Download Safety and effectiveness of olanzapine versus conventional

Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473 – 481
www.elsevier.com/locate/pnpbp
Safety and effectiveness of olanzapine versus conventional antipsychotics
in the acute treatment of first-episode schizophrenic inpatients
Julio Bobesa,*, Juan Gibertb, Antonio Ciudadc, Enrique Alvarezd, Fernando Cañase,
José-Luis Carrascof, Josep Gascóng, Juan-Carlos Gómezc, Miguel Gutiérrezh
a
Facultad de Medicina, Departamento de Psiquiatrı́a, Universidad de Oviedo, Metropolitan Area of Oviedo,
Calle Julián Claverı́a 6, E-33006 Oviedo, Asturias, Spain
b
Psychopharmacology Department, University of Cádiz, Cádiz, Spain
c
Clinical Research, Lilly Research Laboratories, Madrid, Spain
d
Psychiatry Unit, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain
e
Psychiatry Unit, Hospital Psiquiátrico de Madrid, Madrid, Spain
f
Psychiatry Unit, Hospital Fundación Jiménez Dı́az, Madrid, Spain
g
Hospital Mutua de Tarrasa, Barcelona, Spain
h
Psychiatry Unit, Hospital de Cruces, Baracaldo, Spain
Accepted 22 January 2003
Abstract
Objective: To assess the safety and effectiveness of olanzapine compared to typical antipsychotics in the treatment of first-episode
schizophrenics in acute psychiatric inpatient wards. Methods: Data were collected from a prospective, comparative, nonrandomized, open,
observational study of 904 inpatients with schizophrenia. One hundred and fifty-eight patients fulfilled the criteria for first-episode
schizophrenia, defined as (1) the International Classification of Diseases: Mental and Behavioral Disorders, 10th ed. (ICD-10) diagnosis of
schizophrenia, (2) antipsychotic naı̈ve, and (3) a course of illness of less than 5 years. Eighty-nine (56.3%) of these patients were assigned to
the olanzapine treatment group (OLZ) and 69 (43.7%) to the control group that received treatment with conventional antipsychotics (CON).
Safety was evaluated in terms of the spontaneous adverse events reported and a specific questionnaire for extrapyramidal symptoms (EPS).
Clinical status was measured by means of the Brief Psychiatric Rating Scale (BPRS) and Clinical Global Impression of Severity (CGI-S).
Clinical response was defined as the baseline-endpoint decrease in BPRS > 40% plus an endpoint BPRS < 18 or an endpoint CGI 3.
Results: The rate of clinical response to treatment in the OLZ was 76.7%, compared to 54.4% in the CON (c2 = 8.48; P=.003). Olanzapine
was significantly more effective than conventional antipsychotics in lowering the total BPRS score ( P=.0003), as well as each of the
following BPRS subscales: positive symptoms ( P=.0019), negative symptoms ( P < .0001), depression ( P=.018), and agitation ( P=.007),
even after mean scores were adjusted for their baseline value and disease duration. Olanzapine also proved to be significantly superior to
conventional antipsychotics in lowering mean CGI scores ( P=.013). The frequency with which new EPS appeared, or previously existing
ones worsened, was significantly greater in the CON than in the OLZ (55.1% vs. 13.5%; P < .001). Anticholinergics were needed more
frequently in the CON than in the OLZ (58.0% vs. 6.7%; P < .0001). Conclusions: The results of this observational, naturalistic study show
that olanzapine is safe and effective in a nonselected sample of acute, first-episode schizophrenic inpatients.
D 2003 Elsevier Science Inc. All rights reserved.
Keywords: Effectiveness; First episode; Inpatients; Olanzapine; Safety; Schizophrenia
Abbreviations: ANCOVA, analysis of covariance; BPRS, Brief Psychiatric Rating Scale; CGI-S, Clinical Global Impression of Severity; CON,
conventional antipsychotics; EPS, extrapyramidal symptoms; ICD-10,
International Classification of Diseases: Mental and Behavioral Disorders,
10th edition; LOCF, Last Observation Carried Forward; NOSIE, Nurses’
Observation Scale for Inpatient Evaluation; OLZ, olanzapine treatment
group; OR, odds ratio; UKU, Udvalg for Kliniske UndersØgelser (Side
Effect Rating Scale); WHO, World Health Organization.
* Corresponding author. Tel./fax: +34-985-10-35-53; mobile: +34-62987-78-00.
E-mail address: [email protected] (J. Bobes).
1. Introduction
Schizophrenia is a severe, incapacitating psychiatric disorder that, in most cases, leads to progressive personal
deterioration in patients, as well as to a significant burden
on family members and society in general (Torrey, 1995).
Despite the fact that the overwhelming majority of patients
recover from their first episode of schizophrenia (Tohen et
al., 1992), they run a high risk of subsequent relapse and the
0278-5846/03/$ – see front matter D 2003 Elsevier Science Inc. All rights reserved.
doi:10.1016/S0278-5846(03)00035-6
474
J. Bobes et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473–481
chronic persistence of residual positive and negative symptoms, cognitive deterioration, as well as social and employment difficulties (Hegarty et al., 1994; Szymanski et al.,
1995).
Longitudinal studies have demonstrated that schizophrenia progresses most during the first 5 years following the
initial episode of the disease (McGlashan, 1998). Nevertheless, there is currently great controversy regarding the exact
progression both of psychotic symptoms and the cognitive
deficits associated with schizophrenia (Keefe et al., 1999).
In spite of this controversy, early intervention at the very
beginning of the disease appears to be one of the most
important prognostic factors for the subsequent course of
schizophrenia (Loebel et al., 1992; Wyatt, 1995; Linszen et
al., 1998; Linszen and Dingemans, 2002; McGlashan et al.,
2002; Woods and McGlashan, 2002).
Conventional antipsychotics have been the mainstay of
the treatment of schizophrenia for several decades. Their
introduction in the 1950s allowed for the treatment of this
disease to move out of the hospital setting, where it had
been administered in the majority of cases, and into the
community, reserving hospitalization for those situations in
which the severity of psychotic or behavioral symptoms
did not allow the patient to undergo treatment in his/her
usual environment. Conventional neuroleptics have proven
their usefulness in rapidly reducing the intensity of psychotic symptoms, allowing for prompt hospital discharge, as
well as establishing the basis for long-term management of
the disease in these patients. However, conventional antipsychotics present a series of limitations in when treating
patients suffering from a psychotic episode. On the one
hand, a high percentage of schizophrenic patients respond
poorly when treated with this kind of drug (Brenner et al.,
1990) with a relapse rate of up to 60% during the year
following admission (Kane, 1996). In addition, conventional antipsychotics have the disadvantage of causing
many adverse effects, especially extrapyramidal symptoms
(EPS) including parkinsonism, dystonia, akathisia, and
tardive dyskinesia (Levinson et al., 1990) that, due to the
discomfort they cause, increase the likelihood that patients
will abandon medication (Weiden et al., 1986), thereby
increasing the risk of relapse. These drugs are also the
cause of the so-called ‘‘neuroleptic dysphoria,’’ a behavioral syndrome characterized by flat affect and avolition
that make social and occupational rehabilitation more
difficult in these patients.
Bearing this in mind, the emergence of the atypical
antipsychotics represented a ray of hope in the treatment
of schizophrenia, in particular because of their improved
tolerability profile compared with that of conventional
antipsychotics. In this regard, clozapine, the first atypical
antipsychotic to be put on the market, proved its superior
clinical efficacy in the treatment of refractory schizophrenic
patients with a minimal risk of provoking adverse extrapyramidal effects (Weiden et al., 1986). However, its poor
safety profile (due to the possibility of causing agranulocy-
tosis) has restricted its use, precisely to those patients in
whom the disease has been resistant to other types of
treatment. For this reason, there are no studies that bear
out its use in patients with non-refractory schizophrenia.
Olanzapine is a tienobenzodiazepine with a chemical
structure similar to clozapine. Receptor-affinity studies have
also shown that its affinity characteristics are quite similar to
those of clozapine (Bymaster et al., 1996; Moore et al.,
1993), while recent pharmacogenetic studies have also
discovered a great similarity between the two drugs (Mata
et al., 2001). Based on several controlled clinical trials,
olanzapine’s clinical efficacy has been shown to be at least
similar to (and in some cases even greater than) that of
conventional antipsychotics, with a much lesser risk of
producing extrapyramidal adverse effects (Beasley et al.,
1996a,b, 1997; Tollefson and Sanger, 1997, Tollefson et al.,
1997a,b; Tran et al., 1997).
A total of 1996 patients (Tollefson et al., 1997b) were
enrolled in a multicenter, double blind, 6-week clinical trial
comparing olanzapine with haloperidol; of them, those
patients suffering their first episode of schizophrenia were
selected. Olanzapine was seen to have a better efficacy
profile than haloperidol with a lesser risk of causing adverse
effects in this subgroup of patients, concluding that olanzapine should be considered a treatment of choice for firstepisode psychoses (Sanger et al., 1999).
However, the experimental conditions in which clinical
trials are conducted are such that the results cannot always
be entirely extrapolated to normal clinical practice. The
experimental nature of clinical trials makes it impossible
to enroll patients with little understanding of their disease,
with a co-morbid physical or psychiatric condition, or with
the concomitant use of other psychotropic drugs. If we take
into consideration the fact that when facing most schizophrenic patients, we must deal with one or more of the
aforementioned circumstances, it becomes clear that naturalistic studies that corroborate a drug’s efficacy in terms of
effectiveness in daily clinical practice are a necessity.
The objectives of this study were to determine olanzapine’s safety and effectiveness compared to conventional
antipsychotics in a cohort of schizophrenic patients with a
short course of illness and who had never received prior
treatment with antipsychotics.
2. Methodology
2.1. Study design
The subjects that participated in this study were taken a
phase IV, multicenter, observational, prospective, nonrandomized, comparative pharmacoepidemiological study
to assess the effectiveness and safety profile of olanzapine
in comparison with other antipsychotics in the hospital
setting. The study was conducted in Spain with the participation of 83 centers (Álvarez et al., 2003). Between
J. Bobes et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473–481
January and September of 1999, a total of 910 patients
diagnosed with schizophrenia according to the International
Classification of Diseases: Mental and Behavioral Disorders, 10th ed. (ICD-10) (F.20 of ICD10, World Health
Organization (WHO), 1994) that had been hospitalized for
an acute psychotic episode were included. Patients were
included in the study when, after admission, treatment was
started with oral olanzapine or any conventional antipsychotic drug. Those patients that were participating in any
other clinical trial or that were being treated with an atypical
antipsychotic other than olanzapine (for example, clozapine,
risperidone, sertindol, or quetiapine) were excluded, as were
those in whom antipsychotic treatment was contraindicated.
Indication for treatment was determined solely by clinical
criteria, and there was no restriction placed on the clinical
management of the patients, which was left entirely in the
hands of each psychiatrist. Patients were allowed to voluntarily drop out of the study at any time.
For the purposes of this study, those patients that met the
following conditions were selected from the aforementioned
sample: (1) they had never received antipsychotic treatment
prior to admission, and (2) the investigator considered that
the course of the disease had a duration of no more than 5
years. A subgroup of ‘‘naı̈ve’’ schizophrenic patients with a
course of disease of less than 5 years was thus defined.
The study protocol was developed by the sponsor and an
external advisory group created for this study (Drs. Alvarez,
Bobes, Carrasco, Cañas, Gascón, Gibert, and Gutierrez).
The protocol was submitted to the Spanish National Pharmacovigilance Department in compliance with Spanish
legislation applicable to nonexperimental observational
studies. In line with this regulation, no approval by Ethics
Committees of the participating centers nor patients signed
informed consent were required to be obtained prior to the
commencement of the study. Nevertheless, investigators
informed the patients about the objectives of the study,
and then oral consent to participate was obtained. Patients’
confidentiality was kept as no details were reflected in the
study documentation.
2.2. Evaluation instruments
The instruments used for evaluation were administered at
baseline at the beginning of treatment and weekly thereafter,
until the patient was discharged or medication was withdrawn.
The severity of psychotic symptoms was clinically
assessed by means of the Clinical Global Impression of
Severity (CGI-S) (Guy, 1976)) that gives a score ranging
from 1 (disease-free) to 7 (greatest severity possible), and
the Brief Psychiatric Rating Scale (BPRS) (Woerner et al.,
1988), which is a Likert-type scale comprised of 18 items
that are evaluated by the interviewer, in which each item
can be given a score of seven different values ranging from
0 (absence of symptom) to 6 (extreme severity). Apart from
the total score, scores were obtained for the subscales of
475
positive symptoms (conceptual disorganization, suspiciousness, hallucinations, thought disorders), negative symptoms
(emotional barriers, motor slowness, flat affect), agitation
(anxiety, tension, hostility, lack of collaboration, excitement), and depression (guilt feelings, depressed mood,
somatic complaints, anxiety). Patients’ behavior was
assessed by means of the Nurses’ Observation Scale for
Inpatient Evaluation (NOSIE) (Honigfeld and Klett, 1965).
Finally, EPS were assessed by means of a simple questionnaire based on the section of EPS of the Udvalg for
Kliniske Unders;gelser (UKU; Side Effect Rating Scale)
scale (Lingjaerde et al., 1987) that evaluated the presence
or absence of dystonia, hypertonia, hypokinesia, tremor,
dyskinesia, and akathisia. All adverse events spontaneously
reported by patients or taken down by each psychiatrist
were recorded and coded.
2.3. Statistical analysis
The department of biometrics at Phoenix International
carried out the statistical analysis. The SAS computer
program, version 8.1 for Windows (SAS Institute, Cary,
NC), was used for data processing and statistical analysis.
Investigators were asked to include patients consecutively
and naturalistically. In that way, two treatment groups were
defined: the olanzapine treatment group (OLZ), including
patients treated with olanzapine in monotherapy or in combination with conventional antipsychotics, and the control
group (CON), that included patients receiving one or more
conventional antipsychotic drug. In order to minimize selection bias, the participating psychiatrists were asked to include
six patients in each treatment group. As soon as a patient was
enrolled in the study, treatment could be modified as seen fit
by each psychiatrist, and all medication changes were
recorded. Patients could be switched from one group to the
other in the event of adverse effects, lack of efficacy, or for
any other reason as the psychiatrist saw fit. Each patient was
followed up throughout his/her entire hospital stay, although
in order to prevent distortions in the scores, this study only
includes the analysis of the scores obtained by those patients
that changed treatment group, up until the point at which
those patients were changed.
Analyses were made considering the two treatment
groups previously described: patients treated with olanzapine (OLZ) and patients treated with conventional antipsychotics (CON). Statistical analyses were made following
the principle of ‘‘intent-to-treat,’’ taking into account all
those patients from whom information had been gathered.
The incidence of each adverse event in each of the
treatment groups was calculated by means of the number
of patients that presented the event at any time during the
study with respect to the total number of patients. The
quantitative variables are described in terms of the mean,
median, standard deviation, and range. Discreet variables are
described in terms of frequency and percentage. Parametric
and nonparametric tests were used for the statistical analysis
476
J. Bobes et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473–481
of the continuous variables, based on the fulfillment of
statistical premises (normality and homocedasticity) and on
the nature of the variable. The Wilcoxon test was used to
compare age and duration of the course of illness. The change
on the BPRS and NOSIE scales was analyzed using an
analysis of covariance (ANCOVA) test with the treatment
group as a factor and the baseline score and years since
diagnosis as covariables. The CGI scale was evaluated by
means of a Wilcoxon test, given its ordinal nature. The chisquare test, or Fisher’s exact test in the event that the chisquare test could not be applied, was used to analyze the
discreet variables (sex, type of schizophrenia, incidence of
EPS and other adverse effects, percentage of patients that
responded, dropouts due to adverse effects, and the presence
of concomitant treatment). A bilateral signification level of
.05 was considered for all analyses. No adjustment was made
for those variables in which differences were not found at
baseline, given their lack of clinical relevance.
A Last Observation Carried Forward (LOCF) approximation was made to analyze the changes in the clinical
scales, including those patients with a baseline and at least
one post-baseline evaluation.
3. Results
The sociodemographic and baseline clinical data of the
158 patients that made up the sample appear in Table 1.
Randomization was not carried out due to the naturalistic
design of the study. Thus, the data pertaining to the 89
(56.3%) patients assigned to the OLZ were compared
with the 69 (43.7%) patients assigned to the control
group that was treated with conventional antipsychotics
(CON). No clinically significant differences were detected
between the two groups in any of the sociodemographic
or baseline clinical characteristics. Although more male
than female patients were included (73.4% vs. 26.6%),
distribution in the two groups was similar. The course of
the disease was of less than 1 year in 47 patients (24 in the
OLZ group vs. 23 in the CON group), between 1 and 2
years in 37 patients (24 vs. 13), between 2 and 3 years in 31
patients (16 vs. 15), between 3 and 4 years in 22 patients
(12 vs. 10), and between 4 and 5 years in 21 patients (13
vs. 8).
The percentage of patients that had received some kind
of pharmacological treatment since admission but before the
initial evaluation was similar in both groups (80.9% vs.
82.6%). Nevertheless, a greater percentage of patients in the
control group had received parenteral treatment upon admission, as compared to the olanzapine group (26.3% vs.
13.9%, P=.076).
At the time of discharge, a greater percentage of patients
from the olanzapine group were receiving concomitant
treatment with benzodiazepines (44.9% vs. 33.3%) and
antidepressants (4.5% vs. 0.0%). However, anticholinergics
were more frequently used in the control group made up of
patients treated with conventional antipsychotics as opposed
to the olanzapine group (58.0% vs. 6.7%, P < .0001).
Upon discharge, 13 (14.6%) of the patients treated with
olanzapine were receiving concomitant treatment with highpotency antipsychotics, whereas 6 (6.7%) were receiving
low-potency antipsychotics. In the group of patients treated
with conventional antipsychotics, although all had received
high-potency antipsychotics at some point during hospit-
Table 1
Sociodemographic and baseline clinical characteristics of the sample
Characteristic
Age
mean (S.D.)
median
range
Sex (% males)
Time of evolution (years)
Schizophrenia subtype (%)
Number of prior admissions
Baseline CGI score
Baseline BPRS score
BPRS Positive Symptom Subscale
BPRS Negative Symptom Subscale
BPRS Agitation Subscale
BPRS Depression Subscale
Baseline NOSIE score
Baseline extrapyramidal symptoms (%)
a
b
c
Wilcoxon test.
Chi-square test.
Fisher’s exact test.
mean (S.D.)
median
paranoid
undifferentiated
disorganized
residual
mean (S.D.)
median
mean (S.D.)
mean (S.D.)
mean (S.D.)
mean (S.D.)
mean (S.D.)
mean (S.D.)
mean (S.D.)
Olanzapine (n = 89)
Control group (n = 69)
P-value
28.6 (9.4)
27
16 – 74
77.5
1.6 (1.4)
1
80.7
11.4
6.8
1.1
1 (1.6)
0
5 (0.9)
43.6 (12.8)
14.3 (4.3)
7.3 (4.1)
13.7 (6.2)
8.1 (4.2)
44.2 (12.6)
27.0
26.1 (8.4)
24
17 – 68
68.1
1.5 (1.4)
1
78.3
8.7
11.6
1.4
0.7 (1.2)
0
5.3 (0.7)
46.6 (13.4)
15.9 (3.9)
7.6 (4.5)
14.8 (5.5)
7.9 (4.3)
44.9 (17.1)
18.8
.042a
.184b
.651a
.735c
.340a
.042a
.214a
.066a
.657a
.183a
.739a
.891a
.232b
J. Bobes et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473–481
Table 2
Initial, mean, final, and modal doses of olanzapine and haloperidol during
study period
Initial doses (mg)
Mean doses (mg)
Final doses (mg)
Modal doses (mg)
mean (S.D.)
median
range
mean (S.D.)
median
range
mean (S.D.)
median
range
mean (S.D.)
median
range
Olanzapine
(n = 89)
Haloperidol
(n = 61)
13.8 (6.2)
10
5 – 30
16.4 (6.0)
16.7
5 – 36
17.0 (6.1)
20
5 – 30
15.8 (6.2)
15
5 – 30
14.3 (7.5)
12
2 – 30
15.5 (8.1)
12
3.5 – 42.5
13.6 (7.8)
10.5
2.5 – 30
15.3 (11.7)
12
1.5 – 75
alization, 52 (75.4%) of them were discharged while receiving only the high-potency antipsychotic drugs, 13 (18.8%)
were taking only low-potency antipsychotics, and 4 (5.8%)
were on a combination of both. Haloperidol was the most
frequently prescribed antipsychotic in the control group,
with 60 (87%) patients having received this drug at some
point during hospitalization, and 46 (66.7%) were receiving
it as treatment upon discharge. The doses of olanzapine and
haloperidol used throughout the study are shown in Table 2.
The initial dose refers to the one prescribed at baseline,
whereas the mean dose was calculated based on the mean
dose that each patient received throughout the study period.
The final dose refers to the one prescribed at the time of
discharge or at dropout. Finally, the modal dose is defined as
the daily dose of the drug most frequently prescribed
throughout the study period.
Only one of the 89 patients that were assigned to the
olanzapine group at the beginning of the study was
changed to the treatment group receiving conventional
antipsychotics during the course of hospitalization due to
insufficient clinical efficacy. However, of the 69 patients
that had started the study in the conventional antipsychotics
treatment group, 13 (18.8%) were changed to the OLZ by
the investigator. The reasons for switching these patients
were the emergence of secondary effects in 10 cases and
insufficient clinical efficacy in the remaining 3 cases.
A total of 10 (11.2%) patients from the olanzapine
group prematurely discontinued their participation in the
study, versus a total of 7 (10.1%) in the conventional
antipsychotics treatment group. The reasons were the
patient’s decision (one case in each treatment group), the
investigator’s decision (five cases in the OLZ group; four
cases in the CON group), and loss to follow-up (four cases
in the OLZ group; two cases in the CON group).
477
the patients in the olanzapine group and those in the control
group (27% vs. 18.8%). Considering EPS individually,
statistically significant differences were observed only with
regard to dystonia, which was observed in four (5.8%)
patients belonging to the control group, as compared to
none of the patients assigned to the olanzapine group
(Fisher’s exact test, P=.035).
The frequency with which new EPS appeared, or with
which preexisting ones worsened, was significantly
greater in the control group as opposed to the olanzapine
group (55.1% vs. 13.5%, P < .001). When examined
individually, all the EPS considered in this study were
significantly more frequent in the control group (conventional antipsychotics) than in the group that was treated
with olanzapine, with the exception of dyskinesia that did
not attain a level of statistical significance (Table 3).
There was a higher percentage of patients with at least
one adverse effect in the control group compared to the
olanzapine group (60.9% vs. 19.1%, P < .0001). Therefore, the risk of suffering some type of adverse effect in
the olanzapine-treated group was three times less than for
the group treated with conventional antipsychotics. Not a
single case of leukopenia associated with the use of
olanzapine was found.
3.2. Effectiveness
Olanzapine was significantly superior to conventional
antipsychotics in lowering both the total BPRS score, as
well as each one of the subscales (positive symptoms,
negative symptoms, depression, and agitation), even when
the mean values were adjusted for the baseline value and
duration of the course of illness. Olanzapine was also shown
to be significantly superior to conventional antipsychotics in
lowering the mean score on the CGI scale (Clinical Global
Impression). With respect to the NOSIE, olanzapine was
once again superior to conventional antipsychotics, although
Table 3
Presence of extrapyramidal symptoms associated with treatmenta
Olanzapine (n = 88) Control group (n = 68) P-value
N
%
N
%
13.6
38
55.9
< .001y
0.0
5.7
3.4
3.4
3.4
1.1
2.3
10
12
22
17
17
2
2
14.7
17.6
32.4
25.0
25.0
2.9
2.9
< .001y
.021y
< .001y
< .001y
< .001y
.581y
1.0a
3.1. Safety
Extrapyramidal
12
symptoms
Individual symptoms
Dystonia
0
Rigidity
5
Hypokinesia
3
Tremor
3
Akathisia
3
Dyskinesia
1
Others
2
No significant differences were observed with respect to
the presence of EPS at the beginning of the study between
a
There is one case in each treatment group that has been omitted owing
to insufficient data.
y
Fisher’s exact test.
478
J. Bobes et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473–481
in this case the differences observed did not reach a level of
statistical significance (Table 4).
A more robust measure of treatment efficacy can be
obtained by calculating ‘‘clinical response,’’ defined a priori
in our study as a decrease of at least 40% in the BPRS score
plus a CGI scale score of 3 or less, or a final BPRS score of
less than 18. In keeping with these criteria, the treatment
response rate in the olanzapine group was 76.7%, compared
with a rate of 54.4% in the control group. The difference in
the response rate between both groups was statistically
significant (c2 = 8.48, P=.003). When applying a logistics
regression model adjusted for the baseline CGI and BPRS
values and time elapsed since diagnosis, the differences
between both treatment groups continue to be significant
( P=.044), with a response rate between two and three times
greater for the group that was treated with olanzapine (odds
ratio, OR = 2.7).
Effectiveness was also measured by the number of
patients that improved their BPRS scores at different levels.
Thus, 73 patients (84.9%) of the patients in the olanzapine
group were seen to have received total BPRS scores that
were more than 40% lower, as compared to 46 (67.6%)
Table 4
Change in CGI, BPRS, and NOSIE scores in each treatment group
Olanzapine
(n = 89)
CGI mean
improvement
(S.D.)
Total BPRS mean
improvement
(S.D.)
BPRS Positive
Symptom Subscale
mean improvement
(S.D.)
BPRS Negative
Symptom Subscale
mean improvement
(S.D.)
BPRS Depression
Subscale mean
improvement
(S.D.)
BPRS Agitation
Subscale mean
improvement
(S.D.)
NOSIE mean
improvement
(S.D.)
a
Control group
(n = 69)
P-value
.013a
2.0 (1.2)
1.6 (1.1)
30.0 (14.3)
24.7 (16.3)
30.8
10.0 (5.1)
23.6
8.9 (5.5)
.0003b
10.5
3.9 (3.4)
8.3
2.0 (3.7)
.0019b
4.0
5.2 (3.5)
1.9
4.2 (3.8)
< .0001b
5.2
10.2 (6.2)
4.2
9.2 (6.0)
.018
10.5
20.3 (13.2)
8.8
17.4 (15.7)
.007b
20.6
16.9
.0671b
Wilcoxon Test.
ANCOVA with mean values adjusted for baseline value and duration
of course of illness.
b
patients in the control group. This finding was confirmed at
reduction levels of 60% (69.8% of the OLZ group versus
45.6% of the CON group) and 80% (34.9% of the OLZ
group compared to 19.1% of the CON group). Overall, the
differences were statistically significant ( P=.001).
4. Discussion
This study was carried out with a subgroup of patients
that had participated in large prospective, observational
study performed with atypical antipsychotics in the hospital
setting and that had a diagnosis of schizophrenia that had
not previously been treated with antipsychotics and with a
course of disease of less than 5 years. The subgroup of
patients that are the subject this work is the largest one we
are aware of, in terms of the number of patients with firstepisode schizophrenia studied in this way. Certain limitations, inherent in any observational study, should be mentioned before proceeding with the discussion of the most
relevant results: (1) selection bias due to the lack of
randomization, (2) problems in establishing causal relationships due to the heterogeneity of the control group and the
frequent use of concomitant medications, and (3) the possible decrease in the appreciation of adverse effects as
compared to clinical trials. Nevertheless, we must take into
account the fact that it would be logical to assume that the
latter two limitations mentioned occurred equally in both
treatment groups such that the comparisons made between
both treatment groups should not therefore be greatly
affected. On the other hand, the fact that it was not an
experimental clinical trial is, to a certain degree, an advantage, if what we are attempting to study is what truly occurs
in normal clinical practice with patients being treating in an
inpatient setting and without the limitations imposed by a
clinical trial.
The main objective of this study was to assess olanzapine’s safety and effectiveness in comparison with conventional antipsychotics in normal clinical practice with patients
in whom the course of disease was a short one, that were
naı̈ve to antipsychotics, and that were experiencing a first
episode of schizophrenia. A relatively high number of
patients were studied, considering that they were schizophrenics that had never been previously treated and that were
hospitalized in a psychiatric ward. At the same time, we find
that the retention rate was high, over 84%. Another advantage to this study, unlike what has occurred in other observational studies that have been published recently with other
antipsychotics, is precisely the fact that this study included a
control group of patients treated with conventional antipsychotics.
The mean dose of olanzapine prescribed was well above
that prescribed in the only controlled clinical trial with
patients experiencing their first psychotic episode that we
are aware of (Sanger et al., 1999). The modal dose (defined
as the dose most frequently administered to each patient
J. Bobes et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473–481
throughout the study period) of olanzapine in the aforementioned study was 11.6 mg/day, as compared to the 15.8 mg/
day that we obtained in this study. Similarly, in the aforementioned article, the modal dose of haloperidol was 10.8,
whereas in ours, it was 15.3 (for those patients being treated
with this antipsychotic in the control group). We can think
as explanations for this difference between the mean doses
used that the conditions in which a clinical trial take place
require that the investigators keep their patients under strict
protocol limits, with a tendency to administer lower doses of
medication. Nevertheless, we estimate that the value of
these data is that they reflect the routine clinical practice
where antipsychotics are frequently used in higher doses.
This was a naturalistic study that allowed concomitant
medication to be administered according to the psychiatrists’ criteria. A high percentage of patients in both treatment groups received concomitant medication throughout
the time they were hospitalized, especially benzodiazepines.
Insofar as both benzodiazepines and antidepressants or
mood stabilizers are concerned, the differences observed
between both treatment groups were not statistically significant. On the other hand, it was observed that 72.5% of
patients treated with conventional antipsychotics also
received anticholinergic drugs during their hospital stay,
versus a mere 14.6% of those treated with olanzapine.
Only a small number of patients changed treatment group
while they were hospitalized, confirming the psychiatrists’
propensity to not change a patient’s treatment during their
hospital stay. Whereas only 1 (1.1%) of the patients
assigned to the OLZ upon admission had his/her medication
modified, 13 (18.8%) of those that started treatment with
conventional antipsychotics were switched to the olanzapine
group during the course of hospitalization. It is striking to
note that none of the patients that initiated treatment with
olanzapine had to abandon treatment due to the appearance
of adverse effects.
4.1. Safety
One of the essential objectives of treating recent onset
schizophrenics, once the psychotic symptoms have stabilized, is to assure future treatment compliance, as it has been
proven that this is one of the leading prognostic factors in
schizophrenia (Loebel et al., 1992; Linszen et al., 1998). In
our study, none of the patients that were assigned to the
OLZ required a change in medication as the result of
adverse effects, whereas 10 (14.5%) of those that were
assigned to the conventional antipsychotics treatment group
were switched to the OLZ for this reason. There have been
no unexpected safety problems of clinical relevance.
As for the frequency with which any kind of drug-related
adverse effect appeared, a significant difference was also
noted between both treatment groups ( P < .0001). Patients
treated with conventional antipsychotics presented a relative
risk of developing an adverse effect of more than three times
that of patients receiving olanzapine.
479
While no significant differences were detected in terms
of the presence of EPS at the start of the study, it is
noteworthy that the frequency with which they worsened
or new symptoms of this type appeared was significantly
greater in the group of patients treated with conventional
antipsychotics than in the olanzapine-treated group
( P < .0001). If we examine EPS (dystonia, rigidity, hypokinesia, tremor, akathisia, and dyskinesia) individually, all
of these symptoms became worse or appeared for the first
time significantly more frequently in the group that received
treatment with conventional antipsychotics than in the OLZ,
with the exception of dyskinesia. These results are compatible with those observed in clinical trials both with firstepisode schizophrenics (Sanger et al., 1999) and with more
chronic patients (Beasley et al., 1996a,b).
In this study, olanzapine, in comparison to conventional
antipsychotics, showed a better safety profile. This is of
great relevance in treating schizophrenic patients for the first
time, a point at which it is crucial that adverse effects be
minimized in order to assure subsequent compliance with
the medication. Patients never previously treated in general
tend to be more sensitive to antipsychotics, especially as
regards the emergence of EPS (McEvoy, 1991), so that
when treating this type of patient, we should opt for
administering those antipsychotics with a lesser risk of
causing them.
4.2. Effectiveness
It is of the utmost importance that effective treatment be
started as soon as possible in schizophrenia. Many studies
have established a clear association between the duration of
untreated or ineffectively treated psychosis and a poor
prognosis of the disease (Loebel et al., 1992; Wyatt,
1995). In our study, olanzapine demonstrated superior
effectiveness in comparison with conventional antipsychotics in the treatment of schizophrenic patients naı̈ve to
antipsychotics.
If we take the individual measurement instruments into
consideration, the improvement observed in olanzapinetreated patients was greater than that achieved by the
patients treated with conventional antipsychotics. After
making the necessary adjustments for the most relevant
baseline variables, a statistically significant improvement
was seen in patients treated with olanzapine as measured by
all the instruments used.
Clinical treatment response, defined either as a decrease
of 40% on the BPRS plus a CGI score of 3 or less, or a final
BPRS score under 18, was significantly more frequent in the
OLZ than in the CON treatment group ( P=.003). This result
was corroborated when the different levels of improvements
on the BPRS were taken into account.
Our results with respect to clinical effectiveness do not
appear to present artifacts as a result of the patients’ baseline
situation since, on the one hand, no statistically significant
differences were observed between both groups at the start
480
J. Bobes et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473–481
of the study and, on the other hand, the analyses were made
conservatively, adjusting the differences for baseline values.
5. Conclusions
In short, this observational study found a superior effectiveness of olanzapine, both in terms of efficacy and safety,
when compared to conventional antipsychotics in treating
hospitalized patients with schizophrenia that had never been
previously treated with antipsychotics.
Acknowledgements
This work was sponsored by Lilly, Madrid, Spain.
Psychiatrists who participated in the study as investigators:
José Antonio Hernández, Recaredo Cap de Pon, Andrés
Sandoval, Concepción Ros, Marı́a Concepción Sáez,
Francisco Martı́nez, Francisco Gázquez, Antonio Martı́nez,
Ali Abú Taleb, Alfredo de la Rubia, Andrés Simón,
Agustina González, Marı́a Antonia Garcı́a, Juan Luis
Figuerido, Enrique Daniel Vega, Juan Pedro Jiménez, Alicia
Iglesias, Adolfo J. Pellejero, Santiago Parada, Alfonso
Casas, José Marı́a Blanco, Jose Civeira, Juan Carlos Dı́az
del Valle, Manuel Serrano, Ma. Isabel Gómez del Cid, Marı́a
Victoria Ortega, Mercedes Alba, Ernesto Linares, Marina
Gómez, César Azpeleta, Alicia González, Mauro Garcı́a,
Marcelo Camus, Rosario Gil, Pilar Samada, Avelino de
Dios, José Vicente Pozo, Marı́a Remedios Fernández, Julio
César Carazo, Rosa Marı́a Alameda, Juan Francisco
Guardia, Margarita Silvestre, Pablo Calderón, Leopoldo
Elvira, Concepción Merino, Raimundo Muñóz, Leonor Coy,
Manuel Delgado, Carlos de Gregorio, Ana Moro, José
Valle, Patricia Alcinari, Santiago Pérez, Zoilo Fernández,
Rafael Gómez, Federico Wamba, Antonio Sánchez, Jose
Manuel Bertolı́n, Juan Petrel, Francisco Moreno, David
Simón, Manuel Brito, César Rodrı́guez, Angel Trujillo,
Antonio Higueras, Pedro Bustos, Rosario de Arce, Laura
Iglesias, Marı́a Pilar Pérez, Marta Torreblanca, Jesús Garcı́a,
Marı́a Gulbaharian, Pedro Sopelana, Manuel Franco, Jesús
Monforte, Jordi Sanahuja, Maria Elena Alea, Jesús Gómez,
Jesús Valle, Pablo Ramos, Pilar Nieves Ureña, Manuel
Barceló, Concepción Ciscar, Pilar P. de la Fuente, Josep
Ribes, Manuel Camarero, José Vicente Baeza, Enrique
Navarro, Eliseo López, Teresa Guilera, Román Calabuig,
Jesús Morillas, Yolanda Zapico, José Miguel Segovia,
Alfredo Dı́az, Leandro Palicio, Juan José Madrigal, Blanca
Garcı́a, J. Fernández, José Ferrer, Ester Gómez, Marı́a Jesús
Montes, Santiago Escote, Ramón Coronas, Gemma Garcı́a,
César Antón, Pedro Moreno, Carmen Busuldo, Jesús Cobo,
Emily Miró, Enrique del Moral, José Horta, Ana Marı́a
Fargas, Carlos Martı́n Martı́n, Jesús Derecho, Rafael
Iglesias, Imanol Querejeta, Ana Garcı́a, Ana Cristina Sierra,
Edith Garcı́a, Ildefonso Mateo, Ma. Angeles Caballero,
Ginés Palenciano, Federico Dourdill, David Huertas,
Ignacio Tortajada, Jorge Pérez, Javier Sanz, Angel Redondo, José Carlos González, Francisco Ballester, José
González, Salvador Sarro, Carmen Sarrı́, José Marı́a
Sánchez, Luisa Tifón, Vicenc Vallés, Luis Delgado, Enrique
Hernández, Manuel Rodrı́guez, Julio Lequerica, Josefina
Pérez, Luis Torremocha, Lucı́a Pérez, Diego Arenas, José
Luis Benavente, Ana Robador, Marı́a Jesús Luna, Ignacio
Zarranz, José Antonio Sánchez, Javier Casanova, Purificación Salgado, Alejandro David Cabo, Adela Paillissé, Jose
Manuel Jaquotot, Roberto Rodrı́guez, Eduardo Pons, Pilar
Alvarez, Nieves Casas, Ana Yáñez, Ana Rodrı́guez, Ester
Ibarrola, Elena Basurto, José Civeira, Juan Luis Figuerido,
Ma Concepción Sáez, Francisco Martı́nez, Concepción
Merino, Rosario de Arce, Pedro Sopelana, Jesús Gómez,
Juan José Madrigal, Marı́a Jesús Montes, and Carmen
Busuldo.
References
Álvarez, E., Bobes, J., Gómez, J.C., Sacristán, J.A., Cañas, F., Carrasco,
J.L., Gascón, J., Gibert, J., Gutiérrez, M., 2003. Safety of olanzapine
versus conventional antipsychotics in the treatment of patients with
acute schizophrenia: a naturalistic study. Eur. Neuropsychopharm. 13
(1), 39 – 48.
Beasley, C.M., Sanger, T., Satterlee, W., Tollefson, G., Tran, P., Hamilton,
S., 1996a. Olanzapine HGAP Study Group: olanzapine versus placebo:
results of a double-blind, fixed-dose olanzapine trial. Psychopharmacology (Berl.) 124, 159 – 167.
Beasley, C.M., Tollefson, G., Tran, P., Satterlee, W., Sanger, T., Hamilton,
S., 1996b. Olanzapine HGAP Study Group: olanzapine versus placebo
and haloperidol: acute phase results of the North American double-blind
olanzapine trial. Neuropsychopharmacology 14, 105 – 118.
Beasley, C.M., Hamilton, S., Crawford, A.M., Dellva, M.A., Tollefson,
G., Tran, P., Blin, O., Beuzen, J.M., 1997. Olanzapine E003 Study
Group: olanzapine versus haloperidol: acute-phase results of the international double-blind olanzapine trial. Eur. Neuropsychopharmacol. 7,
125 – 137.
Brenner, H.D., Dencker, S.J., Godstein, M.J., Hubbard, J.W., Keegan, D.L.,
Kruger, G., Kulhanek, F., Liberman, R.P., Malm, U., Midha, K.K.,
1990. Defining treatment-refractoriness in schizophrenia. Schizophr.
Bull. 16, 551 – 561.
Bymaster, F.P., Calligaro, D.O., Falcone, J.F., Marsh, R.D., Moore, N.A.,
Tye, N.S., Seeman, P., Wong, D.T., 1996. Radioreceptor binding profile
of the atypical antipsychotic olanzapine. Neuropsychopharmacology
14, 87 – 96.
Guy, E., 1976. Clinical global impressions. Early Clinical Drug Evaluation
(ECDEU) Assessment Manual for Psychopharmacology. National
Institute of Mental Health, Rockville, pp. 218 – 222.
Hegarty, J.D., Baldessarini, R.J., Tohen, M., Waternaux, C., Oepen, G.,
1994. One hundred years of schizophrenia: a meta-analysis of the outcome literature. Am. J. Psychiatry 151, 1409 – 1416.
Honigfeld, G., Klett, J., 1965. The nurses observation scale for in-patient
evaluation. A new scale for measuring improvement in chronic schizophrenia. J. Clin. Psychol. 21, 65 – 71.
Kane, J.M., 1996. Schizophrenia. New Engl. J. Med. 334, 34 – 41.
Keefe, R.S., Silva, S.G., Perkins, D., Lieberman, J.A., 1999. The effects of
atypical antipsychotic drugs on neurocognitive impairment in schizophrenia: a review and meta-analysis. Schizophr. Bull. 25, 201 – 222.
Levinson, D.F., Simpson, G.M., Singh, H., Yadalam, K., Jain, A., Stephanos, M.J., Silver, P., 1990. Fluphenazine dose, clinical response, and
extrapyramidal symptoms during acute treatment. Arch. Gen. Psychiatry 47, 761 – 768.
J. Bobes et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 27 (2003) 473–481
Lingjaerde, O., Ahlfors, U.G., Bech, P., Dencker, S.J., Elgen, K., 1987. The
UKU side effect rating scale: a new comprehensive rating scale for
psychotropic drugs and a cross-sectional study of side effects in neuroleptic-treated patients. Acta Psychiatr. Scand. 334, 1 – 100 (Suppl.).
Linszen, D.H., Dingemans, P.M., 2002. Is early intervention effective?
Schizophr. Res. 53 (3), 14 (Suppl.).
Linszen, D., Lenior, M., De Hann, L., Dingemans, P., Gersons, B., 1998.
Early intervention, untreated psychosis and the course of early schizophrenia. Br. J. Psychiatry 172, 84 – 89 (Suppl.).
Loebel, A.D., Lieberman, J.A., Alvir, J.M., Mayerhoff, D.I., Geisler, S.H.,
Szymanski, S.R., 1992. Duration of psychosis and outcome in firstepisode schizophrenia. Am. J. Psychiatry 149, 1183 – 1188.
Mata, I., Arranz, M.J., Mancama, D.T., Huezo-Diaz, P., Osborne, S., Munro, J., Kerwin, R.W., 2001. Olanzapine response influenced by serotonin genes (abstract). Schizophr. Res. 49, 74 – 75 (Suppl.).
McEvoy, J., 1991. Dosing issues: target symptoms and dose response.
Psychiatr. Ann. 25, 297 – 300.
McGlashan, T.H., 1998. A selective review of recent North American longterm follow-up studies of schizophrenia. Schizophr. Bull. 14, 515 – 542.
McGlashan, T.H., Miller, T.J., Woods, S.W., 2002. Psychosis treatment
prior to psychosis onset: ethical issues. Schizophr. Res. 53 (3), 15
(Suppl.).
Moore, N.A., Calligaro, D.O., Wong, D.T., Bymaster, F.P., Tye, N.C., 1993.
The pharmacology of olanzapine and other new antipsychotic drugs.
Curr. Opin. Invest. Drugs 2, 281 – 293.
Sanger, T.M., Lieberman, J.A., Tohen, M., Grundy, S., Beasley, C., Tollefson, G., 1999. Olanzapine versus haloperidol treatment in first-episode psychosis. Am. J. Psychiatry 156, 79 – 87.
Szymanski, S., Lieberman, J.A., Alvir, J.M., Mayerhoff, D., Loebel, A.,
Geisler, S., Chakos, M., Koreen, A., Jody, D., Kane, J., Woerner, M.,
Cooper, T., 1995. Gender differences in onset of illness, treatment response, course, and biological indexes in first-episode schizophrenic
patients. Am. J. Psychiatry 152, 698 – 703.
481
Tohen, M., Stoll, A.L., Strakowski, S.M., Faedda, G.L., Mayer, P.V., Goodwin, D.C., Kolbrener, M.L., Madigan, A.M., 1992. The McLean firstepisode psychosis project: six-month recovery and recurrence outcome.
Schizophr. Bull. 18, 273 – 282.
Tollefson, G.D., Sanger, T.M., 1997. Negative symptoms: a path analytic
approach to a double-blind, placebo- and haloperidol-controlled clinical
trial with olanzapine. Am. J. Psychiatry 154, 466 – 474.
Tollefson, G.D., Beasley, C.M., Tamura, R.N., Tran, P.V., Potvin, J.H.,
1997a. Blind, controlled, long-term study of the comparative incidence
of treatment-emergent tardive dyskinesia with olanzapine and haloperidol. Am. J. Psychiatry 154, 1248 – 1254.
Tollefson, G.D., Beasley, C.M., Tran, P.V., Street, J.S., Krueger, J.A., Tamura, R.N., Graffeo, K.A., Thieme, M.E., 1997b. Olanzapine versus
haloperidol in the treatment of schizophrenia and schizoaffective and
schizophreniform disorders: results of an international collaborative
trial. Am. J. Psychiatry 154, 457 – 465.
Torrey, E.F. (Ed.), 1995. Surviving Schizophrenia: A Manual for Families,
Consumers, and Providers. Harper Collins, New York, p. 409.
Tran, P., Dellva, M.A., Tollefson, G., Beasley, C.M., Potvin, J.H., Kiesler,
G.M., 1997. Extrapyramidal symptoms and tolerability of olanzapine
versus haloperidol in the acute treatment of schizophrenia. J. Clin.
Psychiatry 58, 205 – 211.
Weiden, P.J., Shaw, E., Mann, J.J., 1986. Causes of neuroleptic noncompliance. Psychiatr. Ann. 16, 571 – 575.
Woerner, M.G., Mannuzza, S., Kane, J.M., 1988. Anchoring the BPRS: an
aid to improved reliability. Psychopharmacol. Bull. 24, 112 – 117.
Woods, S.W., McGlashan, T.H., 2002. Adverse effects of olanzapine in
adolescents and adults. Schizophr. Res. 53 (3), 170 (Suppl.).
World Health Organization (WHO), 1994. International Classification of
Diseases: Mental and Behavioural Disorders, 10th ed. World Health
Organization, Geneva, Switzerland.
Wyatt, R.J., 1995. Early intervention for schizophrenia: can the course of
the illness be altered? Biol. Psychiatry 38, 1 – 3.