You are on page 1of 8

Special Section on CATIE Baseline Data

Baseline Use of Concomitant

Psychotropic Medications to Treat
Schizophrenia in the CATIE Trial
Miranda H. Chakos, M.D.
Ira D. Glick, M.D.
Alexander L. Miller, M.D.
Mark B. Hamner, M.D.
Del D. Miller, M.D.
Jayendra K. Patel, M.D.
Andre Tapp, M.D.
Richard S. E. Keefe, Ph.D.
Robert A. Rosenheck, M.D.

Objective: This study examined the prevalence and correlates of con- he introduction of first-gener-
comitant psychotropic medications and use of anticholinergic drugs to ation antipsychotic drugs in
treat schizophrenia. Methods: Concomitant medication use was studied at the 1950s for the treatment of
baseline for participants in the Clinical Antipsychotic Trials of Interven- schizophrenia was an enormous ad-
tion Effectiveness (CATIE) trial. Results: Of the 1,380 patients with base- vance over the treatments at that
line medication data, 82 percent were taking psychotropic medications. time. Unfortunately for most pa-
Of this group, 6 percent were taking two antipsychotics (one first gener- tients, response was at best partial. As
ation and one second generation); 38 percent, antidepressants; 22 per- such, clinicians struggled to find
cent, anxiolytics; 4 percent, lithium, and 15 percent, other mood stabiliz- strategies to improve outcome by
ers. The strongest predictors of taking several medications were having augmenting the treatment with so-
anxiety or depression, being female, and taking second-generation an- called ancillary or concomitant psy-
tipsychotics. Conversely, African Americans and those with better neu- chotropic medications. These strate-
rocognitive functioning were less likely to be taking several concomitant gies included combining antipsy-
psychotropic medications. In some cases symptoms that were likely tar- chotics or combining an antipsychotic
gets of polypharmacy, such as depression, remained prominent, suggest- with an antidepressant, a mood stabi-
ing only partial response. Conclusions: Concomitant use of psychotropic lizer, anxiolytic agents, or sedatives.
medications to treat people with schizophrenia is common. Empirical This strategy has not changed even
data demonstrating the effectiveness of many of these agents for this pop- with the introduction of second-gen-
ulation are lacking. (Psychiatric Services 57:1094–1101, 2006) eration agents. In fact, polypharmacy
has become the rule rather than the
exception in the United States and
Dr. Chakos is affiliated with the Department of Psychiatry, State University of New York elsewhere (1–10). This change has
Downstate Medical Center, CB 1203, Brooklyn, NY 11572 (e-mail: miranda.chakos@ evolved despite an almost total lack of Dr. Glick is with the Department of Psychiatry, Stanford University School controlled scientific data supporting
of Medicine, Palo Alto, California. Dr. Alexander L. Miller is with the Department of Psy- the practice (10,11). The evidence
chiatry, University of Texas Health Science Center at San Antonio. Dr. Hamner is with the has been mostly anecdotal.
Department of Psychiatry, Medical University of South Carolina, Charleston. Dr. Del D. In addition, there are several risks
Miller is with the Department of Psychiatry, University of Iowa Carver College of Medi- working against the benefits of this
cine, Iowa City. Dr. Patel is with the Department of Psychiatry, University of Massachu-
practice. Concomitant use of psy-
setts Medical School, Worcester, Massachusetts. Dr. Tapp is with the Department of Psy-
chotropic medications may worsen
chiatry, American Lake Veterans Administration Medical Center, Tacoma, Washington. Dr.
Keefe is with the Department of Psychiatry, Duke University Medical Center, Durham, the patient’s quality of life, induce
North Carolina. Dr. Rosenheck is with the Department of Psychiatry, Yale Medical School, side effects, and create drug interac-
West Haven, Connecticut. This article is part of a special section of reports based on data tions that decrease medication effica-
from the CATIE (Clinical Antipsychotic Trials in Intervention Effectiveness) schizophrenia cy. In addition, and important for
trial. Robert A. Rosenheck, M.D., served as guest editor of the special section. many settings and patients, treatment
1094 PSYCHIATRIC SERVICES ♦ ♦ August 2006 Vol. 57 No. 8
costs can increase dramatically. sign of the CATIE study and entry or second-generation antipsychotic
Moreover, the more complex the criteria have been presented else- among those taking any antipsy-
medication regimen is, the lower the where (12,13). This study presents chotics, and a fifth measure repre-
treatment compliance. baseline data collected from 2001 to sented taking both a first- and a sec-
In this study we describe the fre- 2003 from participants with schizo- ond-generation medication. No pa-
quency and the demographic and phrenia before random assignment tients were taking two second-gener-
clinical correlates of concomitant use and initiation of study treatments. ation antipsychotics.
of psychotropic medications among Five additional measures repre-
participants in the Clinical Antipsy- Measures sented each of the other classes of
chotic Trials of Intervention Effec- As part of the baseline assessment, psychotropic medication among pa-
tiveness (CATIE) before they were patients were asked to report all med- tients taking medication. One addi-
randomly assigned to a treatment ications they were currently pre- tional dichotomous measure repre-
group. This study differentiates itself scribed. Study personnel also re- sented use of an anticholinergic med-
from prior studies on use of polyphar- viewed patients’ medical records and ication among patients on any an-
macy to treat schizophrenia in that spoke to treating clinicians (when tipsychotic. A final, continuous meas-
patients in the CATIE study had available) to determine medications ure represented the total number of
much greater exposure to second- that were prescribed. The dependent different agents taken by each patient
generation antipsychotics, whereas variables in this study were a series of (excluding anticholinergics) among
previous studies were conducted with those taking any such medication.
patients who were primarily treated The diagnosis of schizophrenia
with first-generation antipsychotics. was confirmed by the Structured
Thus this study is a more up-to-date Clinical Interview for DSM-IV Axis
evaluation of the impact of second- Use of I Disorders (SCID) (13) for all pa-
generation antipsychotics on the tients, and SCID data were also
practice of polypharmacy. concomitant available on secondary axis I diag-
noses. Symptoms of schizophrenia
Methods psychotropic medications were assessed with the rater-admin-
The CATIE study was designed to istered PANSS (14). Overall quality
compare the effectiveness of current- may worsen quality of of life and functioning were assessed
ly available second-generation an- with the Heinrichs-Carpenter Qual-
tipsychotics with a representative life, induce side effects, and ity of Life Scale (QOLS) (15) and
first-generation antipsychotic, per- the global quality-of-life item meas-
phenazine, through a randomized create drug interactions ured with the Lehman Quality of
clinical trial that involved a large sam- Life Interview (16).
ple of patients treated for schizophre- that decrease Medication side effects were as-
nia at multiple sites, including both sessed with the Barnes scale for
academic and community providers. medication akathisia (17), the Abnormal Involun-
Patients who met criteria for schizoaf- tary Movement Scale (AIMS) for tar-
fective disorder, depressive type (with efficacy. dive dyskinesia (18), and the Simp-
predominant schizophrenic features), son-Angus scale for extrapyramidal
could also participate if approved by side effects (SAEPS) (19). Depres-
the project’s medical officer. Partici- sion was measured with the Calgary
pants provided written informed con- Depression Rating Scale (20). For
sent to participate in protocols ap- dichotomous measures representing this article substance use was defined
proved by local institutional review various classes of prescribed medica- by SCID diagnosis.
boards. Of the 1,460 patients, 77 per- tions. The first measure was a general Neurocognitive functioning was
cent were male, 60 percent were category indicating use of at least one measured by separate test scores, de-
white, 35 percent were African Amer- of the following seven classes of psy- scribed in a previous publication (21),
ican, and 5 percent were of other chotropic medication: first-genera- which were converted to z scores and
racial backgrounds. The mean±SD tion antipsychotics, second-genera- combined to construct five separate
age was 40±11. Approximately 25 tion antipsychotics, antidepressants, scales (processing speed, verbal
percent of patients had an exacerba- anxiolytic agents, sedative-hypnotics, memory, vigilance, reasoning, and
tion of illness in the three months be- mood stabilizers (valproic acid or car- working memory) that were them-
fore study entry. Patients had been bamazepine), and lithium. The group selves averaged to form an overall
taking antipsychotic medications for taking at least one psychotropic med- neurocognitive functioning scale.
approximately 14 years. The average ication was then divided into partici- Higher overall neurocognitive scores
Positive and Negative Syndrome pants taking no antipsychotic as con- indicated better functioning. Neu-
Scale (PANSS) score at baseline was trasted with those taking one or more rocognitive data were missing for 8
75, and the average Clinical Global antipsychotics. The third and fourth percent of the sample (N=117), and
Index was 4. Details of the study de- measures represented use of a first- the missing data were imputed by
PSYCHIATRIC SERVICES ♦ ♦ August 2006 Vol. 57 No. 8 1095
Table 1 Eighty-two percent (N=1,138) of pa-
Frequencies of psychotropic use at baseline among participants in the Clinical tients were taking a psychotropic med-
Antipsychotic Trials of Intervention Effectiveness ication from one of the seven classes.
Table 1 summarizes the frequencies of
Percent of Percent of group on psychotropic use by category.
total group psychotropic med-
Medication variable N (N=1,380) ication (N=1,138)
No psychotropic or no
No psychotropic 242 18 na antipsychotic medication
Psychotropic but no antipsychotic 115 8 10 Twenty-six percent of patients
One second-generation antipsychotic 794 58 70 (N=357) entering the trial were not
One first-generation antipsychotic 154 11 14 taking any antipsychotic medication.
First- and second-generation
antipsychotics 69 5 6
Eighteen percent of the patients
Antidepressant 432 31 38 (N=242) were taking no psychotropic
Anxiolytica 248 18 22 medication at study entry, and 8 per-
Sedative or hypnoticb 217 16 19 cent (N=115) were taking a psy-
Lithium 47 3 4 chotropic medication but no antipsy-
Mood stabilizer (not lithium) 170 12 15
Anticholinergic and antipsychoticc
chotic medication. Participants who
211 15 19
Number of medication classes were not taking psychotropic medica-
(M±SD) 1.88±.92 tion were more likely to be African
Number of psychotropics (M±SD) 2.03±1.10 American and were less likely to have
SCID diagnoses of depression (Table
Includes clonazepam, lorazepam, hydroxyzine, buspirone, alprazolam, diazepam, oxazepam, clo-
razepate dipotassium, and chlordiazepoxide 2). Participants who were taking psy-
b Includes trazodone, zolpidem, temazepam, chloral hydrate, and flurazepam chotropic but not antipsychotic med-
c Twenty-one percent of patients taking antipsychotic medication were also taking anticholinergic ications had higher AIMS scores and
medication. lower scores on the Lehman Quality
of Life Interview (Table 2).

substituting the mean value for the linear regression models were used Numbers and classes of
whole group. to identify correlates of the total concomitant medications
number of medication classes and The mean±SD number of psy-
Statistical analysis agents. Values with a p value of less chotropic medications per patient for
First, a large group of potential pre- than .01 are shown in the tables. those who were taking at least one
dictors were screened with bivariate psychotropic at baseline was 2.03±
Spearman correlation analysis to Results 1.1. The strongest predictors of taking
identify sociodemographic and clini- In the CATIE study, 1,493 participants multiple psychotropic medications
cal correlates of each measure of were randomly assigned to the schizo- were being anxious or depressed, be-
medication use. Measures that were phrenia component of the study. Data ing female, and being treated with a
significant in bivariate analyses at from one site (33 patients) were ex- second-generation antipsychotic.
p<.01 were then included in a subse- cluded from all analyses because of Conversely, African-American pa-
quent series of stepwise multivariate concerns about their integrity, leaving tients and those with better neu-
logistic regression analyses that iden- 1,460 patients at the baseline cut. rocognitive functioning were less
tified independent correlates of each Medication data were available for 95 likely to be taking several psychotrop-
measure of medication use. Stepwise percent (N=1,380) of these patients. ic medications (Table 3).

Table 2
Significant odds ratios for baseline medication groups in the Clinical Antipsychotic Trials of Intervention Effectiveness

Taking any psychotropic Taking antipsychotic medication

medication (for all patients) (for patients taking psychotropics)
(N=1,380) (N=1,138)

Variable OR 95% CI OR 95% CI

African American .42 .33–.54

Structured Clinical Interview for DSM-IV
Axis I Disorders depression 2.18 1.30–3.65
Lehman Quality of Life Interviewa 1.24 1.08–1.42
Abnormal Involuntary Movement Scalea .54 .38–.77
a Odds of taking an antipsychotic per unit increase in global score

1096 PSYCHIATRIC SERVICES ♦ ♦ August 2006 Vol. 57 No. 8

Antipsychotic polypharmacy of first- and second-generation an- Table 3
At baseline, 69 participants (5 per- tipsychotic medications. After includ- Prediction (by stepwise linear
cent) were taking two antipsychotic ing age, gender, race, Barnes aka- regression) of total numbers of baseline
agents, one that was first generation thisia scale ratings, SAEPS, and concomitant psychotropic medications
and the other, second generation. No PANSS negative and positive symp- for patients taking at least one in
patients were taking more than two tom scores in the model, the use of the Clinical Antipsychotic Trials of
antipsychotics. The neurocognitive anticholinergic medications by those Intervention Effectiveness
composite score showed a negative receiving an antipsychotic medication
association with the use of antipsy- was associated with poorer neurocog- Regression
chotic polypharmacy (odds ratio nitive functioning (OR=.55, CI= Variable coefficienta
[OR]=.55; 95 percent confidence in- .37–.82) (Table 4). SCID anxietyb .561
terval [CI]=.37–.82) (Table 4), indi- SCID depression .396
cating that patients with lower neu- Mood stabilizers Calgary Depression
rocognitive scores were more likely to and antidepressants Rating Scale .269
be taking two agents. Four percent (N=47) of patients who Female .209
African American –.269
were taking psychotropic medications Neurocognitive score –.158
Anticholinergics received treatment with lithium, and Second-generation
Of the 1,017 patients taking antipsy- 15 percent (N=170) were receiving antipsychotic .433
chotic medications, 211 patients (21 other mood stabilizers (valproic acid
a Per unit increase in the parameter
percent) were receiving adjunctive or carbamazepine). Female gender b Structured Clinical Interview for DSM-IV
anticholinergic treatment. Anti- predicted adjunctive treatment with Axis I Disorders
cholinergic medications were being lithium. Higher PANSS positive
taken by 93 of the 794 patients (12 symptom scores predicted adjunctive
percent) who were taking a second- treatment with other mood stabilizers
generation antipsychotic, 81 of the (Table 4). SCID diagnoses of major depressive
154 patients (53 percent) who were Thirty-eight percent of patients disorder, obsessive-compulsive disor-
taking a first-generation antipsychot- who were taking psychotropic med- der, and any other anxiety disorder in-
ic, and 37 of the 67 patients (55 per- ications (N=432) were receiving anti- dependently predicted adjunctive
cent) who were taking a combination depressant treatment. Additional treatment with antidepressants. Even

Table 4
Odds ratios and confidence intervals for specific baseline concomitant psychotic medication groups for patients taking at
least one psychotropic in the Clinical Antipsychotic Trials of Intervention Effectiveness

First- and
second-gen- Anticholin-
eration anti- Antide- Other mood ergic with
psychotics pressant Anxiolytic Sedative Lithium stabilizer antipsychotic

Variable OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI

composite score .55 .37–.82 .56 .43–.74
Calgary depressiona 2.14 1.67–2.76 1.49 1.7–1.91
Depression 3.15 2.05–4.83
Obsessive-compulsive 3.22 1.74–5.95
Anxietyc 2.05 1.41–2.98 2.00 1.34–2.95
African American .46 .32–.66
Quality of lifed .85 .77–.93
Female 2.70 1.48–4.91
Age 1.03 1.01–1.04
No high school diploma 1.85 1.32–2.59
Total 1.02 1.00–1.03
Positive 1.05 1.02–1.08
AIMSf 1.67 1.17–2.38
a Calgary Depression Rating Scale
b Structured Clinical Interview for DSM Axis I Disorders
c Includes any anxiety disorder other than obsessive-compulsive disorder
d Heinrichs-Carpenter Quality of Life Scale
e Positive and Negative Syndrome Scale
f Abnormal Involuntary Movement Scale

PSYCHIATRIC SERVICES ♦ ♦ August 2006 Vol. 57 No. 8 1097

after SCID diagnoses for generalized study is relatively low compared with ence these symptoms. A prescriber
anxiety disorder, major depressive the rates in other studies. This may bias is another possible explanation
disorder, and obsessive-compulsive reflect the greater use of second-gen- for this difference. Less frequent use
disorder were included in the model, eration antipsychotics in the treat- of anxiolytics among African-Ameri-
higher scores on the Calgary Depres- ment of patients with schizophrenia. can patients is consistent with find-
sion Rating Scale predicted adjunc- However, patients treated with multi- ings of other investigators (9,34) and
tive antidepressant use. Gender, race, ple antipsychotic medications also may reflect prescriber bias, econom-
akathisia scale, SAEPS, and PANSS may have been less likely to be re- ic factors, or less engagement in the
negative and positive symptom scores ferred to the study. mental health system (34,35).
were not predictors of antidepressant Our findings on the prevalence of Both the number of concomitant
use. use of psychotropic medications, in- psychotropic medications and the use
cluding 31 percent taking antide- of antidepressants and anxiolytics had
Anxiolytics and sedatives pressants, 18 percent taking anxi- no association with positive or nega-
Twenty-two percent (N=248) of pa- olytics, 16 percent taking sedative- tive symptoms of schizophrenia but
tients taking psychotropic medica- hypnotics, and 15 percent taking an- were associated with symptoms of
tions received anxiolytics. Even after ticholinergic medication in combi- anxiety and depression. The lack of an
SCID diagnoses for generalized anxi- nation with an antipsychotic, are association of antidepressant or anxi-
ety, major depressive, and obsessive- similar to the prevalence rates of use olytic use with positive symptoms or
compulsive disorders were included of psychotropic medications among potential confounders of depression,
in the model, higher Calgary depres- patients with schizophrenia report- such as negative symptoms and ex-
sion scores predicted adjunctive anxi- ed by other investigators (24–33). In trapyramidal side effects, suggests
olytic use. African-American patients contrast, the prevalence of use of that symptoms of depression and anx-
were less likely to receive anxiolytic mood stabilizers (3 percent taking iety are components of the illness that
medications, and use of anxiolytic lithium, 12 percent taking other are independent of positive or nega-
medication was associated with worse mood stabilizers) was relatively low tive symptoms or side effects such as
functioning on the Heinrichs-Car- compared with these prior studies. extrapyramidal effects. The increase
penter QOL scale (Table 4). Because patients who entered this in prescribing concomitant psycho-
Nineteen percent (N=217) of pa- study had much greater exposure to tropic medications to patients with
tients who were taking psychotropic second-generation medications than symptoms of anxiety and depression
medications received sedative-hyp- in prior studies of antipsychotic suggests that physicians have become
notics. Older patients and patients polypharmacy, the finding suggests very responsive to treating these
with less than a high school education that clinicians are less likely to pre- symptoms among patients with schiz-
were more likely to receive sedative scribe mood stabilizers with second- ophrenia. This may in part be due to
hypnotics. Patients with a comorbid generation antipsychotics. patients’ complaints about these an-
diagnosis of anxiety disorder and cillary symptoms and patients’ in-
those with higher positive symptom Correlates of concomitant use creased expectations of treatments to
scores were more likely to receive of psychotropic medications relieve them but may also reflect the
sedative hypnotics (Table 4). A broad array of factors was associat- physician’s desire to ameliorate the
ed with concomitant psychotropic feeling of demoralization and anxiety
Discussion treatment of patients with schizo- associated with having to deal with
Prevalence of use of phrenia in this trial. With respect to this chronic illness.
psychotropic medications demographic predictors of concomi- The association of more severe de-
Antipsychotic polypharmacy has in- tant use of psychotropic medications, pressive symptoms with concomitant
creased over the years and is pre- African-American participants re- antidepressant use, in the form of
scribed to 25 percent of outpatients ceived fewer concomitant psy- higher Calgary depression scores,
and more than 50 percent of inpa- chotropic medications and anxiolyt- even after the analysis controlled for
tients with schizophrenia in the Unit- ics, whereas women received more comorbid axis I diagnoses of depres-
ed States and to 13 to 90 percent of concomitant psychotropic medica- sion and anxiety disorders, suggests
patients internationally (1–7,11,22, tions, especially antidepressants and that, at least with respect to depres-
23). Whereas prior examinations of lithium. Female patients may have sive symptoms, concomitant use of
antipsychotic polypharmacy were been taking more concomitant psy- antidepressants as used in this pa-
conducted for patients with schizo- chotropic medications because tient population was not fully effec-
phrenia who had much greater expo- women generally are better able to tive. Because we do not have data on
sure to first-generation antipsychotic express their symptoms of anxiety duration or dose of treatment with
medications, the study reported here and depression to their providers, antidepressants, this lack of efficacy
was conducted for patients who had a which in this case may have resulted may be due to suboptimum dosing or
much greater exposure to second- in an increase in prescriptions of con- inadequate duration of treatment
generation antipsychotic medica- comitant psychotropic medications. with antidepressants. Alternatively,
tions. The 5 percent prevalence rate Alternatively, the female patients the addition of antidepressants to
of antipsychotic polypharmacy in our perhaps were more likely to experi- second-generation antipsychotics may
1098 PSYCHIATRIC SERVICES ♦ ♦ August 2006 Vol. 57 No. 8
not be effective in treating depres- possible that patients with more se- suggests that cognitive functioning is
sion among patients with schizo- vere cognitive problems were more a core part of the pathology of schizo-
phrenia. This issue can be addressed likely to be treated by antipsychotic phrenia (56,57), and those cognitive
only in a randomized controlled clin- polypharmacy, perhaps because of problems may persist even when psy-
ical trial of adjunctive antidepressant more partial adherence problems. chotic symptoms are in remission
use in depressed patients with schiz- Although there is clear evidence (58). These cognitive deficits may
ophrenia who are taking second-gen- that anticholinergic medications are cause more social and vocational dis-
eration antipsychotics. effective in the treatment and pre- ability than positive or negative symp-
The efficacy of antidepressant vention of acute extrapyramidal side toms and may play a role in relapse
treatment among patients with effects (45,46), the need for long- and rehospitalization (57–60).
schizophrenia has been an issue of term anticholinergic use among pa- Many of the concomitant prescrib-
considerable discussion. It is con- tients receiving antipsychotics is less ing patterns in treating patients with
founded by the fact that depressive clear (47,48). Therefore, assessing the schizophrenia with persistent psy-
symptoms among patients with side effect liability of such an inter- chosis, anxiety, depression, and cogni-
schizophrenia may reflect demoral- vention is critical. Our finding of tive deficits are not robustly support-
ization or be subsyndromal. On the greater neurocognitive impairment ed by evidence in the literature.
basis of a stringent definition of Therefore, such prescribing is likely
DSM-IV diagnosis of depression, 60 driven by inadequate response to
percent of patients with schizophre- monotherapy or unmet treatment
nia have an episode of major depres- needs among these patients. The data
sion at some point in their illness presented in this study suggest that
course (36). Depression has been as-
The better treatments for the depression,
sociated with an increased risk of re- anxiety, and cognitive deficits of pa-
lapse, poor social adjustment, and
evidence tients with schizophrenia need to be
suicide (10,36–38). The evidence developed. In addition, the most fre-
with respect to the benefits of
with respect quent concomitant prescribing prac-
adding antidepressants to antipsy- tices should be tested empirically in
chotic medication for treatment of
to the benefits of randomized controlled clinical trials.
depressive symptoms among stable
outpatients with schizophrenia has
adding antidepressants to Methodological limitations
been mixed (10,39–44). The primary limitations of this study
Unlike concomitant use of antide-
antipsychotic medication for are that the data are cross-sectional
pressants or anxiolytics with antipsy- and correlational. Patient assignment
chotic medications, the use of mood
treatment of depressive to drug treatment was naturalistic,
stabilizers other than lithium and the and correlations between various con-
use of sedatives were associated with
symptoms in comitant psychotropic medications
more severe illness as indicated by and clinical status may be due to con-
higher PANSS scores. The association
schizophrenia founding variables. Another major
of use of these mood stabilizers with limitation is that we did not have in-
higher positive symptom scores sug-
is mixed. formation about duration of treatment
gests that these adjunctive medica- or dosage of baseline medications and
tions are being used to treat patients therefore could not draw meaningful
with more positive symptoms. There conclusions about the efficacy of such
are no randomized controlled studies interventions. This could only be done
of the persistent benefits of adjunc- among patients with schizophrenia in randomized controlled clinical tri-
tive divalproex compared with an- who were treated with anticholiner- als of these interventions. In addition,
tipsychotic monotherapy after treat- gics is consistent with prior reports patients willing to enter a clinical trial
ment of an acute exacerbation of that anticholinergic agents impair such as CATIE and clinicians refer-
schizophrenic illness. several domains of cognitive func- ring such patients may not be repre-
A primary finding with respect to tioning, including attention, declara- sentative of all treated individuals with
antipsychotic polypharmacy was that tive memory, verbal memory, and schizophrenia and their clinicians. An
individuals who were treated with spatial working memory (49–55). additional limitation is that for a small
more than one antipsychotic medica- Although we do not know whether percentage of cases, medication histo-
tion had a lower neurocognitive com- antipsychotic polypharmacy is a cause ry may not have been verified by a
posite score. It is possible that the as- or a result of the neurocognitive im- treating clinician or by medical
sociation of antipsychotic polyphar- pairment, the association of poor neu- record. A final limitation of the study
macy with lower neurocognitive rocognitive functioning with both an- is that we did not report concomitant
scores was a treatment side effect, tipsychotic polypharmacy and anti- nonpsychotropic medications, some
perhaps caused by an increase in an- cholinergic medications has impor- of which are commonly used and have
ticholinergic side effects. It is also tant clinical implications. Evidence high atropine equivalence factors.
PSYCHIATRIC SERVICES ♦ ♦ August 2006 Vol. 57 No. 8 1099
Conclusions tipsychotics and adjunctive medications by Neurocognitive assessment in the Clinical
Use of concomitant psychotropic an inner urban community psychiatric serv- Antipsychotic Trials of Intervention Effec-
ice. Australian and New Zealand Journal of tiveness (CATIE) project schizophrenia tri-
medications to treat patients with Psychiatry 33:896–901, 1999 al: development, methodology, and ration-
schizophrenia is prevalent, despite lit- ale. Schizophrenia Bulletin 29:45–55, 2003
6. Rittmannsberger H, Meise U, Schauflinger
tle information in regard to efficacy K, et al: Polypharmacy in psychiatric treat- 22. Ereshefsky L: Pharmacologic and pharma-
and liabilities associated with ment: patterns of psychotropic drug use in cokinetic considerations in choosing an an-
polypharmacologic interventions. Austrian psychiatric clinics. European Psy- tipsychotic. Journal of Clinical Psychiatry
chiatry 14:33–40, 1999 60(suppl 10):20–30, 1999
Concomitant use of psychotropic
medications in the treatment of schiz- 7. Tapp A, Wood AE, Secrest L, et al: Combi- 23. Stahl SM, Grady MM: A critical review of
nation antipsychotic therapy in clinical atypical antipsychotic utilization: compar-
ophrenia is associated with diverse practice. Psychiatric Services 54:55–59, ing monotherapy with polypharmacy and
factors, including age, race, gender, 2003 augmentation. Current Medicinal Chem-
persistent depression, concurrent 8. Hardy P, Payan C, Bisserbe JC, et al: Anxi-
istry 11:313–327, 2004
anxiety and depressive disorder diag- olytic and hypnotic use in 376 psychiatric 24. Ascher-Svanum H, Kennedy JS, Lee D, et
noses, severity of illness, functional inpatients. European Psychiatry 14:25–33, al: The rate, pattern, and cost of use of an-
1999 tiparkinsonian agents among patients treat-
impairment, and neurocognitive im-
ed for schizophrenia in a managed care set-
pairment. 9. Wolkowitz OM, Pickar D: Benzodiazepines
ting. American Journal of Managed Care
in the treatment of schizophrenia: a review
10:20–24, 2004
and reappraisal. American Journal of Psy-
Acknowledgments chiatry 148:714–726, 1991 25. Buchanan RW, Kreyenbuhl J, Zito JM, et al:
This article is based on results from the 10. Levinson DF, Umapathy C, Mustaq M: Relationship of the use of adjunctive phar-
Clinical Antipsychotic Trials of Interven- Treatment of schizoaffective disorder and macological agents to symptoms and level
tion Effectiveness project, supported by schizophrenia with mood symptoms. Amer- of function in schizophrenia. American
contract NO1-MH-90001 from the Na- ican Journal of Psychiatry 156:1138–1148, Journal of Psychiatry 159:1035–1043, 2002
tional Institute of Mental Health 1999
26. Magliano L, Fiorillo A, Guarneri M, et al:
(NIMH). The project was carried out by 11. Stahl SM: Antipsychotic polypharmacy, Prescription of psychotropic drugs to pa-
principal investigators from the Universi- part I: therapeutic option or dirty little se- tients with schizophrenia: an Italian nation-
ty of North Carolina, Duke University, the cret? Journal of Clinical Psychiatry 60:425– al survey. European Journal of Clinical
University of Southern California, the 426, 1999 Pharmacology 60:513–522, 2004
University of Rochester, and Yale Univer- 27. Galletly CA, Tsourtos G: Antipsychotic
12. Lieberman JA, Stroup TS, McEvoy JP, et al:
sity in association with Quintiles, Inc.; the Effectiveness of antipsychotic drugs in pa- drug doses and adjunctive drugs in the out-
program staff of the Division of Interven- tients with chronic schizophrenia. New patient treatment of schizophrenia. Annals
tions and Services Research of NIMH; England Journal of Medicine 353:1209– of Clinical Psychiatry 9(2):77–80, 1997
and investigators from 56 sites in the 1223
28. Citrome L, Jaffe A, Levine J, et al: Use of
United States (CATIE Study Investiga- mood stabilizers among patients with schiz-
13. First MB, Spitzer RL, Gibbon MB, et al:
tors Group). AstraZeneca Pharmaceuti- Structured Clinical Interview for Axes I ophrenia, 1994–2001. Psychiatric Services
cals L.P., Bristol-Myers Squibb Company, and II DSM IV Disorders—Patient Edition 53:1212, 2002
Forest Pharmaceuticals, Inc., Janssen (SCID-I/P). New York, Biometrics Re-
Pharmaceutica Products, L.P., Eli Lilly search Department, New York State Psy- 29. Glick ID, Zaninelli R, Chuanchieh H: Pat-
chiatric Institute, 1996 terns of concomitant psychotropic medica-
and Company, Otsuka Pharmaceutical
tion use during a 2-year study comparing
Co., Ltd., Pfizer, Inc., and Zenith Gold- 14. Kay SR, Fiszbeing, Opler DR: The Positive clozapine and olanzapine for the preven-
line Pharmaceuticals, Inc., provided med- and Negative Syndrome Scale (PANSS) for tion of suicidal behavior. Journal of Clinical
ications for the studies. Details about the schizophrenia. Schizophrenia Bulletin 13: Psychiatry 65:679–685, 2004
CATIE Study Investigators Group are 261–276, 1987
available at 30. Casey DE, Daniel DG, Wassef AA, et al:
15. Heinrichs DW, Hanlon ET, Carpenter WT: Effect of divalproex combined with olanza-
phrenia/locations.html#clinicalsitelocation The Quality of Life Scale: an instrument for pine or risperidone in patients with an
list. rating the schizophrenic deficit syndrome. acute exacerbation of schizophrenia. Neu-
Schizophrenia Bulletin 10:388–398, 1984 ropsychopharmacology 28:182–192, 2003
16. Lehman A: A quality of life interview for 31. Wassef AA, Dott SG, Harris A, et al: Ran-
1. Tempier RP, Pawliuk NH: Conventional, the chronically mentally ill. Evaluation and domized placebo controlled pilot study of
atypical, and combination antipsychotic Program Planning 11:51–62, 1988 divalproex sodium in treatment of acute ex-
prescriptions: a 2-year comparison. Journal acerbation of chronic schizophrenia: clini-
of Clinical Psychiatry 64:673–679, 2003 17. Barnes TR: A rating scale for drug-induced
cal and economic implications. Journal of
akathisia. British Journal of Psychiatry
Clinical Psychopharmacology 20:357–361,
2. Botts S, Hines H, Littrell R: Antipsychotic 154:672–676, 1989
polypharmacy in the ambulatory care set-
ting, 1993–2000. Psychiatric Services 54: 18. Guy W: Abnormal involuntary movements,
in ECDEU Assessment Manual for Psy- 32. Tiihonen J, Hallikainen T, Ryynanen OP, et
1086, 2003 al: Lamotrigine in treatment-resistant
chopharmacology. Edited by Guy W.
3. Fourrier A, Gasquet I, Allicar MP, et al: DHEW ADM-76-338. Rockville, Md, Na- schizophrenia: a randomized placebo-con-
Patterns of neuroleptic drug prescription: a tional Institute of Mental Health, 1976 trolled crossover trial. Biological Psychiatry
national cross-sectional survey of a random 54:1241–1248, 2003
sample of French psychiatrists. Journal of 19. Simpson GM, Angus JWS: A rating scale
for extrapyramidal side effects. Acta Psy- 33. Zisook S, McAdams LA, Kuck J, et al: De-
Clinical Pharmacology 49:80–86, 1999
chiatrica Scandinavica 212:11–19, 1970 pressive symptoms in schizophrenia. Amer-
4. Ito C, Kubota Y, Sato M: A prospective sur- ican Journal of Psychiatry 156:1736–1743,
vey on drug choice for prescriptions for ad- 20. Addington D, Addington J, Maticka-Tyn- 1999
mitted patients with schizophrenia. Psychi- dale E: A depression rating scale for schiz-
atry and Clinical Neurosciences 53(suppl): ophrenics. Schizophrenia Research 3:247– 34. Glazer WM, Morgenster H, Doucette J:
35–40, 1999 251, 1996 Race and tardive dyskinesia among outpa-
tients at a CMHC. Hospital and Communi-
5. Keks NA, Altson K, Hope J, et al: Use of an- 21. Keefe RS, Mohs RC, Bilder RM, et al: ty Psychiatry 45:38–42, 1994

1100 PSYCHIATRIC SERVICES ♦ ♦ August 2006 Vol. 57 No. 8

35. Kreyenbuhl J, Zito JM, Buchanan RW, et al: phrenia. British Journal of Psychiatry 139: and neuropsychological test performance
Racial disparity in the pharmacological 97–101, 1981 in schizophrenia. Psychiatry Research
management of schizophrenia. Schizophre- 37:297–308, 1991
nia Bulletin 29:183–193, 2003 45. Bezchhilbnyk-Butler KZ, Remington GJ:
Antiparkinsonian drugs in the treatment of 53. Tracy JI, Monaco C, Giovannetti T, et al:
36. Martin RL, Cloninger CR, Guze SB, et al: neuroleptic-induced extrapyramidal symp- Anticholinergicity and cognitive processing
Frequency and differential diagnosis of de- toms. Canadian Journal of Psychiatry in chronic schizophrenia. Biological Psy-
pressive syndromes in schizophrenia. Jour- 39:74–84, 1994 chology 56:1–22, 2001
nal of Clinical Psychiatry 46:9–13, 1985
46. Winslow RS, Stiller V, Coons DJ, et al: Pre- 54. McGurk SR, Green MF, Wirshing WC, et
37. Herz MI, Melville C: Relapse in schizo- vention of acute dystonic reactions in pa- al: Antipsychotic and anticholinergic effects
phrenia. American Journal of Psychiatry tients beginning high-potency neuroleptics. on two types of spatial memory in schizo-
137:801–805, 1980 American Journal of Psychiatry 143:706– phrenia. Schizophrenia Research 68:225–
710, 1986 233, 2004
38. Roy A: Depression, attempted suicide, and
suicide in patients with chronic schizophre- 47. Rifkin A, Quitkin F, Kane J, et al: Are pro- 55. Minzenberg MJ, Poole JH, Benton C, et al:
nia. Psychiatric Clinics of North America phylactic antiparkinson drugs necessary? A Association of anticholinergic load with im-
9:193–206, 1986 controlled study of procyclidine withdraw- pairment of complex attention and memory
39. Siris SG, Bermanzohn PC, Mason SE, et al; al. Archives of General Psychiatry 35: in schizophrenia. American Journal of Psy-
Maintenance imipramine therapy for sec- 483–489, 1978 chiatry 161:116–124, 2004
ondary depression in schizophrenia. 48. Caradoc-Davies G, Menkes DB, Clarkson 56. Goldberg TE, Ragland JD, Torrey EF, et al:
Archives of General Psychiatry 51:109–115, HO, et al: A study of the need for anti- Neuropsychological assessment of monozy-
1994 cholinergic medication in patients treated gotic twins discordant for schizophrenia.
40. Siris SG, Morgan V, Fagerstrom R, et al: with long-term antipsychotics. Australian Archives of General Psychiatry 47:1066–
Adjunctive imipramine in the treatment of and New Zealand Journal of Psychiatry 1072, 1990
postpsychotic depression. Archives of Gen- 20:225–232, 1986
57. Gold JM, Harvey PD: Cognitive deficits in
eral Psychiatry 44:533–539, 1987 schizophrenia. Psychiatric Clinics of North
49. Gelenberg AJ, Van Putten T, Lavori PW, et
41. Hogarty GE, McEvoy JP, Ulrich RF, et al: al: Anticholinergic effects on memory: ben- America 16:295–312, 1993
Pharmacotherapy of impaired affect in re- ztropine versus amantadine. Journal of
Clinical Psychopharmacology 9:180–185, 58. Green MF: What are the functional conse-
covering schizophrenic patients. Archives
1989 quences of neurocognitive deficits in schiz-
of General Psychiatry 52:29–41, 1995
ophrenia? American Journal of Psychiatry
42. Prusoff BA, Williams DH, Weissman MM, 50. Tune LE, Strauss ME, Lew MF, et al: 153:321–330, 1996
et al: Treatment of secondary depression in Serum levels of anticholinergic drugs and
impaired recent memory in chronic schizo- 59. Harvey PD, Howanitz E, Parrella M, et al:
schizophrenia. Archives of General Psychi-
phrenic patients. American Journal of Psy- Symptoms, cognitive functioning, and
atry 36:569–575, 1979
chiatry 139:1460–1462 1982 adaptive skills in geriatric patients with life-
43. Becker RE: Implications of the efficacy of long schizophrenia: a comparison across
thiothixene and a chlorpromazine-imipra- 51. Perlick D, Stastny P, Katz I, et al: Memory treatment sites. American Journal of Psy-
mine combination for depression in schizo- deficits and anticholinergic levels in chron- chiatry 155:1080–1086, 1998
phrenia. American Journal of Psychiatry ic schizophrenia. American Journal of Psy-
140:208–211, 1983 chiatry 143:230–232, 1986 60. Bellack AS, Gold JM, Buchanan RW: Cog-
nitive rehabilitation for schizophrenia:
44. Johnson DAW: A double-blind trial of nor- 52. Sweeney JA, Keilp JG, Haas GL, et al: Re- problems, prospects, and strategies. Schiz-
triptyline for depression in chronic schizo- lationships between medication treatments ophrenia Bulletin 25:257–274, 1999

RSS Feeds Now Available for Psychiatric Services

Tables of contents of recent issues and abstracts of recent articles are now avail-
able to Psychiatric Services’ readers via RSS (Really Simple Syndication) feeds.
RSS feeds provide a quick and easy way to review each month’s content, with
quick links to the full text.
Please visit the Psychiatric Services Web site at ps.psychiatry and
click on “RSS” on the lower right-hand corner of the screen. The site offers a
choice of RSS software for free installation, links to tutorials on using RSS feeds,
and a contact for providing feedback on this new online feature of the journal.

PSYCHIATRIC SERVICES ♦ ♦ August 2006 Vol. 57 No. 8 1101