosson, & Milnerowicz, 2016), carbamazepine
(Zhang et al., 2008), and, in combination with anti-
depressants: benzodiazepines (Furukawa, Streiner,
Young, & Kinoshita, 2001), thyroid hormones (Ar-
onson, Oman, Joe, & Naylor, 1996), risperidone

Wojciech
Oronowicz-
Jaśkowiak,
Przemysław Bąbel
4  
(Ostro & Nelson, 1999), and lithium compounds
(Bauer &Döpfmer, 1999). In addition, supplementing
omega-3 fay acids is believed to reduce depression
(Osher &Belmaker, 2009). e evidence for the ef-
fectiveness of these medications, which are supposed
to act independently from the chemical imbalance
theory, may support the hypothesis that pharma-
cological treatment of depression is not specic for
antidepressants.
It should also be noted that empirically supported
methods of depression treatment are not limited to
pharmacological agents (e.g., Rybakowski, 2011), but
also include physical activity (Netz, 2017), cognitive-
behavioral psychotherapy (Hofmann et al., 2012),
mindfulness meditation (Ramel, Goldin, Carmona,
&Mcaid, 2004), short-term psychodynamic ther-
apy (Driessen et al., 2010), and eye movement desen-
sitization and reprocessing (Edmond, Rubin, &Wam-
bach, 1999; Capezzani et al., 2013). It must be noted
that these methods are supposed to have various
mechanisms of action and their characteristics dier.
However, their non-specic eects enable depressive
symptoms to be improved.
Some of the most interesting studies focus on the
eectiveness of homeopathy, which, similarly to pla-
cebo administration, does not introduce active sub-
stances to the patient’s organism (NHMRC, 2015).
e results of a study with 566 patients, some of
whom consulted ahomeopath, suggest that homeop-
athy is equally eective as antidepressants in reduc-
ing depressive symptoms (measured with the PHQ-9
questionnaire) (Viksveen, Relton, &Nicholl, 2017). In-
terestingly, the duration of the study was 12 months,
which is unusually long for this kind of research.
Herbal preparations are also used in treatment or
supportive treatment of major depressive episodes or
aective disorders that are not classied as major de-
pression, Hypericum (St. John’s wort) being anotable
example. Hypericum is believed to improve reuptake
of serotonin, noradrenaline and dopamine (Koszew-
ska, 2003). However, it must be noted that there is
no reliable evidence to support the use of St. John’s
wort for depressive disorder. A study on 340 patients
diagnosed with depression based on DSM-IV criteria
was conducted (Hypericum Depression Trial Study
Group, 2002). e patients were divided into groups
taking an antidepressant (Serotax), placebo or Hyper-
icum. Aer 8 weeks and 6 months of treatment, there
was no dierence in the outcome between groups. In
other words, antidepressant, placebo, and Hypericum
were all equally ineective.
A meta-analysis on depression treatment with Hy-
pericum indicated that it presents atherapeutic eect
greater than placebo and similar to antidepressants
(studies included uoxetine, sertraline, imipramine,
citalopram, paroxetine, maprotiline, and amitripty-
line; Ernst, 1995). Interestingly, the authors of another
meta-analysis comparing the eectiveness of Hyperi-
cum and placebo observed that Ernst’s conclusions
were essentially true, but only in the case of German
patients (Linde, Berner, &Kriston, 2008). e analy-
sis of the participants’ nationality revealed that 8 out
of 11 German clinical trials reported Hypericum to be
more eective than placebo, whereas none of the non-
German studies found Hypericum to be eective. e
combined eectiveness of Hypericum and placebo was
signicantly higher in German-speaking countries.
e results of the analysis may show the role of dier-
ent cultural expectations and its eect on the clinical
outcomes or may indicate adierent methodological
approach of German researchers. On the other hand,
the observed dierence may be due to dierences in
the quality of products distributed in Germany and in
other countries, resulting from using dierent brands
containing varying doses of the supplement (see Kle-
mow et al., 2004). In Germany, there are regulations
concerning, among others, the amount of active in-
gredient in Hypericum supplements.
Particular aention should be drawn to the use
of antidepressants in minor depressive disorder (i.e.
mild, short-term depression). A meta-analysis of
6clinical trials on the eectiveness of uoxetine, am-
itriptyline and isocarboxazid in treatment of minor
depression (Barbui, Cipriani, Patel, Ayuso-Mateos,
&van Ommeren, 2011) showed that treatment with
these substances is no more eective than placebo.
Moreover, according to the guidelines of the National
Institute for Health and Care Excellence (NICE, 2009),
antidepressants should not be used for treatment of
mild and short-term depression, as their eective-
ness compared with placebo is hard to demonstrate.
Nevertheless, the Institute underlines that antide-
pressants may be indicated in treatment of dysthy-
mia and other forms of mild, long-term depression.
It is also suggested that previous analyses prov-
ing the eectiveness of antidepressants are burdened
with ahigh risk of error due to the controversies ac-
companying clinical trials that they were based on
(Deacon &Spielmans, 2017). e re-analysis of the ef-
fects of treatment of major depressive disorders of ad-
olescents with paroxetine and imipramine may serve
as an example (Le Noury et al., 2015). e re-analysis
of the clinical data was performed by a team other
than the team responsible for the primary analysis,
which was sponsored by GSK Pharmaceuticals. It was
proven that the interpretation method of clinical data
that supported the use of paroxetine and imipramine
in treatment of major depressive disorder in adoles-
cents was faulty. In the paper by Keller et al. (2001),
the endpoint of the study was modied aer the trial
had been completed. According to the standards for
conducting clinical trials, the endpoint should be es-
tablished before the onset of the trial to determine
what would be considered asignicant improvement
in the patients’ condition. In this case, without modi-
fying primary data, the nal conclusion changed – it

Twenty years
aer ‘Listening to
Prozac but hearing
placebo’
5
 7(1), 9
was concluded that paroxetine was an eective drug,
whereas the same data with the previous endpoint
would have proven otherwise. It was later shown that
paroxetine and imipramine were equally eective as
pure placebo (Le Noury et al., 2015). Le Noury et al.
(2015) also underlined the need for a greater trans-
parency of clinical trial databases. GSK Pharmaceu-
ticals received amajor ne for marketing paroxetine
as a safe drug for the treatment of major depressive
disorder in adolescents (Reuters, 2015).
An analysis of the results of clinical trials on an-
tidepressants from the years 1987-2013 was conduct-
ed in order to determine whether the dierences in
clinical outcome between placebo and antidepres-
sants varies over time (Khan, Fahl Mar, Fauce, Khan
Schilling, &Brown, 2017). is study was aresponse
to a publication cycle suggesting that over time, the
therapeutic eect of antidepressant increases, but this
eect was to depend exclusively on the increasing
share of the placebo eect in its ecacy. It was prov-
en that over time, the positive outcomes of both anti-
depressants and placebo increase to the same extent.
In this analysis, the aspect of expecting certain side
eects of taking placebo or medication was not dis-
cussed, as the analyzed studies did not report on side
eects perceived by the patients. It is possible that if
the aspect was controlled, the dierence between the
placebo eect and the active substance eect would
increase over timeOn the other hand, Huneke et al.
(2017) published an interesting experimental study, in
which the relationship between emotional process-
ing and taking aplacebo was examined. Emotional
processing is believed to be an adequate new index
for testing antidepressant eectiveness. e partici-
pants were randomized to one of three study groups
(placebo, antidepressant, no treatment) and were as-
sessed using psychological tools evaluating emotional
processing. ere was no dierence between the pure
placebo group and the no treatment group in emo-
tional processing. In this study, the occurrence of
side eects was not controlled, and the authors are
yet to present adetailed comparison of the three ex-
perimental groups and only compared placebo and
no treatment groups. e results seem to contradict
what we know about placebo, and it would be very
interesting to see asimilar study on an active placebo.
If an active placebo aected emotional processing to
agreater extent than apure placebo, it would be an-
other argument for using an active rather than apure
placebo in control groups.
conclusion and further
directions
e arguments for the active placebo response hypoth-
esis are based on direct and indirect evidence. Direct
evidence includes a smaller number of experimental
studies in which an active placebo was used which re-
port no signicant dierence in outcomes of treatment
with antidepressants and an active placebo. Moreover,
it may be assumed that the reduction in depressive
signs and symptoms can be achieved by avariety of
pharmacological agents (including medications other
than antidepressants), as well as by dierent non-
pharmacological treatment methods. It seems that
acommon factor of all these methods is patients’ or
participants’ conviction of their eectiveness.
It is important to conduct experimental studies in
which researchers manipulate participants’ expecta-
tions on the occurrence of side eects when taking
antidepressants. e majority of currently avail-
able studies did not include an active placebo – only
antidepressants and pure placebos were tested, and
additional questionnaires were distributed in order
to collect information on side eects that the partici-
pants had experienced during the trial. e aware-
ness of the fact that when taking part in aclinical
trial, one may be included either in agroup receiv-
ing placebo or receiving a medication, may not be
neutral to the treatment outcomes due to patients’
dierent expectations.
It is necessary to study the frequency of patients
discovering their study group assignment. According
to the active placebo response theory, participants of
studies guess their study group aer having observed
some side eects. e assumption is important for
the theory; however, empirical studies are scarce.
Although the Hamilton Rating Scale for Depres-
sion (Hamilton, 1986), which can be considered avalu-
able diagnostic tool (Kirsch, 2009), has been frequently
used in drug ecacy studies, it seems appropriate
to test the hypothesis with other depression scales.
It cannot be doubted that the Hamilton Rating Scale
for Depression is an adequate tool for measuring de-
pressive signs and symptoms, as it is characterized by
good psychometric properties, and the evaluation is
performed by aclinician who assesses the symptoms
by observing the patient (Gençöz, Gençöz, &Soykan,
2007). However, the inclusion of more tools could thus
support the ecological validity of the active placebo
response hypothesis. Furthermore, it has been proven
that self-rated scales might be more sensitive than
clinician-rated scales in milder forms of depression
(Cuijpers, Li, Hofmann, &Andersson, 2010).
References
American Psychiatric Association (2013). Diagnostic
and Statistical Manual of Mental Disorders (DSM-5).
Washington: APA Publishing.
Angoa-Pérez, M., Kane, M. J., Briggs, D. I., Herrera-
Mundo, N., Sykes, C. E., Francescui, D. M.,
&Kuhn, D. M. (2014). Mice genetically depleted
of brain serotonin do not display adepression-like

Wojciech
Oronowicz-
Jaśkowiak,
Przemysław Bąbel
6  
behavioral phenotype. ACS chemical neuroscience,
5, 908–919. hps://doi.org/10.1021/cn500096g
Aronson, R., Oman, H. J., Joe, R. T., &Naylor, C.D.
(1996). Triiodothyronine augmentation in the treat-
ment of refractory depression: a meta-analysis.
Archives of General Psychiatry, 53, 842–848. hps://
doi.org/10.1001/archpsyc.1996.01830090090013
Barbui, C., Cipriani, A., Patel, V., Ayuso-Mateos, J. L.,
&van Ommeren, M. (2011). Eicacy of antidepres-
sants and benzodiazepines in minor depression: sys-
tematic review and meta-analysis. British Journal of
Psychiatry, 198, 11–16. hps://doi.org/10.1192/bjp.
bp.109.076448
Bauer, M., & Döpfmer, S. (1999). Lithium augmen-
tation in treatment-resistant depression: meta-
analysis of placebo-controlled studies. Journal of
Clinical Psychopharmacology, 19, 427–434. hps://
doi.org/10.1097/00004714-199910000-00006
Bąbel, P. (2006). Placebo i nocebo. Próba integracji
[Placebo and nocebo. An aempt to integrate the
terms]. Przegląd Psychologiczny, 49, 141–156.
Beck, A. T., & Alford, B. A. (2009). Depression: Causes
and treatment. Philadelphia: University of Pennsyl-
vania Press.
Capezzani, L., Ostacoli, L., Cavallo, M., Carleo, S.,
Fernandez, I., Solomon, R., Pagani, M., &Cantel-
mi,T. (2013). EMDR and CBT for cancer patients:
Comparative study of eects on PTSD, anxiety,
and depression. Journal of EMDR Practice and Re-
search, 7, 134–143. hps://doi.org/10.1891/1933-
3196.7.3.134
Cowen, P. J. (2008). Serotonin and depression: patho-
physiological mechanism or marketing myth?
Trends in Pharmacological Sciences, 29, 433–436.
hps://doi.org/10.1016/j.tips.2008.05.004
Cuijpers, P., Li, J., Hofmann, S. G., &Andersson,G.
(2010). Self-reported versus clinician-rated symp-
toms of depression as outcome measures in
psychotherapy research on depression: a meta-
analysis. Clinical Psychology Review, 30, 768–778.
hps://doi.org/10.1016/j.cpr.2010.06.001
Dawes, R. M. (1998). “Listening to Prozac but hear-
ing placebo”: Commentary on Kirsch and Sa-
pirstein. Prevention & Treatment, 1. hps://doi.
org/10.1037//1522-3736.1.1.15c
Deacon, B. J., &Spielmans, G. I. (2017). Is the eicacy
of “antidepressant” medications overrated? In S.
O. Lilienfeld &I. D. Waldman (Eds.), Psychological
science under scrutiny (pp. 250–270). Hoboken, NJ:
John Wiley & Sons, Inc.
Dolińska, B. (2011). Placebo. Dlaczego działa coś, co
nie działa? [Placebo. Why it works when it doesn’t
work?]. Sopot: Smak Słowa.
Driessen, E., Cuijpers, P., de Maat, S. C., Abbass,A.A.,
de Jonghe, F., & Dekker, J. J. (2010). The eicacy of
short-term psychodynamic psychotherapy for de-
pression: a meta-analysis. Clinical Psychology Re-
view, 30, 25–36.
Dydak, K., Sliwinska-Mosson, M., &Milnerowicz, H.
(2016). Psilocybin as an alternative medicine for
patients suering from depression. Psychiatria
i Psychologia Kliniczna, 16, 165–170. hps://doi.
org/10.15557/PiPK.2016.0023
Edmond, T., Rubin, A., &Wambach, K. G. (1999). The
eectiveness of EMDR with adult female survi-
vors of childhood sexual abuse. Social Work Re-
search, 23, 103–116.
Ernst, E. (1995). St. John’s Wort, an anti-depressant?
A systematic, criteria-based review. Phytomedi-
cine, 2, 67–71.
Faria, V., Gingnell, M., Hoppe, J. M., Hjorth, O., Alaie,I.,
Frick, A., Hultberg, S., Wahlstedt, K., Engman, J.,
Månsson, K. N. T., Carlbring, P., Andersson, G.,
Reis, M., Larsson, E. M., Fredrikson, & M., Fur-
mark,T. (2017). Do you believe it? Verbal sugges-
tions influence the clinical and neural eects of
escitalopram in social anxiety disorder: arandom-
ized trial. EBioMedicine, 24, 179–188. hps://doi.
org/10.1016/j.ebiom.2017.09.031
Furukawa, T. A., Streiner, D., Young, L. T., &Kinoshita,Y.
(2001). Antidepressants plus benzodiazepines for major
depression. London: The Cochrane Library.
Gençöz, F., Gençöz, T., &Soykan, A. (2007). Psycho-
metric properties of the Hamilton Depression
Rating Scale and other physician-rated psychi-
atric scales for the assessment of depression in
ESRD patients undergoing hemodialysis in Tur-
key. Psychology, Health & Medicine, 12, 450–459.
hps://doi.org/10.1080/13548500600892054
Greenberg, R. P., Bornstein, R. F., Fisher, S., Zborow-
ski,M.J., &Greenberg, M. D. (1994). A meta-anal-
ysis of fluoxetine outcome in the treatment of de-
pression. The Journal of Nervous and Mental Disease,
182, 547–551. hps://doi.org/10.1097/00005053-
199410000-00003
Gu, Q., Dillon, C. F., &Burt, V. L. (2010). Prescription
drug use continues to increase: US prescription
drug data for 2007-2008. NCHS Data Brief, 42, 1–8.
Hamilton, M. (1986). The Hamilton rating scale for
depression. In M. Hamilton (Ed.), Assessment of
depression (pp. 143–152). Heidelberg: Springer.
Hieronymus, F., Lisinski, A., Nilsson, S., &Eriksson,E.
(2017). Eicacy of selective serotonin reuptake in-
hibitors in the absence of side eects: amega-anal-
ysis of citalopram and paroxetine in adult depres-
sion. Molecular Psychiatry, 23, 1731–1736. hps://
doi.org/10.1038/mp.2017.147
Hofmann, S. G., Asnaani, A., Vonk, I. J., Sawyer, A.T.,
& Fang, A. (2012). The eicacy of cognitive behav-
ioral therapy: A review of meta-analyses. Cogni-
tive Therapy and Research, 36, 427–440.
Hughes, S., Lacasse, J., Fuller, R. R., &Paulding-Giv-
ens, J. (2017). Adverse eects and treatment sat-
isfaction among online users of four antidepres-
sants. Psychiatry Research, 255, 78–86. hps://doi.
org/10.1016/j.psychres.2017.05.021

Twenty years
aer ‘Listening to
Prozac but hearing
placebo’
7
 7(1), 9
Huneke, N. T., Walsh, A. E., Brown, R., Browning, M.,
&Harmer, C. J. (2017). No evidence for an acute
placebo response on emotional processing in
healthy volunteers. Journal of Psychopharmacol-
ogy, 31, 1578–1587.
Hypericum Depression Trial Study Group (2002). Ef-
fect of Hypericum perforatum (St John’s wort)
in major depressive disorder: arandomized con-
trolled trial. Journal of the American Medical As-
sociation, 287, 1807–1814. hps://doi.org/10.1001/
jama.287.14.1807
Jakobsen, J. C., Katakam, K. K., Schou, A., Hellmuth,S.G.,
Stallknecht, S. E., Leth-Møller, K., &Krogh, J. (2017).
Selective serotonin reuptake inhibitors versus pla-
cebo in patients with major depressive disorder.
Asystematic review with meta-analysis and trial se-
quential analysis. BMC Psychiatry, 17, 58. hps://doi.
org/10.1186/s12888-016-1173-2
Jensen, J. S., Bielefeldt, A. Ø., & Hróbjartsson, A.
(2017). Active placebo control groups of pharmaco-
logical interventions were rarely used but merited
serious consideration: A methodological overview.
Journal of Clinical Epidemiology, 87, 35–46. hps://
doi.org/10.1016/j.jclinepi.2017.03.001
Keller, M. B., Ryan, N. D., Strober, M., Klein, R.G.,
Kutcher, S. P., Birmaher, B., Hagino, O. R., Kople-
wicz,H., Carlson, G. A., Clarke, G. N., Emslie,G.J.,
Feinberg, D., Geller, B., Kusumakar, V., Pa-
patheodorou,G., Sack, W. H., Sweeney, M., Wag-
ner, K. D., Weller, E. B., Winters, N. C., Oakes, R.,
&McCaerty, J. P. (2001). Eicacy of paroxetine
in the treatment of adolescent major depression:
arandomized, controlled trial. Journal of the Amer-
ican Academy of Child &Adolescent Psychiatry, 40,
762–772.
Khan, A., Fahl Mar, K., Fauce, J., Khan Schilling, S.,
&Brown, W. A. (2017). Has the rising placebo re-
sponse impacted antidepressant clinical trial out-
come? Data from the US Food and Drug Admin-
istration 1987-2013. World Psychiatry, 16, 181–192.
Kirsch, I. (2014). Antidepressants and the placebo
response. Zeitschri für Psychologie, 222, 128–134.
hps://doi.org/10.1027/2151-2604/a000176
Kirsch, I. (2009). The Emperor’s New Drugs. London:
The Bodley Head.
Kirsch, I., &Sapirstein, G. (1998). Listening to Prozac
but hearing placebo: A meta-analysis of antide-
pressant medication. Prevention & Treatment, 1.
hps://doi.org/10.1037/1522-3736.1.1.12a
Klein, D. F. (1998). Listening to meta-analysis but
hearing bias. Prevention & Treatment, 1. hps://
doi.org/10.1037/1522-3736.1.1.16c
Klemow, K. M., Bilbow, E., Grasso, D., Jones, K., Mc-
Dermo, J., &Pape, E. (2011). Medical aributes
of St. John’s wort (Hypericum perforatum). In
I. F.F. Benzie &S. Wachtel-Galor (Eds.), Herbal
Medicine (2nd ed., pp. 757–780). Boca Raton, FL:
CRC Press/Taylor &Francis.
Koszewska, I. (2003). Dziurawiec zwyczajny w lecze-
niu depresji [St. John’s wort in the treatment of
depression]. Farmakoterapia w Psychiatrii i Neuro-
logii, 1, 45–54.
Kube, T., &Rief, W. (2017). Are placebo and drug-spe-
cific eects additive? estioning basic assump-
tions of double-blinded randomized clinical trials
and presenting novel study designs. Drug Discov-
ery Today, 22, 729–735. hps://doi.org/10.1016/j.
drudis.2016.11.022
Lacasse, J. R., & Leo, J. (2015). Challenging the nar-
rative of chemical imbalance. In B. Probst (Ed.),
Critical Thinking in Clinical Diagnosis and Assess-
ment (pp. 275–282). New York: Springer.
Le Noury, J., Nardo, J. M., Healy, D., Jureidini, J.,
Raven, M., Tufanaru, C., &Abi-Jaoude, E. (2015).
Restoring Study 329: eicacy and harms of par-
oxetine and imipramine in treatment of major de-
pression in adolescence. British Medical Journal,
351, h4320. hps://doi.org/10.1136/bmj.h4320
Linde, K., Berner, M. M., & Kriston, L. (2008). St
John’s wort for major depression. Cochrane
Database of Systematic Reviews, 4. hps://doi.
org/10.1002/14651858.CD000448.pub3
Lund, K., Vase, L., Petersen, G. L., Jensen, T. S., &Finn-
erup, N. B. (2014). Randomised controlled trials
may underestimate drug eects: balanced placebo
trial design. Public Library of Science One, 9, e84104.
hps://doi.org/10.1371/journal.pone.0084104
Moncrie, J., Wessely, S., & Hardy, R. (2004). Ac-
tive placebos versus antidepressants for depres-
sion. Cochrane Database of Systematic Reviews, 1.
hps://doi.org/10.1002/14651858.CD003012.pub2
Netz, Y. (2017) Is the comparison between exercise
and pharmacologic treatment of depression in
the clinical practice guideline of the American
College of Physicians evidence-based? Frontiers
in Pharmacology, 8, 257. hps://doi.org/10.3389/
fphar.2017.00257
NHMRC (2015). NHMRC Statement: Statement on
Homeopathy. Retrieved from www.nhmrc.gov.au
[accessed 09.2017]
Osher, Y., &Belmaker, R. H. (2009). Omeg-3 fay ac-
ids in depression: A review of three studies. CNS
Neuroscience &Therapeutics, 15, 128–133.
Ostro, R. B., &Nelson, J. C. (1999). Risperidone aug-
mentation of selective serotonin reuptake inhibi-
tors in major depression. The Journal of Clinical
Psychiatry, 60, 256–259.
Rabkin, J. G., Markowitz, J. S., Stewart, J., McGrath,P.,
Harrison, W., itkin, F. M., &Klein,D.F. (1986).
How blind is blind? Assessment of patient and
doctor medication guesses in aplacebo-controlled
trial of imipramine and phenelzine. Psychiatry
Research, 19, 75–86. hps://doi.org/10.1016/0165-
1781(86)90094-6
Ramel, W., Goldin, P. R., Carmona, P. E., &Mcaid,J.R.
(2004). The eects of mindfulness meditation on cog-

Wojciech
Oronowicz-
Jaśkowiak,
Przemysław Bąbel
8  
nitive processes and aect in patients with past de-
pression. Cognitive Therapy and Research, 28, 433–455.
hps://doi.org/10.1023/B:COTR.0000045557.15923.96
Reuters (2015). Analysis of GSK’s Seroxat antidepres-
sant finds key data was held back. Retrieved from
www.reuters.com [accessed March 16, 2017]
Rutherford, B. R., Wall, M. M., Brown, P. J., Choo, T.H.,
Wager, T. D., Peterson, B. S., Chung, S., Kirsch, I.,
&Roose, S. P. (2016). Patient expectancy as amedi-
ator of placebo responses in antidepressant clinical
trials. American Journal of Psychiatry, 174, 135–142.
hps://doi.org/10.1176/appi.ajp.2016.16020225
Rybakowski, J. (2011). Leki stosowane w chorobach
afektywnych [Pharmacological treatment in aec-
tive disorders]. In M. Jarema &J. Rabe-Jabłońska
(Eds.), Psychiatria: podręcznik dla studentów medy-
cyny [Psychiatry for medical students] (pp. 457–
534). Warszawa: Wydawnictwo Lekarskie PZW.
Siwak, S., Oronowicz-Jaśkowiak, W., & Oronowicz-
Jaśkowiak, A. (2017). Efekt leczenia antydepresan-
tami może sprowadzać się do wzmocnionego efek-
tu placebo. Przegląd literatury [The eect of treat-
ment with antidepressant drugs is equivalent to
the enhanced placebo eect. The empirical support
of the Irving Kirsch’s theory]. In E. Bajek &B. A.
Nowak (Eds.), Choroby XXI w. – analiza współczes-
nej sytuacji zdrowotnej [Diseases of the 21st cen-
tury - analysis of the current health situation] (pp.
133–141). Lublin: Wydawnictwo Naukowe Tygiel.
Sneed, J. R., Rutherford, B. R., Rindskopf, D., Lane,D.T.,
Sackeim, H. A., &Roose, S. P. (2008). Design makes
a dierence: a meta-analysis of antidepressant
response rates in placebo-controlled versus com-
parator trials in late-life depression. The American
Journal of Geriatric Psychiatry, 16, 65–73. hps://doi.
org/10.1097/JGP.0b013e3181256b1d
Stahl, S. (1994). 5HT1A receptors and pharmacother-
apy. Is serotonin receptor down-regulation linked
to the mechanism of action of antidepressant
drugs? Psychopharmacology Bulletin, 30, 39–43.
Viksveen, P., Relton, C., & Nicholl, J. (2017). De-
pressed patients treated by homeopaths: a ran-
domised controlled trial using the “cohort multi-
ple randomised controlled trial” (cmRCT) design.
Trials, 18, 299. hps://doi.org/10.1186/s13063-017-
2040-2
Zhang, Z. J., Tan, Q. R., Tong, Y., Li, Q., Kang, W. H.,
Zhen, X. C., &Post, R. M. (2008). The eectiveness
of carbamazepine in unipolar depression: adouble-
blind, randomized, placebo-controlled study. Jour-
nal of Aective Disorders, 109, 91–97. hps://doi.
org/10.1016/j.jad.2007.11.004
.