Antipsychotic Treatment of Behavioral and
Psychological Symptoms of Dementia in
Geropsychiatric Inpatients
William S. Edell, Ph.D.
Sandra L. Tunis, Ph.D.

Behavioral/psychological symptoms of dementia (BPSD) affect caregiver burden and
transition from home to hospital or long-term care. The authors examined change in
BPSD for dementia patients (from hospital admission to discharge) who were prescribed haloperidol (n‫ ,)982 ס‬olanzapine (n‫ ,)902ס‬or risperidone (n‫ .)005ס‬Olanzapine was associated with significantly greater overall improvement in BPSD (based
on the Psychogeriatric Dependency Rating Scale total score) than risperidone or haloperidol. Olanzapine was significantly superior on measures of active-, verbal-, and
passive-aggression and delusions/hallucinations to risperidone or haloperidol, and,
on manipulative behavior and noisiness, to risperidone. Results support the effectiveness of olanzapine in improving several BPSD in hospitalized dementia patients. (Am
J Geriatr Psychiatry 2001; 9:289–297)

B

ehavioral disturbances such as agitation, physical
aggression, and noisy vocalizations, and symptoms
such as delusions and hallucinations, are commonly
noted in elderly individuals with dementia. It is estimated that such behavioral and psychological symptoms of dementia (BPSDs) are observed in at least 50%
of those with Alzheimer’s Disease (AD),1 and in up to
70%–80% of those with any type of dementia during the
course of their illness.2–5 The deleterious impact of
BPSD on the individual’s functioning and on caregiver
burden is well documented.4,6–9 Indeed, it is the behavioral disturbances, rather than cognitive decline, that
often precipitates the transition from home care to hospital and to long-term institutional-care settings such as
nursing homes.10–13

First-generation antipsychotic medications, such as
the neuroleptic haloperidol, have traditionally been the
most common pharmacologic intervention for BPSD in
nursing homes and long-stay institutions.14–19 Efficacy
with such agents has varied from 25% to 75%,20 as compared with an average response rate to placebo of about
37%.21 These agents have been temporally associated
with the frequent occurrence of adverse events (e.g.,
orthostatic hypotension, cardiac toxicity, anticholinergic effects, and daytime sedation associated with lowpotency drugs; extrapyramidal symptoms such as parkinsonism, akathisia, and tardive dyskinesia with
high-potency drugs).22
Second-generation, or atypical, antipsychotic
agents, such as olanzapine and risperidone, have been

Received July 6, 2000; revised September 25, 2000; accepted October 30, 2000. From Mental Health Outcomes, Lewisville, Texas, and Eli Lilly
and Company, Lilly Corporate Center, Indianapolis, Indiana. Address correspondence to Dr. Edell, Mental Health Outcomes, 1500 Waters Ridge
Drive, Lewisville, TX 75057. e-mail: William_edell@horc.com
Copyright ᭧ 2001 American Association for Geriatric Psychiatry

Am J Geriatr Psychiatry 9:3, Summer 2001

289

Antipsychotic Treatment in Dementia
found to possess efficacy at least equal to haloperidol,
with more favorable adverse-event profiles in patient
populations including elderly individuals with dementia.23–26
Encouraging data have been published on the efficacy and relative safety of the second-generation atypical
antipsychotic agents in the treatment of BPSD.23,25,27–30
Data on the comparative effectiveness of these agents are
needed to assist in optimizing treatment for dementia
patients with disruptive and costly BPSD.22,31
The purpose of this study is to compare the shortterm effectiveness of three antipsychotic monotherapy
regimens (haloperidol, olanzapine, and risperidone) for
the treatment of BPSD in elderly inpatients with a primary DSM-IV hospital discharge diagnosis of Dementia
Disorder.

METHODS
Data Source
Data were retrospectively obtained (October 1,
1996 to September 30, 1998) from the CQI‫ם‬SM Outcomes Measurement System developed by Mental
Health Outcomes in Lewisville, Texas. This system
tracks patients admitted to psychiatric inpatient units in
acute-care medical/surgical facilities. A total of 2,747 patients had a principal DSM-IV32 discharge diagnosis of
Dementia Disorder (ICD-9-CM codes 290.0–290.3;
290.40– 290.43; 294.1; 294.8), representing 23.3% of
the larger national population of 11,801 inpatients, ages
55 and older, treated across 73 geropsychiatric programs. Participating programs invite the first 17 patients
admitted per month (about 200 patients/year) to complete standardized questionnaires at time of admission
(within 72 hours), discharge (within 48 hours), and follow-up (3 months post-discharge). Staff at each site receives extensive training on the administration, scoring,
and interpretation of all instruments. Forms are entered
into the database with high-speed scanners. Weekly
data-integrity reports are provided to maintain high levels of compliance in form completion and accuracy.
Assessments
Demographic information, recent medical resource
utilization, and “baseline” functioning were captured on

290

an admissions questionnaire. This was completed by an
informant (typically a family member), with assistance
from staff as necessary, at time of program admission.
To examine treatment group comparability, several
measures of illness severity (at hospital admission) with
demonstrated validity in the elderly dementia population were examined. The Global Assessment of Functioning (GAF) Scale is based on a modification of the
Global Assessment Scale33 for Axis V of the DSM-IV multiaxial assessment. It provides a clinician’s estimate of
overall level of functioning based on psychological, social, and occupational status.
The Scale of Instrumental and Physical Activities of
Daily Living,34 used in the Older Americans Resources
and Services Program (OARS) Multidisciplinary Functional Assessment Questionnaire (MFAQ) developed at
Duke University, measures an individual’s ability (as
rated by an informant) to carry out a variety of physical
ADLs (e.g., eat, dress, walk) and instrumental ADLs
(e.g., use telephone, prepare meals, take medications).
The Mini-Mental State Exam (MMSE)35 is a widely used,
11-item, clinician-administered measure of general cognitive functioning with demonstrated reliability and validity across diverse populations.36 The collateral-source
version of the Geriatric Depression Scale (GDS-C),37 a
30-item, “Yes/No” questionnaire, is designed specifically for rating depression in the past week in elderly
subjects. Medication utilization was documented by a
medication usage questionnaire. This questionnaire captures information on the specific psychotropic agents,
as well as the types of non-psychotropic agents, used
by the patient at admission to the program and at discharge. Hospital length of stay was obtained directly
from the medical record.
The tool used to define BPSD was the behavioral
section of the Psychogeriatric Dependency Rating Scale
(PGDRS).38,39 The items were developed around the
concept of dependency, defined as “nursing time demanded by the patient.” The PGDRS has demonstrated
both interrater reliability and validity.38
A total of 16 behavioral/symptom items are rated
according to frequency of occurrence. The 3-point scale
for each behavior or symptom includes the following
anchors: Never (i.e., not occurring on any of the past
five shifts), Occasionally (i.e., occurring on one or two
of the past five shifts), and Frequently (i.e., occurring
on three or more of the past five shifts).
Specific problems measured include disruptive behavior, manipulative behavior, wandering, socially ob-

Am J Geriatr Psychiatry 9:3, Summer 2001

Edell and Tunis
jectionable behavior, demanding interaction, communication difficulties, noisiness, active aggression,
passive aggression, verbal aggression, restlessness, destructiveness (self), destructiveness (property), elated
affect, delusions/hallucinations, and incoherent speech
content. Definitions of each item are provided on the
form to facilitate consistency of meaning across raters.
Clinical and nursing staff familiar with the patient completed the PGDRS shortly after admission to and shortly
before discharge from the program.
Identification of Antipsychotic Medication
Groups
In order to compare outcomes (admission-to- discharge changes in BPSD) with respect to specific antipsychotic medication therapy, patients were identified
for whom only one antipsychotic medication (i.e., haloperidol, olanzapine, or risperidone) was prescribed at
discharge. A total of 998 patients (representing 36.6%
of the total Dementia Disorder cohort) were identified
as receiving antipsychotic “monotherapy” (i.e., taking
only one of the three antipsychotic agents at discharge).
Of this 998, 289, or 29.0%, were treated with haloperidol; 209, or 20.9%, were treated with olanzapine; and
500, or 50.1%, were treated with risperidone.
Patients on FDA-indicated cognitive enhancers
were excluded, but those on other forms of psychotropic medications (e.g., antidepressants; anxiolytics;
mood stabilizers) were not excluded from these antipsychotic monotherapy groups. Also, patients may have
been taking other, non-psychotropic medications.
Data Analyses
Change in PGDRS scores by antipsychotic treatment
group. These analyses examined the relationship of
each “monotherapy” antipsychotic medication group to
change in the frequency of occurrence of each of the
16 specific BPSD measured from time of admission to
time of discharge. They allowed for a full range of outcomes (i.e., improvement, decline, or no change) in
each behavior by medication group.
Analysis of variance (ANOVA) methods were used
to compare medication groups on change in each maladaptive behavior or symptom rating. Kruskal-Wallis
tests were also performed in case the response variable
violated the assumptions necessary for the appropriate
use of ANOVA. Tukey pairwise multiple comparisons

Am J Geriatr Psychiatry 9:3, Summer 2001

were conducted only when the overall ANOVA and
Kruskal-Wallis tests were statistically significant at the
0.05 level or better.
Comparability of medication groups. To examine
“baseline” comparability of the three treatment groups,
we compared the haloperidol-, olanzapine-, and risperidone-treated patients on demographics, level of functioning, previous resource utilization, severity of illness,
and dementia subtype assessed at admission. We also
compared the percentage of patients in each of the
three antipsychotic treatment groups who were taking
anticholinergic, anxiolytic, antidepressant, and moodstabilizer medications at time of discharge. Medications
prescribed on a prn basis were recorded only if taken
regularly (i.e., most of the days in the previous week,
including day of discharge). Finally, we compared
length of current hospital stay across medication
groups.
For continuous variables, one-way analyses of variance (ANOVAs) were performed to determine whether
the mean values at admission differed across the three
groups. The Tukey multiple comparison procedure was
used to determine the nature of any obtained differences. The analyses comparing drug groups on categorical variables consisted of chi-square tests of homogeneity. All statistical tests were two-tailed. When
significant baseline differences were identified, analyses
of covariance (ANCOVAs) and correlational analyses
were conducted to determine the effects the initial differences may have had on the changes in PGDRS items.
We then conducted a re-analysis to determine the impact on the original univariate results.
Patient Characteristics
In the total sample of geropsychiatric patients with
Dementia Disorder (N‫ ,)747,2ס‬about two-thirds
(65.7%) were female, with an average age of 81 (range:
56–108). Age distributions were 55–64 years (2%), 65–
74 years (17%), 75–84 years (47%), 85–94 years (29%),
and 95‫ ם‬years (5%). The most common diagnosis was
Dementia, Alzheimer’s type (DAT; 65.4%), followed by
Vascular Dementia (VD; 15.8%), Dementia NOS
(12.4%), and Dementia due to General Medical Condition (DGMC; 6.4%). Most patients were Caucasian
(86.9%), with 10.7% African American, and 1.2% Hispanic. Almost half (49.7%) had come to the hospital
from their private residence, whereas 42.4% had come

291

Antipsychotic Treatment in Dementia
from a long-term care facility such as a nursing home,
residential treatment, or board-and-care facility.
Clinical and Functional Status at Hospital
Admission
At time of admission, indices of illness severity
based on informant report, as well as clinician ratings,
suggested significant clinical, functional, and behavioral
impairment. The mean GAF score was 23, suggesting
that patients’ behaviors, on average, were considerably
influenced by delusions or hallucinations, serious impairment in communication or judgment, and inability
to function in almost all areas. Half (50.4%) met threshold for clinical depression (i.e., scores of at least 21 on
the GDS-C37).
The mean score on the MMSE35 at admission was
11.6, suggesting, on average, a severe degree of cognitive impairment. Almost three-quarters of the sample
(73.3%) met threshold for severe impairment; 18.6%
were rated as having moderate impairment, and 8.2% as
having mild cognitive impairment.
Informants reported that more than half of the patients were completely unable to perform a variety of
instrumental activities of daily living (e.g., handle
money; prepare meals; take medications), with significant proportions also unable to complete basic physical
activities of daily living (e.g., 31% could not take a bath/
shower unassisted; 19% could not dress self).
Defined by PGDRS ratings, a wide variety of BPSD
were observed in this dementia cohort at time of admission. Across the defined monotherapy groups, substantial proportions of the patients had at least occasional occurrence for 13 of the 16 items, ranging from
35% (manipulative behaviors) to 72% (communication
difficulties).

RESULTS
Change in PGDRS Scores from Admission to
Discharge by Medication Group
Of note, patients taking olanzapine experienced a
significantly greater improvement in overall PGDRS
score (mean change‫ ,)0.5ס‬as compared with those taking haloperidol (mean change‫ ;1.3ס‬PϽ0.01) or risperidone (mean change‫ ;8.2ס‬PϽ0.001).
An analysis of change in specific PGDRS items from

292

admission to discharge, by medication groups, revealed
that olanzapine was associated with significantly greater
improvement in active aggression (PϽ0.001), passive
aggression (PϽ0.001), verbal aggression (PϽ0.001), delusions/hallucinations (PϽ0.005), manipulative behavior (PϽ0.05), and noisy behavior (PϽ0.02), as compared with risperidone (see Figure 1). Olanzapine also
significantly exceeded haloperidol in change in active
aggression (PϽ0.05), passive aggression (PϽ0.001),
verbal aggression (PϽ0.001), and delusions/hallucinations (PϽ0.01; see Figure 2).
There were no statistically significant differences
between treatments for other PGDRS items. However,
there was a trend for olanzapine to be associated with
greater improvement in “demanding interaction,” as
compared with both risperidone (PϽ0.08) and haloperidol (PϽ0.07). No significant differences between
haloperidol and risperidone were found (see Figure 3).
Relationship of medication-prescribing to patient
characteristics. There were no significant differences
among the three antipsychotic medication groups on
the majority of baseline patient characteristics examined. Only one of nine demographic characteristics differentiated groups: a greater proportion of non-white
patients were treated with haloperidol (20.0%) or olanzapine (17.3%), as compared with risperidone (9.6%),
at discharge (v2[2]‫ ;39.91ס‬PϽ0.001).
With regard to overall group comparability, there
were no significant differences in global functioning,
mental status, activities of daily living, number of comorbid physical illnesses, recent life stressors, or history of suicide attempts. Also, no significant differences
in lifetime or past-6-months medical resource utilization
were found across medication groups.
Regarding illness characteristics, patients taking
olanzapine at discharge were more likely to be in the
VD group (28.1%) as compared with patients taking
haloperidol (11.4%) or risperidone (13.1%; PϽ0.05 for
both). Also, they were less likely to be in the DAT
group (56.7%), as compared with the risperidone group
(69.5%), and less likely to be in the DGMC/NOS
group (15.2%), as compared with the haloperidol group
(23.9%; PϽ0.05 for both).
Dementia patients given olanzapine at discharge
had also been rated by informants as somewhat more
depressed at admission (mean‫ ,)2.12ס‬compared with
those given haloperidol (mean‫ )2.91ס‬at discharge
(PϽ0.005). Medication groups also differed significantly

Am J Geriatr Psychiatry 9:3, Summer 2001

Edell and Tunis
in overall score on the PGDRS at admission (F[2,
98]‫ ;30.7ס‬PϽ0.01). Specifically, on a scale from 16
(complete absence of all BPSD) to 48 (frequent display
of all BPSD), patients taking olanzapine at discharge had
a significantly higher overall PGDRS score at admission
(mean‫ ;2.72ס‬median‫ ,)72ס‬as compared with patients
taking risperidone (mean‫ ;2.52ס‬median‫ ;42ס‬PϽ0.01).
The haloperidol group (mean‫ ;1.62ס‬median‫ )62ס‬did
not differ significantly from the other two groups.
The three medication groups did not differ significantly with respect to the use of anticholinergic medication at hospital discharge (haloperidol, 12.1%; olanzapine, 7.7%; risperidone, 9.4%; [P‫ .)]032.0 ס‬However,
medication groups differed significantly in concurrent
use of antidepressants (olanzapine, [64.5%] Ͼ risperidone [50.8%] Ͼ haloperidol [40.8%]; PϽ0.05 for each),
anxiolytics (haloperidol [40.8%] Ͼ risperidone [29.3%];
PϽ0.05), and mood stabilizers (olanzapine [26.8%] Ͼ
haloperidol [14.5%] Ͼ risperidone [9.8%]; PϽ0.05 for
each). Medication groups were found to differ in average length of stay, with patients taking olanzapine havFIGURE 1.

ing a longer length of stay (18.2 days) as compared with
those taking haloperidol (15.8 days) or risperidone
(16.4 days; PϽ0.05 for both).
Statistical control for initial group differences. When
the relationships between initial level of depression and
change in each specific PGDRS item were examined,
only the relationship between depression and change
in self-destructive behavior was significant (r‫;60.0ס‬
PϽ0.05), accounting for less than 1% of the response
variance. As expected, the subsequent multivariate result was consistent with the original (univariate) results
displayed in Figure 1, Figure 2, and Figure 3. Similarly,
adjusting for differences in admission severity of BPSD
entirely replicated the significant differences obtained
in the original model.
Controlling for dementia subtype in the analysis
was particularly important in that patients with VD
showed significantly greater improvement from admission to discharge on the overall PGDRS score and on
multiple items (i.e., demanding interaction, noisiness,

Psychogeriatric Dependency Rating Scale behavior changes in occurrence from admission to discharge (olanzapine
vs. risperidone)
Disruptive

NS
Olanzapine
Risperidone

P<0.05

Manipulative
Wandering

NS

Socially Objectionable

NS

P<0.08 (Trend)

Demanding Interaction
Communication Difficulty

NS

Noisy

P<0.02
P<0.001

Active Aggression

P<0.001

Passive Aggression
Verbal Aggression

P<0.001

Restless

NS

Destructive (self)

NS

Destructive (property)

NS

Affect-Elated

NS

Delusions/Hallucinations

P<0.005

Speech Content

NS
0

0.1

Am J Geriatr Psychiatry 9:3, Summer 2001

0.2

0.3
0.4
0.5
0.6
Change in Behavior Occurrences (Score)

0.7

0.8

0.9

1

293

Antipsychotic Treatment in Dementia
haloperidol and risperidone groups were all replicated,
with one exception; patients taking haloperidol showed
significantly greater improvement on noisiness than
those taking risperidone (PϽ0.05).
Use of concurrent antidepressant, anxiolytic, or
mood-stabilizer medications was found to have minimal
impact on change in PGDRS scores from admission to
discharge. Antidepressant use had no significant impact
on change scores on any PGDRS item. Concurrent use
of anxiolytics was associated with significantly greater
improvement in one item (socially objectionable) and
lesser improvement on one other (delusions/hallucinations). Use of mood stabilizers was associated with significant improvement on only one behavior (self-destructiveness). Results of the analyses controlling for
these effects once again completely replicated the originally reported differences between antipsychotic medication groups. Finally, the correlation between length
of stay and change in PGDRS score from admission to
discharge was negligible (r‫ ;900.0ס‬NS), suggesting that
it had little impact on the findings reported.

passive aggression, verbal aggression, and delusions/hallucinations) as compared with patients with DAT or
DGMC/NOS. They also exceeded patients with DAT on
noisiness, active aggression, restlessness, and destructiveness to self and property.
Importantly, the ANCOVA model controlling for dementia subtype both replicated and strengthened the
univariate findings. The reported differences between
the olanzapine and risperidone groups were all replicated at the same or stronger significance level (e.g.,
changed from PϽ0.05 to PϽ0.01). Also, the trend for
significance on demanding interaction (PϽ0.08) became statistically significant (PϽ0.02). Similarly, all of
the significant differences obtained between the haloperidol and olanzapine groups were replicated or
strengthened, with one trend (demanding interaction,
PϽ0.07) becoming statistically significant (PϽ0.02),
and one non-significant item (manipulative) now approaching significance favoring olanzapine (PϽ0.06).
Finally, the non-significant differences between the
FIGURE 2.

Psychogeriatric Dependency Rating Scale behavior changes in occurrence from admission to discharge (olanzapine
vs. haloperidol)
Disruptive

NS

Manipulative

Olanzapine
Haloperidol

NS

Wandering

NS

Socially Objectionable

NS

P<0.07 (Trend)

Demanding Interaction
Communication Difficulty

NS
NS

Noisy
Active Aggression

P<0.05
P<0.001

Passive Aggression
Verbal Aggression

P<0.001

Restless

NS

Destructive (self)

NS

Destructive (property)

NS

Affect-Elated

NS

P<0.01

Delusions/Hallucinations
Speech Content

NS
0

294

0.1

0.2

0.3
0.4
0.5
0.6
Change in Behavior Occurrences (Score)

0.7

0.8

0.9

1

Am J Geriatr Psychiatry 9:3, Summer 2001

Edell and Tunis
DISCUSSION
The comprehensive management of behavioral disturbances and psychotic symptoms of dementia is likely to
require an array of treatment modalities, including behavioral interventions, environmental modification,
family therapy and support, and medications.9 Pharmacologic interventions that can relatively quickly and
effectively reduce such behaviors for the patient, as well
their impact on family and caregivers, could potentially
reduce the need for expensive inpatient or long-term
care treatment. The present study utilized a large national database of geropsychiatric inpatients with DSMIV Dementia Disorder to identify three antipsychotic
monotherapy regimens and to examine change in BPSD
from admission to discharge associated with haloperidol, olanzapine, or risperidone therapy.
All three antipsychotic agents were associated with
some degree of improvement. However, olanzapine was
associated with significantly greater overall improveFIGURE 3.

ment from admission to discharge compared with the
other two medications. This result was attributable to
differential treatment effects on four distinct behaviors
(active-, verbal-, and passive-aggression, and delusions/
hallucinations) as compared with both risperidone and
haloperidol, and on two additional behaviors (manipulative behavior, noisiness), as compared with risperidone.
Significantly, differential improvement cannot be
accounted for by differences in examined baseline patient or illness characteristics or by the use of additional
psychotropic medications. Although the patients
treated with the antipsychotic monotherapies were
comparable in many important ways, it was necessary
to control for several potential confounds. These analyses replicated and, in some instances, strengthened the
univariate findings. The reported improvements took
place, on average, in a relatively brief time period (16–
18 days from admission to discharge). Some of the treatment-responsive symptoms and behaviors (e.g., active
aggression, delusions, disruptive vocal behavior) have

Psychogeriatric Dependency Rating Scale behavior changes in occurrence from admission to discharge (risperidone
vs. haloperidol)
Disruptive

NS
Haloperidol
Risperidone

NS

Manipulative
Wandering

NS

Socially Objectionable
Demanding Interaction

NS
NS

Communication Difficulty

NS

Noisy
Active Aggression

NS
NS

Passive Aggression
Verbal Aggression

NS

Restless

NS

Destructive (self)

NS

Destructive (property)

NS

Affect-Elated

NS

Delusions/Hallucinations

NS

Speech Content

NS
0

0.1

Am J Geriatr Psychiatry 9:3, Summer 2001

0.2

0.3
0.4
0.5
0.6
Change in Behavior Occurrences (Score)

0.7

0.8

0.9

1

295

Antipsychotic Treatment in Dementia
been found to be particularly pervasive and persistent,
as well as disturbing and burdensome to caregivers.40
Although no medication dosages were available for analysis, the relatively small percentages of patients on anticholinergics suggest that they were being treated with
doses at levels low enough not to warrant treatment for
extrapyramidal symptoms.
Still largely unknown is the relationship between
specific BPSD and underlying neuropathologic and neurochemical correlates such as neurotransmitter
changes41 associated with dementia.42 Bymaster et al.43
have recently pointed out that olanzapine and the atypical antipsychotic clozapine interact with multiple neuronal receptors, setting these drugs apart from other
atypical antipsychotic medications. They suggest that
olanzapine and similar agents belong to a new class of
antipsychotic agents called multi-acting-receptor-targeted antipsychotics (MARTA) and that agents with a
multi-acting binding profile may have efficacy and adverse-event profile advantages over compounds with selectivity for just one or two receptors.44,45 Research to
link psychopharmacology, clinical response, and behavioral/functional outcomes will play an important role in
understanding the value of antipsychotic treatment for
dementia.
This study provides evidence supporting the effectiveness of olanzapine in improving several problematic
symptoms and behaviors associated with dementia
from hospital admission to discharge. It should be
noted, however, that the retrospective, naturalistic
study design utilized in this investigation has clear limitations. Patients were not randomly assigned to treat-

ment conditions, nor were the patients, their families,
or the health care providers blind to the medications
prescribed. Comparative data on conventional and atypical antipsychotic agents, including dosage, obtained
from rigorously controlled, prospective, randomized,
double blind studies using structured diagnostic tools
and outcome scales relevant to BPSD are clearly needed
to replicate and extend the findings reported here.31
Finally, the generalizability of these findings to less
intensive treatment settings and to longer-term care facilities, such as nursing homes, has yet to be established. Research to determine longer-term antipsychotic medication effects (e.g., past the point of
hospital discharge) as has been undertaken in younger,
schizophrenic patient populations,46 is also warranted.
This work would represent an important next step in
determining the value of antipsychotic medications in
the treatment of BPSD.
The two authors contributed equally to the research and manuscript.
Portions of the work have been presented at the
51st Institute on Psychiatric Services, New Orleans, LA,
October 31, 1999, and at the American Society of Consultant Pharmacists Annual Meeting, St. Louis, MO,
November 12, 1999.
We thank the following individuals for their assistance with project implementation and data analyses:
Kristina L. Greenwood, Ph.D.; Chimene W. Kellis, M.S.;
and Bryan E. Adams, Ph.D. John S. Kennedy, M.D., and
Ralph Swindle, Ph.D., provided valuable scientific input, for which we are grateful.
This research was funded by Eli Lilly and Company, Indianapolis, IN.

References
1. Loreck DJ, Bylsma FW, Folstein MF: The Dementia Signs and
Symptoms Scale: a new scale for comprehensive assessment of
psychopathology in Alzheimer’s disease. Am J Geriatr Psychiatry
1994; 2:60–74
2. Cohen-Mansfield J: Agitated behaviors in the elderly, II: preliminary results in the cognitively deteriorated. J Am Geriatr Soc
1986; 34:722–727
3. Lyketsos CG, Steele C, Galik E, et al: Physical aggression in dementia patients and its relationship to depression. Am J Psychiatry 1999; 156:66–71
4. Rabins PV, Mace NL, Lucas MJ: The impact of dementia on the
family. JAMA 1982; 248:333–335
5. Rovner BW, Kafonek S, Filipp L, et al: Prevalence of mental illness
in a community nursing home. Am J Psychiatry 1986; 143:1446–
1449
6. Brodaty H: Caregivers and behavioral disturbances: effects and
interventions. Int Psychogeriatr 1996; 8:455–458
7. Deimling GT, Bass DM: Symptoms of mental impairment among

296

elderly adults and their effects on family caregivers. J Gerontol
1986; 41:778–784
8. Gilleard CJ, Belford H, Gilleard JE, et al: Emotional distress
amongst the supporters of the elderly mentally infirm. Br J Psychiatry 1984; 145:172–177
9. Dyck G: Management of geriatric behavior problems. Psychiatr
Clin North Am 1997; 20:165–180
10. Colerick EJ, George LK: Predictors of institutionalization among
caregivers of patients with Alzheimer’s disease. J Am Geriatr Soc
1986; 34:492–498
11. O’Donnell BF, Drachman DA, Barnes HJ, et al: Incontinence and
troublesome behaviors predict institutionalization in dementia. J
Geriatr Psychiatry Neurol 1992; 5:45–52
12. Ritchie K: Behavioral disturbances of dementia in ambulatory
care settings. Int Psychogeriatr 1996; 8:439–442
13. Steele C, Rovner B, Chase GA, et al: Psychiatric symptoms and
nursing home placement of patients with Alzheimer’s disease.
Am J Psychiatry 1990; 147:1049–1051

Am J Geriatr Psychiatry 9:3, Summer 2001

Edell and Tunis
14. Kunik ME, Yudofsky SC, Silver JM, et al: Pharmacologic approach
to management of agitation associated with dementia. J Clin Psychiatry 1994; 55:13–17
15. Gilleard CJ, Morgan K, Wade BE: Patterns of neuroleptic use
among the institutionalized elderly. Acta Psychiatr Scand 1983;
68:419–425
16. Barnes R, Veith R, Okimoto J, et al: Efficacy of antipsychotic medications in behaviorally disturbed dementia patients. Am J Psychiatry 1982; 139:1170–1174
17. Lohr JB, Jeste DV, Harris MJ, et al: Treatment of disordered behavior, in Clinical Geriatric Psychopharmacology. Edited by Salzman C. Baltimore, MD, Williams & Wilkins, 1992, pp 79–113
18. Raskin M: Alzheimer’s disease: treatment of noncognitive behavioral abnormalities, in Psychopharmacology: The Fourth Generation of Progress. Edited by Kupfer FBD. New York, Raven, 1995,
pp 1427–1436
19. Tune LE, Steele C, Cooper T: Neuroleptic drugs in the management of behavioral symptoms of Alzheimer’s disease. Psychiatr
Clin North Am 1991; 14:353–373
20. Devanand DP: Antipsychotic treatment in outpatients with dementia. Int Psychogeriatr 1996; 8:355–361
21. Schneider LS: Meta-analysis of controlled pharmacologic trials.
Int Psychogeriatr 1996; 8:375–379
22. Finkel SI: Managing the behavioral and psychological signs and
symptoms of dementia. Int Clin Psychopharmacol 1997; 12:25–
28
23. Kumar V, Brecher M: Psychopharmacology of atypical antipsychotics and clinical outcomes in elderly patients. J Clin Psychiatry 1999; 60:5–9
24. Madhusoodanan S, Brenner R, Cohen CI: Role of atypical antipsychotics in the treatment of psychosis and agitation associated
with dementia. Central Nervous System Drugs 1999; 12:135–150
25. Glazer WM: Olanzapine and the new generation of antipsychotic
agents: patterns of use. J Clin Psychiatry 1997; 58:18–21
26. Sciolla A, Jeste DV: Use of antipsychotics in the elderly. International Journal of Psychiatry in Clinical Practice 1998; 2:27–34
27. Irizarry MC, Ghaemi SN, Lee-Cherry ER, et al: Risperidone treatment of behavioral disturbances in outpatients with dementia. J
Neuropsychiatry Clin Neurosci 1999; 11:336–342
28. Katz IR, Jeste DV, Mintzer JE, et al: Comparison of risperidone
and placebo for psychosis and behavioral disturbances associated
with dementia: a randomized, double-blind trial (Risperidone
Study Group). J Clin Psychiatry 1999; 60:107–115
29. Madhusoodanan S, Suresh P, Brenner R, et al: Experience with
the atypical antipsychotics: risperidone and olanzapine in the
elderly. Ann Clin Psychiatry 1999; 11:113–118
30. Street J, Clark WS, Gannon KS, et al: Olanzapine treatment of
psychotic and behavioral symptoms in patients with Alzheimer

Am J Geriatr Psychiatry 9:3, Summer 2001

disease in nursing care facilities: a double-blind, randomized, placebo-controlled trial. Arch Gen Psychiatry 2000; 57:968–976
31. Davidson, M, Weiser, M, Soares, K: Novel antipsychotics in the
treatment of psychosis and aggression associated with dementia:
a meta-analysis of randomized controlled clinical trials. Int Psychogeriatr 2000; 12: 271–277
32. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American
Psychiatric Press, 1994
33. Endicott J, Spitzer RL, Fleiss JL, et al: The Global Assessment
Scale. a procedure for measuring overall severity of psychiatric
disturbance. Arch Gen Psychiatry 1976; 33:766–771
34. Fillenbaum GG: Multidimensional functional assessment of older
adults: The Duke Older Americans Resources and Services Procedures. Hillsdale, NJ, Lawrence Erlbaum Associates, 1988
35. Folstein MF, Folstein SE, McHugh PR: Mini-Mental State: a practical method for grading the cognitive state of patients for the
clinician. J Psychiatr Res 1975; 12:189–198
36. Tombaugh TN, McIntyre NJ: The Mini-Mental State Exam: a comprehensive review. J Am Geriatr Soc 1992; 40:922–935
37. Nitcher RL, Burke WJ, Roccaforte WH, et al: A collateral-source
version of the Geriatric Depression Rating Scale. Am J Geriatr
Psychiatry 1993; 1:143–152
38. Wilkinson IM, Graham-White J: Psychogeriatric Dependency Rating Scale (PGDRS): a method of assessment for use by nurses. Br
J Psychiatry 1980; 137:558–565
39. Wilkinson I, Zissler LM: Standardized assessments for the elderly:
clinical applications. Nursing Times 1984; 80:36–37
40. Cooler YC: Perspectives of a new clinical trials research nurse.
Int Psychogeriatr 1996; 8:465–468
41. Gottfries CG: Neurochemistry and neurotransmitters. Int Psychogeriatr 1996; 8:225–231
42. Zubenko GS: Clinicopathologic and neurochemical correlates of
major depression and psychosis in primary dementia. Int Psychogeriatr 1996; 8:219–223
43. Bymaster FP, Moore NA, Nakazawa T: Review of the preclinical
pharmacology of olanzapine: a MARTA-class antipsychotic. Japanese Journal of Clinical Psychopharmacology 1999; 2:885–911
44. Bymaster FP, Breier A, Tollefson GD, et al: Olanzapine belongs to
a new, multi-acting, receptor-targeted antipsychotic (MARTA)
class of antipsychotics, in Olanzapine (Zyprexa): A Novel Antipsychotic. Edited by Tran PV, Bymaster FP, Tye N, et al. Philadelphia, PA, Lippincott, Williams & Wilkins Healthcare, 2000, pp
IX–XII
45. Bymaster FP, Calligaro DO, Falcone JF, et al: Radioreceptor-binding profile of the atypical antipsychotic olanzapine. Neuropsychopharmacology 1996; 14:87–96
46. Tran PV, Dellva MA, Tollefson GD, et al: Oral olanzapine vs. oral
haloperidol in the maintenance treatment of schizophrenia and
related psychoses. Br J Psychiatry 1998; 172:499–505

297