Evidence Table 15Active-control trials in patients with behavioral and psychological symptoms of dementia

Author, year (Quality score)NDurationStudy design
Eligibility criteriaExclusion criteriaInterventions (drug, dose)Run-in/washout periodAllowed other medications/interventionsAge
Other population characteristics (diagnosis, etc)Number screened/eligible/enrolledNumber withdrawn/lost to fu/analyzedOutcome measuresMethod of outcome assessment and timing of assessmentResultsMethod of adverse event assessmentAdverse events
Olanzapine vs. haloperidol
Moretti, 2005 (FAIR)34612 monthsOpen-label, non-randomized controlled trialMen and women, age 71–92 years with MMSE scores of at least 14 and satisfying DSM-IV criteria for dementia. Probable vascular dementia according to NINDS-AIREN criteria.Signs of normal pressure hydrocephalus; previous psychiatric illness or CNS disorders, alcoholismOlanzapine 2.5–7.5 mg/day; mean dose 4.23 mg/day (SD 2.12)

Typical antipsychotics: promazine 4%, up to 10 drops tid; mean dose 1.65 mg/day (SD 0.23) or haloperidol 0.2%, up to 10 drops tid; mean dose 1.65 mg (0.23
One week to discontinue benzodiazepines or other neurolepticsAllowed to continue previous therapy (e.g., cholinesterase inhibitors, antihypertensive, antidyslipidemic, antidiabetic drugs)Mean age 76.78 (SD 4.01)
44.4% female
Race/ethnicity not reported
Subcortical vascular dementia:
Multi-infarct dementia: 88.4%
NR/NR/3460/0/346Clinical Dementia Rating Scale
Barthel Index
Instrumental ADL
Tinetti scale for equilibrium/balance and gait
Cumulative Illness Rating Scale
Hachinski Ischemic score
Matthew's Stroke Scale
Caregiver Burden Inventory
Hachinski Ischemic score and Matthew's Stroke Scale at first and last visit, others at every visitMean change from baseline, olanzapine vs typical antipsychotics
Clinical Dementia Rating Scale: +0.4 vs +0.51 (NS)
NPI: −12.1 vs −9.74 (NS)
Barthel Index: −6.4 vs −13.45 (p<0.05)
Instrumental ADL: −1.7 vs −2.4 (NS)
TINETTI equilibrium: −1.3 vs −5.7 (<0.01)
TINETTI gait: −2.7 vs −7.4 (<0.01)
TINETTI total: −4.0 vs −13.1 (<0.01)
Clinical Insight Rating Scale: +1.4 vs+2.7 (<0.05)
Caregiver Burden Inventory: −10.2 vs +2.7 (<0.05)
Not reported2 deaths in olanzapine group (1.15%: MI and pneumonia); 3 in haloperidol group (1.73%: pulmonary embolism, MI, fracture complications)
Olanzapine group: 5 new angina pectoris (2.89%), 2 (1.15%) diagnosed with diabetes, 1 peripheral arteriopathy, 1 renal failure, 1 fall.
Haloperidol group: 4 (2.31%) angor episodes, 2 (1.15%) MI, 3 (1.73%) diagnosed with diabetes, 13 falls
Mean weight increase:
olanzapine: 5.65 kg (SD 1.45)
haloperidol: 4.89 kg (SD 2.32)
Quetiapine vs. haloperidol
Savaskan, 2006
Switzerland (POOR)
225 weeksOpen-label, randomized, single center; inpatientsConfirmed diagnosis of Alzheimer's disease, behavioral symptoms (at least 3 of the following: aggression, psychotic symptoms, sleep-wake cycle disturbances, agitation, restlessness or sundowning), permanent medical or social care available during the study, written informed consent and over age 65.Known sensitivity to study drugs, evidence of chronica and/or severe renal, hepatic, cardiovascular, pulmonary of gastrointestinal impairment or cancer, other antipsychotic medication than the study drugs, participation in any other drug trial and contraindications as detailed in the country-specific prescribing information for the study drugs.Quetiapine mean dose 125 mg

Haloperidol mean dose 1.9 mg
Maximum 7-day run-in period (not described)Concomitant medication was continued and documented. All patients received a cholinesterase inhibitor (galantamine 2 X 8 mg)Mean age 82
68.2% female
Race/ethnicity not reported
All had Alzheimer's Disease, no prior history of psychiatric diagnosisNR/NR/304/0/22NPI
Consortium to Establish a Registry for
Alzheimer's Disease (CERAD)
neuropsychological test battery
Nurses' Observation Scale for Geriatric Patients (NOSGER)
Baseline during run-in, 1 day prior to commencing study drugs, and end of week 5Results reported graphically only
NPI: similar effects for both treatment groups for delusions and agitation.
CERAD: both groups improved in word recall
MMSE: no significant differences from baseline
NOSGER: quetiapine improved instrumental ADL
Not reportedquetiapine: 1 discontinuation for postural hypotonia and 1 for MI
haloperidol: 1 discontinuation for EPS and 1 for TIA Other adverse events:
quetiapine: 1 reversible syncope, 1 gastroenteritis
haloperidol: 1 infection of unknown origin, 1 arterial hypertonia
Tariot, 2006: full publication (Replaces Tariot 2004, previously available as a poster only) (FAIR)28410 weeksDouble-blind, multicenter, 47 nursing homesMen and women, age 65 and older, not bedridden, residing in nursing homes for at least 2 weeks; DSM-IV diagnosis of dementia or National Institute of Neurological and Communicative Disorders & Stroke-Alzheimer's Disease (NINCDS) diagnosis of Alzheimer's Disease; BPRS score 24 or higher, CGI-Severity score 4 or higher.Other clinically significant medical conditions, history of drug-induced agranulocytosis, acute orthostasis, clinically significant abnormal electrocardiogram, or concurrent other Axis I DSM-IV diagnosis.quetiapine: median of mean daily
dose 96.9 mg; maximum 125.0 mg

haloperidol: median of mean
daily dose 1.9 mg; maximum 2.0 mg
No placebo run-in; antipsychotics discontinued for at least 48 hours.Psychotropics permitted: chloral hydrate, zolpidem, lorazepam for sleep/agitation; anti-EPS medication (but not prophylactically), cholinesterase inhibitors if stable dose for >6 weeks prior to entry.Mean age 83.9
73% female
89% white, 8% black, 2% Hispanic, <1% Asian.
100% Alzheimer's dementia501/378/284104 withdrawn/1 lost to followup/265 analyzedBPRS-Total score, agitation factor subscale (tension, hostility, uncooperativeness, and excitement items) and anergia factor subscale (emotional withdrawal, motor retardation, blunted affect, disorientation)
NPI-NH Agitation + Hallucinations + Delusions (NPI-3)
Multidimensional Observation Scale for Elderly Subjects (MOSES)
Physical Self-Maintenance Scale (PSMS)
Screening, baseline, weeks 2, 4, 6, and 10All drug treatment groups improved from baseline to LOCF on BPRS total score and on the NPI-3 (Data presented graphically only)
Quetiapine group had statistically significantly better functional status as assessed by the MOSES, PSMS, AND BPRS anergia factor compared with haloperidol (comparison to placebo not reported, data presented graphically only)
Quetiapine and haloperidol groups had significantly more improvement than placebo patients on the BPRS agitation subscale (change from baseline, quetiapine −2.4 [p=0.033], haloperidol −2.9 [p=0.001], placebo −1.1)
Quetiapine patients' scores on MMSE not significantly different from placebo; haloperidol results not reported.
Simpson-Angus Scale and AIMS at baseline, weeks 2, 4, 6, 8, and 10.Withdrawals due to AEs:
11% quetiapine
18% haloperidol
13% placebo

AEs with >10% incidence of which were statistically significantly different from placebo: somnolence, infection, rash, pain, conjunctivitis, vomiting, headache, cough increased, postural hypotension, dizziness, weight gain, weight loss, accidental injury. Of treatment-emergent adverse events, somnolence occurred statistically more often for quetiapine and haloperidol than for placebo.
Zeneca Pharmaceuticals (5077IL/0049)1146 weeksDB RCT
Multicentre (18 centres)
Male or female, aged 65 years or over; satisfaction of the ICD-10 research diagnostic criteria for dementia in Alzheimer’s disease, with the presence of predominantly delusional or hallucinatory symptoms; a score of at least 3 (mildly ill) on the CGI-S item; a score between 10 and 26 from the Mini Mental State Examination.Evidence of any significant clinical disorder or laboratory finding for this age group; patients with a history or clinical evidence on ECG of myocardial infarction within the last 3 months, or any clinically significant ECG result; total white blood cell count less than the lower limit of the reference range of the laboratory used for hematological monitoring; history of drug-induced agranulocytosis; satisfaction of diagnostic criteria for delirium superimposed on dementia.Quetiapine (50, 75, 100, 200, or 300 mg/day) or haloperidol (1, 1.5, 2, 4, or 6 mg/day)NRNR77.7% are >75 years old
75.9% female
99.1% Caucasian
Quetiapine vs haloperidol

late onset dementia: 83.6% vs 78.9%
Alzheimer’s Disease Assessment Scale: 30.8 vs 28.7
CGI-S: 4 vs 3.9
140/NR/11434/NR/108Neuropsychiatric Inventory, MADRS, CGI Severity of Illness, and Mini Mental State Examination, Cognitive assessments using the Alzheimer’s Disease Assessment Scale, CGI Global Improvement (Day 42)Baseline, Day 15 and Day 42No differences between both groups in changes in Neuropsychiatric Inventory and MADRS total scores (P=0.156 and P=0.065 respectively)SAR-S at baseline and at Days 15 and 42; the use of anticholinergic medication to treat EPS and adverse events related to EPS were recorded; all adverse events, and routine clinical laboratory tests, vital signs measurements, and ECGs were also performed.Quetiapine vs haloperidol

Death: 3 (5.6) vs 1 (1.8)
Withdrawals due to AEs: 9 (16.7) vs 11 (19.6)

No statistically significant differences between groups in proportion of patients developing clinically significant EPS (P=0.598), who experienced AEs related to EPS (P=0.447), and who received anticholinergic mediation to treat EPS (P=0.254) No clinically significant changes in laboratory data, blood pressure or pulse rate
Risperidone vs. haloperidol
Chan et al, 2001
Hong Kong (FAIR)
5812 weeksDouble-blind, multicenter (3 centers)Age 55 or older and met DSM-IV criteria for Dementia of Alzheimer's Type with behavioral disturbance, vascular dementia with behavioral disturbance or a combination of the two. Active behavioral symptoms, as evidenced by a frequency score of at least 4 on one and at least 3 on another item of the Cohen-Mansfield Agitation Inventory (CMAI). Symptoms present for at least 2 weeks. Score of at least 8 on Behavioral Pathology in Alzheimer's Disease Rating Scale (BEHAVE- AD).No additionalrisperidone 0.5 mg vs haloperidol 0.5 mg to start. Titrated by increments of 0.5 mg no faster than every other day. Target dose 1 mg per day, could be stepped up to 2 mg per day if symptoms poorly controlled.7- to 14-day washout during which all psychotropic and antiparkinsonian drugs were stopped.Medications permitted not reported, but report patients taking benzodiazepines (4 haloperidol, 4 risperidone), chloral hydrate (1 risperidone), benzhexol (2 haloperidol), donepezil (1 haloperidol), and donepezil (1 haloperidol).Mean 80.5 (sd 8.2)
72% female
100% Chinese
79% Alzheimer's dementia, 21% vascular dementiaNumber screened, eligible not reported, 58 enrolled3 withdrew (1 haloperidol, 2 risperidone), 55 analyzed.CMAI total score, BEHAVE-AD subscale scores, Functional Assessment Staging Rating Scale (FAST), Cantonese version of Mini-Mental State Examination (CMMSE).Baseline, weeks 4, 8, and 12.
Additional CMAI ratings at weeks 2, 6, and 10.
Mean change from baseline to endpoint, risperidone vs haloperidol CMAI total:
−8.1 vs −10 (p=0.95)
BEHAVE-AD (Psychosis):
−1.1 vs −0.6 (p=0.91)
BEHAVE-AD (Activity disturbances):
−0.8 vs −0.7 (p=0.16)
BEHAVE-AD (Aggressiveness):
−1.3 vs −1.3 (p=0.56)
BEHAVE-AD (Diurnal rhythm disturbances):
−0.4 vs −0.3 (p=0.36)
BEHAVE-AD (Affective disturbances):
−0.2 vs 0 (p=0.11)
BEHAVE-AD (Anxieties and phobia):
0 vs −0.1 (p=0.19)
AIMS, Simpson-Angus Extrapyramidal Symptoms Scale, Barnes Akathisia ScaleWithdrawals due to Aes:0 risperidone; 3% haloperidol (somnolence)

risperidone: no significant increase from baseline on Simpson-Angus, Barnes, or AIMS.
haloperidol: significant increase in Simpson-Angus Scale (p<0.001)
De Deyn et al, 1999
Multiple European countries (FAIR)
Engelborghs (subanalysis)
34412 weeksDouble-blind, placebo-controlled, multicenterAge 55 or older, institutionalized, diagnosis of primary degenerative dementia of the Alzheimer's type, vascular dementia, or mixed dementia according to the DSM-IV. Scores of 4 or greater on Functional Assessment Staging (FAST); 23 or greater on Mini-Mental Status Examination (MMSE); 1 or greater on the BEHAVE-AD global rating; and 8 or greater on the BEHAVE-AD total score.Other conditions that diminish cognitive function; other psychiatric disorders; clinically relevant organic or neurologic disease; ECG or laboratory abnormalities; administration f a depot neuroleptic within one treatment cycle of Visit 1; history of allergic reaction to neuroleptics or history of neuroleptic malignant syndrome; participation in clinical trial(s) with investigational drugs during the 4 weeks preceding this trial.risperidone 0.5 mg vs haloperidol 0.5 mg to start. Titrated by increments of 0.5 mg every 4 days if indicated, to 2 mg. Could be increased up to 4 mg per day if symptoms poorly controlled and no EPS.1-week single-blind washout phase during which all psychotropic medications were discontinued.Use of antipsychotics, antidepressants, lithium, carbamazepine, and valproic acid not permitted. Lorazepam permitted if limited to 4 days per week for the first 4 weeks of treatment. If needed beyond week 4, patient discontinued from study.Mean 81 (range 56–97)
56% female
99% white, <1% black, <1% Asian
74% Alzheimer's dementia, 33% Vascular Dementia (7% had both diagnoses)Number screened not reported/371 eligible/344 enrolled (27 dropped out during washout)344 analyzedBEHAVE-AD, Cohen-Mansfield Agitation Inventory (CMAI), and Clinical Global Impression (CGI)Evaluations at selection, baseline, weeks 1, 2, 4, 6, 8, 10, 12.Mean change from baseline to endpoint, risperidone vs haloperidol vs placebo
BEHAVE-AD (Total): −5.2 vs −6.6 vs −4.2
BEHAVE-AD (Aggressiveness): −1.7 vs −1.6 vs −0.8
CMAI (Total aggressive): −3.9 vs −3.3 vs −1.6
CMAI (Physical aggressive): −2.7 vs −2.3 vs −0.7
CMAI (Verbal aggressive): −1.2 vs −1.0 vs −0.8 (No significant differences between risperidone and haloperidol)

Mean change from baseline to week 12, risperidone vs haloperidol vs placebo
BEHAVE-AD (Total): −8.6 vs −7.5 vs −6.2 (p NS for risperidone vs haloperidol)
BEHAVE-AD (Aggressiveness): −2.9 vs −1.8 vs −1.5 (p=0.05 for risperidone vs haloperidol; post hoc analysis)
CMAI (Total aggressive): −8.3 vs −3.6 vs −4.9 (p=0.02 for risperidone vs haloperidol; post hoc analysis)
CMAI (Physical aggressive): −5.9 vs −2.8 vs −3.5 (p NS for risperidone vs haloperidol)
CMAI (Verbal aggressive): −2.5 vs −0.8 vs −1.4 (p=0.01 for risperidone vs haloperidol; post hoc analysis)
Extrapyramidal Symptom Rating ScaleWithdrawals due to AEs:
18% total, no significant differences between groups.

Mean change in Extrapyramidal Symptoms Rating Scale score:
risperidone 0.5 to 2 mg: −0.3
haloperidol 0.5 to 2 mg: +1.6
placebo: −1.4
(p <0.05 for risperidone vs haloperidol, NS for risperidone vs placebo)
Suh et al, 2004
South Korea (FAIR)

Suh 2006 [post hoc analyses]
12018 weeks (1 week washout, 8 weeks active treatment, 1 week washout, 8 weeks crossover treatment)Double-blind, crossover, single centerAge 65 or older, diagnosis of dementia of the Alzheimer's type with behavioral disturbance, vascular dementia with behavioral disturbance, or a combination of the two, according to DSM-IV criteria. Score of 4 or higher on the Functional Assessment Staging Test, a total score of 8 or higher on the Korean version of the BEHAVE-AD, and a score of more than 3 on any two items of the Korean version of the CMAI.Other conditions that diminish cognitive function (e.g., Lewy-body dementia, hypothyroidism), other psychiatric disorders that might contribute to the psychotic symptoms (e.g., schizophrenia, delusional disorder), clinically relevant organic or neurologic disease, unstable medical conditions (e.g., poorly controlled hypertension, angina, or diabetes), abnormal electrocardiograms as diagnosed by a cardiologist or laboratory tests, a history of allergic reaction to antipsychotic treatment, and a history of neuroleptic malignant syndrome.risperidone or haloperidol 0.5 mg to 1.5 mg (target dose was 1 mg). Dose could be titrated up or down; dosing regimen and intervals between dose titrations were individualized for each patient.1-week washout period during which all psychotropic medications were discontinued.Concomitant use of antipsychotic drugs, antidepressants, and mood stabilizers was not permitted. Lorazepam permitted if limited to 4 days/week for the first 4 weeks of treatment.Mean age 80.9 (SD 8.2, range 65–97)
80% female
Ethnicity not reported (trial conducted in South Korea)
65.8% Alzheimer's dementia
28.3% vascular dementia
5.8% mixed
280 screened/# eligible not reported/120 enrolled6 withdrawn/0 lost to followup/114 analyzedBEHAVE-AD-K, CMAI-K, AND CGI-CPatients assessed weekly during the first 4 weeks and then every 2 weeks (twice) until the end of the final (8th week)Mean change from baseline to endpoint, risperidone vs haloperidol
BEHAVE-AD-K (Total): − 7.2 vs − 4.7 (p=0.004)
BEHAVE-AD-K (Psychosis): − 3.7 vs − 2.0 (p=0.582)
BEHAVE-AD-K (Activity Disturbances)
− 1.1 vs − 0.8 (p=0.858): BEHAVE-AD-K (Aggressiveness)
− 1.1 vs − 0.9 (p=0.002)
BEHAVE-AD-K (Diurnal Rhythm Disturbances): − 0.5 vs − 0.2 (p=0.038)
BEHAVE-AD-K (Affective Disturbance): − 0.5 vs − 0.2 (p=0.248)
BEHAVE-AD-K (Anxieties and Phobias): − 0.3 vs + 0.1 (p<0.0001)
BEHAVE-AD-K (Wandering): − 0.3 vs + 0.1 (p<0.0001)*
BEHAVE-AD-K (Agitation): − 0.3 vs + 0.1 (p<0.0001)*
BEHAVE-AD-K (Godot syndrome): − 0.3 vs + 0.1 (p<0.0001)*
BEHAVE-AD-K (Other anxieties): − 0.3 vs + 0.1 (p<0.0001)*

CMAI-K (Total): − 14.2 vs − 5.9 (p<0.0001)
CMAI-K (Aggressive Behavior): − 4.0 vs − 3.3 (p=0.001)
CMAI-K (Physical Non-Aggressive Behavior): − 2.4 vs − 1.0 (p=0.024)
CMAI-K (Verbally Agitated Behavior): − 1.1 vs − 0.5 (p=0.002)
CMAI-K (Physical sexual advances): − 1.1 vs − 0.5 (p=0.002)*
CMAI-K (Pace, aimless wandering): − 1.1 vs − 0.5 (p=0.002)*
CMAI-K (Intentional falling): − 1.1 vs − 0.5 (p=0.002)*
CMAI-K (Hoarding): − 1.1 vs − 0.5 (p=0.002)*
CMAI-K (Performing repetitious mannerisms): − 1.1 vs − 0.5 (p=0.002)*
CMAI-K (Repetitive sentence or questions): − 1.1 vs − 0.5 (p=0.002)*
CMAI-K (Complaining): − 1.1 vs − 0.5 (p=0.002)*
CMAI-K (Negativism): − 1.1 vs − 0.5 (p=0.002)*

CGI-C: − 0.1 vs + 0.2 (p=0.001)
*post hoc analysis from Suh 2006
All reported adverse events were recorded, and the severity of EPS was assessed by use of the ESRS.Withdrawals due to AEs:
7% risperidone
3% haloperidol
Mean change from baseline on ESRS, risperidone vs haloperidol:
Total: +4.8 vs +13.8 (p=0.0001)
Parkinsonism: +3.5 vs +10.4 (p=0.0001)
Dystonia: +1.0 vs +2.5 (p=0.6503)
Dyskinetic movement: +0.5 vs +0.9 (p=0.4144)
Risperidone vs. citalopram
Pollock 2007
10312 weeksDB RCT
Single center - University of Pittsburgh Medical Center
Dementia of the Alzheimer type (DAT), vascular dementia, dementia with Lewy bodies, mixed dementia, or dementia not otherwise specified.Current or past diagnosis of schizophrenia, schizoaffective disorder, delusional disorder, psychotic disorders not otherwise specified, bipolar disorder, mental retardation, cognitive deficits following head trauma, or a current diagnosis of delirium, substance-induced persisting dementia, Parkinson disease, drug/alcohol abuse, or dependence; major depressive episode within the past 6 months or clinically significant depressive symptoms with a rating of 12 or higher on the Cornell Scale for Depression in Dementia ; unstable physical illness; creatinine 2.0 mg/100 mL; aspartate aminotransferase or bilirubin more than twice the upper limit of normal; potentially reversible cause of dementia; treatment with a depot neuroleptic drug within 2 months or fluoxetine within 4 weeks; or a history of allergy or intolerance to citalopram or risperidone.Mean (SD) final doses
Citalopram 29.4 (9.9) mg/day
Risperidone 1.25 (0.51) mg/day
NRLorazepam and cholinesterase inhibitors or memantineCitalopram vs. Risperidone
Age (years)
82.0 (7.3) vs. 81.5 (9.2)
Female, % (N)
47.2 (25) vs. 76.0 (38)
White, % (N)
81.1 (43) vs. 82.0 (41)
Citalopram vs. Risperidone
NBRS total score
60.3 (16.8) vs. 53.6 (15.6)
NBRS agitation score
10.1 (4.8) vs.8.9 (4.5)
NBRS psychosis score
5.9 (3.5) vs..6.0 (4.5)
UKU total score
13.7 (5.2) vs..12.1 (4.1)
408/111/10558/0/103NBRS, CSDD, Neuropsychiatric Inventory (NPI), Udvalg for Kliniske Undersogelser (UKU)
side effect scale, Mini-Mental State Examination (MMSE), Severe Impairment Battery (SIB), and
Cumulative Illness Rating Scale–Geriatrics
Assessed at the time of enrollment; baseline ; after receiving study medication for 3 days, 7 days, then weekly for 5 weeks, then every 2 weeks; and at the time of discharge from the hospital or termination if the study was terminated early.Citalopram vs. Risperidone
NBRS agitation score (items 8, 11, and 14)
Post – pre 1.26 (4.58) 0.73 (4.91) P = 0.57

NBRS psychosis score (items 16, 18, and 20)
Post – pre 1.90 (4.49) 2.16 (4.68) P = 0.79
UKUCitalopram vs. Risperidone
UKU total score
Post – pre 0.49 (4.81) 2.33 (6.05) P = 0.011
UKU psychic subscale score
Post – pre 0.91 (3.42) vs. 1.06 (3.77) P = 0.007
UKU neurologic subscale score
Post – pre 0.15 (1.42) vs. 0.29 (2.36) P = 0.27
UKU autonomic subscale score
Post – pre 0.02 (1.73) vs. 0.10 (2.34) P = 0.84
UKU other subscale score
Post – pre 0.57 (1.08) vs. 0.76 (1.45) P = 0.46

From: Evidence Tables

Cover of Drug Class Review: Atypical Antipsychotic Drugs
Drug Class Review: Atypical Antipsychotic Drugs: Final Update 3 Report [Internet].
McDonagh M, Peterson K, Carson S, et al.
Portland (OR): Oregon Health & Science University; 2010 Jul.
Copyright © 2010 by Oregon Health & Science University, Portland, Oregon 97239. All rights reserved.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.