Institut für Gesundheitsökonomie und Klinische Epidemiologie, Köln, Germany
The nocebo effect: A reason for patients’ non-adherence
to generic substitution?
J. Weissenfeld, S. Stock, M. Lüngen, A. Gerber
Received August 19, 2009, accepted December 4, 2009
Jörn Weißenfeld, Institut für Gesundheitsökonomie und Klinische Epidemiologie, Gleueler Str. 176-178 50935 Köln,
Pharmazie 65: 451–456 (2010) doi: 10.1691/ph.2010.9749
Substituting generic formulations of the same chemical agent is a common practice in German health care
on the basis of so called rebate contracts. The substitution of a medication may affect the patients’ adher-
experiencing a nocebo effect can subsequently become non-adherent or even discontinue an appropiate
therapy, the aim of this article is to study patients’ adherence to generic substitution and the extent of
the nocebo effect. In MEDLINE and EMBASE, a search was carried out for articles which were published
between March 25th, 1989 and March 25th, 2009 by using the following search terms: generic substitu-
tion, adherence, non-adherence, non-persistence, rebate contracts, patients’ attitude, nocebo, negative
placebo effects, placebo adverse reactions, placebo induced side effects and negative placebo responses.
In addition a manual search was performed in the reference lists of the articles retrieved. 14 studies met
the inclusion and exclusion criteria and were included in this article. The generic substitution was gen-
erally accepted by over two thirds of the study populations. But up to 34% of patients being treated for
psychological diseases confronted with a change of their medication had additional adverse events. On the
basis of the studies analysed, the conclusion can be drawn that the nocebo effect can play a crucial role
in the treatment of psychological diseases. Therefore, physicians and pharmacists should be responsible
to prevent the nocebo effect through adequately educating the patients.
Substituting generic formulations of the same chemical agent
is common practice in the German Statutory Health Insurance
(SHI) on the basis of so called rebate contracts. SHI serves
around 90% of the German population. The rebate contracts are
organizations of funds that traditionally serve the same segment
of population and pharmaceutical companies. Aside from prob-
adherence (Palagyi and Lassanova 2008).
Himmel et al. (2005), for instance, found that 36.7% of all
patients held the view that inexpensive products are infe-
rior to or different from the brand-name drugs. 13.2% of
patients who already had experience with a generic substi-
tute reported adverse effects that had not been observed with
the brand drug. In a survey of GfK-Marktforschung (GfK
Market Research), 43% of patients mentioned that the substitu-
encountered adverse events after the switch (http://www.nav-
accessed on February 11, 2008).
In order to improve patient care, it would be crucial to evalu-
ate the patients’ attitudes towards and experiences with generic
The latter are explained by the so called nocebo effect.
The term nocebo was coined as the opposite of placebo that is
to differentiate the positive and the negative effects of a placebo
(Kennedy 1961; Hahn 1985, 1997). According to their defini-
substance. Yet, for the purpose of our study a wider under-
standing of the nocebo effect is applied: The nocebo effect also
denotes any adverse effect that is caused by a medication, but is
not a pharmacological effect.
The existence of the nocebo effect is strongly influenced by a
appearance (Liccardi et al. 2004). The relevance of patients’
expectations can be demonstrated by the results of the Framing-
ham Study (Eaker et al. 1992): Women who believed that they
had a higher risk to die from a heart attack suffered as a matter
of fact from a heart attack 3.7 times more often than women
who did not believe that they had such a risk. The nocebo effect
usually surfaces as subjectively felt symptoms such as nausea,
headache or itching. But tachykardia, skin rashes or vomiting
as objective signs can also be observed (Liccardi et al. 2004).
Pharmazie 65 (2010) 451
The objective of this study was to evaluate patients’ adher-
ence despite generic substitution. Furthermore, the size and the
sequelae of the nocebo effect were assessed in order to demon-
strate whether the nocebo effect could be a reason for patients’
non-adherence to generics.
2. Inclusion and exclusion criteria for studies and
sion of studies were set up:
(1) Publication between March 25th, 1989 and March 25th,
(2) The patients’ expectations should not have been influ-
enced verbally before the onset of the study.
Additional criteria for the studies concerning the nocebo effect
(2) At least one medication had to be given in the study.
(3) Any side effects not considered a pharmacological effect
should have been analysed separately.
2.1. Search strategies
MEDLINE and EMBASE were searched for the time period
indicated. The search terms used were combinations of the
following keywords: generic substitution, adherence, non-
adherence, non-persistence, rebate contracts and patients’
attitude respectively nocebo, negative placebo effects, placebo
adverse reactions, placebo-induced side effects and negative
placebo responses. All hits were analysed on the basis of the
title, the abstract and the key words. In addition, we performed
a manual search in the reference lists of the articles that fulfilled
all inclusion criteria.
3. Patients’ acceptance of generic substitution
Eight studies (see Table 1) from the hits which were found in
MEDLINE and EMBASE could be included into the review
based on the inclusion and exclusion criteria.
jects were defined as substituted if they had been switched from
their prescribed brand-name product to a generic formulation
for the first time. The patients’ adherence was measured by cal-
was determined by MPR below 80%. 63 out of 463 substi-
tuted patients (13.6%) versus 111 out of 595 non-substituted
patients (18.7%) were considered to be non-adherent (Crude
OR 0.69; 95% CI 0.49 to 0.96). None of the patients of the
substituted group discontinued the therapy directly after the
switch. There was no difference in hospitalization for cardio-
vascular disease between the groups in the 6 months after the
Honrubia Alujer et al. (2007) examined the acceptance of the
769 patients. The patients were asked to substitute their pre-
scribed drug by a generic product when the prescribed one was
not available. 698 out of 769 patients (90.8%) agreed to the
substitution of their medication (95% CI; 88.5% to 92.7%).
The main reason for non-acceptance of the replacement was the
fact that the patients did not want any changes in their current
Palagyi and Lassanova (2008) analyzed the patients’ attitudes
to and their experiences with generic drugs in Slovakia. 2000
questionnaires with five questions related to generic prod-
ucts and generic substitution were handed out to patients
via eleven pharmaceutical companies representing the Slovak
Generic Assosiation (GENAS). Responses were received from
1777 out of 2000 patients (88.85%). 61.1% of the patients
did not have any distrust with taking generic drugs. A sig-
nificant difference could be detected with reference to the
patients’ age (p<0.001): The highest acceptance was found in
the age category up to 30 years (65.7%). 17.5% (311 out of
1775) of the patients preferred being prescribed a brand name
drug albeit being asked a higher co-payment. 56.5% (1003
out of 1775) patients preferred medications with lower co-
The attitudes to and experiences with generic substitution of
prescribed drugs among patients 50 years old and over were
evaluated by Ringuier et al. (2008) with a self-questionnaire.
440 patients were included in the study. 67% of the patients
had already received a generic by their physicians but only 45%
got additional information on the substitution. Elderly patients
(age category 75 and more) reported more adverse effects after
age group (20% versus 9%; p=0.027). 72% of the patients
were satisfied with the generic substitution but 57% felt the
need to obtain more information on the substitution. 85% of
them explicitly asked to receive this information from their
Patients’ attitudes to and experiences of generic substitution
were also assessed by Kjoenniksen et al. (2006) in Norway.
They included 386 patients who received eight or more differ-
ent drugs and age-adjusted controls who received three to seven
drugs in their study. The study was based on questionnaires
which were mailed to the participants. The response rate was
generic substitution from their physicians while 53% received
this information from the pharmacy staff (p<0.001). Patients
who were informed on generic substitution were more likely
to be switched (p<0.001). The substitution of their prescribed
drug with a product of a lower price was actually effectuated
by 138 (49%) of the patients. 41% preferred not to change their
medication without at the same time seeing financial savings.
Compared to the control group the patients with polypharmacy
were 2.6 times more likely to change to a generic formulation.
Respondents younger than 50 years old were 3.7 times more
likely to effectuate a generic substitution than patients 70 years
and older. 50 of the 138 participants of the study who had been
experiences with the generic product.
Shrank et al. (2007) analyzed different factors related to the
utilization of generic drugs. 5399 patients who filled a new pre-
cription in at least 1 of 5 classes of chronic drugs with generic
alternatives were enrolled in the study. 1262 (23.4%) of the
filled prescriptions were generics. Another 606 (14.9%) of the
patients who started their treatment with a brand-name prod-
uct switched to a generic formulation in the subsequent year.
Shrank reported that patients living in high-income zip codes
were more likely to initiate a therapy with generic drugs than
patients in low-income zip codes (p=0.02). Males in the age
category over 55 years were over 7.5 times more likely to be
substituted with generic formulations than male patients being
less than 25 years old (p=0.04). The customers of mail-order
pharmacies were 65% more likely to switch to a generic drugs
than those of independent pharmacies (p=0.003). Participants
of a three-tiered benefit plan were 2.61 times more likely to
switch to generic formulations than those enrolled in a 1- or
2-tier plan (p=0.03).
452Pharmazie 65 (2010)
Table 1: Overview of the studies concerning generic substitution
Response RateMean age of subjectsGender Focus on generic substitution
Kjoenniksen et al.
386 73.0% (281)65.8 167 females
50 of 138 patients who switched
to a generic formulation reported
13.6% of the patients who
switched to a generic
formulation were non-adherent
61.1% of the patients did not
have any distrust to use generic
Van Wijk et al.
463— 60.5 235 females
2000 88.9% (1777) Up to 18 years 5%
Up to 30 years 24.5%
Up to 45 years 32.9%
Up to 60 years 25.8%
Above 60 years 11.7%
50–65 years 45.0%
66–75 years 34.5%
Above 75 years 20.5%
Up to 20 years 4.8%
20–39 years 21.6%
40–59 years 29.4%
Above 60 years 44.2%
Up to 25 years 10.6%
25–39 years 23.0%
40–55 years 44.5%
Above 55 years 21.9%
1 no information
Ringuier et al.
440— 269 females
72.0% of the patients were
satisfied with the generic
90.8% of the patients agreed to
the substitution of their
et al. (2007)
769— 474 females
Shrank et al. (2007)
5399— 3261 females
Patients were 2.61 times more
likely to switch to a generic
formulation if they were enrolled
in three-tiered benefit plans
The comfort with the generic
substitution and the
communication with physicians
or pharmacists about generic
products were significantly
associated with the use of
13.2% of the patients who
switched to a generic formulation
reported additional side effects
Shrank et al. (2009)
2202 48.0% (1047)706 females
76 drop outs
Himmel et al. (2005)804 51.9%NANA
The relationship between patients’ beliefs or communication
about generic drugs was also evaluated by Shrank et al. (2009)
by using factor analysis to develop 5 multi-item scales from
1054 patient survey responses. In the fully adjusted model only
the comfort with the generic substitution (p=0.021) and the
communication with physicians or pharmacists about generic
products (p=0.012) were significantly associated with the use
of generic drugs.
222 out of a total sample of 804 patients (27.6%) included in a
a sceptical attitude towards the substitution. Nearly 30% were
not satisfied with the information that they had received from
their physicians. 12.2% of the patients reported that the generic
product showed a lower effectiveness than the brand-name for-
mulation. 13.2% complained about additional side effects and
28.9% needed to get used to the new colour or shape of the
4. Nocebo effect
Six studies (see Table 2) could be included into the review after
applying all inclusion and exclusion criteria.
Liccardi et al. (2004) included 600 patients in their study
who already had complained about medication adverse events.
Participants received capsules or oral solutions with various
concentrations of alternative drugs (vs. placebo), different in
structure from those suspected to have caused the previous
adverse effects. 162 of the patients (27%) mentioned adverse
effects with the placebo. 50/162 had objectively measurable
symptoms such as tachycardia, cough or skin lesions whereas
effects, however, differed from previous adverse events in more
than two thirds of cases. The prevalence of the nocebo effect
was significantly higher in women than in men (30% vs. 19%,
The interaction of the nocebo effect with the participants’ per-
sonality was analysed in a study by Drici et al. (1995) on 52
were assessed as type A (competitive and aggressive, N=16,
31%) or type B (N=36, 69%). Over a period of seven days,
they were given a drug-containing solution in one eye and a
of type A participants and 17% (N=6) of type B participants
reported adverse effects after placebo administration (p=0.03).
Participants in both groups who exhibited the nocebo effect
had higher mean Bortner rating scores than the non-responders
of 52) exhibited the nocebo effect.
Ströhle (2000) compared the effects of placebo and sodium lac-
disorder and in a control group with 23 otherwise healthy par-
ticipants. Participants got infusions of 0.5M sodium lactate or
0.9% saline over a 20min period. With the Acute Panic Inven-
tory that collects symptoms of a spontaneous as well as a lactate
induced panic attack the severity of symptoms were assessed
before the infusion and after 10, 20 and 30min. A score of 20
or more points or an increase of 14 points over the preinjection
score is considered a manifest panic attack (Dillon et al. 1987;
Pharmazie 65 (2010)453
Table 2: Overview of the studies concerning the nocebo effect
Number of participants Number of drop outsMean age of subjects GenderFocus on nocebo
Liccardi et al.
Drici et al. (1995)
600— 42 418 females
Nocebo effect occurred in 162
out of 600 patients (27%)
Nocebo effect occurred in 8 out
of 16 patients (50%) with
behaviour pattern A and 6 out of
36 (17%) type B subjects
Evidence for an increased
nocebo response in female
patients with panic disorder
53— 33.2 (female patients)
35.8 (male subjects)
29.4 (female control)
35.5 (male control)
60 (group 1)
61 (group 2)
Mondaini et al.
12013120 males 43.6% of group 2 patients
(informed on sexual side effects)
reported side effects as compared
to 15.3% in group 1 (no
information about sexual side
3.1% of the patients in group A
(no information), 15.6% in group
B (informed on given drug) and
31.2% (informed on given drug
and side effects) complained
about erectile dysfunction
Percentage of patients reported
side effects increased from 26%
to 60% after the switch of the
Silvestri et al.
96— 52 (group A)
52 (group B)
53 (group C)
Uhlenhuth et al.
541 40.8 28 females
patients and 21.7% of the controls had panic attacks (p<0.05)
while none was observed among the placebo group. The female
participants with panic disorder of the placebo group showed a
significant soaring of their API scores while this did not hap-
pen in the male patients nor in the controls (p<0.05). “Female
patients with panic disorder had more subthreshold panic anx-
iety as measured with the API score. The data give evidence
for an increased nocebo response in female patients with panic
disorder.” (Ströhle 2000, p. 439.)
Mondaini et al. (2007) detected nocebo induced sexual dys-
functions among 120 (13 drop outs) men with benign prostatic
of patients (N=52) got no information that sexual dysfunctions
may appear with the medication. The second group (N=55)
were handed out a brochure that contained a passage with the
information about the sexual adverse events of finansteride. In
the second group, the adverse effects could be found signif-
icantly more often than in the first group (43.6% vs. 15.3%,
p=0.03). Erectile dysfunction, decreased libido, and ejacula-
tion disorders could be observed in 9.6%, 7.7%, and 5.7% of
group 1 and in 30.9% (p=0.02), 23.6% (p=0.04), and 16.3%
(p=0.06) in group 2.
A similar result was stated by Silvestri et al. (2003). They
included 96 patients newly diagnosed with a cardiovascular dis-
ease who had no erectile dysfunction into a two-phase study.
32 patients (group A) were prescribed the beta-blocker atenolol
32 patients (group B) were informed about the medication, but
not about its side effect, and a third group C (32 patients) were
well informed about the chances of an erectile dysfuntion. After
three months of therapy, 3.1% (1 patient) in group A, 15.6%
(5 patients) in group B, and 31.2% (10 patients) in group C
(p<0.01) complained about erectile dysfuntions. The 16 men
with the erectile dysfunction were randomly assigned to silde-
nafil or placebo (= phase 2 of the study). In all but one of the
sixteen, Sildenafil and placebo had the same effect in reversing
the erectile dysfunction.
Uhlenhuth et al. (1998) analyzed the switch from alprazolam
to extended release alprazolam among 54 patients (1 drop out)
with an anxiety disorder stabilized on alprazolam. After the first
two weeks with their usual product, patients were switched for
two weeks to the extended release formulation. While 26% of
patients complained about adverse effects during the first part,
this went up to 60% after the change. After the switch, 48% of
a sedative effect. Caregiving physicians, however, refered to
only 24% of anxiety related effects as potentially caused by
the medication whereas 74% of the sedative effect might have
been caused by the alteration. Study data may not support the
existence of a pharmacological effect among the anxiety related
adverse effects, yet this may not be fully ruled out since plasma
for the anxiety related adverse effects that may not be explained
by the switch in the formula.
The studies included into our review show that generic substi-
tution is accepted by over two thirds of the patients (Honrubia
Alujer et al. 2007; Palagyi and Lassanova 2008; Ringuier et
al. 2008). Factors that contributed to a higher acceptance of a
switch are financial benefits (e.g., reduction of co-payment),
age (30–50 years old) or information on the substitution from
a physician or pharmacist (Kjoenniksen et al. 2006; Palagyi
and Lassanova 2008; Ringuier et al. 2008; Shrank et al. 2007,
Yet the findings indicate that the generic substitution can also
make patients non-adherent or trigger additional adverse events
(Himmel et al. 2005; Kjoenniksen et al. 2006; Ringuier et al.
switch of a formula yielded an increase of adverse effects by
454 Pharmazie 65 (2010)
34% in one study (Uhlenhuth et al. 1998). These adverse events
seemed to be related to the nocebo effect.
for physicians and pharmacists in the patient contact we would
like to dwell on strengths and limitations of the current study.
The study is based upon a literature search complemented by a
manual search in the reference lists of the studies that met our
inclusion criteria. Because of the heterogeneity among partici-
pants and the diversity of the endpoints of the various studies no
meta-analysis could be performed.
A general problem with the detection of the nocebo effect is
that adverse effects cannot be attributed to pharmacological or
non-pharmacological sequelae with certainty. Moreover, there
can be ethical reasons that forbid to perform research on the
nocebo effect and its size. It is possible to produce sequelae
such as asthmatic symptoms (Mc Fadden et al. 1969), allergic
symptoms (Jewett et al. 1990) or hyperalgesia (Benedetti et al.
2003) by inducing patients’ expectations. Yet, this is such a
stressful and terrifying experience for patients that may even
end in real adverse effects for the patients (Benedetti et al.
The fact that patients who are informed about generic substi-
tution are more likely to be switched (Kjoenniksen et al. 2006;
Palagyi and Lassanova 2008; Shrank et al. 2009) demonstrates
that it remains crucial in any patient-physician encounter that
the physician explains the substitution of a medication. Hence,
pharmacist should either attract the patient’s attention to differ-
ences with reference to the prior medication or assure that the
chemical agent is absolutely identical, albeit from a different
manufacturer. Since the rebate contracts result in financial ben-
efits for the statutory health funds, physicians and pharmacists
should be compensated for their additional effort.
The results of this review on the nocebo effect demonstrate
that the nocebo effect often occurs in patients being treated for
psychological diseases (Uhlenhuth et al. 1998). Patient charac-
teristics may contribute to nocebo adverse events (Barsky et al.
2002; Drici et al. 1995). Depression, anxiety, and somatisation
of active drugs and to be associated with nocebo-like symptoms
(Andrykowski and Redd 1987; Wolf and Pinsky 1954).
With the 2007 Bill on the Enhancement of Competition in the
SHI statutory health funds were given the right to conclude
rebate contracts with pharmaceutical companies in Germany
(http://bundesrecht.juris.de/sgb 5/, last accessed on November
08, 2009). Pharmacists are obliged to comply with rebate con-
tracts and have to hand out the medications to patients from the
various companies, unless there are pharmaceutical concerns.
Substitution of drugs, hence, is common and daily routine in
Yet a substitution is not recommended for all active ingre-
dients. In the guidelines “Gute Substitutionspraxis” (Good
practice of drug substitution), the “Deutsche Pharmazeutische
Gesellschaft” recommends to substitute antiepileptics or neu-
roleptics only if reliable data of bioequivalence are available for
the particular product to be substituted. (http://www.dphg.de/
lib/dphg leitlinie01 gsp 02-1.pdf, last accessed on November
08, 2009). Nevertheless, several health insurance funds con-
cluded rebate contracts on active pharmaceutical ingredients
like amisulpride, melperone or citalopram (http://www.aok-
html, last accessed on November 08, 2009).
If an antidepressant or neuroleptic is substituted, this may result
in a patient’s non-adherence caused by the nocebo effect or in
the termination of the therapy as the physician would see the
need to switch to a different drug or drug class.
In order to avoid consequences for the patients being treated for
psychological diseases actions should be taken early to prevent
non-pharmacological adverse effects. This is a task that should
be in the physicians’ and pharmacists’ responsibility. Even the
words physicians’ and pharmacists’use play an important role
in the daily practice because some words can trigger nocebo
responses in patients and need to be avoided (Benedetti 2002;
Unless a patient’s doubts could be dispelled in the patient-
physician or patient-pharmacist encounter, substitution should
treated for psychological diseases. In order to avoid discon-
tinuation of therapeutic regimens and economic consequences,
physicians and pharmacists should be committed to adequately
educate their patients and inform them on the nocebo effect.
The article “The nocebo effect: A reason for patients’ non-adherence to
generic substitution?” is part of Jörn Weissenfeld’s Ph.D. thesis on non-
brand substitution of medication in Germany. Neither the first author nor
the co-authors have received any financial contributions while working on
there are no conflicts of interest to declare.
of anticipatory nausea. J Consult Clin Psychol 55: 36–41.
side effects and the nocebo phenomenon. JAMA 287: 622–627.
Benedetti F (2002) How the doctor’s words affect the patient’s brain. Eval
Health Prof 25: 369–386.
Benedetti F, Pollo A, Lopiano L, Lanotte M, Vighetti S, Rainero I (2003)
and hormonal placebo/nocebo responses. J Neurosci 23: 4315–4323.
Benedetti F, Lanotte M, Lopiano L, Colloca L (2007) When words are
painful: Unraveling the mechanisms of the nocebo effect. Neuroscience
ment of lactate-induced panic and anxiety. Psychiatry Res 20: 97–105.
of the behaviour pattern on the nocebo response of healthy volunteers. Br
J Clin Pharmacol 39: 204–206.
death among women: psychosocial predictors from a 20-year follow-up
of women in the Framingham Study. Am J Epidemiol 135: 854–864.
Hahn RA (1985) A sociocultural model of illness and healing In: White
L, Tursky B, Schwatz GE (eds), Placebo: Theory, Research, and Mecha-
nisms, Guilford, New York, pp. 332–350.
Hahn RA (1997) The nocebo phenomenon: concept, evidence, and impli-
cation for public health. Prev Med 26: 607–611.
MM, Gleiter CH, Hummers-Pradier E (2005) What do primary care
patients think about generic drugs? Int J Clin Pharmacol Ther 43: 472–
Honrubia Alujer F, Carbajal de Lara JA, Cebrián Picazo C, Cuéllar Bolas B,
Silvestre Molina P, Merino Campos P, López-Torres Hidalgo R, Gascón
Escribano MJ, Martínez Tebar E, Cano Verdejo T, Andrés Cifuentes
E, Salar Pomares F, Gómez Escribano E, Giménez Benítez E (2007)
Aceptación de la sustitución por medicamentos genéricos en la oficina de
farmacia. Aten Primaria 39: 81–85.
Jewett DL, Fein G, Greenberg MH (1990) A double-blind study of symp-
tom provocation to determine food sensitivity. N Engl J Med 323:
Kennedy WP (1961) The nocebo reaction. Med World 95: 203–205.
Kjoenniksen I, Lindbaek M, Granas AG (2006) Patients’ attitudes towards
Liccardi G, Senna G, Russo M, Bonadonna P, Crivellaro M, Dama
A, D’Amato M, D’Amato G, Canonica GW, Passalacqua G (2004)
Pharmazie 65 (2010) 455
Evaluation of the nocebo effect during oral challenge in patients
with adverse drug rections. J Invest Allergol Clin Immunol 14:
Mc Fadden ER Jr, Luparello T, Lyons HA, Bleecker E (1969) The mech-
anism of action of suggestion in the induction of acute asthma attacks.
Psychosom Med 31: 134–143.
Mondaini N, Gontero P, Giubilei G, Lombardi G, Cai T, Gavazzi A, Bar-
toletti R (2007) Finasteride 5mg and sexual side effects: How many
of these are related to a nocebo phenomenon? J Sex Med 4: 1708–
Palagyi M, Lassanova M (2008) Patients attitudes towards experience with
use of generics in Slovakia, performance of generic substitution. Bratisl
Lek Listy 109: 324–328.
Ringuier R, Rouquette A, Dagorne C, Garnier F, Fanello S (2008) Connais-
sance et perceptions des médicaments génériques après 50 ans. Thérapie
Schenk PW (2008) “Just breathe normally”: Words choices that trigger
nocebo responses in patients. Am J Nurs 108: 52–57.
Shrank WH, Stedman M, Ettner SL, DeLapp D, Dirstine J, Brookhart MA,
Fischer MA, Avorn J, Asch SM (2007) Patient, physician, pharmacy, and
pharmacy benefit design factors related to generic medication use. J Gen
Intern Med 22: 1298–1304.
Shrank WH, Cadarette SM, Cox E, Fischer MA, Mehta J, Brookhart AM,
Avorn J, Choudhry NK (2009) Is there a relationship between patient
Med Care 2009 47: 319–325.
Silvestri A, Galetta P, Cerquetani E, Marazzi G, Patrizi R, Fini M, Rosano
is related to patient knowledge of side effects and is reversed by placebo.
Eur Heart J 24: 1928–1932.
Ströhle A, Kellner M, Yassouridis A, Holsboer F, Wiedemann K (1998)
Effect of flumazenil in lactate-sensitive patients with panic disorder. Am
J Psychiatry 155: 610–612.
Ströhle A (2000) Increased response to a putative panicogenic nocebo
administration in female patients with panic disorder. J Psychiatr Res
Ströhle A, Holsboer F, Rupprecht R (2000) Increased ACTH con-
centrations associated with cholecystokinin tetrapeptide-induced panic
attacks in patients with panic disorder. Neuropsychopharmacology 22:
tek AM (1998) Medication side effects in anxious patients: Negative
placebo responses? J Affect Disord 47: 183–190.
Van Wijk BL, Klungel OH, Heerdink ER, de Boer A (2006) Generic
substitution of antihypertensive drugs: does it affect adherence? Ann
Pharmacother 40: 15–20.
Weissenfeld J, Lüngen M, Stock S, Drabik A, Gerber A (2009) Substitution
of theophylline slow-release formulations according to the rebate con-
tracts in the German statutory health insurance. Arzneimittelforschung
of toxic reactions. J Am Med Assoc 155: 339–341.
456 Pharmazie 65 (2010)