Odprn.ca

Atypical antipsychotics for the behavioural and psychological symptoms of dementia in the elderly Systematic Review and Network Meta-Analysis George A. Wells, Shannon Kelly, Amy Johnston, Annie Bai, Li Chen, Ahmed Kotb, Becky Skidmore 30 Bond Street, Toronto ON, M5B 1W8 Table of Contents Ontario Drug Policy Research Network Ontario Drug Policy Research Network List of Exhibits Ontario Drug Policy Research Network Ontario Drug Policy Research Network Background
Dementia is a syndrome characterized by a decline in cognitive capacities that results in impairment in
function, but not in alertness or attention. Common types of dementia are Alzheimer's Disease (AD),
Lewy body dementia (DLB) and vascular or frontotemporal dementia (1). It is largely seen in individuals
later in life, and more specifically in populations aged 65 years or older where prevalence is estimated to
be 9 to 13% (2). A recent cohort study of Ontario, Canada home care clients revealed that as many as
22% had a diagnosis of dementia (3). Behavioural and psychological symptoms of dementia (BPSD) may
include depression/dysphoria, anxiety, irritability/lability, agitation/aggression, apathy, aberrant motor
behavior, sleep disturbance and appetite/eating disturbance, delusions and hallucinations, and
disinhibition and elation/euphoria (4).
Significant financial, physical or emotional burden may accompany the care of people with dementia, and most forms have no known cure (5). Drug and non-drug treatment of dementia focuses on maintenance of function and the well-being of patients and their caregivers (6). A variety of drugs are used on and off-label for individuals with behavioural and psychological symptoms of dementia, yet clinicians continue to struggle to find the ‘right' pharmacologic treatment. Atypical antipsychotics (AAP) are increasingly being used to control symptoms of dementia in both community and long-term care settings, however, the use of these medications is controversial given that benefit has not been definitely established and any perceived efficacy may need to be counterbalanced by a potential increase in adverse events. In addition, as many as eight of the nine AAPs licensed for use in Canada may be prescribed off-label to combat symptoms of dementia, and there is concern about the appropriate use of this class of drugs in this vulnerable population. Objectives
The objective of this review is to summarize comparative data on the efficacy, effectiveness, and safety
of atypical antipsychotics in both community and long-term care settings. The review is intended for a
broad audience and aims to inform provincial funding agencies, clinicians, patients and their care
providers so that they make informed choices about the use of AAPs in older adults with BPSD.
Research Questions
1. What is the efficacy and safety of AAPs for the treatment of the behavioural and psychological symptoms of dementia in older adults? 2. Does the efficacy or safety of atypical antipsychotics differ in those who live in community settings when compared to those in long-term care? Inclusion Criteria
Studies were eligible for inclusion in the systematic review if they satisfied specific Population,
Intervention, and Comparators of interest (the PIC portion of the PICO Statement) describ
Ontario Drug Policy Research Network Ontario Drug Policy Research Network including the study design of interest. Studies were not excluded based on outcomes reported; however, data were only extracted for those studies reporting outcomes of interesto language restrictions were set. Ontario Drug Policy Research Network Exhibit 1: Eligibility Criteria for Study Inclusion
Study Population
Older adults (≥ 65 years of age) with BPSD; all forms of diagnosed dementia were included including Alzheimer's Disease, Vascular, Lewy Body and Mixed Dementia. Index Node
Comparisons
AAPs: aripiprazole, asenapine, clozapine, lurasidone, olanzapine, paliperidone, quetiapine,
risperidone, ziprasidone.
AAPs vs. placebo Head to head comparisons of the AAPs Active-controlled trials comparing AAPs to any other medication Fixed and flexible doses as well as all routes of administration Study Design
Randomized controlled trials (RCTs). No limits were placed on study duration or patient follow-up. Exclusions
• RCTs with < 10 participants • Studies not conducted in humans • Herbal comparators • Discontinuation studies Only data from the first period of crossover designs were included and, similar to the parallel studies, analysis of results from the first period of the crossover studies was conducted where data were available. Exhibit 2: Efficacy and Safety Outcomes of Interest
Outcomes of Interest
EFFICACY: Scales addressing the following
SAFETY: The following four (4) safety and
five (5) categories were considered: adverse event outcomes were considered: • All-cause mortality** • Psychosis, agitation, and • Individual treatments and pooled aggression subscales • Caregiver burden • Global measures/Impressions • Extrapyramidal Symptoms (EPS) • Weight change • Activities of Daily Living * NMAs of BPSD subscales were part of sensitivity analyses ** A pooled all-cause mortality NMA was completed as part of a sensitivity analysis Ontario Drug Policy Research Network Literature Search Strategy
The strategy used to build the evidence base for the efficacy and safety of AAPs in older adults with
BPSD consisted of three fundamental steps:
1) We searched for a comprehensive, well-conducted and recently published (within 5 years) evidence synthesis that met the PICO requirements laid out in our inclusion criteria; 2) The search strategies of eligible evidence syntheses were made available to an experienced medical information specialist (IS) who appraised each for appropriateness of literature sources, breadth, potential bias, and barriers that could limit the ability for updating the search. In parallel, two clinical reviewers assessed the quality of eligible evidence syntheses using the "Assessment of Methodological Quality of Systematic Reviews" (AMSTAR) instrument and critically appraised the methodology of particular components of the review in further detail to ensure completeness of included studies. Reviews that both the IS and the clinical review team deemed to be comprehensive in both search and review methodology were used to form the existing evidence base for the systematic review. Included studies from selected evidence syntheses were identified and imported in full-text format into an online systematic review software tool, DistillerSR (7). Use of the online tool by the review team maximizes efficiency in the review process and facilitates consistency across reviewers for literature screening, selection and data extraction. Individual studies were independently screened using the project eligibility criteria prior to inclusion in this review. 3) A medical IS experienced in evidence syntheses, and specifically with network meta-analyses, designed an search strategy to augment the available randomized evidence extracted from the evidence syntheses. Using the OVID platform, we searched Ovid MEDLINE®, Ovid MEDLINE® In-Process & Other Non-Indexed Citations, Embase Classic+Embase and PsycINFO on October 10, 2014. CENTRAL in the Cochrane Library on Wiley on the same date was also searched. Strategies utilized included a combination of controlled vocabulary (e.g., Dementia, Clozapine, and Risperidone) and keywords (e.g., Alzheimer, atypical antipsychotics, Clozaril). Vocabulary and syntax were adjusted across the databases. A validated filter to identify randomized controlled trials was also employed during the search. Additional references were sought through hand-searching of appendices and reference lists. A grey literature search of relevant databases and web sites was also performed using resources listed in CADTH's Grey Matters Light All citations were imported into electronic reference management software (EndNote X7) (8). The complete search strategy is reported in Appendix A. Eligibility and Study Selection
Selection eligibility criteria were applied to each title and abstract by two independent review authors in
Ontario Drug Policy Research Network a standardized method using electronic tools customized to the project in DistillerSR. Any uncertainties were resolved by discussion and, if required, consensus was reached with a third review author (SK or GAW). Attempts were made to obtain all studies that meet the selection criteria in full-text format. The eligibility crit were then applied, and a final decision was made for inclusion. The reviewers were not blinded as to the study authors or centre of publication prior to study selection because this can complicate the review process and only weak evidence suggests that this would improve the results. Exhibit 3: Eligibility Criteria for Full-Text Article Screening
CATEGORY
INCLUSION
EXCLUSION
Databases
OVID Medline, Ovid MEDLINE® In-Process and Other databases not recommended by IS
other non-indexed citations, Embase Classic,
Embase, and PsychINFO
Clinicaltrials.com, Health Canada and US Food Other grey literature sources Literature
and Drug Administration Center for Drug Evaluation and Research website for medical and statistical reviews of individual drug products, CADTH Grey Matters Light. Languages
Populations
Adults ≥ 65 years of age years with BPSD Psychosis or symptoms unrelated to BPSD It is possible that RCTs will include a broader Other psychiatric conditions. age group of dementia patients. In this case, RCTs will meet the population inclusion Population included in study arms with a mean criteria if the mean age of participants is ≥ 65. age under 65. Study Design
Letters, editorials, or any other publications that have no original data. RCTs, Health Canada, US Food and Drug Administration reports, labels and warnings Publication of original study data in abstract Crossover RCTs must report first period treatment results for efficacy. If not, they will be formally included but no study data will be analyzed. Ontario Drug Policy Research Network As specified in the PICO statement All interventions or comparators not explicitly identified in the PICO. Behavioural or non-drug interventions Complex interventions where interventions and comparators are only one component of many. Duration
No exclusions on study duration. Outcomes
Any – Inclusion decisions for primary studies No exclusions based on outcome reporting will not be made based on the outcomes reported. Studies reporting the safety and efficacy outcomes identified in the study protocol will be analyzed. Data Management and Extraction
All information was extracted using standardized data extraction forms developed specifically for the
review in consultation with clinical experts. All forms were piloted in a calibration exercise prior to data
extraction using a random sample of five articles. Data was extracted by a single review author and
checked for accuracy by a second independent review author. Specifically, the following data were
extracted from included RCTs:
1. Study characteristics (Design, setting, funding); 2. Population Characteristics (Age, sex, ethnicity, diagnosis); 3. Eligibility and exclusion criteria 4. Study medications: a. Interventions (dose, durations, route of administration) b. Comparators(dose, durations) c. Concomitant medications allowed 5. Numbers screened, eligible, enrolled, lost-to-follow-up; 6. Methods of outcome ascertainment; and, 7. Results for each outcome of interest at end of study. The primary publication for each unique RCT was used as the principle source for data extraction. Where companion publications were located for a unique RCT, the most recently adjudicated data for each outcome specified a priori was extracted. Studies included from existing evidence syntheses went through de novo data extraction process following identical methods and procedures as articles identified in the updated literature search. Data from all studies, regardless of source, were abstracted into a single dataset Ontario Drug Policy Research Network Extraction of Efficacy and Safety Outcomes
Data were extracted for seven continuous outcomes of interest (BPSD overall and three subscales
Caregiver Burden, Global Measures/Impressions, Cognition, Activities of Daily Living, EPS, Weight
Change) and two dichotomous outcomes (all-cause mortality, falls).
Continuous data was extracted in the form of change scores (from baseline to endpoint for each trial arm) along with the associated measure of variation (standard deviation or error). Dichotomous data were extracted by the frequency of events for all-cause mortality and the number of patients who experienced at least one event for falls. For safety outcomes (EPS, weight change, all-cause mortality, falls), we extracted the safety population (number of patients who took at least one dose of medication). If a sample size for a change score or safety population was not provided, the number randomized was extracted. If studies reported a change from baseline score using more than one scale within a single outcome category, we preferentially extracted scales reporting changes in ‘total' scores from baseline to endpoint over any sub-scale value. Data Synthesis
The data were first summarized descriptively. In preparation for statistical analysis, we imputed missing
values using methods described in the Cochrane Handbook for Systematic Reviews of Interventions (6).
Specifically, when included studies did not provide a change score standard deviation, it was imputed
using a correlation coefficient of ρ=0.8.
Clinical experts were consulted to determine a strategy for combining data reported by different measurement scales across single outcome categories. Experts advised on the appropriateness of merging data from various outcome measurement scales and recommended a methodology based on the outcome most representative of older adults with BPSD given the data reported by the included studies. In order to limit the level of clinical diversity, trials that were very short (e.g. 24 hours) and very long (e.g. 6 months) were eliminated from all primary analyses (81% of trials were between five and twelve weeks long, inclusive). Study arms comparing various fixed doses of the same drug were pooled together to form a single composite ‘flexible-dose' arm for analysis. For example, a four-arm trial comparing placebo to three fixed doses of risperidone (0.5, 1.0 and 2.0 mg/d) was collapsed into two arms for analysis: placebo (Arm 1) and risperidone, oral tablet, 0.5 -2mg/d (Arm 2). Routes of administration were not analysed separately. Statistical Analysis
When data were available, sufficiently similar, and of sufficient quality, Bayesian network meta-analyses
(NMAs) were conducted using WinBUGS software (MRC Biostatistics Unit, Cambridge, UK) (9, 10) for
Ontario Drug Policy Research Network each efficacy and safety outcome, specified a priori. Both fixed- and random-effects NMAs were conducted. Model fit for Bayesian analyses was based on the Deviance Information Criterion (DIC) and comparison of residual deviance to number of unconstrained data points (11-14). Selection of the model/measure depended on the outcome of interest and the availability of data. Heterogeneity across trials in terms of patient characteristics, trial methodologies, and treatment protocols was carefully assessed. We also considered sensitivity analyses including removal of studies from the network of therapies that were not scored as being of high quality. We formally (14) and informally assessed consistency between direct and indirect evidence by comparing direct estimates obtained from pair-wise meta-analysis with estimates from the Bayesian network meta-analysis (12, 13). Model diagnostics including trace plots and the Brooks-Gelman-Rubin statistic were also assessed to ensure model convergence. At least two chains will be fit in WinBUGS for each analysis, each employing at least 40,000 iterations, with a burn-in of at least 20,000 iterations (10, 12). For continuous outcomes, the effect size was expressed in terms of the mean difference (MD) when change scores from the same scale were analyzed together (e.g., cognition) and standardized mean difference (SMD) when change scores from a variety of scales were pooled together for analyses. Effect estimates for dichotomous outcomes (all-cause mortality, falls) are reported using odds ratios with 95% credible intervals. The methods and procedures followed were those developed by the Canadian Collaboration for Drug Safety, Effectiveness and Network Meta-Analysis (ccNMA), funded by the Drug Safety and Effectiveness Network (DSEN) of the Canadian Institute of Health Research. A protocol was developed using guidance
from the PRISMA Statement and following the methods and procedures outlined in the Cochrane
Handbook for Systematic Reviews for Interventions. It was peer-reviewed by clinical experts and those
with expertise in pharmacology, statistics, and systematic review methodology. The protocol was posted
publicly to solicit stakeholder feedback.
Assessment of Heterogeneity
We assessed data for both clinical and methodological diversity. Clinical diversity was assessed by
checking that the participants, interventions, and comparators were not too different from each other
such that combining them was not appropriate. Methodological diversity was assessed by checking that
the studies were similar in terms of study design.
Staircase Diagrams
Staircase diagrams have been assembled to present results for mean differences (continuous outcomes)
and odds ratios (dichotomous outcomes) generated by the network meta-analyses (random-effects
model) of the various treatment strategi
provides a guide to the interpretation of the results in each staircase diagram presented in the Ontario Drug Policy Research Network results section.

Exhibit 4: Interpretation of Results Presented in a Staircase Diagram of Mean Differences and Odds
Ratios
Treatment
Mean difference or odds Treatment
ratio of treatment 2 compared to treatment 1 Mean difference or odds Mean difference or odds Treatment
ratio of treatment 3 ratio of treatment 3 compared to treatment 1 compared to treatment 2 Mean difference or odds Mean difference or odds Mean difference or odds Treatment
ratio of treatment 4 ratio of treatment 4 ratio of treatment 4 compared to treatment 1 compared to treatment 2 compared to treatment 3
Mean difference= MD (standard deviation), Odds ratio= OR (95% Credible Interval)
Note: Bolded numbers in the table indicate statistical significance


Ontario Drug Policy Research Network Selection of Previously Published Evidence Syntheses
We identified two comprehensive, well-conducted and recently published (2010 and 2011) syntheses of
available randomized evidence on the efficacy and safety of AAPs in older adults with BPSD (15, 16):
1. Drug Class Review: Atypical Antipsychotic Drugs: Final Update 3. Produced by the Oregon
Health & Science University (A review completed for the Drug effectiveness review Program (DERP) available at: http://www-ncbi-nlm-nih gov.proxy.bib.uottawa.ca/books/NBK50583/pdf/TOC.pdf. 2. Off-Label Use of Atypical Antipsychotics: An Update. Comparative Effectiveness Review
Number 43. Prepared by the Southern California Evidence-based Practice Centre for the Agency for Healthcare Research and Quality (AHRQ). Available at: Both evidence syntheses reported on efficacy and safety of atypical antipsychotics across a wide range of indications (e.g., schizophrenia, dementia, eating disorders), however, only studies meeting our eligibility criteria were included in this review (i.e., those reporting on populations of patients with BPSD). Based on the results of the appraisal by the IS and the clinical review team, the AHRQ report was chosen as the primary evidence synthesis to form the basis for the existing RCTs. A total of 42 unique RCTs and seven companion articles were imported directly into DistillerSR for screening of full text. To ensure robustness of the evidence base, we additionally incorporated 11 unique and 12 companion articles from the DERP review (16). Search Results
The literature search update (2009 to present) returned a total of 435 unique database abstracts and 68
grey literature documents (Exhibit 4). Following a review of the titles and abstracts 378 were excluded
as they did not meet our eligibility criteria or were duplicates. A total of 129 articles were assessed using
the full-text publication: Thirty-five articles from the database search, 72 articles identified from the
AHRQ and DERP reviews and 22 abstracts that could not be located that were moved to the full-text
screening level to ensure comprehensiveness. Thirteen articles became available during various stages
of the literature selection process, and were reviewed as they were obtained. Nine articles remained
unavailable for assessment. No grey literature documents were included.
Following a detailed assessment of the full text publications, a total of 37 unique RCTs were included along with 26 companion publications. Ontario Drug Policy Research Network Exhibit 5: Literature Search and Study Selection Process
Characteristics of Included Studies
All studies were published between 1995 and 2014. Study and patient characteristics for 32 studies that
reported at least one outcome of interest are reported below. A summary of study characteristics is
located in Appendix B.
Five (17-21) of the 37 included studies either failed to report an efficacy or safety outcome of interest,
or failed to report data from the first phase of a cross-over study. No data were extracted for these
studies (Appendix C).
Duration of Treatment
The most common study duration (baseline to last day of double-blind treatment) was six weeks (9/32
or 28%) (22-30). Five studies each lasted eight (31-35), ten (36-40) or twelve (41-45) weeks (16%). Two
studies (46, 47) were five weeks long. Three studies had a duration that lasted more than one month:
Ontario Drug Policy Research Network one lasted six months (48), another lasted six and a half months (49), and the longest RCT included for review lasted 9 months (50). Three other studies had a duration of two weeks or less: two studies lasted 24 hours (51, 52), and one study lasted two weeks (53). Interventions and Comparators of Interest
Of the nine AAPs of interest, only four were identified in the 32 included studies reporting outcomes of
interest (aripiprazole, olanzapine, quetiapine, and Risperidone). Risperidone, quetiapine, and olanzapine
intervention arms were distributed equally across the included studies and aripiprazole was least
frequently used as an intervention.
AAPs were compared to placebo in 19 RCTs (23, 26, 27, 29, 30, 35-41, 43, 44, 48-52), to another AAP in 14 RCTs (23, 25, 27, 30, 31, 34, 36, 37, 39, 43, 50-53), and to an active comparator in 13 RCTs (24, 28, 33, 34, 38, 41, 42, 45-47, 49, 51, 54). Seven studies compared the efficacy and safety of different fixed doses of the same type of AAP. Four compared risperidone to olanzapine (23, 34, 36, 53), and one study each compared risperidone to quetiapine (31) and risperidone to olanzapine and quetiapine (25). A single study compared the efficacy and safety of different formulations of the same drug (25). When an AAP was compared to an active comparator, eight unique drugs were identified across the 32 included studies reporting outcome data. Five compared an AAP to haloperidol (38, 41, 42, 46, 47), three compared an AAP to one of three selective serotonin reuptake inhibitors (SSRIs) (citalopram (45), escitalopram (24), and fluvoxamine (33)). Other active comparators included rivastigmine (n=2) (28, 49), the anticonvulsant topiramate (n=3) (54), benzodiazepine (lorazepam, n=1) (51), or a typical/first generation antipsychotic (promazine, n=1) (34). Doses and Routes of Administration
AAPs were administered in a number of different doses that fell within the categorization of (titration up
to) ‘fixed' dose, or ‘flexible' dose ranges. Evidence was available for the following methods of
administration: oral tablet, capsule, solution, and intramuscular injection (IM). The route of
administration was not specified in six of the included RCTs (23, 29, 30, 36, 40, 49)summarizes
the clinical doses and routes employed in the AAP intervention arms of the 32 studies that reported
efficacy and safety outcomes of interest.
Exhibit 6: Dose Categorization for AAPs in the Treatment of BPSD in the Elderly
FIXED DOSES REPORTED
FLEXIBLE DOSES REPORTED
(Routes of administration)
(Routes of administration)
Risperidone
0.5 mg/d (oral tablet)
0-2.5 mg/d (oral capsule)
1 mg/d (oral tablet, k=2; oral capsule)
0.5-1.5 mg/d (oral tablet)
2.0 mg/d (oral tablet)
0.5–2 mg/d (oral tablet, k=2; oral capsule, k=2;
oral, k=2; oral solution; route unspecified) 0.5-4 mg/d (oral, oral solution, and route
unspecified)
1-2 mg/d (oral)
Ontario Drug Policy Research Network FIXED DOSES REPORTED
FLEXIBLE DOSES REPORTED
(Routes of administration)
(Routes of administration)
Quetiapine
100 mg/d (oral tablet)
0-200 mg/d (oral capsule)
200 mg/d (oral tablet)
25-225 mg/d (oral)
25-400 mg/d (oral)
25- 600 mg/d (oral capsule)
50-100 mg/d (route unspecified)
50-300 mg/d (oral tablet, k=4; two of which
were either extended release (XR) and
immediate release (IR) oral tablets)
50-400 mg/d (oral)
Olanzapine
1 mg/d (oral capsule)
0-17.5 mg/d (oral capsule)
2.5 mg/d (oral capsule; IM after 2 hours)
2.5-10 mg/d (oral; route unspecified, k=2)
4.71 mg (average oral capsule dose at start of
5-10 mg/d (oral)
study period II) 5 mg/d (oral capsule; tablet; IM after 2 hours)
7.5 mg/d (oral capsule)
10 mg/d (oral tablet)
15 mg/d (oral tablet)
2 mg/d (route unspecified)
2-15 mg/d (route unspecified)
5 mg/d (IM; route unspecified)
10 mg/d (IM; route unspecified)
15 mg/d (IM)
1.75 mg (average oral capsule dose at start of
0.5-2 mg/d (oral tablet)
Comparator:
study period II) 0.5-4 mg/d (oral; oral solution)
Haloperidol
0.5- 12 mg/d (oral capsule)
1-6 mg/d (oral capsule; oral tablet)
All Other
Rivastigmine, 6 mg/d (oral capsule)
Rivastigmine, 6 - ≥ 9 mg/d (route unspecified)
Escitalopram, 10 mg/d (oral tablet)
Citalopram, 10-40 mg/d (oral capsule)
Comparators
Lorazepam, 1.0 mg (after 2 hrs IM)
Topiramate, 25-50 mg/d (oral tablet)
Fluvoxamine, 25-200 mg/d (oral)
Promazine, 50-100 mg/d (oral)
Study Participants and Setting
Analysis of the patient characteristics from the 32 included studies reporting outcome data of interest
revealed an elderly, mostly female, patient population. Overall, 68% of participants were female and the
mean age of all study participants was 80 ± 8.5 (SD) years.
As illustrated in Error! Reference source not found., the three most common types of dementia
diagnosed in study participants were Alzheimer's Disease (AD), vascular and mixed dementia (a
combination of AD and VD). Most of the participants had AD (87.2% of participants).
Ontario Drug Policy Research Network Exhibit 7: Breakdown of Dementia Diagnosis across included studies
* Includes a diagnosis of viral encephalitis, subdural hematoma, multi-infarct dementia, fronto-temporal lobe dementia, dementia syndrome, or Lewy Body disease (number of patients in each group could not be determined) One study (23) reported that all 29 participants had been diagnosed with AD, however, the criteria used to form that diagnosis was not reported. Of the remaining 31 studies that did report this information, five different types of criteria were used. These included: 1) the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV) and its ‘text revision' form (DSM-IV-TR), the 2) National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA), 3) the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10), as well as the 4) National Institute of Neurological Disorders and Stroke and Association Internationale pour la Recherché et l'Enseignement en Neurosciences. Finally, one study (35) confirmed a diagnosis of AD in its study participants using criteria as defined by Jest and Finke (2000) (55). Just under half of included studies (49%) used the DSM-IV exclusively to make a dementia diagnosis, while six (19%) studied used criteria set out by both the DSM-IV and the NINCDS-ADRDA to assess this type of baseline diagnosis in participants. In two studies (22, 46) the ICD-10 criteria were used exclusively to form a dementia diagnosis in participants. Across all studies, participants were based in three types of care settings: Long term care (nursing homes, assisted living facilities, and long term care facilities), hospital (inpatients), and the community (outpatients, participants living with a caregiver, or those living in their own homes). The most common RCT setting was long term care (34%). Three studies each reported that participants resided only in hospitals (24, 33, 46), or the community (31, 32, 48). Four studies did not define the care setting (22, 23, 26, 34). Ontario Drug Policy Research Network Results Overview: Efficacy and Safety Outcomes
A set of network meta-analyses were conducted for five efficacy outcomes (including three subscale
BPSD outcomes), and four safety outcomes.
Efficacy
Network meta-analyses were conducted for BPSD, three BPSD subscale measures (psychosis, aggression,
and agitation), Global Measures/Impressions, Cognition, Activities of Daily Living, and Caregiver Burden.
The choice of these outcomes for NMA was based on their importance and the sufficiency of the data
available to derive robust network models.
provides an overall summary of the NMA results for both head-to-head and placebo comparisons across the AAPs and haloperidol for the five efficacy outcomes at end of study. Exhibit 8: Comparisons Across AAPs, Placebo, and Haloperidol for the Efficacy Outcomes: BPSD, Global
Measures/Impressions, Cognition, Activities of Daily living, and Caregiver Burden at End of Study
Risperidone
Olanzapine
Quetiapine
Haloperidol
Risperidone
Olanzapine
Quetiapine
Haloperidol
The five contiguous circles correspond, from LEFT to RIGHT (respectively) to five efficacy outcomes: BPSD (overall), Global Measures/
Impressions, Cognition, Activities of Daily Living, and Caregiver Burden

A green circle indicates that the "row" AAP is significantly (statistically) better compared with the "column" AAP
A red circle indicates that the "row" AAP is significantly (statistically) worse compared with the "column" AAP
An open circle indicates that there is no statistically significant difference between the "row" and "column" AAP
A missing circle indicates that the outcome was not available for analysis
Results for each of the five outcomes reportedare discussed separately below. Behavioural and Psychological Symptoms of Dementia (BPSD): Overall and Sub-scale Analyses
Two NMAs were completed for BPSD: 1) efficacy of individual AAPs as measured by total BPSD score on
the outcome scale, and 2) efficacy of individual AAPs using three BPSD subscales: psychosis, agitation,
and aggression.
The results for each are presented below.
Behavioural and Psychological Symptoms of Dementia (BPSD): Overall Analyses
Nearly all studies (30/32 or 94%) that reported an outcome of interest also reported at least one
outcome scale measuring change in BPSD. A variety of scales were reported; however, to preserve
Ontario Drug Policy Research Network clinical homogeneity, only BPSD outcomes reported using th Neuropsychiatric Inventory (NPI), Neuropsychiatric Inventory- Nursing Home version (NPI-NH), and NPI-NH version 2 were considered for the NMA. Of the eighteen studies reporting these scales, only data from fifteen studies were analyzed. One study was excluded because of study duration (14 days) (53), and another due to insufficient data (22). Data from a five week trial were also excluded because of the intervention (immediate vs. extended release quetiapine (25)). Following the NMA for BPSD, the standardized mean difference scores generated were converted back to the NPI-NH scale for refshows the most frequent comparator was placebo and the most frequent AAP was risperidone. Exhibit 9: Geometry of the Evidence Network for BPSD (Total)
NMA results for BPSD Ontario Drug Policy Research Network show that haloperidol was the only treatment to significantly decrease NPI-NH scores when compared to placebo (random-effects model) (-5.46 (SD 2.37)). Lower scores indicate total improvement in BPSD. None of the other treatments, when compared against each other, showed significant improvement in behavioural symptoms. Ontario Drug Policy Research Network Exhibit 10: BPSD- Mean Differences (SD) in NPI-NH Scores for All Treatment Comparisons (Random-
Effects Model)
Risperidone
Olanzapine
Quetiapine
Haloperidol
Fluvoxamine
Topiramate

Behavioural and Psychological Symptoms of Dementia (BPSD): Sub-scale Analyses
Three separate NMAs were conducted for psychosis, agitation, and aggression subscales. Results are
presented below for each.
Behavioural and Psychological Symptoms of Dementia (BPSD): Psychosis
Fourteen studies (44%) reported patients' change from baseline on at least one psychosis subscale. As the most common subscales reporting psychosis outcomes were the NPI and NPI-NH, our NMA was limited to studies reporting psychosis using these subscales (k=6). Importantly, while one additional study (47) reported psychosis outcomes using the NPI-psychosis subscale, the questions that formed the basis of their measurement differed slightly from the other six. Accordingly, in an effort to preserve clinical homogeneity, data from this study was excluded from the analysis. The geometry of the evidence network for psychosis is shown in Ontario Drug Policy Research Network Exhibit 11: Geometry of the Evidence Network for BPSD- Psychosis
The NMA for psychosis showed no significant mean differences in any treatment compared to placebo,
or when AAPs were individually compared head-to-head (random effects model) ).
Exhibit 12: BPSD- Psychosis- Mean Differences (SD) in NPI/NPI-NH Scores for All Treatment
Comparisons (Random-Effects Model)
Risperidone
Olanzapine
Quetiapine
Behavioural and Psychological Symptoms of Dementia (BPSD): Aggression
A total of six studies (19%) reported a change in participants' level of aggression from baseline to end of study using at least one type of aggression subscale. Only data from studies using the BEHAVE-AD- Ontario Drug Policy Research Network aggression subscale, when sufficient for analysis, were analyzed within our NMA for this outcome (k=4). The most frequent comparator was placebo and the most frequent AAP was risperidone . Exhibit 13: Geometry of the Evidence Network for BPSD- Aggression
provides a summary of the NMA results for aggression. None of treatments, when compared against each other, showed significant improvement in aggression symptoms. Ontario Drug Policy Research Network Exhibit 14: BPSD-Aggression- Mean Differences (SD) in the BEHAVE-AD Aggressiveness Subscale for All
Treatment Comparisons (Random-Effects Model)
Risperidone
Haloperidol
Behavioural and Psychological Symptoms of Dementia (BPSD): Agitation
Fifty-three percent of studies (k=17) reported a change in participants' level of agitation from baseline to end of study, as measured by at least one overall scale or subscale. Of the scales reported, the CMAI was the most common. As such, only data from those studies using the CMAI that was ‘complete' (incorporable within the NMA), were analyzed in the baseline analysis (k=6). The geometry of the evidence network for agitation is shown in Ontario Drug Policy Research Network
Exhibit 15: Geometry of the Evidence Network for BPSD- Agitation
NMA results for agitation Ontario Drug Policy Research Network show that rivastigmine was the only treatment to significantly increase CMAI scores (equivalent to an increase in agitation) when compared to risperidone [(random effects model, MD 10.13 (SD 5.09)]. Caution is advised when interpreting these results, as data for rivastigmine was available from only one study (49) and effect variations may exist that are difficult to account for. None of the other treatments, when compared against each other, showed significant improvement in, or worsening of, agitation symptoms. Ontario Drug Policy Research Network Exhibit 16: BPSD- Agitation- Mean Differences (SD) in the CMAI Scale for All Treatment Comparisons
(Random-Effects Model)
Risperidone
Quetiapine
Haloperidol
Global Measures/Impressions
Fourteen studies (44%) reported at least one global impression outcome measure. To preserve clinical homogeneity, only change scores reported using the Clinical Global Impression - Severity (CGI-S) subscale were considered (7 RCTs). Two studies were excluded from analysis. One study (53) was excluded due to short trial duration (14 days) and another due to the interventions used (immediate vs. extended release quetiapine (25)). As shown the most frequent comparator was placebo. Exhibit 17: Geometry of the Evidence Network for Global Measures/Impressions
The NMA results for Global Measures/Impressions showed no significant mean differences in any treatment compared to placebo or any other active comparators in the random-effects model. Ontario Drug Policy Research Network Exhibit 18: Global Measures/Impressions: Mean Differences (SD) in CGI-S Scores for All Treatment
Comparisons (Random Effects Model)
Risperidone
Olanzapine
Quetiapine
Haloperidol
Ontario Drug Policy Research Network Cognition
Twenty studies reported at least one cognition outcome of interest. The Mini Mental State Examination
(MMSE) was the most commonly reported scale for this outcome (n=17 RCTs) and the analysis was
limited to studies reporting this scale . Data from thirteen studies were included for analysis.
One study was excluded because of short duration (24 hours) (52) and two others were excluded due to
insufficient data (31, 43). A third study (25) was excluded due to the interventions used (immediate vs.
extended release quetiapine).
Exhibit 19: Geometry of the Evidence Network for Cognition
The NMA results for the Cognition outcome Ontario Drug Policy Research Network showed no significant mean differences in any treatment compared to placebo or any other active comparators in the random-effects model. Ontario Drug Policy Research Network Exhibit 20: Cognition: Mean Differences (SD) in MMSE Scores for All Treatment Comparisons
(Random- Effects Model)
Risperidone
Olanzapine
Quetiapine
Haloperidol
Fluvoxamine
Topiramate
Activities of Daily Living
Four studies (33, 38, 46, 50) reported baseline to end of study differences in Activities of Daily Living.
Data were reported using four different scales: the Physical Self Maintenance Scale (PSMS), the
Alzheimer's Disease Cooperative Study- Activities of Daily Living Inventory (ADCS-ADL), the Instrumental
Activities of Daily Living Dimension of the Nurses' Observation Scale for Geriatric Patients (NOSGER-
IADL), as well as the Functional Independence Measure (FIM). Given that these scales are unique, we
highlight the ADL items they are each said to measure:
 The FIM assesses thirteen items related to basic activities of daily living (e.g.: eating, grooming, and
bathing, as well as toileting and stair climbing), but also includes an assessment of five items related to socio-cognition (comprehension, expression, social interaction, problem solving, and memory).  The PSMS measures physical functioning across six different basic ADL items: toileting, feeding,
dressing, grooming, physical ambulation, and bathing  The ADCS-ADL is a twenty-three item scale that covers patients' abilities to perform basic (e.g.:
eating, bathing, and toileting) to more complex (e.g. shopping and using the telephone) ADL activities.  The NOSGER-IADL is a five-item subscale of the NOSGER, which measures patients' ability to
perform what can be described as ‘instrumental ADL outcomes'. These include: following favorite radio or television programs, attempting to keep his/her room tidy, shopping for small items (e.g. groceries), enjoyment of certain events (e.g. visits or parties), and whether or not the patient is orientated when in unusual surroundings. Ontario Drug Policy Research Network
The evidence network diagram for this outcome is shown in
Exhibit 21: Geometry of the Evidence Network for Activities of Daily Living
provides a summary of study level data for the four studies reporting ADL outcomes. Generally, patients treated with AAPs risperidone and olanzapine and active comparators haloperidol and fluvoxamine experienced worsening functional outcomes over the course of study treatment. In one study (38), quetiapine significantly improved ADL outcomes (PSMS scale). In another study (50), patients treated with placebo showed significant improvement in functional symptoms when compared to olanzapine. Ontario Drug Policy Research Network Exhibit 22: Study Level Data: Activities of Daily Living
Study Scale Used to
Change from Baseline (SD)
Study Significance
by Treatment
Teranishi et. al. FIM
Risperidone, oral, 0.5-2.0 mg/d:
No significant within- (2013) (33)
group changes from Fluvoxamine oral, 25- 200 mg/d:
baseline were found. Tariot et. al. PSMS
Placebo, oral capsule:
Those treated with (2006) (38)
Quetiapine, oral capsule,
significantly worse 25mg/day - 600mg/day:
functional outcomes compared to quetiapine Haloperidol, oral capsule,
0.5mgday - 12mg/day:
-1.59 (3.06)
Schneider et. al. ADCS-ADL
Placebo, oral capsule:
(2006) (50)
significantly worse than Risperidone (Risperdal), oral capsule,
placebo at improving 0 -2.5mg/day:
ADL (p< 0.001) -1.1 (8.8)
Quetiapine (Seroquel), oral capsule,
0 - 200mg/day:
-1.0 (7.7)
Olanzapine (Zyprexa), oral capsule,
0 - 17.5mg/day:
-6.1 (8.2)
Savaskan et. al. NOSGER-IADL Quetiapine, oral, 25 - 225mg/day:
Significance not (2006) (46)
discussed. Those treated Haloperidol, oral, 0.5 - 4mg/day:
with quetiapine showed functional improvement. * Higher scores in PSMS and NOSGER-IADL indicate worsening of functional ability, whereas higher scores in ADCS-ADL and FIM indicate improvement. For the sake of comparison, all change scores shown in this table have been adjusted such that higher (more positive) scores indicate improvement in functional ability. Data from each of these four scales (PSMS, FIM, and ADCS-ADL), and one subscale (NOSGER-IADL) were combined together within the NMA performed for this outcome. Preliminary NMA results from this analysis generated standardized mean difference scores from an NMA that were converted back to the PSMS scale for reference. Briefly, the random-effects model NMA for Activities for Daily Living showed that olanzapine significantly decreased PSMS scores (indicating improvement) compared with placebo, risperidone, and quetiapine. PSMS scores increased significantly (indicating a worsening of symptoms) with haloperidol when compared to patients taking olanzapine (mean difference: 5.23, SD 1.85). Ontario Drug Policy Research Network Given the content diversity present across these four scales, as well as the small number of studies that reported data for this outcome, the results of this NMA should be interpreted with caution. The mechanism of action leading to these differences is unclear, and this may be a statistical anomaly and we continue to investigate. Caregiver Burden
Six studies (27, 37, 39, 44, 50, 54) reported baseline to end of study differences in Caregiver Burden.
Four studies (27, 37, 39, 54) reported this outcome using the NPI Part 2 ‘Occupational Disruptiveness of
Caregivers' subscale, one study used the Caregiver Activity Survey (50), another study (44) administered
the Modified Nursing Care Assessment Scale (Total Strain Domain). Only data from the four studies
reporting a caregiver burden outcome using the Occupational Disruptiveness subscale of the NPI were
eligible for the NMA. During the data preparation process, the results from one study (54) were found to
be disconnected from the network, and thus, were excluded from analysis.
The evidence network diagram for this outcome is shown in
Exhibit 23: Geometry of the Evidence Network for Caregiver Burden
The NMA results for the Caregiver Burden show no significant mean differences in any of the comparisons under the random-effects model. Ontario Drug Policy Research Network Exhibit 24: Caregiver Burden: Mean Differences (SD) in the NPI-NH Occupational Disruptiveness Scale
for All Treatment Comparisons (Random-Effects Model)
Olanzapine
Quetiapine
Safety
A set of network meta-analyses were conducted for two continuous (Extrapyramidal Symptoms, or EPS,
and Weight Change) and two dichotomous (falls and all-cause mortality) safety outcomes. Measures of
treatment effect were reported using mean differences and their corresponding standard deviations for
the continuous outcomes scales. The relative effect estimates for the dichotomous outcomes are
presented as odds ratios and 95% credible intervals.
provides an overall summary of the NMA results for both head-to-head and placebo
comparisons across the AAPs and haloperidol for the four safety outcomes at end of study.
Previous reviews have found an association between AAPs and mortality; however the data from those
reports were not readily available and could not be integrated into the NMA until verified. Results here
should be interpreted with caution until these data are verified and appropriately incorporated into the
network.

Exhibit 25: Comparisons Across AAPs, Placebo, and Haloperidol for the Safety Outcomes: Mortality,
Falls, EPS, and Weight Change
Risperidone
Olanzapine
Quetiapine
Haloperidol
Risperidone
Olanzapine
Quetiapine
Haloperidol
The five contiguous circles correspond, from LEFT to RIGHT (respectively) to five efficacy outcomes: Mortality (individual treatments), Falls,
EPS and Weight Change outcomes.

A green circle indicates that the "row" AAP is significantly (statistically) better compared with the "column" AAP
A red circle indicates that the "row" AAP is significantly (statistically) worse compared with the "column" AAP
An open circle indicates that there is no statistically significant difference between the "row" and "column" AAP
Results for each of the four safety outcomes reported inare discussed separately below. Ontario Drug Policy Research Network All-Cause Mortality: Individual Treatment Effects and Pooled Analysis
Two NMAs were conducted to evaluate the safety of AAPs for mortality. The first addressed individual
treatment effects by comparing individual treatments to each other. For the second NMA, a pooled
analysis was completed. Specifically, all AAPs, active comparators, and placebos were grouped together
to form three broad comparator groups.
The results of each of these two NMAs are discussed separately in the following two sub-sections that follow directly. All-Cause Mortality: Individual Treatments Seventeen studies (53%) reported all-cause mortality; however, six studies (25, 29, 49-52) were excluded from analysis. Four studies were excluded due to short or long duration of treatment. Two additional studies were excluded from analysis due to the interventions used (immediate vs. extended release quetiapine (25), and because of a study design change (29). Six of the eleven studies included in the treatment effects NMA reported all deaths occurring within the
trial period (22, 31, 33, 36, 38, 44). Five trials (35, 37, 39, 40, 43) reported all deaths occurring during
treatment period or within 30 days after trial completion. The geometry of the evidence network for this
outcome is sho
Exhibit 26: Geometry of the Evidence Network for All-cause Mortality (Individual Treatments)
Results for the all-cause mortality analysis by all treatments showed no significant increase in mortality in patients treated with placebo, any AAP, or Ontario Drug Policy Research Network other active comparator. Exhibit 27: All-cause Mortality (Individual Treatments): Odds Ratios (95% Credible Intervals) for All
Treatment Comparisons (Random-Effects Model)
Risperidone
Olanzapine
Quetiapine
Haloperidol
Fluvoxamine
All-Cause Mortality: Pooled Analysis In the pooled analysis for all-cause mortality, risperidone, olanzapine, quetiapine, and aripiprazole were grouped together as a single AAP treatment group. Likewise, haloperidol and fluvoxamine were combined to form a single active comparator group. Rainer et al. (2007) (31) was excluded from this analysis as it completed a head-to-head randomized controlled trial comparing the efficacy and safety of risperidone to quetiapine. The combination of these two arms (as individual AAPs) left no comparator group, and thus, data that was unincorporable within the pooled mortality NMA. The most frequent AAP comparator was placebo Ontario Drug Policy Research Network Exhibit 28: Geometry of the Evidence Network for All-cause Mortality (Pooled Analysis)
Results of the pooled analysis for all-cause mortality show a significant increase in the odds of mortality in patients treated with any AAP when compared to placebo. None of the other treatments, when compared against each other, showed a significant increase in the odds of mortality. Exhibit 29: All-Cause Mortality (Pooled Analysis): Odds Ratios (95% Credible Intervals) for All
Treatment Comparisons (Random-Effects Model)
(1.19, 3.16)
Any Active Comparator
Falls
Nineteen studies reported data on falls and eleven were included in the NMA. Three of the eight studies
excluded from the analysis (25, 41, 42) reported falls as an outcome without presenting data that could
be used in the analysis (e.g. "Adverse events occurring in ≥ 10% of patients in any one group were
Ontario Drug Policy Research Network falls…") (41). Three studies (22, 27, 50) reported falls as part of a composite outcome (e.g. all ‘accidental
injuries' reported, which included falls or fractures) and the exact number of falls occurring in each
study arm could not be elucidated. Two studies (52, 53) were excluded due to a short study duration (24
hours (52) and 14 days (53). The geometry of the evidence network for this outcome is sho
Exhibit 30: Geometry of the Evidence Network for Falls
NMA results for falls showed no significant differences in falls in patients treated with placebo, AAP, or any other active comparators under the random-effects model. Exhibit 31: Falls: Odds Ratios (95% Credible Intervals) for All Treatment Comparisons (Random-Effects
Model)
Risperidone
Olanzapine
Quetiapine
Haloperidol
Fluvoxamine
Ontario Drug Policy Research Network Extrapyramidal Symptoms (EPS)
Twenty-two studies reported at least one outcomes scale measuring EPS. A total of six different scales
measuring baseline to endpoint change scores in EPS were reported. Of these, the Simpson-Angus Scale
(SAS) was the most frequently reported (11 RCTs) and studies reporting the SAS were included in the
NMA for this outcome. Data from eight studies were included . Data from two studies (53,
56) were excluded due to duration of study (24 hours and 2 weeks) and one study (31) was excluded
due to insufficient data. The geometry of the evidence network for this outcome is shown in
Exhibit 32: Geometry of the Evidence Network for EPS
Compared to placebo, SAS scores increased significantly in patients treated with haloperidol (MD 3.57 (SD 0.78) . Positive scores indicate worsening EPS while negative scores indicate improvement. SAS scores also increased significantly in the haloperidol group when compared to risperidone and quetiapine. Olanzapine showed significant improvement in EPS when compared to haloperidol (MD-2.86 (SD 0.98)). Exhibit 33: EPS: Mean Differences (SD) in the SAS scale for All Treatment Comparisons (Random-
Effects Model)
Risperidone
Olanzapine
Quetiapine
Ontario Drug Policy Research Network Haloperidol
Weight Change
Nine studies reported a change in participants' mean weight (in kilograms (kg)) from
baseline to end of study and seven studies were included in the NMA. Two studies were excluded from
the analysis, one due to differences in the interventions studied (immediate vs. extended release
quetiapine) (25) and the other (50) due to insufficient data.
Exhibit 34: Geometry of the Evidence Network for Weight Change

The NMA results for weight change showed no significant differences in any AAP when compared to placebo or any other active comparators under the random-effects model. Exhibit 35: Weight Change (Kg): Mean Differences (SD) for All Treatment Comparisons (Random-
Effects Model)
Ontario Drug Policy Research Network Risperidone
Olanzapine
Quetiapine
Haloperidol
Ontario Drug Policy Research Network DISCUSSION
Bayesian NMA allows for inferences into the comparative effectiveness of interventions that have been directly and indirectly evaluated against each other. A meta-analysis, by contrast, allows only for inferences to be made about the comparative effectiveness of interventions that have been directly evaluated against each other. Previous reviews have employed meta-analyses (15, 82-84) in examination of research questions similar to those investigated here. To our knowledge, this is the first NMA providing a comprehensive evaluation of currently available evidence on the efficacy and safety of AAPs in elderly men and women with dementia and BPSD. We conducted a series of NMAs to compare AAPs for their efficacy and safety across nine different outcomes (five efficacy and four safety). Our analyses included data from placebo controlled, head-to-head, and active-controlled trials. While other reviews of multiple AAPs (15, 82-84) have limited their analyses to placebo-controlled trial data, head-to-head trials, or trials with haloperidol as an active control, we included all available active comparators in our analyses, including combination treatments. Efficacy
Behavioural and Psychological Symptoms of Dementia: Overall Analysis
Our results showed that haloperidol significantly reduced behavioural symptoms when measured by
overall NPI scores and compared to placebo. None of the other treatments evaluated, when compared
against each other, showed significant improvement in overall behavioural and psychological symptoms
as assessed using NPI scores.
Evidence relating to the efficacy of AAPs on BPSD differs across existing reviews. At least one drug class review (15) found that risperidone had a significant effect on reducing BPSD in elderly populations with dementia. This review also concluded that aripiprazole was no more effective in reducing BPSD than placebo, whereas Ma et al. (82) found a significant difference in the efficacy of aripiprazole for BSPD versus placebo. We found no significant difference in the efficacy of aripiprazole compared to placebo for BPSD. Further, our analysis revealed no significant difference with regard to the efficacy of quetiapine when compared to placebo in reducing BPSD - supporting conclusions reached by AHRQ. The study by Ma et al. (82), by contrast, concluded that quetiapine was more efficacious in reducing behavioural symptoms in elderly patients with BPSD when compared to placebo. It is worth noting that our analysis approach for BPSD (overall) differed from those of previous reviews. Our analysis of BPSD symptoms was based entirely on overall (or total) NPI scores - the most widely used neuropsychiatric scale measuring behaviour in AD (85). Others (AHRQ included), combined more than one overall BPSD scale in their analyses (e.g. BEHAVE-AD total scores were used in combination with NPI total scores). Ontario Drug Policy Research Network Behavioural and Psychological Symptoms of Dementia: Subscale Analyses
Although BPSD subscales may not be psychometrically-tested, there is a suggestion through related
recommendations from Canadian clinical practice guidelines and expert opinion (86) that these
subscales are likely more sensitive to differences in the individual agents (as opposed to the overall
score).
In general, we found no significant mean differences in any AAP treatment compared to placebo, or
when AAPs were themselves compared head-to-head across psychosis, aggression, or agitation
outcomes. The only significant finding was with respect to agitation, where rivastigmine was found to
significantly increase CMAI scores (indicating a worsening of agitation symptoms) when compared to
risperidone (random effects model, MD 10.13 (SD 5.09). As this finding is based on data from a single
study, it may not be generalizable to the wider study population.
Ma et. al. (2014) (82) meta-analyzed CMAI change scores from placebo controlled AAP trials and found
that CMAI scores for patients taking risperidone significantly improved compared to those who took
placebo. It is worth noting that our analysis approach for this outcome differed from Ma et al. (2014) in
that their analysis incorporated CMAI- aggression subscale change scores within their analysis when
overall scores were not provided by the primary article. By contrast, we restricted our NMA to overall
CMAI scores, which could account for the differences in conclusions reached.
AHRQ (15) completed meta-analyses of placebo controlled AAP trials measuring mean change (from
baseline to study endpoint) in participants' agitation and psychosis symptoms. Again, whereas AHRQ
incorporated a variety of scales and subscales within each of their analyses, we limited our analyses to a single, representative, scale for each outcome. The AHRQ report found that patients who received risperidone showed significantly improved agitation and psychosis symptoms compared to those who took placebo. Our analysis, in contrast, showed no such differences. Indeed, the contrasting approaches taken to perform the analyses could account for the differences in findings across these subscale outcomes. Global Measures/Impression
Our NMA results for global measures/impression showed no significant mean differences in any agent
compared to placebo or any other active comparator. There are two distinct categories of global
impression measurements: severity and degree of change, as judged by clinician interview. In an effort
to preserve clinical homogeneity, we limited our base-case NMA for this outcome to data reported using
the CGI-S (severity scale). This decision resulted in a small number of studies (n=5) being included in our
analysis.

Our results for this outcome may be attributable to differences in included studies when compared to
previous reports of this outcome; however, we believe that the small number of studies best reflects the
evidence base for CGI-S. Ma et al. (82) analyzed both change and severity scales of the CGI for this
outcome compared to placebo and found significant improvement in patients assigned to aripiprazole
Ontario Drug Policy Research Network and risperidone in five included studies. AHRQ (15) did not report this outcome. We included only 3 of the 5 studies included in the Ma et al. analysis, in addition to 2 other studies not included in Ma et al. One study (30) was excluded from analysis due to study design (switch to open label) and another (43) due to missing end-of-study data for the CGI-S outcome. Cognition
The results of our NMA showed no significant mean differences in any treatment compared to placebo
or any other active comparators in the random-effects model of 12 included studies. Cognition was not
studied in many previous reviews (15, 82-83); however, Schneider (2006) (84) meta-analyzed MMSE
scores in seven placebo controlled RCTs of four AAPs. They concluded that aripiprazole, olanzapine, and
risperidone significantly worsened cognition as measured by MMSE, although findings for aripiprazole
were based on a single RCT and two studies of risperidone.
Activities of Daily Living and Caregiver Burden
The NMA for Activities for Daily Living is based on a composite pooling of outcomes across four unique
ADL scales. Results showed that olanzapine significantly decreased PSMS scores (indicating
improvement) compared with placebo, risperidone, and quetiapine. PSMS scores increased significantly
(indicating a worsening of symptoms) with haloperidol when compared to patients taking olanzapine
(mean difference: 5.23, SD 1.85).
Results for both ADL and caregiver burden should be interpreted with caution. The outcome scales are
diverse and have differing psychometric properties (ADL). Further, a small number of studies were
available for analyses (ADL and caregiver burden). The mechanism of action leading to differences in
ADL for olanzapine is unclear and we are unsure if these results may be a statistical anomaly. All-cause Mortality: Individual Effects and Pooled Analysis
The increased risk of mortality associated with AAPs in this population is well documented in the
literature, and noted in the individual product monographs. Further, Health Canada has also issued
warnings for these treatments (89). Our pooled analysis of all-cause mortality in patients who took any
AAP compared to those who took placebo or any active comparator showed a similar trend. Specifically,
we found that participants who took any AAP had 1.9 times the odds of death from any cause (95% Crl
1.19, 3.16) compared to those who took placebo.
Previously completed pooled analyses of mortality in patients who took any AAP compared to those who took placebo showed a similar risk increase: OR 1.54, 95% CI 1.06 to 2.23 (83) and OR 1.52, 95% CI 1.06, 2.18 (82). A meta-analysis of RCTs by the US Food and Drug Administration (using data not in the public domain) also suggested a significant increase in mortality (OR 1.7) (88). With respect to individual treatment effects, our NMA showed no significant increase in mortality across patients treated with placebo, any individual AAPs, or active comparator. Ma (82) and Schneider (83) Ontario Drug Policy Research Network also examined all-cause mortality with respect to individual treatment comparisons and found no significant increase in the odds of mortality across any comparison of aripiprazole, olanzapine, quetiapine or risperidone to placebo. Differences were noted, however, in event data reported in studies analyzed in both Ma (2014) (82) and Schneider (2005) (83) that we continue to explore. Specifically, Ma et al. (82) included a study by De Deyn (1999) (41) reporting one death in the risperidone group, and five deaths in the placebo group. Three independent clinical reviewers examined the publication and found no deaths were directly reported in the primary publication. Following a thorough reference check, we located a related pooled analysis (87) reporting mortality data in the De Deyn et al. (41) study; however, we could not confirm results based on the data supplied in the pooled analysis. Both Ma and Schneider et al. (82-83) recorded six deaths in the olanzapine group and no deaths in the placebo group of an RCT by Street (2000) (27). This study was not included in our analysis as mortality data was not explicitly reported in the primary publication or any of its companion articles. References for the mortality data pointed to conference proceedings no longer in print (2004). We continue to endeavor to clarify these results by following up with industry and study authors before we consider including them in our analysis. Falls
We found no significant differences in the number of falls in elderly patients with BPSD and dementia
who were treated with placebo, AAP, or any other active compactor under the random-effects model.
Similar reviews (82) for this outcome also found no important differences in risk of falls in elderly adults
taking olanzapine, quetiapine, or risperidone when compared to placebo. Their pooled analysis of all AAPs compared to placebo also revealed no significant increase in falls. Similarly, our NMA revealed no significant differences across all drug comparisons. It may be worth noting that although our findings agree with regard to significance, we noted that falls data used in Ma et al. (82) were similar to ours with the exception of one study, Deberdt (2005) (36). Specifically, zero falls were recorded in the risperidone group, four in the olanzapine group and two in the placebo group. These numbers do not match with the data extracted from the same study based on percentages provided in the primary publication. Although an increase in falls has been documented in large, observational cohorts of dementia patients, we were unable to confirm these findings with evidence from RCTs. Although RCTs protect well against sources of bias that can be found in their non-randomized counterparts, reporting of adverse events is uneven, and limited by the duration and focus of the studies (90). Additionally, although we identified 19 studies that reported falls, data from only 11 of those studies could be included in the NMA. For example, some studies (25, 41, 42) reported falls as an outcome without presenting data that could be incorporated within our analysis (e.g. "Adverse events occurring in ≥ 10% of patients in any one group were falls…") (41). Three additional studies (22, 27, 50) reported falls as part of a composite outcome Ontario Drug Policy Research Network (e.g. all ‘accidental injuries' reported, which included falls or fractures) thus, the exact number of falls occurring in each study arm could not be elucidated. Extrapyramidal Symptoms
We examined the safety of AAPs with regard to their role in the development of EPS by analyzing
changes from baseline to end of study on the Simpson Angus Scale (SAS). This ten-item instrument
assesses a variety of symptoms, such as: gait, head dropping, salivation, and tremor. We identified three
previously published systematic reviews that also analyzed the relationship between AAPs and EPS (15,
82, 84)), however, these studies used binary or adverse ‘event' data to complete each of their meta-
analyses for this outcome.
Our NMA results for EPS showed no significant differences in any AAP when compared to placebo under
the random-effects model. When compared to placebo, however, SAS scores increased significantly in
patients treated with haloperidol (MD 3.57 (SD 0.78)). SAS scores also increased significantly in the
haloperidol group when compared to risperidone and quetiapine, and olanzapine showed significant
improvement in EPS when compared to haloperidol (MD-2.86 (SD 0.98)). These results are not
unexpected given that the incidence of EPS symptoms resulting in use first generation antipsychotics is
well-documented (91).
Due to differences in the type of data considered and analysis approach, it is difficult to directly compare
our results with those in previous reviews.
Weight Change
Our NMA results for weight change showed no significant differences in any AAP when compared to placebo or any other active comparators under the random-effects model. Our approach using continuous outcome data (mean increase or decrease in weight over the trial period) differed from the AHRQ drug class review (15), which addressed weight using binary event data (number of patients whose weight increased or decreased). As a result, we can only compare our results at a very high level. AHRQ noted that olanzapine significantly increased appetite or weight in patients who took olanzapine compared to those who took placebo. The authors also found a similar significant trend when comparing risperidone to placebo, while we did not reach the same conclusion. We present results for weight change as a safety outcome; however, it is difficult to discuss weight without specifically addressing subgroups of elderly patients who may find either benefit or harm with a change in weight. In the frail elderly, even a small weight loss can be dangerous to their health and validates a safety concern. Conversely, weight gain may be beneficial in this same population. Further weight gain may be a safety issue in obese individuals or those with other medical complications. Additional investigation into subgroups may be warranted for future reviews to clarify the impact of weight change on elderly patients with dementia. Ontario Drug Policy Research Network Community Setting Subgroup Analysis
We continue to explore opportunities for meaningful subgroup analyses with respect to community
setting (long term care and community). The findings from such analyses will be described in our full
technical report.
Conclusions
Efficacy evidence from this review suggests that there is limited benefit from the use of AAPs in elderly
patients with BPSD, although further investigation is warranted to explore differences in the individual
symptom scales, notably for BPSD. Given the concern over increased risk of death noted in the product
monographs and in warnings from Health Canada and the US FDA, clinical judgment is required when
applying these research findings to every day practice.
Ontario Drug Policy Research Network KEY MESSAGES
Efficacy
Behavioural and Psychological Symptoms of Dementia:
 In general, there were no significant differences in the improvement of BPSD with the AAPs (risperidone, quetiapine, olanzapine, aripiprazole) when compared to placebo.  Haloperidol was the only agent significantly better than placebo in improving BPSD (MD: -5.46, SD 2.37) on the Neuropsychiatric Inventory-Nursing Home edition Scale.  In elderly patients with dementia and BPSD, none of the AAPs compared showed significant overall symptom improvements when compared to each other or haloperidol.  Rivastigmine is significantly worse than risperidone in reducing agitation (MD: 10.13, SD 5.09). As data for rivastigmine was available from only one study, caution is advised when interpreting these results.  Generally, there were no significant differences in the improvement of agitation or psychosis with the AAPs (risperidone, quetiapine, olanzapine, aripiprazole) when compared to placebo, each other, or with any other active comparator. Global Impressions/Impressions, Cognition, and Caregiver Burden:
 There were no significant differences amongst the AAPs in the improvement of Global Measures/Impressions, Cognition, or Caregiver Burden outcomes when compared to placebo or any other active comparator. Activities of Daily Living:
 Olanzapine is significantly better than placebo, risperidone, quetiapine and haloperidol for improving Activities of Daily Living outcomes in elderly patients with dementia. The mechanism of action leading to these differences is unclear, and may be a statistical anomaly. We continue to investigate this result. Safety
All-cause Mortality, Falls, and Weight Change:
 In general, there are no significant differences in all-cause mortality, falls or weight change when comparing AAPs to placebo or any other active comparator.  There is an increased risk of all-cause mortality in the elderly with dementia and BPSD who take any AAP compared to those who take placebo (OR: 1.9, CI 1.19, 3.16). Extrapyramidal Symptoms:
 Haloperidol was the only agent to significantly increase EPS symptoms in elderly patients with dementia experiencing BPSD compared to placebo, risperidone, quetiapine, or olanzapine. Ontario Drug Policy Research Network Appendix A: Literature Search Strategy

AAP – Dementia – Elderly
2014 Oct 10 – Final Strategy
Database: Embase Classic+Embase <1947 to 2014 October 09>, Ovid MEDLINE(R) In-Process & Other Non-Indexed
Citations and Ovid MEDLINE(R) <1946 to Present>, PsycINFO <1806 to October Week 2 2014> Search Strategy:
--------------------------------------------------------------------------------
1 exp Dementia/ (420833)
2 (dement* or amentia$1 or pseudodement*).tw. (225806)
3 alzheimer*.tw. (261150)
4 (progressive adj2 aphasi*).tw. (3381)
5 PPA syndrome*.tw. (45)
6 (senile or senility).tw. (43078)
7 (Mesulam* adj1 syndrome*).tw. (7)
8 binswanger*.tw. (1723)
9 (spongiform encephalopath$3 adj1 (subacute or sub-acute)).tw. (392)
10 "Kosaka-Shibayama".tw. (6)
11 ("diffuse neurofibrillary" adj tangle$1 adj5 calcif*).tw. (76)
12 ((frontotemporal or fronto-temporal) adj lobar degeneration).tw. (4967)
13 FTLD.tw. (3945)
14 DDPAC.tw. (12)
15 (Pick$2 adj1 disease*).tw. (6302)
16 (FLDEM or FTD-GRN or FTD-PGRN or FTDP-17 or FTLD-17 GRN or FTLD-TDP or HDDD1 or HDDD2).tw. (1622) 17 "Wilhelmsen-Lynch".tw. (2) 18 ((brain or lobar) adj2 atroph*).tw. (11166) 19 Huntington*.tw. (32829) 20 ("Kluver-Bucy" or "Kleuver-Bucy").tw. (604) 21 (temporal lobectomy behavio?r adj2 syndrome*).tw. (0) 22 (Lewy bod$3 adj2 disease*).tw. (3728) 23 CADASIL.tw. (2253) 24 ((mental* or cognit*) adj2 (declin* or degenerat* or deteriorat* or loss* or losing or lost)).tw. (51865) 25 or/1-24 (585723) 26 (atypical antipsychotic* or atypical anti-psychotic*).tw. (25430) 27 (new generation antipsychotic* or new generation anti-psychotic*).tw. (311) 28 (second generation antipsychotic* or second generation anti-psychotic*).tw. (6293) 29 (2nd generation antipsychotic* or 2nd generation anti-psychotic*).tw. (75) 30 (novel antipsychotic* or novel anti-psychotic*).tw. (2121) 31 (atypical neuroleptic* or new generation neuroleptic* or second generation neuroleptic* or 2nd generation neuroleptic* or novel neuroleptic*).tw. (3997) 32 (aripiprazole or Abilify or Abilitat or Discmelt or HSDB 7320 or Opc 14597 or OPC 31 or OPC-14597 or UNII-82VFR53I78).tw. (8553) 33 aripiprazole.rn. (9825) Ontario Drug Policy Research Network 34 (Asenapine or Org 5222 or EINECS 265-829-4 or UNII-JKZ19V908O or Saphris or Sycrest).tw. (879) 35 Asenapine.rn. (707) 36 Clozapine/ (37170) 37 (Clozapine or "BRN 0764984" or CCRIS 9171 or Alemoxan or Azaleptin or Clopine or Clopsine or Clorazil or Clozapin or Clozapina or Clozapinum or Clozaril or Denzapine or Dorval or Dozapine or EINECS 227-313-7 or Elcrit or Fazaclo or HF 1854 or HF1854 or HSDB 6478 or Iprox or Lapenax or Leponex or Lozapin or Lozapine or LX 100-129 or Sizopin or UNII-J60AR2IKIC or Versacloz or W-801 or Zapen or Zaponex).tw. (28879) 38 Clozapine.rn. (31923) 39 (Lurasidone or Latuda or MK 3756 or MK3756 or SM-13496 or SM13496 or SMP-13496 or SMP13496 or UNII-22IC88528T).tw. (590) 40 Lurasidone.rn. (426) 41 (Olanzapine or Anzatric or Lanopin or Lanzac or LY 170053 or LY170053 or Meltolan or Midax or Olace or Oladay or Olan or Olandus or Olanex or Olansek or Olapin or Olazax or Oleanz or Olexar or Oltal or Olzap or Onza or Ozapin or Psychozap or Relprevv or UNII-N7U69T4SZR or Zalasta or Zelta or Zypadhera or Zydis or Zyprexa*).tw. (22648) 42 Olanzapine.rn. (27851) 43 (Paliperidone or Invega or 9-hydroxyrisperidone or 9-hydroxy-risperidone or 9-OH-risperidone or R 76477 or RO76477 or UNII-838F01T721 or xeplion).tw. (2572) 44 Paliperidone.rn. (1763) 45 (quetiapine or Seroquel or "ICI 204,636" or ICI 204636 or Co-Quetiapine or HSDB 7557 or Socalm or Tienapine or UNII-BGL0JSY5SI).tw. (12555) 46 Quetiapine.rn. (17048) 47 Risperidone/ (35954) 48 (Risperidone or Risperdal or Risperidal or Riperidon or "R-64,766" or R-64766 or Belivon or Consta or Neripros or Noprenia or Pexidone or RN 4891881 or HSDB 7580 or Psychodal or R 64 766 or "R 64,766" or R 64766 or Risolept or Rispen or Risperidona or Risperidonum or Risperin or Rispid or Rispolept or Rispolet or Rispolin or Rizodal or Sequinan or Spiron or UNII-L6UH7ZF8HC or Zargus or Zofredal).tw. (24245) 49 Risperidone.rn. (30703) 50 (Ziprasidone or CP 88059 or CP 88059-01 or "CP-88,059" or "CP-88,059-01" or "CP-88,059-1" or Geodon or HSDB 7745 or UNII-6UKA5VEJ6X or Zeldox or Zeldrox or Ziprazidone or Zipsydon).tw. (5381) 51 Ziprasidone.rn. (7551) 52 or/26-51 (111217) 53 25 and 52 (7414) 54 (controlled clinical trial or randomized controlled trial).pt. (482489) 55 clinical trials as topic.sh. (175785) 56 (randomi#ed or randomly or RCT$1 or placebo*).tw. (1609346) 57 ((singl* or doubl* or trebl* or tripl*) adj (mask* or blind* or dumm*)).tw. (327622) 58 trial.ti. (328390) 59 or/54-58 (2038136) 60 53 and 59 (1165) 61 exp Animals/ not (exp Animals/ and Humans/) (9168750) 62 60 not 61 (1157) 63 (comment or editorial or interview or letter or news).pt. (2894548) 64 62 not 63 (1137) Ontario Drug Policy Research Network 65 64 use prmz (349) 66 exp dementia/ (420833) 67 (dement* or amentia$1 or psuedodement*).tw. (225478) 68 alzheimer*.tw. (261150) 69 (progressive adj2 aphasi*).tw. (3381) 70 PPA syndrome*.tw. (45) 71 (senile or senility).tw. (43078) 72 (Mesulam* adj1 syndrome*).tw. (7) 73 binswanger*.tw. (1723) 74 (spongiform encephalopath$3 adj1 (subacute or sub-acute)).tw. (392) 75 "Kosaka-Shibayama".tw. (6) 76 ("diffuse neurofibrillary" adj tangle$1 adj5 calcif*).tw. (76) 77 ((frontotemporal or fronto-temporal) adj lobar degeneration).tw. (4967) 78 FTLD.tw. (3945) 79 DDPAC.tw. (12) 80 (Pick$2 adj1 disease*).tw. (6302) 81 (FLDEM or FTD-GRN or FTD-PGRN or FTDP-17 or FTLD-17 GRN or FTLD-TDP or HDDD1 or HDDD2).tw. (1622) 82 "Wilhelmsen-Lynch".tw. (2) 83 ((brain or lobar) adj2 atroph*).tw. (11166) 84 Huntington*.tw. (32829) 85 ("Kluver-Bucy" or "Kleuver-Bucy").tw. (604) 86 (temporal lobectomy behavio?r adj2 syndrome*).tw. (0) 87 (Lewy bod$3 adj2 disease*).tw. (3728) 88 CADASIL.tw. (2253) 89 ((mental* or cognit*) adj2 (declin* or degenerat* or deteriorat* or loss* or losing or lost)).tw. (51865) 90 or/66-89 (585536) 91 exp atypical antipsychotic agent/ (75226) 92 (atypical antipsychotic* or atypical anti-psychotic*).tw. (25430) 93 (new generation antipsychotic* or new generation anti-psychotic*).tw. (311) 94 (second generation antipsychotic* or second generation anti-psychotic*).tw. (6293) 95 (2nd generation antipsychotic* or 2nd generation anti-psychotic*).tw. (75) 96 (novel antipsychotic* or novel anti-psychotic*).tw. (2121) 97 (atypical neuroleptic* or new generation neuroleptic* or second generation neuroleptic* or 2nd generation neuroleptic* or novel neuroleptic*).tw. (3997) 98 aripiprazole/ (10335) 99 (aripiprazole or Abilify or Abilitat or Discmelt or HSDB 7320 or Opc 14597 or OPC 31 or OPC-14597 or UNII-82VFR53I78).tw. (8553) 100 aripiprazole.rn. (9825) 101 asenapine/ (750) 102 (Asenapine or Org 5222 or EINECS 265-829-4 or UNII-JKZ19V908O or Saphris or Sycrest).tw. (879) 103 Asenapine.rn. (707) 104 clozapine/ (37170) 105 (Clozapine or "BRN 0764984" or CCRIS 9171 or Alemoxan or Azaleptin or Clopine or Clopsine or Clorazil or Clozapin or Clozapina or Clozapinum or Clozaril or Denzapine or Dorval or Dozapine or EINECS 227-313-7 or Elcrit Ontario Drug Policy Research Network or Fazaclo or HF 1854 or HF1854 or HSDB 6478 or Iprox or Lapenax or Leponex or Lozapin or Lozapine or LX 100-129 or Sizopin or UNII-J60AR2IKIC or Versacloz or W-801 or Zapen or Zaponex).tw. (28879) 106 Clozapine.rn. (31923) 107 lurasidone/ (512) 108 (Lurasidone or Latuda or MK 3756 or MK3756 or SM-13496 or SM13496 or SMP-13496 or SMP13496 or UNII-22IC88528T).tw. (590) 109 Lurasidone.rn. (426) 110 olanzapine/ (28320) 111 (Olanzapine or Anzatric or Lanopin or Lanzac or LY 170053 or LY170053 or Meltolan or Midax or Olace or Oladay or Olan or Olandus or Olanex or Olansek or Olapin or Olazax or Oleanz or Olexar or Oltal or Olzap or Onza or Ozapin or Psychozap or Relprevv or UNII-N7U69T4SZR or Zalasta or Zelta or Zypadhera or Zydis or Zyprexa*).tw. (22648) 112 Olanzapine.rn. (27851) 113 paliperidone/ (2117) 114 (Paliperidone or Invega or 9-hydroxyrisperidone or 9-hydroxy-risperidone or 9-OH-risperidone or R 76477 or RO76477 or UNII-838F01T721 or xeplion).tw. (2572) 115 paliperidone.rn. (1763) 116 quetiapine/ (17897) 117 (quetiapine or Seroquel or "ICI 204,636" or ICI 204636 or Co-Quetiapine or HSDB 7557 or Socalm or Tienapine or UNII-BGL0JSY5SI).tw. (12555) 118 quetiapine.rn. (17048) 119 risperidone/ (35954) 120 (Risperidone or Risperdal or Risperidal or Riperidon or "R-64,766" or R-64766 or Belivon or Consta or Neripros or Noprenia or Pexidone or RN 4891881 or HSDB 7580 or Psychodal or R 64 766 or "R 64,766" or R 64766 or Risolept or Rispen or Risperidona or Risperidonum or Risperin or Rispid or Rispolept or Rispolet or Rispolin or Rizodal or Sequinan or Spiron or UNII-L6UH7ZF8HC or Zargus or Zofredal).tw. (24245) 121 Risperidone.rn. (30703) 122 ziprasidone/ (7131) 123 (Ziprasidone or CP 88059 or CP 88059-01 or "CP-88,059" or "CP-88,059-01" or "CP-88,059-1" or Geodon or HSDB 7745 or UNII-6UKA5VEJ6X or Zeldox or Zeldrox or Ziprazidone or Zipsydon).tw. (5381) 124 Ziprasidone.rn. (7551) 125 or/91-124 (125637) 126 90 and 125 (8100) 127 randomized controlled trial/ or controlled clinical trial/ (971067) 128 exp "clinical trial (topic)"/ (118674) 129 (randomi#ed or randomly or RCT$1 or placebo*).tw. (1609346) 130 ((singl* or doubl* or trebl* or tripl*) adj (mask* or blind* or dumm*)).tw. (327622) 131 trial.ti. (328390) 132 or/127-131 (2177494) 133 126 and 132 (1447) 134 exp animal experimentation/ or exp models animal/ or exp animal experiment/ or nonhuman/ or exp vertebrate/ (38770859) 135 exp humans/ or exp human experimentation/ or exp human experiment/ (29332576) 136 134 not 135 (9439953) Ontario Drug Policy Research Network 137 133 not 136 (1438) 138 (editorial or letter).pt. (2564173) 139 137 not 138 (1410) 140 139 use emczd (885) 141 exp dementia/ (420833) 142 (dement* or amentia$1 or pseudodement*).tw. (225806) 143 alzheimer*.tw. (261150) 144 (progressive adj2 aphasi*).tw. (3381) 145 PPA syndrome*.tw. (45) 146 (senile or senility).tw. (43078) 147 (Mesulam* adj1 syndrome*).tw. (7) 148 binswanger*.tw. (1723) 149 (spongiform encephalopath$3 adj1 (subacute or sub-acute)).tw. (392) 150 "Kosaka-Shibayama".tw. (6) 151 ("diffuse neurofibrillary" adj tangle$1 adj5 calcif*).tw. (76) 152 ((frontotemporal or fronto-temporal) adj lobar degeneration).tw. (4967) 153 FTLD.tw. (3945) 154 DDPAC.tw. (12) 155 (Pick$2 adj1 disease*).tw. (6302) 156 (FLDEM or FTD-GRN or FTD-PGRN or FTDP-17 or FTLD-17 GRN or FTLD-TDP or HDDD1 or HDDD2).tw. (1622) 157 "Wilhelmsen-Lynch".tw. (2) 158 ((brain or lobar) adj2 atroph*).tw. (11166) 159 Huntington*.tw. (32829) 160 ("Kluver-Bucy" or "Kleuver-Bucy").tw. (604) 161 (temporal lobectomy behavio?r adj2 syndrome*).tw. (0) 162 (Lewy bod$3 adj2 disease*).tw. (3728) 163 CADASIL.tw. (2253) 164 ((mental* or cognit*) adj2 (declin* or degenerat* or deteriorat* or loss* or losing or lost)).tw. (51865) 165 or/141-164 (585723) 166 (atypical antipsychotic* or atypical anti-psychotic*).tw. (25430) 167 (new generation antipsychotic* or new generation anti-psychotic*).tw. (311) 168 (second generation antipsychotic* or second generation anti-psychotic*).tw. (6293) 169 (2nd generation antipsychotic* or 2nd generation anti-psychotic*).tw. (75) 170 (novel antipsychotic* or novel anti-psychotic*).tw. (2121) 171 (atypical neuroleptic* or new generation neuroleptic* or second generation neuroleptic* or 2nd generation neuroleptic* or novel neuroleptic*).tw. (3997) 172 aripiprazole/ (10335) 173 (aripiprazole or Abilify or Abilitat or Discmelt or HSDB 7320 or Opc 14597 or OPC 31 or OPC-14597 or UNII-82VFR53I78).tw. (8553) 174 (Asenapine or Org 5222 or EINECS 265-829-4 or UNII-JKZ19V908O or Saphris or Sycrest).tw. (879) 175 clozapine/ (37170) 176 (Clozapine or "BRN 0764984" or CCRIS 9171 or Alemoxan or Azaleptin or Clopine or Clopsine or Clorazil or Clozapin or Clozapina or Clozapinum or Clozaril or Denzapine or Dorval or Dozapine or EINECS 227-313-7 or Elcrit or Fazaclo or HF 1854 or HF1854 or HSDB 6478 or Iprox or Lapenax or Leponex or Lozapin or Lozapine or LX 100- Ontario Drug Policy Research Network 129 or Sizopin or UNII-J60AR2IKIC or Versacloz or W-801 or Zapen or Zaponex).tw. (28879) 177 (Lurasidone or Latuda or MK 3756 or MK3756 or SM-13496 or SM13496 or SMP-13496 or SMP13496 or UNII-22IC88528T).tw. (590) 178 olanzapine/ (28320) 179 (Olanzapine or Anzatric or Lanopin or Lanzac or LY 170053 or LY170053 or Meltolan or Midax or Olace or Oladay or Olan or Olandus or Olanex or Olansek or Olapin or Olazax or Oleanz or Olexar or Oltal or Olzap or Onza or Ozapin or Psychozap or Relprevv or UNII-N7U69T4SZR or Zalasta or Zelta or Zypadhera or Zydis or Zyprexa*).tw. (22648) 180 (Paliperidone or Invega or 9-hydroxyrisperidone or 9-hydroxy-risperidone or 9-OH-risperidone or R 76477 or RO76477 or UNII-838F01T721 or xeplion).tw. (2572) 181 quetiapine/ (17897) 182 (quetiapine or Seroquel or "ICI 204,636" or ICI 204636 or Co-Quetiapine or HSDB 7557 or Socalm or Tienapine or UNII-BGL0JSY5SI).tw. (12555) 183 risperidone/ (35954) 184 (Risperidone or Risperdal or Risperidal or Riperidon or "R-64,766" or R-64766 or Belivon or Consta or Neripros or Noprenia or Pexidone or RN 4891881 or HSDB 7580 or Psychodal or R 64 766 or "R 64,766" or R 64766 or Risolept or Rispen or Risperidona or Risperidonum or Risperin or Rispid or Rispolept or Rispolet or Rispolin or Rizodal or Sequinan or Spiron or UNII-L6UH7ZF8HC or Zargus or Zofredal).tw. (24245) 185 (Ziprasidone or CP 88059 or CP 88059-01 or "CP-88,059" or "CP-88,059-01" or "CP-88,059-1" or Geodon or HSDB 7745 or UNII-6UKA5VEJ6X or Zeldox or Zeldrox or Ziprazidone or Zipsydon).tw. (5381) 186 or/166-185 (110389) 187 165 and 186 (7366) 188 clinical trials/ (54654) 189 (randomi#ed or randomly or RCT$1 or placebo*).tw. (1609346) 190 ((singl* or doubl* or trebl* or tripl*) adj (mask* or blind* or dumm*)).tw. (327622) 191 trial.ti. (328390) 192 or/188-191 (1809091) 193 187 and 192 (1137) 194 exp Animals/ not (exp Animals/ and Humans/) (9168750) 195 193 not 194 (1128) 196 195 use prmz (297) 197 195 use emczd (608) 198 195 not (196 or 197) (223) 199 65 or 140 or 198 (1457) 200 limit 199 to yr="2009-current" (554) 201 remove duplicates from 200 (431) [TOTAL UNIQUE HITS] 202 201 use prmz (89) [MEDLINE UNIQUE HITS] 203 201 use emczd (323) [EMBASE UNIQUE HITS] 204 201 not (202 or 203) (19) [PSYCINFO UNIQUE HITS] *************************** Search Name: AAPs Elderly - Dementia Date Run: 10/10/14 17:20:55.349 Ontario Drug Policy Research Network Ottawa Heart Institute - 2014 Oct 10 [mh Dementia] 3831 (dement* or amentia* or pseudodement*):ti,ab,kw alzheimer*:ti,ab,kw (progressive near/2 aphasi*):ti,ab,kw 8 (PPA next syndrome*):ti,ab,kw 0 (senile or senility):ti,ab,kw (Mesulam* near/1 syndrome*):ti,ab,kw binswanger*:ti,ab,kw 6 (spongiform next encephalopath*) near/1 (subacute or sub-acute):ti,ab,kw "Kosaka-Shibayama":ti,ab,kw 0 ("diffuse neurofibrillary" next tangle*) near/5 calcif*:ti,ab,kw 0 ((frontotemporal or fronto-temporal) next lobar degeneration):ti,ab,kw 14 (Pick* near/1 disease*):ti,ab,kw (FLDEM or FTD-GRN or FTD-PGRN or FTDP-17 or FTLD-17 GRN or FTLD-TDP or HDDD1 or "Wilhelmsen-Lynch":ti,ab,kw 0 (brain or lobar) near/2 atroph*:ti,ab,kw 171 Huntington*:ti,ab,kw 265 "Kluver-Bucy" or "Kleuver-Bucy":ti,ab,kw ("temporal lobectomy" next behavio*r) near/2 syndrome*:ti,ab,kw (Lewy next bod*) near/2 disease*:ti,ab,kw CADASIL:ti,ab,kw (mental* or cognit*) near/2 (declin* or degenerat* or deteriorat* or loss* or losing or lost):ti,ab,kw 987 #25 atypical next (antipsychotic* or anti-psychotic*):ti,ab,kw "new generation" next (antipsychotic* or anti-psychotic*):ti,ab,kw "second generation" next (antipsychotic* or anti-psychotic*):ti,ab,kw 303 "2nd generation" next (antipsychotic* or anti-psychotic*):ti,ab,kw novel next (antipsychotic* or anti-psychotic*):ti,ab,kw 152 (atypical or "new generation" or "second generation" or "2nd generation" or novel) next neuroleptic*:ti,ab,kw 119 #32 (aripiprazole or Abilify or Abilitat or Discmelt or HSDB 7320 or OPC 14597 or OPC 31 or OPC- 14597 or UNII-82VFR53I78):ti,ab,kw (Asenapine or Org 5222 or EINECS 265-829-4 or UNII-JKZ19V908O or Saphris or Sycrest):ti,ab,kw Ontario Drug Policy Research Network [mh Clozapine] 424 (Clozapine or "BRN 0764984" or CCRIS 9171 or Alemoxan or Azaleptin or Clopine or Clopsine or Clorazil or Clozapin or Clozapina or Clozapinum or Clozaril or Denzapine or Dorval or Dozapine or EINECS 227-313-7 or Elcrit or Fazaclo or HF 1854 or HF1854 or HSDB 6478 or Iprox or Lapenax or Leponex or Lozapin or Lozapine or LX 100-129 or Sizopin or UNII-J60AR2IKIC or Versacloz or W-801 or Zapen or Zaponex):ti,ab,kw (Lurasidone or Latuda or MK 3756 or MK3756 or SM-13496 or SM13496 or SMP-13496 or SMP13496 or UNII-22IC88528T):ti,ab,kw (Olanzapine or Anzatric or Lanopin or Lanzac or LY 170053 or LY170053 or Meltolan or Midax or Olace or Oladay or Olan or Olandus or Olanex or Olansek or Olapin or Olazax or Oleanz or Olexar or Oltal or Olzap or Onza or Ozapin or Psychozap or Relprevv or UNII-N7U69T4SZR or Zalasta or Zelta or Zypadhera or Zydis or Zyprexa*):ti,ab,kw (Paliperidone or Invega or 9-hydroxyrisperidone or 9-hydroxy-risperidone or 9-OH-risperidone or R 76477 or RO76477 or UNII-838F01T721 or xeplion):ti,ab,kw (quetiapine or Seroquel or "ICI 204,636" or ICI 204636 or Co-Quetiapine or HSDB 7557 or Socalm or Tienapine or UNII-BGL0JSY5SI):ti,ab,kw [mh Risperidone] (Risperidone or Risperdal or Risperidal or Riperidon or "R-64,766" or R-64766 or Belivon or Consta or Neripros or Noprenia or Pexidone or RN 4891881 or HSDB 7580 or Psychodal or R 64 766 or "R 64,766" or R 64766 or Risolept or Rispen or Risperidona or Risperidonum or Risperin or Rispid or Rispolept or Rispolet or Rispolin or Rizodal or Sequinan or Spiron or UNII-L6UH7ZF8HC or Zargus or Zofredal):ti,ab,kw (Ziprasidone or CP 88059 or CP 88059-01 or "CP-88,059" or "CP-88,059-01" or "CP-88,059-1" or Geodon or HSDB 7745 or UNII-6UKA5VEJ6X or Zeldox or Zeldrox or Ziprazidone or Zipsydon):ti,ab,kw #25 and #43 Publication Year from 2009 to 2014 65 DSR – 6 (did not download – RCT search only) DARE – 2 (did not download – RCT search only) CENTRAL – 55 HTA – 2 (did not download – RCT search only) Ontario Drug Policy Research Network Appendix B: Included Study List
Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
De Deyn, 1999
Setting unclear:
Dementia:
Placebo, oral solution:
51 centers in 8
countries
‘institutionalized' Placebo, oral solution:
(unspecified) Risperidone, oral solution, 0.5-4
Scores of ≥ 4 on FAST, >1 BEHAVE-AD global rating, and ≥8 on the BEHAVE- Risperidone, oral solution, 0.5-4
mg/d: n=115
Haloperidol, oral solution, 0.5-4
Haloperidol, oral solution, 0.5-4
mg/d: n=115
Ballard, 2005
National: Care
Long term care:
Dementia: NINCDS-ADRDA
Placebo, route unspecified:
North East of
BPSD: "clinically significant agitation
Placebo, route unspecified:
reported by a member of staff or a Quetiapine, route unspecified,
50-100mg/d: n=8/31 (26%)
Quetiapine, route unspecified,
CMAI total score >39 Rivastigmine, route
unspecified, 6 to ≥ 9mg/d:
Rivastigmine, route
unspecified, 6 to ≥ 9mg/day:
Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
National: 3
Hospital:
Dementia: DSM-IV
Risperidone, oral tablet, 0.5-
Risperidone, oral tablet, 0.5-
2001 (42)
Inpatients from the 2mg/day: n=29/58
2mg/day: n=2/29 (7%)
Hong Kong,
BPSD: "Active behavioral symptoms,
Departments of three as evidenced by a frequency score of Haloperidol, oral tablet, 0.5-
Haloperidol, oral tablet, 0.5-
Hong Kong hospitals at least 4 on one and at least 3 on 2mg/day n= 29/58
2mg/day n= 1/29 (3%)
another item of the CMAI" Long-term care:
A total score of at least 8 on BEHAVE- Community:
Outpatients from the
Psychogeriatric
Departments of three
Hong Kong hospitals

Note: 31 inpatients and
19 outpatients accounted
for 88% of the total study
population.

Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
De Deyn, 2004
Long-term care:
Dementia diagnosed by:
Placebo, oral capsule:
Placebo, oral capsule: n=38/129
Residents of nursing NINCDS-ADRDA and DSM-IV-TR Australia,
Olanzapine, oral capsule, fixed
Olanzapine, oral capsule, fixed
BPSD Criteria for Inclusion: "All
dose 1.0mg/day: n=129/649
dose 1.0mg/day: n=44/129
Lebanon, and
exhibited clinically significant South Africa
Hospital:
psychotic symptoms (delusions or Olanzapine, oral capsule, fixed
hallucinations) due to AD". dose 2.5mg/day: n=134/649
Olanzapine, oral capsule, fixed
dose 2.5mg/day: n=33/134
Olanzapine, oral capsule, fixed
dose 5.0mg/day: n=125/649
Olanzapine, oral capsule, fixed
Olanzapine, oral capsule, fixed
dose 5.0mg/day: n=31/125
dose 7.5mg/day: n=132/649
*Note: 652 patients were
Olanzapine, oral capsule, fixed
randomized; data for only 649 were dose 7.5mg/day: n=38/132
Fontaine,
Location not
Long-term care:
Dementia diagnosed by:
Risperidone (Risperdal), oral,
Risperidone (Risperdal), oral,
2003 (53)
defined:
All subjects resided in 0.5-2mg/day: n=19/39
0.5-2mg/day: n=19/39 (11%)
BPSD Criteria for Inclusion: ADCS
Olanzapine (Zyprexa), oral, 2.5-
Olanzapine (Zyprexa), oral, 2.5-
United States
agitation screening scale score ≥ 25 10mg/day: n=20/39
10mg/day: n=20/39 (20%)
with 6 points on the delusions, hallucinations, physical aggression, or verbal aggression subscales. Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Holmes, 2007
Location not
Long-term care:
Dementia diagnosed by:
Risperidone, oral capsule,
Discontinuations were not defined:
All participants resided 1mg/day: n=12/27
in nursing homes BPSD Criteria for Inclusion: CMAI
Rivastigmine, oral capsule,
score > 39 points for at least 6 weeks 6mg/day: n=15/27
Kingdom and
National:
Long-term care:
Dementia diagnosed by:
Placebo, oral tablet: n=163/625
Placebo, oral tablet: n=44/163
1999 (43)
Residents of nursing United States
Risperidone (Risperdal), oral
BPSD Criteria for Inclusion: BEHAVE-
tablet, 0.5 mg/day: n=149/625
Risperidone (Risperdal), oral
Grossman,
AD total score ≥ 8 and BEHAVE-AD tablet, 0.5 mg/day: n=32/149
2004 (57)
global rating ≥ 1 Risperidone (Risperdal), oral
Hospital:
tablet, 1.0 mg/day: n=148/625
Inpatients in a chronic Risperidone (Risperdal), oral
disease hospital Risperidone (Risperdal), oral
tablet, 1.0 mg/day: n=45/148
tablet, 2.0 mg/day: n=165/625
Risperidone (Risperdal), oral
tablet, 2.0 mg/day: n=69/165
(42%)
Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Deberdt, 2005
National: 64
Long-term care:
Dementia diagnosed by:
Placebo, route unspecified:
Placebo, route unspecified:
Residents of assisted- NINCDS-ADRDA or DSM-IV United States
BPSD Criteria for Inclusion:
Risperidone, route unspecified,
Risperidone, route unspecified,
Community:
"All patients exhibited clinically 0.5–2 mg/day: n=196/494
0.5–2 mg/day: n=61/196 (31%)
significant psychotic symptoms…" Olanzapine, route unspecified,
Olanzapine, route unspecified,
Note: All but two
2.5- 10 mg/day: n=204/494
2.5- 10 mg/day: n=77/204
patients fell into the scores of ≥6 (severity x frequency) on above categories Hallucinations + Delusions subscales Gareri, 2004
Location not
Setting not defined
Dementia diagnosed by:
Risperidone, oral, 1-2mg/day:
Discontinuations were not defined:
based in Italy
BPSD Criteria for Inclusion:
Olanzapine, oral, 5-10mg/day:
A score of 24 or more on the NPI Promazine, oral, 50-
100mg/day: n=20/60
Meehan, 2002
Long-term care:
Dementia diagnosed by:
Placebo, IM: n=67/272
Discontinuations were not NINCDS-ADRDA or DSM-IV reported separately by study the United
Olanzapine, IM, Fixed 2.5mg
period (hour 2 was end of States (n=33),
BPSD Criteria for Inclusion: Score of ≥
(after 2 hrs): n=71/272
Eli Lily,
Russia (n=2),
Hospital: Hospitalized
14 on the Excited Component of 2005 (58)
and Romania
PANSS; at least one individual PANSS Olanzapine, IM, fixed dose
item score ≥ 4 on a scale of 1-7 5.0mg (after 2 hrs): n=66/272
(24 hours; NOTE:
A diagnosis of "clinically significant Lorazepam, IM, fixed dose
2 hours was agitation for which treatment with a 1.0mg (after 2 hrs): n=68/272
treated as end of parenteral agent is indicated" study period due to a change in study design (crossover) and how the data were reported) Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Mintzer, 2006
National:
Long-term care:
Dementia diagnosed by:
Placebo, oral tablet: n=238/473
Placebo, oral tablet: n=59/238
Residents of nursing As specified in Jeste and Finkel, 2008 home and long-term Risperidone, oral tablet, 0.5-
United States
1.5mg/day: n=235/473
Risperidone, oral tablet, 0.5-
BPSD Criteria for Inclusion: Score of ≥
1.5mg/day: n=60/235 (26%)
2 on any item of BEHAVE-AD psychosis subscale Tariot, 2006
National: 47
Patients > 64 Long-term care:
Dementia diagnosed by:
Placebo, oral capsule: n=99/284
Placebo, oral capsule: n=36/99
sites across the NINCDS-ADRDA or DSM-IV United States
residents who were Quetiapine, oral capsule, 25-
BPSD Criteria for Inclusion:
600mg/day: n=91/284
Quetiapine, oral capsule, 25-
600mg/day: n=29/91 (32%)
"The presence of psychosis was Haloperidol, oral capsule, 0.5-
required, defined as BPRS scores ≥ 24 12mg/day: n=94/284
Haloperidol, oral capsule, 0.5-
and CGI-S scores ≥4 at screening and 12mg/day: n=39/94 (41%)
baseline, scores of ≥3 on two or more of the following BPRS items: 4, conceptual disorganization; 11, suspiciousness; 12, hallucinatory behavior; 15, unusual thought content; and frequency scores of ≥3 on at least one of two psychosis items (delusions or hallucinations) of the NPI-NH" Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Verhey, 2006
National: 6
Long-term care:
Dementia diagnosed by:
Olanzapine, oral capsule,
Nine (9) patients discontinued Patients living in 4.71mg/day (mean dose at
from the study prematurely start of study period II):
Netherlands
BPSD Criteria for Inclusion: "A level
N=4 patients discontinued of agitation that was clinically judged Haloperidol, oral capsule,
because of severe aggressive Community: Patients
to represent a clinical problem 1.75mg/day (mean dose at
behavior, n=2 because of severe living in their own requiring antipsychotic treatment for start of study period II):
motor behavior, and n=2 homes (out-patients) a behavioral disorder" because of refusal. Authors do not advise which group(s) these "A score of at least 45 on the CMAI" Note: 59 patients were
patients were randomized to. randomized; authors excluded one patient because of too many missing data but did not state which group to which they were randomized Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Zhong, 2007
National: 53
Long-term care:
Dementia diagnosed by:
Placebo, oral tablet: n=92/333
Placebo, oral tablet: n=32/92
Residents of nursing NINCDS-ADRDA or DSM-IV the United
homes and assisted Quetiapine, oral tablet, fixed
living facilities BPSD Criteria for Inclusion:
dose 100mg/d: n=124/333
Quetiapine, oral tablet, fixed
"Documented clinical symptoms of dose 100mg/d: n=43/124 (35%)
2009 (22)
agitation that did not result directly Quetiapine, oral tablet, fixed
from the participant's medical dose 200mg/d: n=117/333
Quetiapine, oral tablet, fixed
condition and required treatment dose 200mg/d: n=43/117 (37%)
with antipsychotic medication on in the opinion of the investigator" A score of ≥14 on the PANSS-EC and ≥4 on one of the 5 PANSS-EC items (hostility, tension, uncooperativeness, excitement, poor impulse control) Brodaty, 2003
Long-term care:
Dementia diagnosed by:
Placebo, oral liquid: n=170/337
Placebo, oral liquid: n=56/170
Australia and
Risperidone, oral liquid, 0.5-
New Zealand
BPSD Criteria for Inclusion: A
2ml/day: n=167/337
Risperidone, oral liquid, 0.5-
Brodaty, 2005
minimum aggression score on CMAI 2ml/day: n=45/167 (27%)
"of ≥ 4 on at least 1 aggressive item, Note: 345 patients were
or a score of 3 on at least 2 aggressive randomized; only 337 patients items, or a score of 2 on at least 3 received a drug intervention. 2004 (60)
aggressive items, or 2 aggressive items occurring at a frequency of 2 and 1 at a frequency of 3" Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
De Deyn, 2005
Location not
Long-term care:
Dementia diagnosed by:
Placebo, route unspecified:
Placebo, route unspecified:
defined:
Residents of assisted living facilities or adult BPSD Criteria for Inclusion:
Aripiprazole, route unspecified,
Aripiprazole, route unspecified,
Belgium,
"Symptoms of delusions or 2-15 mg/d: n=106/208
2-15 mg/d: n= 18/106 (17%)
United States,
Community: Patients
hallucinations present (at least and Japan
living with a caregiver intermittently) for 1 month or A score of ≥ 6 on delusions or hallucinations items on the NPI Mintzer, 2007
Long-term care:
Dementia diagnosed by: DSM-IV
Placebo, unknown route:
Placebo, unknown route:
Residents of nursing homes and residential BPSD Criteria for Inclusion: Persistent
United States,
assisted-living facilities or intermittent "delusions, Aripiprazole, unknown route,
Aripiprazole, unknown route,
Australia,
hallucinations or both for at least one Fixed Dose, 2mg/d: n=118/487
Fixed Dose, 2mg/d: n=41/118
2004 (61)
Canada, South
Africa, and
Aripiprazole, unknown route,
Argentina
Fixed Dose, 5mg/d: n=122/487
Aripiprazole, unknown route,
Fixed Dose, 5mg/d: n=49/122
(6 weeks, NOTE:
Aripiprazole, unknown route,
this was the end Fixed Dose, 10mg/d: n=126/487
Aripiprazole, unknown route,
Fixed Dose, 10mg/d: n= 57/126
Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Mowla, 2010
National: The
Community:
Dementia diagnosed by:
Risperidone, oral tablet, 0.5-
Risperidone, oral tablet, 0.5-
Abolfazl Clinic; Outpatients "referred 2mg/day: n=23/48
2mg/day: n=3/23 (13%)
to Abolfazl Clinic, affiliated with Bushehr BPSD Criteria for Inclusion: A
Topiramate, oral tablet, 25-50
Topiramate, oral tablet, 25-50
University of Medical "complaint of behavioral mg/day: n=25/48
mg/day: n=4/25 (16%)
Sciences (Iran)
A NPI Part 1 score of > 1 in delusions, hallucinations, agitation/aggression, and irritability/ liability subscales Paleacu, 2008
Location not
Setting unspecified
Dementia diagnosed by:
Placebo, oral tablet: n=20/40
Placebo, oral tablet: n= 5/20
defined:
Quetiapine, oral tablet, 50-
BPSD Criteria for Inclusion: A score of
300mg/day: n=20/40
Quetiapine, oral tablet, 50-
> 6 on any item included in the NPI 300mg/day: n= 8/20 (40%)
Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Pollock, 2007
National: The
Hospital:
Dementia diagnosed by:
Risperidone, oral capsule, 0.5-
Risperidone, oral capsule, 0.5-
"Participants were 2.0mg/day: n=50/103
2.0mg/day: n=30/50 (60%)
admission to the gero- BPSD Criteria for Inclusion:
Citalopram, oral capsule, 10-
Citalopram, oral capsule, 10-
(United States)
psychiatric unit of an "Target symptoms had to be of 40mg/day: n=53/103
40mg/day: n=28/53 (53%)
Dombrovski,
academic hospital" moderate or higher severity as 2010 (62)
evidenced by the need for Long-term care:
hospitalization and a rating of 3 or Culo, 2010
If patients improved higher (moderate to severe) on at sufficiently, some least one of the agitation items were "discharged to (aggression, agitation, hostility) or nursing homes or… psychosis items (suspiciousness, personal care homes… hallucinations, or delusions)" on the Community:
If patients improved sufficiently, some were discharged to "residential homes for continued treatment under double-blind conditions" Rainer, 2007
National: 6
Community:
Dementia diagnosed by:
Risperidone, oral, 0.5-4
Risperidone, oral, 0.5-4
Patients lived with DSM-IV and ICD-10 mg/day: n=34/72
mg/day: n=3/34 (9%)
Austria;
duration of the study BPSD Criteria for Inclusion:
Quetiapine, oral, 50-400
Quetiapine, oral, 50-400
(out-patients) or had "Suffer from behavioral disturbances" mg/day: n=38/72
mg/day: n=4/38 (11%)
substantial daily A NPI Part 1 score in "delusions, hallucinations, agitation/aggression, disinhibition and aberrant motor behavior" Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Rappaport,
National: 16
Setting unclear:
Dementia diagnosed by:
Placebo, IM: n=26/129
One (n=1) discontinuation was 2009 (52)
Patients resided in recorded; treatment arm was the United
"healthcare facilities" Aripiprazole, IM, Fixed dose
BPSD Criteria for Inclusion:
5mg/d: n=12/129
A PANSS-EC score "≥ 15 and ≤ 32 with ≥1 of the 5 items with a score ≥ 4" Aripiprazole, IM, Fixed dose
10mg/d: n=78/129
Aripiprazole, IM, Fixed dose
15mg/d: n=13/129
Savaskan,
National: A
Hospital:
Dementia diagnosed by:
Quetiapine, oral, 25-
Four (n=4) "patients dropped 2006 (46)
single centre in Participants were 225mg/day: n=11/30
out in the course of the study"; Switzerland
treatment arm was unspecified "hospitalized on the BPSD Criteria for Inclusion:
Haloperidol, oral, 0.5-4mg/day:
gerontopsychiatric "Behavioral symptoms (at least three of the following: aggression, psychotic symptoms, sleep wake cycle Note: 30 patients "entered the
disturbances, agitation, restlessness study"; # randomized not Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Schneider,
National: 42
Community:
Dementia diagnosed by:
Placebo. oral capsule:
Placebo. oral capsule:
2006 (50)
sites across the Patients that were NINCDS-ADRDA (72) or DSM-IV United States
ambulatory and living BPSD Criteria for Inclusion:
Risperidone (Risperdal), oral
Risperidone (Risperdal), oral
Note: Trial was
Delusions, hallucinations, aggression, capsule, 0-2.5mg/day:
capsule, 0-2.5mg/day: n=66/85
Schneider,
conducted at Long-term care:
or agitation that developed after the 2001 (64)
45 sites, but 3 Patients living in onset of dementia and was severe enough to disrupt their functioning Olanzapine (Zyprexa), oral
Olanzapine (Zyprexa), oral
Sultzer,
facilities. Ten patients and, in the opinion of the study capsule, 0-17.5mg/day:
capsule, 0-17.5mg/day:
2008 (65)
assign patients physicians, to justify treatment with homes during phase 1 antipsychotic drugs" (36 weeks; end of Quetiapine (Seroquel), oral
Quetiapine (Seroquel), oral
2011 (66)
Phase I. Note:
At least a moderate rating for capsule, 0-200mg/day:
capsule, 0-200mg/day: n=77/94
This was treated conceptual disorganization, as the end of Schneider,
suspiciousness, or hallucinatory study period due 2003 (67)
to changes in behavior on the BPRS scale. NOTE: Phase I discontinuation
study design and Alternatively, "a frequency rating of rates only Mohamed,
how the data ‘often' or ‘more frequently' and a 2012 (68)
were reported) severe rating of at least ‘moderate' were required for delusions, hallucinations, agitation, or ‘aberrant 2009 (69)
motor behavior'" in the NPI. Schneider,
2009 (70)
Schneider,
2006 (50)
Schneider,
2009 (71)
Sultzer,
2009 (70)
Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Street, 2000
National: 28
Long-term care:
Dementia diagnosed by:
Placebo, oral tablet: n=47/206
Placebo, oral tablet: n=11/47
sites across the Residents of nursing NINCDS-ADRDA (72) United States
Olanzapine, oral tablet, fixed
BPSD Criteria for Inclusion: A score of
dose 5mg/d: n=56/206
Olanzapine, oral tablet, fixed
"3 or higher on any of the Agitation/ dose 5mg/d: n=11/56 (20%)
Cummings,
Aggression, Hallucinations, or Olanzapine, oral tablet, fixed
2002 (73)
Delusions items" of the NPI-NH. dose 10mg/d: n=50/206
Olanzapine, oral tablet, fixed
dose 10mg/d: n=14/50 (28%)
Kennedy,
Olanzapine, oral tablet, fixed
2001 (74)
dose 15mg/d: n=53/206
Olanzapine, oral tablet, fixed
dose 15mg/d: n=18/53 (34%)
Mintzer,
2001 (75)
Clark,
2001 (76)
Street,
2001 (77)
Streim, 2008
National: 35
Long-term care:
Dementia diagnosed by:
Placebo, route unspecified:
Note: due to the change in study
Patients residing in design at Week 6 for some the United
nursing homes and patients, and authors' style of residential assisted- BPSD Criteria for Inclusion:
Aripiprazole, route unspecified,
reporting, only data to week 6 living facilities "Psychotic symptoms of delusions or 2-15 mg/day: n=131/256
was extracted from this study. (6 weeks, NOTE:
hallucinations (at least intermittently) Number of discontinuations 2004 (78)
this was the end for ≥ 1 month" before or at week 6 could not be discerned. Paleacu,
A score of ≥ 6 on either the delusions 2010 (79)
or hallucinations items of the NPI-NH Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Astra Zeneca
Location not
Setting not defined
Dementia diagnosed by:
Quetiapine (Seroquel), oral
Quetiapine (Seroquel), oral
defined: Study
tablet, 50-300mg/day:
tablet, 50-300mg/day: n=17/55
ceuticals,
2009 (22)
BPSD Criteria for Inclusion: "Presence
of predominantly delusional or Haloperidol, oral capsule and
Haloperidol, oral capsule and
hallucinatory symptoms" tablet, 1-6mg/day: n=57/112
tablet, 1-6mg/day: n=17/57
Herz, 2002
Location not
Setting not defined
Dementia diagnosed by:
Placebo, route unspecified:
Placebo, route unspecified:
defined:
only) (23)
BPSD Criteria for Inclusion: Baseline
Risperidone, route unspecified,
Risperidone, route unspecified,
United States
agitation: rating of Moderate on both 0.5-4mg/day: n=14/29
0.5-4mg/day: n=0/14 (0%)
CGS and one of ADAS agitation items or the BPRS Tension or Excitement Olanzapine, route unspecified,
Olanzapine, route unspecified,
2.5-20mg/day: n=7/29
2.5-20mg/day: n=1/7 (14%)

Note: 29 patients are reported
on; # randomized was not
defined.
Barak, 2011
National: The
Hospital:
Dementia diagnosed by:
Risperidone, oral tablet,
Risperidone, oral tablet,
A psychiatric inpatient 1mg/day: n=20/40
1mg/day: n=9/20 (45%)
setting in Israel; Center in Israel
"Patients were BPSD Criteria for Inclusion: "Patients
Escitalopram , oral tablet,
Escitalopram , oral tablet,
hospitalized for the had to be admitted to our psychiatric 10mg/day: n=20/40
10mg/day: n=4/20 (20%)
duration of the study" center… because of signs and symptoms of psychosis, aggression or agitation that were severe enough to warrant hospitalization" Ontario Drug Policy Research Network Last Name of
Location and #
Indication for
Scale(s) Used for Dementia
Total No. Randomized,
No. (%) Patients
First Author,
of Centers
Treatment
Diagnosis and BPSD Criteria for
Group description and No.
Discontinued by Treatment
(Treatment
randomized to group
Inclusion
duration)
Teranishi,
National: The
Hospital:
Dementia diagnosed by:
Risperidone, oral, flex dose,
Risperidone, oral, flex dose,
2013 (33)
Sato Psychiatric Inpatients at Sato DSM-IV; dementia subtypes were 0.5-2.0 mg/d: n=27/55
0.5-2.0 mg/d: n=1/27 (4%)
Psychiatric Hospital in diagnosed based on NINCDS- Japan. Patients ADRDA(72), NINDS-AIREN (80), Fluvoxamine oral, flex dose, 25-
Fluvoxamine oral, flex dose, 25-
entered hospital Consensus guidelines for the clinical 200 mg/d: n=28/55
200 mg/d: n=2/28 (7%)
because their BPSD and pathologic diagnosis of dementia symptoms could no with Lewy Bodies (81) Note: The Yukukansan (herbal
longer be managed by treatment) caregivers or nursing BPSD Criteria for Inclusion: At least 1
trial arm has been excluded symptom score of > 4 in NPI-NH from this study; n=55 were subscales at trial start randomized to two of the three trial arms of interest. De Deyn, 2012
Long-term care:
Dementia diagnosed by:
Quetiapine XR, oral tablet, 50-
Quetiapine XR, oral tablet, 50-
Participants resided in DSM-IV and ICD-10 revision Research 300 mg/day: n=68/100
300 mg/day: n=9/68 (13%)
Belgium,
"nursing homes or Diagnostic Criteria Quetiapine IR, oral tablet, 50-
Quetiapine IR, oral tablet, 50-
Australia,
BPSD Criteria for Inclusion: NPI (Part
300 mg/day: n=32/100
300 mg/day: n=1/32 (3%)
Canada, and
1) score of ≥3 for any of: "agitation, South Africa
delusions and hallucinations; stable general health appropriate for age; willingness and ability to comply with the safety monitoring guidelines and to adhere to the schedule of assessments" National: 1
Community:
Dementia diagnosed by:
Not taking Quetiapine: n=25/51
Discontinuations were not 2014 (48)
Participants were NINCDS-ADRDA (72) or DSM-IV Shanghai, China
Quetiapine, oral, 25-
BPSD Criteria for Inclusion: Patients
400mg/day: n=26/51
had daily or intermittent hallucinations, delusions, agitation, and aggression daily or intermittently for at least 4 weeks prior to visiting doctors Ontario Drug Policy Research Network Appendix C: Included Studies with No Extracted Data

Suh GH, Son HG, Ju YS, Jcho KH, Yeon BK, Shin YM, et al. A randomized, double-blind, crossover
comparison of risperidone and haloperidol in Korean dementia patients with behavioral disturbances.
Am J Geriatr Psychiatry. 2004;12(5):509-16 (17)
Moretti R, Torre P, Antonello RM, Cattaruzza T, Cazzato G. Olanzapine as a possible treatment of
behavioral symptoms in vascular dementia: risks of cerebrovascular events. A controlled, open-label
study. J Neurol. 2005;252(10):1186-93 (18)
Satterlee W, Reams S, Burns P, Hamilton S, Tran PV, GD T. A clinical update on olanzapine treatment in
schizophrenia and in elderly Alzheimer's disease patients. Psychopharmacol Bulletin. 1995;31:534 (19)
Galeotti F, Vanacore N, Gainotti S, Izzicupo F, Menniti-Ippolito F, Petrini C, et al. How legislation on decisional capacity can negatively affect the feasibility of clinical trials in patients with dementia. Drugs and Aging. 2012;29(8):607-14 (20) Mulsant BH, Gharabawi GM, Bossie CA, Mao L, Martinez RA, Tune LE, et al. Correlates of anticholinergic activity in patients with dementia and psychosis treated with risperidone or olanzapine. J Clin Psychiatry. 2004;65(12):1708-14 (21) Ontario Drug Policy Research Network Ontario Drug Policy Research Network References

1.
Rabins P, Blass D. In the Clinic. Dementia. Ann Intern Med. 2014;161(3 ):ITC1. Brookmeyer R, Gray S, Kawas C. Projections of Alzheimer's disease in the United States and the public health impact of delaying disease onset. Am J Public Health. 1998 Sep;88(9):1337-42. Vu M, Hogan DB, Patten SB, Jette N, Bronskill SE, Heckman G, et al. A comprehensive profile of the sociodemographic, psychosocial and health characteristics of Ontario home care clients with dementia Chronic Dis Inj Can. 2014 34(2-3):132-44. Mirakhur A, Craig D, Hart DJ, McLlroy SP, Passmore AP. Behavioural and psychological syndromes in Alzheimer's disease. Int J Geriatr Psychiatry 2004;19(11 ):1035-9. Addressing global dementia. Lancet 2014 383(9936 ):2185. Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions. West Sussex: Wiley-Blackwell, A John Wiley & Sons, Ltd.; 2008. DistillerSR. Ottawa, Canada: Evidence Partners; 2014. EndNote. Philadelphia, PA: Thomson Reuters; 2014. Ntzoufras I. Bayesian modeling using WinBUGS (Wiley series on computational statistics). Hoboken, NJ: John Wiley & Sons; 2009. Spiegelhalter D, Thomas A, Best N, Lunn D. WinBUGS User Manual. Version 1.42003 [cited 2011 June 1]. Available from: http://www.mrc-bsu.cam.ac.uk/bugs/winbugs/manual114.pdf. Dias S, Sutton A, Welton N, Ades A, Golfinopoulos V, Kyrgiou M, et al. NICE DSU Technical Support Document 1: Introduction to evidence synthesis for decision making.May 2011 [cited 2012 June 13]:[1-24 pp.]. Available from: http://www.nicedsu.org.uk/TSD6%20Software.final.08.05.12.pdf. Dias S, Sutton A, Welton N, Ades A, Golfinopoulos V, Kyrgiou M, et al. NICE DSU Technical Support Document 2: Generalised Linear Modelling Framework for Pairwise and Network Meta-Analysis of Randomised Controlled Trials.May 2011 [cited 2012 June 13]:[1-96 pp.]. Available from: http://www.nicedsu.org.uk/TSD6%20Software.final.08.05.12.pdf. Dias S, Sutton A, Welton N, Ades A, Golfinopoulos V, Kyrgiou M, et al. NICE DSU Technical Support Document 4: Inconsistency in Networks of Evidence Based on Randomised Controlled Trials.May 2011 [cited 2012 June 13]:[1-39 pp.]. Available from: http://www.nicedsu.org.uk/TSD6%20Software.final.08.05.12.pdf. Dias S, Sutton A, Welton N, Ades A, Golfinopoulos V, Kyrgiou M, et al. NICE DSU Technical Support Document 3: Heterogeneity: Subgroups, meta-regression, bias and bias-adjustmentMay 2011 [cited 2012 June 13]:[1-24 pp.]. Available from: http://www.nicedsu.org.uk/TSD6%20Software.final.08.05.12.pdf. Maglione M, Ruelaz Maher A, Hu J, Wang Z, Shanman R, Shekelle PG, et al. Off-Label Use of Atypical Antipsychotics: An Update. Comparative Effectiveness Review No. 43. (Prepared by the Southern Ontario Drug Policy Research Network Ontario Drug Policy Research Network California Evidence-based Practice Center under Contract No. HHSA290-2007-10062-1). Rockville, MD: Agency for Healthcare Research and Quality, 2011. McDonagh M, Peterson K, Carson S, Fu R, Thakurta S. Drug Class Review Atypical Antipsychotic Drugs Final Update 3 Report. Portland, ORE: Oregon Evidence-based Practice Center, 2010. Suh GH, Son HG, Ju YS, Jcho KH, Yeon BK, Shin YM, et al. A randomized, double-blind, crossover comparison of risperidone and haloperidol in Korean dementia patients with behavioral disturbances. Am J Geriatr Psychiatry. 2004;12(5):509-16. Moretti R, Torre P, Antonello RM, Cattaruzza T, Cazzato G. Olanzapine as a possible treatment of behavioral symptoms in vascular dementia: risks of cerebrovascular events. A controlled, open-label study. J Neurol. 2005;252(10):1186-93. Satterlee W, Reams S, Burns P, Hamilton S, Tran PV, GD T. A clinical update on olanzapine treatment in schizophrenia and in elderly Alzheimer's disease patients. Psychopharmacol Bulletin. 1995;31:534. Galeotti F, Vanacore N, Gainotti S, Izzicupo F, Menniti-Ippolito F, Petrini C, et al. How legislation on decisional capacity can negatively affect the feasibility of clinical trials in patients with dementia. Drugs and Aging. 2012;29(8):607-14. Mulsant BH, Gharabawi GM, Bossie CA, Mao L, Martinez RA, Tune LE, et al. Correlates of anticholinergic activity in patients with dementia and psychosis treated with risperidone or olanzapine. J Clin Psychiatry. 2004;65(12):1708-14. A multicenter, double-blind, randomized comparison of the efficacy and safety of quetiapine fumarate (SEROQUEL[TM]) and placebo in the treatment of agitation associated with dementia [Internet]. AstraZeneca. 2009. Available from: www.clinicaltrials.gov. Herz L, Frankenburg F, Colon S, Kittur S. A 6-week, double-blind comparison of olanzapine, risperidone, and placebo for behavioral disturbances in Alzheimer's disease. J Clin Psychiatry. 2002;63(11):1065. Barak Y, Plopski I, Tadger S, Paleacu D. Escitalopram versus risperidone for the treatment of behavioral and psychotic symptoms associated with Alzheimer's disease: a randomized double-blind pilot study. Int Psychogeriatr. 2011;23(9):1515-9. De Deyn PP, Eriksson H, Svensson H, Study 115 investigators. Tolerability of extended-release quetiapine fumarate compared with immediate-release quetiapine fumarate in older patients with Alzheimer's disease with symptoms of psychosis and/or agitation: a randomised, double-blind, parallel-group study. International journal of geriatric psychiatry. 2012;27(3):296-304. Paleacu D, Barak Y, Mirecky I, Mazeh D. Quetiapine treatment for behavioural and psychological symptoms of dementia in Alzheimer's disease patients: a 6-week, double-blind, placebo-controlled study. International journal of geriatric psychiatry. 2008;23(4):393-400. Street JS, Clark WS, Gannon KS, Cummings JL, Bymaster FP, Tamura RN, et al. Olanzapine treatment of psychotic and behavioral symptoms in patients with Alzheimer disease in nursing care facilities: a double-blind, randomized, placebo-controlled trial. The HGEU Study Group. Archives of general psychiatry. Ontario Drug Policy Research Network Ontario Drug Policy Research Network 2000;57(10):968-76. Holmes C, Wilkinson D, Dean C, Clare C, El-Okl M, Hensford C, et al. Risperidone and rivastigmine and agitated behaviour in severe Alzheimer's disease: a randomised double blind placebo controlled study. International journal of geriatric psychiatry. 2007;22(4):380-1. Streim JE, Porsteinsson AP, Breder CD, Swanink R, Marcus R, McQuade R, et al. A randomized, double-blind, placebo-controlled study of aripiprazole for the treatment of psychosis in nursing home patients with Alzheimer disease. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2008;16(7):537-50. Mintzer JE, Tune LE, Breder CD, Swanink R, Marcus RN, McQuade RD, et al. Aripiprazole for the treatment of psychoses in institutionalized patients with Alzheimer dementia: a multicenter, randomized, double-blind, placebo-controlled assessment of three fixed doses. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2007;15(11):918-31. Rainer M, Haushofer M, Pfolz H, Struhal C, Wick W. Quetiapine versus risperidone in elderly patients with behavioural and psychological symptoms of dementia: efficacy, safety and cognitive function. European psychiatry : the journal of the Association of European Psychiatrists. 2007;22(6):395-403. Mowla A, Pani A. Comparison of topiramate and risperidone for the treatment of behavioral disturbances of patients with Alzheimer disease: a double-blind, randomized clinical trial. Journal of clinical psychopharmacology. 2010;30(1):40-3. Teranishi M, Kurita M, Nishino S, Takeyoshi K, Numata Y, Sato T, et al. Efficacy and tolerability of risperidone, yokukansan, and fluvoxamine for the treatment of behavioral and psychological symptoms of dementia: a blinded, randomized trial. Journal of clinical psychopharmacology. 2013;33(5):600-7. Gareri P, Cotroneo A, Lacava R, Seminara G, Marigliano N, Loiacono A, et al. Comparison of the efficacy of new and conventional antipsychotic drugs in the treatment of behavioral and psychological symptoms of dementia (BPSD). Archives of gerontology and geriatrics Supplement. 2004(9):207-15. Mintzer J, Greenspan A, Caers I, Van Hove I, Kushner S, Weiner M, et al. Risperidone in the treatment of psychosis of Alzheimer disease: results from a prospective clinical trial. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2006;14(3):280-91. Deberdt WG, Dysken MW, Rappaport SA, Feldman PD, Young CA, Hay DP, et al. Comparison of olanzapine and risperidone in the treatment of psychosis and associated behavioral disturbances in patients with dementia. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2005;13(8):722-30. De Deyn PP, Carrasco MM, Deberdt W, Jeandel C, Hay DP, Feldman PD, et al. Olanzapine versus placebo in the treatment of psychosis with or without associated behavioral disturbances in patients with Alzheimer's disease. International journal of geriatric psychiatry. 2004;19(2):115-26. Tariot PN, Schneider L, Katz IR, Mintzer JE, Street J, Copenhaver M, et al. Quetiapine treatment of psychosis associated with dementia: a double-blind, randomized, placebo-controlled clinical trial. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Ontario Drug Policy Research Network Ontario Drug Policy Research Network Psychiatry. 2006;14(9):767-76. Zhong KX, Tariot PN, Mintzer J, Minkwitz MC, Devine NA. Quetiapine to treat agitation in dementia: a randomized, double-blind, placebo-controlled study. Current Alzheimer research. 2007;4(1):81-93. De Deyn P, Jeste DV, Swanink R, Kostic D, Breder C, Carson WH, et al. Aripiprazole for the treatment of psychosis in patients with Alzheimer's disease: a randomized, placebo-controlled study. Journal of clinical psychopharmacology. 2005;25(5):463-7. De Deyn PP, Rabheru K, Rasmussen A, Bocksberger JP, Dautzenberg PL, Eriksson S, et al. A randomized trial of risperidone, placebo, and haloperidol for behavioral symptoms of dementia. Neurology. 1999;53(5):946-55. Chan WC, Lam LC, Choy CN, Leung VP, Li SW, Chiu HF. A double-blind randomised comparison of risperidone and haloperidol in the treatment of behavioural and psychological symptoms in Chinese dementia patients. International journal of geriatric psychiatry. 2001;16(12):1156-62. Katz IR, Jeste DV, Mintzer JE, Clyde C, Napolitano J, Brecher M. Comparison of risperidone and placebo for psychosis and behavioral disturbances associated with dementia: a randomized, double-blind trial. Risperidone Study Group. The Journal of clinical psychiatry. 1999;60(2):107-15. Brodaty H, Ames D, Snowdon J, Woodward M, Kirwan J, Clarnette R, et al. A randomized placebo-controlled trial of risperidone for the treatment of aggression, agitation, and psychosis of dementia. The Journal of clinical psychiatry. 2003;64(2):134-43. Pollock BG, Mulsant BH, Rosen J, Mazumdar S, Blakesley RE, Houck PR, et al. A double-blind comparison of citalopram and risperidone for the treatment of behavioral and psychotic symptoms associated with dementia. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2007;15(11):942-52. Savaskan E, Schnitzler C, Schroder C, Cajochen C, Muller-Spahn F, Wirz-Justice A. Treatment of behavioural, cognitive and circadian rest-activity cycle disturbances in Alzheimer's disease: haloperidol vs. quetiapine. The international journal of neuropsychopharmacology / official scientific journal of the Collegium Internationale Neuropsychopharmacologicum (CINP). 2006;9(5):507-16. Verhey FR, Verkaaik M, Lousberg R. Olanzapine versus haloperidol in the treatment of agitation in elderly patients with dementia: results of a randomized controlled double-blind trial. Dementia and geriatric cognitive disorders. 2006;21(1):1-8. Shen LL, Xie F, Yao PF, Li X, Lu WH, Chen WZ, et al. Long-term quetiapine therapy on the cognitive function in patients with Alzheimer's disease. Journal of Shanghai Jiaotong University (Medical Science). 2014;34(3):177-80. Ballard C, Margallo-Lana M, Juszczak E, Douglas S, Swann A, Thomas A, et al. Quetiapine and rivastigmine and cognitive decline in Alzheimer's disease: randomised double blind placebo controlled trial. BMJ (Clinical research ed). 2005;330(7496):874. Schneider LS, Tariot PN, Dagerman KS, Davis SM, Hsiao JK, Ismail MS, et al. Effectiveness of atypical antipsychotic drugs in patients with Alzheimer's disease. The New England journal of medicine. Ontario Drug Policy Research Network Ontario Drug Policy Research Network 2006;355(15):1525-38. Meehan KM, Wang H, David SR, Nisivoccia JR, Jones B, Beasley CM, Jr., et al. Comparison of rapidly acting intramuscular olanzapine, lorazepam, and placebo: a double-blind, randomized study in acutely agitated patients with dementia. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2002;26(4):494-504. Rappaport SA, Marcus RN, Manos G, McQuade RD, Oren DA. A randomized, double-blind, placebo-controlled tolerability study of intramuscular aripiprazole in acutely agitated patients with Alzheimer's, vascular, or mixed dementia. Journal of the American Medical Directors Association. 2009;10(1):21-7. Fontaine CS, Hynan LS, Koch K, Martin-Cook K, Svetlik D, Weiner MF. A double-blind comparison of olanzapine versus risperidone in the acute treatment of dementia-related behavioral disturbances in extended care facilities. The Journal of clinical psychiatry. 2003;64(6):726-30. Mowla A, Pani A. Comparison of topiramate and risperidone for the treatment of behavioral disturbances of patients with Alzheimer disease: a double-blind, randomized clinical trial. Journal of clinical psychopharmacology. 2010;30(1):40-3. Jeste D, Finkel, SI. Psychosis of Alzheimer's disease and related dementias. Diagnostic criteria for a distinct syndrome. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2000 8(1):29-34. Rappaport SA, Marcus RN, Manos G, McQuade RD, Oren DA. A randomized, double-blind, placebo-controlled tolerability study of intramuscular aripiprazole in acutely agitated patients with Alzheimer's, vascular, or mixed dementia. Journal of the American Medical Directors Association. 2009;10(1):21-7. Grossman F, Okamoto A, Turkoz I, Gharabawi G. Risperidone in the treatment of elderly patients with psychosis of Alzheimer's disease and related dementias. Journal of the American Geriatrics Society. 2004;52(5):852-3. A Double-blind, Placebo-Controlled Comparison of the Efficacy and Safety of Short-Acting Intramuscular Olazapine, Short-Acting Intramuscular Lorazepam, and Intramuscular Placebo in Treating Agitation in Patients with Dementia of the Alzheimer's Type, Vascular Dementia, and Mixed Dementia. Eli Lilly and Company, 2005 April 7, 2005. Report No.: ID. Brodaty H, Ames D, Snowdon J, Woodward M, Kirwan J, Clarnette R, et al. Risperidone for psychosis of Alzheimer's disease and mixed dementia: results of a double-blind, placebo-controlled trial. International journal of geriatric psychiatry. 2005;20(12):1153-7. Frank L, Kleinman L, Ciesla G, Rupnow MF, Brodaty H. The effect of risperidone on nursing burden associated with caring for patients with dementia. Journal of the American Geriatrics Society. 2004;52(9):1449-55. Breder CS RM, R. et al. Dose-ranging study of ariprprazole in patients with Alzheimer's dementia. 9th International Conference on Alzheimer's Desease and Related Disorders; July 2004; Philadelphia, PA: Neurobiology of Aging; 2004. p. S190. Dombrovski AY, Mulsant BH, Ferrell RE, Lotrich FE, Rosen JI, Wallace M, et al. Serotonin transporter Ontario Drug Policy Research Network Ontario Drug Policy Research Network triallelic genotype and response to citalopram and risperidone in dementia with behavioral symptoms. Int Clin Psychopharmacol. 2010;25(1):37-45. Culo S, Mulsant BH, Rosen J, Mazumdar S, Blakesley RE, Houck PR, et al. Treating neuropsychiatric symptoms in dementia with lewy bodies: a randomized controlled-trial. Alzheimer Dis Assoc Disord. 2010;24(4):360-4. Schneider LS, Tariot PN, Lyketsos CG, Dagerman KS, Davis KL, Davis S, et al. National Institute of Mental Health Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE): Alzheimer disease trial methodology. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2001;9(4):346-60. Sultzer DL, Davis SM, Tariot PN, Dagerman KS, Lebowitz BD, Lyketsos CG, et al. Clinical symptom responses to atypical antipsychotic medications in Alzheimer's disease: phase 1 outcomes from the CATIE-AD effectiveness trial. The American journal of psychiatry. 2008;165(7):844-54. Vigen CL, Mack WJ, Keefe RS, Sano M, Sultzer DL, Stroup TS, et al. Cognitive effects of atypical antipsychotic medications in patients with Alzheimer's disease: outcomes from CATIE-AD. The American journal of psychiatry. 2011;168(8):831-9. Schneider LS, Ismail MS, Dagerman K, Davis S, Olin J, McManus D, et al. Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE): Alzheimer's disease trial. Schizophrenia bulletin. 2003;29(1):57-72. Mohamed S, Rosenheck R, Lyketsos CG, Kaczynski R, Sultzer DL, Schneider LS. Effect of second-generation antipsychotics on caregiver burden in Alzheimer's disease. The Journal of clinical psychiatry. 2012;73(1):121-8. Zheng L, Mack WJ, Dagerman KS, Hsiao JK, Lebowitz BD, Lyketsos CG, et al. Metabolic changes associated with second-generation antipsychotic use in Alzheimer's disease patients: the CATIE-AD study. Am JPsychiatry. 2009;166(5):583-90. Sultzer DL, Davis SM, Tariot PN, Dagerman KS, Lebowitz BD, Lyketsos CG, et al., editors. Antipsychotic medication treatment response in Alzheimer's disease (CATIE-AD): Patient symptoms, function, and life quality, and caregiver well-being. American Journal of Geriatric Psychiatry Annual Meeting 2009; Honolulu, Hawaii, United States. Schneider LS, Mack W, Dagerman K, Hsiao JK, Lebowitz BD, Lyket CG, et al., editors. Second-generation antipsychotics for Alzheimer's disease patients and metabolic abnormalities: The CATIE-AD study. American Journal of Geriatric Psychiatry Annual Meeting; 2009; Honolulu, HI United States. McKhann G, Drachman, D., Folstein, M., Katzman, R., Price, D., Stadlan, E.M. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology. 1984 34(7):939-44. Cummings JL, Street J, Masterman D, Clark WS. Efficacy of olanzapine in the treatment of psychosis in dementia with lewy bodies. Dementia and geriatric cognitive disorders. 2002;13(2):67-73. Kennedy JS, Zagar A, Bymaster F, Nomikos G, Trzepacz PT, Gilmore JA, et al. The central cholinergic system profile of olanzapine compared with placebo in Alzheimer's disease. International journal of geriatric Ontario Drug Policy Research Network Ontario Drug Policy Research Network psychiatry. 2001;16 Suppl 1:S24-32. Mintzer J, Faison W, Street JS, Sutton VK, Breier A. Olanzapine in the treatment of anxiety symptoms due to Alzheimer's disease: a post hoc analysis. International journal of geriatric psychiatry. 2001;16 Suppl 1:S71-7. Clark WS, Street JS, Feldman PD, Breier A. The effects of olanzapine in reducing the emergence of psychosis among nursing home patients with Alzheimer's disease. The Journal of clinical psychiatry. 2001;62(1):34-40. Street JS, Clark WS, Kadam DL, Mitan SJ, Juliar BE, Feldman PD, et al. Long-term efficacy of olanzapine in the control of psychotic and behavioral symptoms in nursing home patients with Alzheimer's dementia. International journal of geriatric psychiatry. 2001;16 Suppl 1:S62-70. Streim JE BC, Swanink R, McQuade, RD, Iwamoto, T, Carson, WH, et al. Flexible dose aripiprazole in psychosis of alzheimer's dementia. American Psychiatric Association Annual Meeting; New York, New York, United States: American Psychiatric Association; 2004. p. 143. Paleacu D, Plopsky I, Tadger S, Barak Y. Designing a comparative study of escitalopram versus risperidone for the treatment of behavioral and psychological symptoms in Alzheimer's disease. 10th International Forum on Mood and Anxiety Disorders Vienna, Austria2010. p. 31-2. Roman G, Tatemichi, TK, Erkinjuntti, T, et al. Vascular dementia: diagnostic criteria for research studies. Report of the NINDS-AIREN International Workshop. Neurology. 1993;43(2):250-60. McKeith I, Galasko, D, Kosaka, K, et al. Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): Report of the consortium on DLB international workshop. Neurology. 1996;47(5):1113-24. Ma H, Huang Y, Cong Z, Wang Y, Jiang W, Gao S. The Efficacy and Safety of Atypical Antipsychotics for the Treatment of Dementia: A Meta-Analysis of Randomized Placebo-Controlled Trials. J Alzheimers Dis. 2014; 42(3): 915-937. Schneider LS, Dagerman K, Insel P. Risk of death with atypical antipsychotic drug treatment for dementia: meta-analysis of randomized placebo-controlled trials. JAMA. 2005;294(15):1934-43. Schneider LS, Dagerman K. Efficacy and adverse effects of atypical antipsychotics for dementia: meta-analysis of randomized, placebo-controlled trials. Am J Geriatr Psychiatry. 2006;14(3):191-210. Robert P, Ferris S, Gauthier S, Winblad B, Tennigkeit F. Review of Alzheimer's disease scales: is there a need for a new multi-domain scale for therapy evaluation in medical practice? Alzheimer Res. Ther. 2010; 2:24. Tariot PN. Do atypical antipsychotic drugs attenuate behavioural and psychological symptoms of dementia in elderly people? A Commentary. Evid Based Mental Health 2005 ; 8: 16. Cited by Lee PE, Gill SS, Freedman M, et al. Atypical antipsychotic drugs in the treatment of behavioural and psychological symptoms of dementia: systematic review. BMJ 2004; 329:75–78. De Deyn PP, Katz IR, Brodaty H, Lyons B, Greenspan A, Burns A. Management of agitation, aggression, and Ontario Drug Policy Research Network Ontario Drug Policy Research Network psychosis associated with dementia: A pooled analysis including three randomized, placebo-controlled double-blind trials in nursing home residents treated with risperidone. Clin Neurol Neurosur 2005; 107:497-508. Information for Healthcare Professionals: Conventional Antipsychotics Silver Spring, MD: U.S. Food and Drug Administration; 2008 [cited 2015 March 10]. Available from: http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm124830.htm. Atypical Antipsychotic Drugs and Dementia- Advisories, Warnings and Recalls for Health Professionals Ottawa, ON: Health Canada; 2005 [cited 2015 March 10]. Available from: http://www.healthycanadians.gc.ca/recall-alert-rappel-avis/hc-sc/2005/14307a-eng.php. Fraser L, Liu K, Naylor K, Hwang J, Dixon S, Shariff S, et al. Falls and Fractures with Atypical Antipsychotic Medication Use: A Population-Based Cohort Study. JAMA : the journal of the American Medical Association. 2015;175(3):450-2. Llorca P, Chereau I, Bayle F, Lancon C. Tardive dyskinesias and antipsychotics: a review. European psychiatry : the journal of the Association of European Psychiatrists. 2002;17(3):129-38. Ontario Drug Policy Research Network Ontario Drug Policy Research Network

Source: http://odprn.ca/wp-content/uploads/2015/06/Antipsychotics-_-systematic-review-Final-Report_1.pdf