Key Question 6: Efficacy and Harms of Prescription Drug Treatment Versus Placebo for Behavioral and Psychological Symptoms of Dementia

Howard A. Fink; Laura S. Hemmy; Eric J. Linskens; Pombie C. Silverman; Roderick MacDonald; J. Riley McCarten; Kristine M.C. Talley; Priyanka J. Desai; Mary L. Forte; Margaret A. Miller; Michelle Brasure; Victoria A. Nelson; Brent C. Taylor; Weiwen Ng; Jeannine M. Ouellette; Nancy L. Greer; Kerry M. Sheets; Timothy J. Wilt; Mary Butler

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

Fink HA, Hemmy LS, Linskens EJ, et al. Diagnosis and Treatment of Clinical Alzheimer’s-Type Dementia: A Systematic Review [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2020 Apr. (Comparative Effectiveness Review, No. 223.)

Cover of Diagnosis and Treatment of Clinical Alzheimer’s-Type Dementia: A Systematic Review

Diagnosis and Treatment of Clinical Alzheimer’s-Type Dementia: A Systematic Review [Internet].

Show details

Contents

< Prev Next >

Appendix HKey Question 6: Efficacy and Harms of Prescription Drug Treatment Versus Placebo for Behavioral and Psychological Symptoms of Dementia

Antipsychotics Versus Placebo

Appendix Table H.1Characteristics of eligible studies: antipsychotics versus placebo

Study Characteristics:

Author/Year

Design

Country

Risk of Bias

Population:

AD Severity

Mean Age

% Female

% White

Education (mean yrs.)

Baseline Cognition

Intervention: Intervention mode Components Frequency Duration

Comparison: Comparison mode Components Frequency Duration

Outcome Timing

Outcome

Domain [Instrument]

Ballard 2018²⁵⁷

Ballard 2019²⁵⁸

RCT UK

Medium

181

Severity NR

Severe Psychosis

(Subgroup analysis NPI-NH≥12)

Mean Age 86

72% Female

85% White

Education NR

MMSE 10

Pimavanserin 34mg/day

Placebo

6, 12 weeks

Agitation

CMAI-SF + subscales

NPI-NH

Psychosis

NPI-NH psychosis subscale

Harms

SAEs

Withdrawal due to AEs

Mortality

Falls

Ballard 2009²⁵⁹

RCT UK

High

165

Severity NR

Mean age 85

77% Female

Race NR

Education NR

Standardized MMSE 11

Severe Impairment Battery 72

Antipsychotics (thioridazine, chlorpromazine, haloperidol, trifluoperazine, or risperidone, doses NR)

Placebo

12 months, up to 54 months

Harms

Mortality

Mintzer 2007²⁶⁰

RCT

Multinational

High

487

Severity NR

Mean Age

82% Female

88% White

Education NR

MMSE 12

Aripiprazole, 2, 5, 10mg/day

Placebo

6 weeks

Psychosis

NPI-NH Psychosis

BPRS

CMAI

Harms

SAEs

Zhong 2007²⁶¹

RCT

High

333

Severity NR

Mean Age 83

74% Female

84% White

Education NR

MMSE 5

Quetiapine, 100, 200mg/day

Placebo

10 weeks

Agitation

CMAI

NPI-NH

PANSS-EC

Harms

SAEs

Mintzer 2006 ²⁶² ²⁶⁷ ²⁶⁷ ²⁶⁷ ²⁶⁷

RCT

Location NR

High

473

Severity NR

Mean Age 83

77% Female

80% White

Education NR

MMSE 13

Risperidone, 1.0-1.5mg/day

Placebo

8 weeks

Agitation

BEHAVE-AD

Harms

SAEs

CATIE Trial

Schneider 2006²⁶³

Sultzer 2008²⁶⁴

Zheng 2009²⁶⁵

Ozawa 2017²⁶⁶

Nagata 2018²⁶⁷

RCT US

High

416

Severity NR

Mean Age 78

56% Female

79% White

24% No School Diploma

34% School Diploma

21% <4yrs college

17% >4yrs college

MMSE 15

Olanzapine (2.5, 5mg), Quetiapine (25, 50mg), Risperidone (0.5, 1mg)

Placebo

12 weeks for efficacy, 36 weeks for harms

Agitation

NPI (agitation, hallucination, delusions subscales)

BPRS

Harms

SAEs

Tariot 2006²⁶⁸

RCT

High

179

Severity NR

Mean Age 83

73% Female

91% White

Education NR

MMSE 13

Quetiapine, 25mg/day

Placebo

Additional active comparator arm

10 weeks

Agitation

BPRS + subscales

NPI-NH14 + subscales

Harms

SAEs

Ballard 2005²⁶⁹

RCT UK

Medium

Severe AD

Mean Age 84

82% Female

Race NR

Education NR

Severe Impairment Battery 64

Quetiapine, 25-50mg/twice daily for 12 weeks, 50mg/twice daily after

No concomitant psychosocial intervention reported

Placebo

Additional active comparator arm

6, 26 weeks

Agitation

CMAI

Harms

SAEs

Mortality

Deberdt 2005²⁷⁰

RCT

Multinational

High

494

Severity NR

Mean Age 78

64% Female

84% White

Education NR

MMSE 14

Olanzapine (2.5-10mg/day), Risperidone (0.5-2mg/day)

Placebo

Additional active comparator arm

10 weeks

Agitation

BPRS

CMAI

NPI Total

NPI-Psychosis

Harms

SAEs

De Deyn 2005²⁷¹

RCT

Multinational

Medium

208

Severity NR

Mean Age 82

72% Female

98% White

Education NR

MMSE 14

Aripiprazole, 2-15mg/day

No concomitant psychosocial intervention reported

Placebo

10 weeks

Agitation

BPRS

NPI Total

NPI-Psychosis

Harms

SAEs

De Deyn 2004²⁷²

RCT

Multinational

High

649

Severity NR

Mean Age 77

75% Female

99% White

Education NR

MMSE 14

Olanzapine, 1, 2.5, 5, 7.5mg/day

Placebo

10 weeks

Agitation

BPRS

NPI-NH

Harms

SAEs

Street 2000²⁷³

Kennedy 2001²⁷⁴

Mintzer 2001²⁷⁵

RCT

High

206

Severity NR

Mean Age 83

62% Female Race NR

Education NR

MMSE 7

Olanzapine, 5, 10, 15mg/day

Placebo

6 weeks

Agitation

BPRS

NPI-NH + subscales

Harms

SAEs

Teri 2000²⁷⁶

RCT

High

Severity NR

Mean Age 75

73% Female

86% White

13 Years Education

MMSE 13

Haloperidol, 0.5mg/day

Placebo

Additional active comparator and behavioral intervention arms

16 weeks

Agitation

ABID

CMAI

RMBPC

Harms

SAEs

De Deyn 1999²⁷⁷ RCT

Multinational

High

229

Severity NR

Mean Age 82

56% Female

99% White

Education NR

MMSE 8

Risperidone, 0.5-4mg/day

Placebo

Additional active comparator arm

12 weeks

Agitation

BEHAVE-AD

CMAI

Harms

SAEs

Devanand 1998²⁷⁸

RCT

Medium 6 weeks

High 12 weeks

Severity NR

Mean Age 72

65% Female

56% White

Education NR

Modified MMSE 19

(scoring range 0-57)

Haloperidol, 0.5-0.75mg/day

No concomitant psychosocial intervention reported

Placebo

Haloperidol, 2-3mg/day

6 weeks

Agitation

BPRS

SADS (psychosis and disorganization items) Behavioral Syndromes Scale for Dementia

Harms

SAEs

Auchus 1997²⁷⁹

RCT

Canada

High

Severity NR

Mean Age 76

67% Female

Race NR

12 Years Education

MMSE 15

Haloperidol, 3mg/day

Placebo

Additional active comparator arm

3, 6 weeks

Agitation

BEHAVE-AD

CMAI

Harms

SAEs

: Abbreviations: ABID=Agitated Behavior in Dementia Scale; AD=Alzheimer’s Disease; BEHAVE-AD=Behavioral Pathology in Alzheimer’s Disease Rating Scale; BPRS=Brief Psychiatric Rating Scale; BSSD=Behavioral Syndromes Scale for Dementia; C=control; CI=confidence interval; CMAI=Cohen-Mansfield Agitation Inventory; CMAI-SF=Cohen-Mansfield Agitation Inventory –Short Form; I=intervention; MG=milligrams; MMSE=Mini Mental State Exam; NPI=Neuropsychiatric Inventory; NPI-NH=Neuropsychiatric Inventory Nursing Home Version; NR=not reported; NS=not statistically significant; PANSS-EC=Positive and Negative Syndrome Scale; RCT=randomized controlled trial; RMBPC=Revised Memory and Behavior Problems Checklist; RoB=Risk of Bias; SADS=Schedule for Affective Disorders and Schizophrenia; SAEs=Serious Adverse Events; SD=standard deviation; UK=United Kingdom; US=United States

Appendix Table H.2Risk of bias ratings: antipsychotics versus placebo

Study	Time	Selection Bias	Attrition Bias	Performance Bias	Detection Bias	Reporting Bias	Overall Rating^* Justification
Ballard 2018²⁵⁷ Ballard 2019²⁵⁸	6, 12 weeks	Medium	Medium 6 weeks Medium 12 weeks	Low	Low	Low	Medium
Ballard 2009²⁵⁹	1 year	Low	High	Low	Low	Low	High 22% participants lost before trial started. Adherence at 1 year 36%.
Mintzer 2007²⁶⁰	8 weeks	Medium	High	Medium	Low	Low	High
Zhong 2007²⁶¹	10 weeks	Low	High	Low	Low	Low	High
Mintzer 2006²⁶²	8 weeks	Medium	High	Low	Low	Low	High
CATIE Trial Schneider 2006²⁶³ Sultzer 2008²⁶⁴	2, 4, 8, 12 weeks	Medium	Medium 2 weeks High 4+ weeks	Low	Low	Low	High No efficacy outcomes reported before 12 weeks
Tariot 2006²⁶⁸	10 weeks	Medium	High	Medium	Low	Low	High
Ballard 2005²⁶⁹	6 weeks, 26 weeks	Low	Medium 6 weeks Likely high 26 weeks	Low	Low	Low 6 weeks High 26 weeks	Medium 6 weeks High 26 weeks
Deberdt 2005²⁷⁰	10 weeks	Medium	High	Medium	Low	Low	High
De Deyn 2005²⁷¹	10 weeks	Medium	Medium	Medium	Low	Low	Medium
De Deyn 2004²⁷²	10 weeks	Medium	High	Low	Low	Low	High
Street 2000²⁷³ Kennedy 2001²⁷⁴ Mintzer 2001²⁷⁵	6 weeks	Low	High	Low	Low	Low	High
Teri 2000²⁷⁶	16 weeks	Low	High	Medium	Low	Low	High
De Deyn 1999²⁷⁷	12 weeks	Medium	High	Low	Low	Low	High
Devanand 1998²⁷⁸	6, 12 weeks	Medium	Medium 6 weeks High 12 weeks	Low	Low	Low	Medium 6 weeks High 12 weeks
Auchus 1997²⁷⁹	3 weeks	Medium	High	Low	Low	Low	High

*: Justifications provided when overall risk of bias rating deviating from guidance provided in tool (Appendix B)

Appendix Table H.3Primary outcomes summary low and medium risk of bias studies: antipsychotics versus placebo

Drug Comparison	AD Severity	Study Followup RoB	General Behavior	Agitation	Aggression	Psychosis	Disinhibited sexual behavior	Harms
Aripiprazole vs. placebo	NR	De Deyn				BPRS-Core Mean Change from Baseline (SD) I: -3.9 (NR) C: -2.7 (NR) p=0.042 BPRS-Psychosis Mean Change from Baseline (SD) I: -1.93 (NR) C: -1.27 (NR) p=0.029 NPI-Psychosis Mean Change from Baseline (SD) I: -6.55 (NR) C: -5.52 (NR) p=0.169 Standardized mean difference could not be calculated.		SAEs I: 15% (16/106) C: 9% (9/102) p=NR Injurious falls I: 8% (8/106) C: 5% (5/102) Somnolence I: 8% (n=NR) C: 1% (n=NR) Withdrawal due to AEs I: 9% (10/106) C: 7% (7/102) Mortality I: 4% (4/106) C: 0 (0/102) Extrapyramidal Symptoms: Simpson-Angus Scale Mean Change from Baseline (SD) I: 0.71 (NR) C: 0.03 (NR) p=0.109 Abnormal Involuntary Movement Scale Mean Change from Baseline (SD) I: -0.13 (NR) C: -0.01 (NR) p=0.617 Barnes Akathisia Rating Scale Mean Change from Baseline (SD) I: -0.09 (NR) C: -0.06 (NR) p=0.470
Haloperidol vs. placebo I1: 2-3mg I2: 0.5-0.75mg	NR	Devanand		BSSD-Psychomotor Agitation Mean Change from Baseline (SD) I1: -1.0^* (NR) I2: -0.1^* (NR) C: -0.25^* (NR) I1 vs. C: p<0.03 Favors I1 I2 vs. C: NR Standardized mean difference I1 vs. C: -0.59 (95% CI -1.22, 0.05) I2 vs. C: 0.12 (95% CI -0.05, 0.74) 25% Reduction BSSD Psychomotor Agitation I1 55% (11/20) I2 25% (5/20) C 30% (6/20) I1 vs. C: p=0.11 I2 vs. C: p=NR I1 vs. C: RR 1.83 (95% CI 0.84, 3.99) I2 vs. C: RR 0.83 (95% CI 0.30, 2.29)	BPRS-Hostile-Suspicious Mean Change from Baseline (SD) I1: -2.5^* (NR) I2: -1.95^* (NR) C: -1.35^* (NR) I1 vs. C: p=NR (NS) I2 vs. C: NR Standardized mean difference I1 vs. C: -0.42 (95% CI -1.05, 0.21) I2 vs. C: -0.23 (95% CI -0.85, 0.39) BSSD-Physical Aggression Mean Change from Baseline (SD) I1: -0.75^* (NR) I2: -0.3^* (NR) C: -0.25^* (NR) I1 vs. C: p=NR (NS) I2 vs. C: p=NR Standardized mean difference I1 vs. C: -0.40 (95% CI -1.03, 0.23) I2 vs. C: -0.04 (95% CI -0.66, 0.58)	BPRS Mean Change from Baseline (SD) 1: -5.95^* (NR) I2: -3.0^* (NR) C: -2.95^* (NR) I1 vs. C: p=NR Significance NR I2 vs. C: NR BPRS-Psychosis Mean Change from Baseline (SD) I1: -2.0^* (NR) I2: -0.85^* (NR) C: -0.85^* (NR) I1 vs. C: p<0.02 Favors I1 I2 vs. C: NR 25% Reduction BPRS-Psychosis I1 60% (12/20) I2 30% (6/20) C 30% (6/20) I1 vs. C: p<0.06 I2 vs. C: p=NR I1 vs. C: RR 2.00 (95% CI 0.94, 4.27) I2 vs. C: RR 1.00 (95% CI 0.39, 2.58) SADS Mean Change from Baseline (SD) I1: -3.35^* (NR) I2: -1.6^* (NR) C: -1.85^* (NR) I1 vs. C: p=NR (NS) I2 vs. C: p=NR 25% Reduction SADS Target Symptoms I1 55% (11/20) I2 35% (7/20) C 25% (5/20) I1 vs. C: p<0.06 I2 vs. C: NR I1 vs. C: RR 2.20 (95% CI 0.93, 5.18) I2 vs. C: RR 1.4 (95% CI 0.53, 3.68)		Extrapyramidal Symptoms I1 vs. C: p=0.08 I2 vs. C: NR Treatment Emergent Symptoms Scale I1 vs. C: p=NR (NS) I2 vs. C: NR
Pimavanserin vs. placebo NR Severe psychosis subgroup (NPI-NH≥12)	Ballard 2018²⁸⁰ 6 weeks efficacy 12 weeks harms Medium		CMAI-SF Total Score Full study sample Adjusted mean change from baseline (95% CI) 0·30 (− 2·04 to 2·63) Severe psychosis subgroup Adjusted mean change from baseline I: -5.22 C: -3.97 p=0.618 CMAI-SF Verbally Agitated Behavior Full study sample Adjusted mean change from baseline (95% CI) −0·17 (−1·35 to 1·02) Severe psychosis subgroup Adjusted mean change from baseline I: -3.23 C: -1.87 p=0.230 NPI-NH Agitation/Aggression Full study sample Adjusted mean change from baseline (95% CI) −0·66 (− -1·80 to 0·48) Severe psychosis subgroup Adjusted mean change from baseline I: -2.38 C: -2.12 p=0.829	CMAI-SF Aggressive Behavior Full study sample Adjusted mean change from baseline (95% CI) 0·30 (−0·52 to 1·11) Severe psychosis subgroup Adjusted mean change from baseline I: -1.11 C: -1.02 p=0.919	NPI-NH Psychosis Full study sample Adjusted mean change from baseline (SE) I: -3.76 (0.65) C: -1.93 (0.63) Delta -1.84 (95% CI -3.64 to -0.04) Cohen’s d -0.32 p=0.045 (SMD, -0.30, [95% CI, -0.59 to -0.01])^* ≥20% Decrease in NPI-NH Psychosis Score I: 59% (53/90) C: 46% (42/91) p=0.094 ≥30% Decrease in NPI-NH Psychosis Score I: 55% (48/90) C: 37% (34/91) p=0.016 ≥50% Decrease in NPI-NH Psychosis Score I: 51% (46/90) C: 34% (31/91) p=0.024 ≥75% Decrease in NPI-NH Psychosis Score I: 28% (25/90) C: 17% (15/91) p=0.066 ≥100% Decrease in NPI-NH Psychosis Score I: 13% (12/90) C: 10% (9/91) p=0.55 AA Severe psychosis subgroup Mean change from baseline (95% CI) I: -10.15 (-12.50 to -7.80) C: -5.72 (-8.14 to -3.30) Delta -4.43 (95% CI -7.81 to -1.04) Cohen’s d -0.73 p=0.011 In severe psychosis subgroup: ≥20% Decrease in NPI-NH Psychosis Score I: 96.3% (n/N NR) C: 53.5% p<0.001 ≥30% Decrease in NPI-NH Psychosis Score I: 88.9% (n/N NR) C: 43.3% p<0.001 ≥50% Decrease in NPI-NH Psychosis Score I: 77.8% (n/N NR) C: 43.3% p<0.008 ≥75% Decrease in NPI-NH Psychosis Score I: 40.7% (n/N NR) C: 16.7% p=0.038 100% Decrease in NPI-NH Psychosis Score I: 11.1% (n/N NR) C: 10.0% p=0.884		SAEs Full study sample I: 17% (15/90) C: 11% (10/91) RR^* 1.25 (95% CI 0.87 to 1.79) Withdrawal due to AEs Full study sample I: 9% (8/90) C: 12% (11/91) RR^* 0.83 (95% CI 0.48 to 1.44)	SAEs Full study sample I: 17% (15/90) C: 11% (10/91) RR^* 1.25 (95% CI 0.87 to 1.79) Withdrawal due to AEs Full study sample I: 9% (8/90) C: 12% (11/91) RR^* 0.83 (95% CI 0.48 to 1.44) Mortality Full study sample I: 4% (4/90) C: 4% (4/91) RR^* 1.01 (95% CI 0.26 to 3.92) Falls Full study sample I: 23% (21/90) C: 23% (21/91) RR^* 1.01 (95% CI 0.60 to 1.72)
Quetiapine vs. placebo	Severe AD	Ballard 2005²⁶⁹ 6 weeks Medium		CMAI Mean change from baseline (SD) I: -4.0 (15.4) C: -6.2 (17.6) Mean difference 3.5 (95% CI -3.7, 10.8) p=0.3 Standardized mean difference 0.26 (95% CI -0.27, 0.79)				SAEs I: 0 (0/31) C: 3% (1/31) Mortality I: 6% (2/31) C: 0 (0/31)

*: Calculated by EPC
: Abbreviations: AD=Alzheimer’s Disease; BPRS=Brief Psychiatric Rating Scale; BSSD=Behavioral Syndromes Scale for Dementia; C=control; CI=confidence interval; CMAI=Cohen-Mansfield Agitation Inventory; I=intervention; NPI=Neuropsychiatric Inventory; NR=not reported; NS=not statistically significant; RoB=Risk of Bias; SADS=Schedule for Affective Disorders and Schizophrenia; SAEs=Serious Adverse Events; SD=standard deviation

Appendix Table H.4Primary outcomes summary low and medium risk of bias studies: antipsychotics dose response

Drug Comparison

AD Severity

Study

Followup

RoB

General Behavior

Agitation

Aggression

Psychosis

Depression

Anxiety

Disinhibited sexual behavior

Harms

Haloperidol high vs. low dose

I1: 2-3mg

I2: 0.5-0.75m

Devanand 1998²⁷⁸

6 weeks

Medium

BSSD-Psychomotor Agitation

Mean Change from Baseline (SD)

I1: -1.0^* (NR)

I2: -0.1^* (NR)

p<0.02

Favors I1

25% Reduction BSSD Psychomotor Agitation

I1 55% (11/20)

I2 25% (5/20)

p<0.06

BPRS-Hostile-Suspicious

Mean Change from Baseline (SD)

I1: -2.5^* (NR)

I2: -1.95^* (NR)

p=NR (NS)

Standardized mean difference -0.20 (95% CI -0.82, 0.42)

BSSD-Physical Aggression

Mean Change from Baseline (SD)

I1: -0.75^* (NR)

I2: -0.3^* (NR)

p=NR (NS)

Standardized mean difference -0.36 (95% CI -0.98, 0.27)

BPRS

Mean Change from Baseline (SD)

I1: -5.95^* (NR)

I2: -3.0^* (NR)

p=NR

BPRS-Psychosis

Mean Change from Baseline (SD)

I1: -2.0^* (NR)

I2: -0.85^* (NR)

p=0.05

Standardized mean difference -0.45 (95% CI -1.08, 0.17)

25% Reduction BPRS-Psychosis

I1 60% (12/20)

I2 30% (6/20)

p<0.06

RR 2.0 (95% CI 0.94, 4.27)

SADS Mean Change from Baseline (SD)

I1: -3.35^* (NR)

I2: -1.6^* (NR)

p=NR (NS)

Standardized mean difference -0.68 (95% CI -1.32, -0.04)

25% Reduction SADS Target Symptoms

I1 55% (11/20)

I2 35% (7/20)

p=0.20

RR 1.57 (95% CI 0.77, 3.22)

Extrapyramidal Symptoms

p<0.08 (NS)

Treatment Emergent Symptoms Scale

p=NR (NS)

*: Calculated by EPC
: Abbreviations: AD=Alzheimer’s Disease; BPRS=Brief Psychiatric Rating Scale; BSSD=Behavioral Syndromes Scale for Dementia; C=control; I=intervention; NR=not reported; NS=not statistically significant; RoB=Risk of Bias; SADS=Schedule for Affective Disorders and Schizophrenia; SAEs=Serious Adverse Events; SD=standard deviation

Appendix Table H.5Summary of strength of evidence: aripiprazole versus placebo

Outcome	AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
NPI	NR	10 weeks	1 RCT (n=203)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient
BPRS	NR	10	1 RCT (n=195)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient
BPRS-Core	NR	10	1 RCT (n=198)	Favors aripiprazole	Medium	Unknown	Direct	Precise	Insufficient
BPRS Psychosis	NR	10	1 RCT (n=192)	Favors aripiprazole	Medium	Unknown	Direct	Precise	Insufficient
NPI Psychosis	NR	10	1 RCT (n=203)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient
Withdrawal SAEs	NR	10	1 RCT (n=208)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient
SAEs		10	1 RCT (n=208)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: AD=Alzheimer’s Disease; BPRS=Brief Psychiatric Rating Scale; NPI=Neuropsychiatric Inventory; RCT=randomized controlled trial; SAEs=serious adverse events

Appendix Table H.6Summary of strength of evidence: haloperidol versus placebo

Outcome	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
BSSD-Psychomotor Agitation NR	6 weeks	1 RCT (n=40)	Favors Haloperidol	Medium	Unknown	Direct	Precise	Insufficient
BPRS Hostile Suspicious NR	6 weeks	1 RCT (n=40)	No Difference	Medium	Unknown	Direct	Imprecise	Insufficient
BSSD Physical Aggression NR	6 weeks	1 RCT (n=40)	No Difference	Medium	Unknown	Direct	Imprecise	Insufficient
BPRS NR	6 weeks	1 RCT (n=40)	Not gradable
BPRS Psychosis NR	6 weeks	1 RCT (n=40)	Favors	Medium	Unknown	Direct	Precise	Insufficient
SADS NR	6 weeks	1 RCT (n=40)	No Difference	Medium	Unknown	Direct	Imprecise	Insufficient (No Difference)

: Abbreviations: AD=Alzheimer’s Disease; BPRS=Brief Psychiatric Rating Scale; BSSD=Behavioral Syndromes Scale for Dementia; NPI=Neuropsychiatric Inventory; RCT=randomized controlled trial; SADS=Schedule for Affective Disorders and Schizophrenia; SAEs=serious adverse events

Appendix Table H.7Summary of strength of evidence: pimavanserin versus placebo

Outcome	AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
CMAI-SF Agitation	NR	6 weeks	1 RCT (n=181)	No Difference	Medium	Direct	Imprecise	Insufficient	CMAI-SF Agitation NR
CMAI-SF Aggression	NR	6 weeks	1 RCT (n=181)	No Difference	Medium	Direct	Precise	Insufficient	CMAI-SF Aggression NR
NPI-NH Psychosis	NR	6 weeks	1 RCT (n=181)	Favors Pimavanserin	Medium	Unknown	Direct	Imprecise	Insufficient
SAEs	Severe AD	6 weeks	1 RCT (n=181)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient
Withdrawal Due to AEs		12 weeks	1 RCT (n=181)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: AD=Alzheimer’s Disease; CMAI=Cohen-Mansfield Agitation Inventory; CMAI-SF=Cohen-Mansfield Agitation Inventory Short Form; NPINH=Neuropsychiatric Inventory for Nursing Homes; RCT=randomized controlled trial; SAES=serious adverse events

Appendix Table H.8Summary of strength of evidence: quetiapine versus placebo

Outcome	AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
CMAI	Severe AD	6 weeks	1 RCT (n=54)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient
SAEs	Severe	6 weeks	1 RCT (n=54)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: AD=Alzheimer’s Disease; CMAI=Cohen-Mansfield Agitation Inventory; RCT=randomized controlled trial; SAES=serious adverse events

Appendix Table H.9Summary of strength of evidence: haloperidol standard dose (2-3 mg) versus low dose (0.5-0.75 mg)

Outcome	AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
BSSD-Psychomot or Agitation	NR	6 weeks	1 RCT (n=40)	Favors Standard Dose	Medium	Unknown	Direct	Precise	Insufficient
BPRS-Hostile Suspicious	NR	6 weeks	1 RCT (n=40)	No Difference	Medium	Unknown	Direct	Imprecise	Insufficient
BSSD-Physical Aggression	NR	6 weeks	1 RCT (n=40)	No Difference	Medium	Unknown	Direct	Imprecise	Insufficient
BPRS	NR	6 weeks	1 RCT (n=40)	Not gradable
BPRS-Psychosis	NR	6 weeks	1 RCT (n=40)	No Difference	Medium	Unknown	Direct	Imprecise	Insufficient
SADS	NR	6 weeks	1 RCT (n=40)	No Difference	Medium	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: AD=Alzheimer’s Disease; BPRS=Brief Psychiatric Rating Scale; BSSD=Behavioral Syndromes Scale for Dementia; NPI=Neuropsychiatric Inventory; RCT=randomized controlled trial; SADS=Schedule for Affective Disorders and Schizophrenia; SAEs=serious adverse events

Antidepressants Versus Placebo

Appendix Table H.10Characteristics of eligible studies: antidepressants versus placebo

Study Characteristics:

Author/Year

Design

Country

Risk of Bias

Population:

AD Severity

Mean Age

% Female

% White

Education (mean yrs.)

Baseline Cognition

Baseline BPSD

Intervention: Intervention mode Components Frequency Duration

Comparison: Comparison mode Components Frequency Duration

Outcome Timing

Outcome

Domain [Instrument]

Zhou 2019²⁸¹

RCT

China

Medium

Severity NR

Mean Age 71

59% Female

Race NR

6 Years Education

MMSE 15

NPI 35

Citalopram, 30 mg/day

All treatment groups received memantine (~20 mg/day)

Placebo

All treatment groups received memantine (~20 mg/day)

12 weeks

Agitation

NPI Agitation

Psychosis

NPI Delusions

NPI Hallucinations

Porsteinsson 2014²⁸²

RCT

USA/Canada

Low

Leonpacher 2016²⁸³

RCT

USA

Medium

186

Probable AD

Mean Age 78

46% Female

65% White

23% Received High School Diploma

Baseline Cognition: MMSE 15.7

Baseline CMAI: 28.2

Citalopram, 30 mg/day

All treatment groups received concomitant psychosocial intervention

Placebo

All treatment groups received concomitant psychosocial intervention

9 weeks

Agitation

CMAI

mADCS-CGIC

NBRS-A

NPI-agitation subscale

Psychosis

NPI-delusion subscale

Banerjee 2011²⁸⁴

RCT

Medium

219

Probable or possible AD

Mean Age 79

68% Female

93% White

Education not reported

Baseline Cognition:

MMSE 18.1

Baseline CSDD: 13.06

Mirtazapine, 45/mg day

No concomitant psychosocial intervention reported

Placebo

39 weeks

Behavior

Depression

Quality of life

Banerjee 2011²⁸⁴

RCT

Medium

218

Probable or possible AD

Mean Age 79

68% Female

93% White

Education not reported

Baseline Cognition: MMSE 18.1

Baseline CSDD: 13.21

Sertraline, 150 mg/day

No concomitant psychosocial intervention reported

Placebo

39 weeks

Behavior

Depression

Quality of life

Finkel 2004²⁸⁵

RCT

USA

Medium

244

Probable or possible AD

Mean Age 76

61% Female

Race not reported

Education not reported

Baseline Cognition:

MMSE 17.8

Baseline NPI: 30.8

Sertraline, 25-200 mg/day

Donepezil 5-10 mg/day for both arms

No concomitant psychosocial intervention reported

Placebo

12 weeks

Agitation

CMAI

Levkovitz 2001²⁸⁶

RCT

Israel High

Unspecified AD

Mean Age 78

45% Female

Race not reported

Education not reported

BPRS ≥ 18

Fluvoxamine, 50 mg/day

Placebo plus

Perphenazine, 4mg 3x/day

7 weeks

Psychosis

BPRS

: Abbreviations: AD=Alzheimer’s Disease; BPRS=Brief Psychiatric Rating Scale; CMAI=Cohen-Mansfield Agitation Inventory; CSDD=Cornell Scale for Depression in Dementia; DEMQOL=DEM Quality of Life; EQ5D=EuroQol 5D. GHQ-12=General Health Questionnaire; HDRS=Hamilton Depression Rating Scale; mADCS-CGIC=modified Alzheimer Disease Cooperative Study-Clinical Global Impression of Change; NBRS-A=Neurobehavioral Rating Scale; NPI=Neuropsychiatric Inventory; RoB=Risk of Bias; SAEs=Serious Adverse Events

Appendix Table H.11Risk of bias ratings: antidepressant versus placebo

Study	Time	Selection Bias	Attrition Bias	Performance Bias	Detection Bias	Reporting Bias	Overall Rating
Zhou 2019²⁸¹	12 weeks	Medium	Low	Low	Low	Low	Medium
Porsteinsson 2014²⁸² Leonpacher 2016²⁸³	9 weeks	Low	Low	Low	Medium	Medium	Medium
Banerjee 2011²⁸⁴	39 weeks	Low	Medium	Medium	Low	Medium	Medium
Finkel 2004²⁸⁵	12 weeks	Medium	Medium	Low	Medium	Medium	Medium
Levkovitz 2001²⁸⁶	7 weeks	Medium	Low	Medium	High	High	High

Appendix Table H.12Primary outcomes summary low and medium risk of bias studies: antidepressant versus placebo

Drug Comparison	AD Severity	Study Characteristics: Author/Year Followup Risk of Bias	General Behavior	Agitation	Aggression	Psychosis	Depression	Anxiety	Disinhibited Sexual Behavior	Harms	Quality of Life
Citalopram vs. Placebo	Probable AD	Porsteinsson 2014 ²⁸² Leonpacher 2016 ²⁸³ 9 weeks Medium	NR	CMAI Estimated treatment effect, mean (95% CI) -2.38 (-4.13 to -0.63) mADCS-CGIC Estimated treatment effect, mean (95% CI) 2.13 (1.23 to 3.69)’ Moderate or marked improvement I: 40% C: 26% NBRS-A Estimated treatment effect, mean (95% CI) -0.93 (-1.80 to -0.06) NPI-agitation subscale Estimated treatment effect, mean (95% CI) -0.78 (-1.77 to 0.21) ≥50% improvement in domain score C: 64% I: 43%	NR	NPI-delusions subscale ≥50% improvement in domain score C: 54% I: 38%	NR	NR	NR	Confusion I: 76.7 % (69/90) C: 83.7% (72/86) p=0.24 Falls I: 16.7% (15/90) C: 11.6% (10/86) p=0.34 Somnolence I: 52.2% (47/90) C: 48.8% (42/86) p=0.65
Citalopram vs. Placebo	Zhou 2019²⁸¹ 12 weeks Medium		NPI Agitation/Aggression Mean change from baseline (SD) I: -2.77 (1.22) C: -2.03 (0.96) p=0.004 (SMD, 0.67 [95% CI, 0.22 to 1.13])^*	See adjacent cell	NPI Hallucinations Mean change from baseline (SD) I: -0.08 (0.35) C: -0.10 (0.45) p=0.78 (SMD, -0.05 [95% CI, -0.49 to 0.39])^* NPI Delusions Mean change from baseline (SD) I: -0.08 (0.27) C: -0.05 (0.22) p=0.649 (SMD, 0.12, [95% CI, -.032 to 0.57])^*				SAEs “Similar” rates between groups, no data reported.
Mirtazapine vs. Placebo	Probable or possible AD	Banerjee 2011²⁸⁴ 39 weeks Medium	NPI Estimated treatment effect, mean (95% CI) -1.51 (-6.25 to 3.24)	NR	NR	NR	CSDD Mean difference from placebo (95% CI) -0.66 (-2.12 to 0.79)	NR	NR	Overall Number of participants (number of events) I: 44 (96) C: 29 (58) p=0.031	DEMQOL Mean treatment effect: -0.03 [95% CI -3.80, 3.75] p=0.99 EQ5D Mean difference from placebo -1.18 [95% CI -9.25, 6.89] p=0.78 (self-report); -1.11 [95% CI -7.44, 5.21] p=0.73 (carer-report)
Sertraline vs. Placebo	Probable or possible AD	Banerjee 2011²⁸⁴ 39 weeks Medium	NPI Estimated treatment effect, mean (95% CI) 2.02 (-2.94 to 6.97)	NR	NR	NR	CSDD Mean difference from placebo (95% CI) 0.37 (-1.12 to 0.87)	NR	NR	Overall Number of participants (number of events) I: 46 (86) C: 29 (58) p=0.01	DEMQOL: Mean treatment effect: -1.76 [95% CI -5.75, 2.23] p=0.39 EQ5D Mean difference from placebo -4.34 [95% CI -12.56, 3.88] p=.30 (self-report); -0.27 [95% CI -6.77, 6.24] p=0.94 (carer-report)
Donepezil + Sertraline vs. Donepezil + Placebo	Probable or possible AD	Finkel 2004 ²⁸⁵ 12 weeks Medium	NR	CMAI Mean change score I: -3.6 ± 1.4 C: -2.7 ± 1.2		NR	NR	NR	NR	NR

*: Calculated by EPC
: Abbreviations: AD=Alzheimer’s Disease; RoB=Risk of Bias; C=Control; CMAI=Cohen-Mansfield Agitation Inventory; CSDD=Cornell Scale for Depression in Dementia; I=Intervention; GHQ-12=General Health Questionnaire 12; mADCS-GCIC=modified Alzheimer Disease Cooperative Study-Clinical Global Impression of Change; NBRAS-A=Neurobehavioral Rating Scale Agitation subscale; NPI=Neuropsychiatric Inventory; SE=Standard Error; SAEs=Serious Adverse Events

Appendix Table H.13Secondary outcomes summary low and medium risk of bias studies: antidepressant versus placebo

Drug Comparison	AD Severity	Study Characteristics: Author/Year Followup Risk of Bias	Change in Caregiver or Staff Outcomes
Citalopram vs. Placebo	Probable AD	Porsteinsson 2014²⁸², Leonpacher 2016²⁸³ 9 weeks Medium	Addition of Citalopram compared with placebo significantly reduced caregiver distress (NPI-caregiver distress subscore estimated treatment effect -2.70 (-4.94 to -0.47), p=0.02; favors drug.
Citalopram vs. Placebo	Probable AD	Zhou 2019²⁸¹ 12 weeks Medium	Addition of citalopram compared with placebo significantly reduced caregiver distress (NPI caregiver distress subscore mean change from baseline 3.18 [SD 1.57] vs. 1.59 [SD 1.35], p<0.001).
Mirtazapine vs. Placebo	Probable or possible AD	Banerjee 2011 ²⁸⁴ 39 weeks Medium	Mirtazapine vs. placebo favored scores associated with carer burden (Zarit) and physical life quality (SF-PCS 12; physical). Mirtazapine vs. placebo worsened scores associated with carer mental health (GHQ) and carer mental life quality (SF-12 MCS; mental).
Sertraline vs. Placebo	Probable or possible AD	Banerjee 2011 ²⁸⁴ 39 weeks Medium	No significant difference for caregiver burden outcomes (carer burden (Zarit), CHQ, SF-12 MCS, SF-12 PCS).
Donepezil + Sertraline vs. Donepezil + Placebo	Probable or possible AD	Finkel 2004²⁸⁵ 12 weeks Medium	No significant between-treatment group efficacy differences for caregiver burden outcomes (CBQ). Mean change score from baseline CBQ was -1.3±0.9 and 0.3±0.8 (p=0.12) for the sertraline + donepezil and donepezil + placebo group, respectively.

: Abbreviations: AD=Alzheimer’s Disease; NPI=Neuropsychiatric Inventory; CBQ=Caregiver Burden Questionnaire; CGI-S=Clinical Global Impression Severity scale; GHQ=General Health Questionnaire; PCS=Physical Composite Score; MCS-Mental Composite Score; NR=Not Reported; RoB=Risk of Bias; SF=Short-Form Health Survey

Appendix Table H.14Summary of strength of evidence: citalopram versus placebo

Antidepressant vs. Placebo Outcome AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
Agitation Probable AD	9 weeks	1 RCT (n=186)	Favors citalopram	Medium	Unknown	Direct	Imprecise	Insufficient
Agitationy^*	12 weeks	Agitation/Aggression	Favors citalopram	Medium	Unknown	Direct	Precise	Insufficient
Psychosis^*	12 weeks	Psychosis	No difference	Medium	Unknown	Direct	Imprecise	Insufficient

*: Citalopram studies not pooled as all participants in 12-week study (Zhou 2019) received memantine.
: Abbreviations: AD=Alzheimer’s Disease; RCT=Randomized Controlled Trial

Appendix Table H.15Summary of strength of evidence: donepezil and sertraline versus donepezil and placebo

Antidepressant vs. Placebo Outcome AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
Agitation Probable or possible AD	12 weeks	1 RCT (n=244)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: AD=Alzheimer’s Disease; RCT=Randomized Controlled Trial

Appendix Table H.16Summary of strength of evidence: mirtazapine versus placebo

Antidepressant vs. Placebo Outcome AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
General behavior Probable or possible AD	39 weeks	1 RCT (n=218)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient
Depression Probable or possible AD	39 weeks	1 RCT (n=218)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: AD=Alzheimer’s Disease; RCT=Randomized Controlled Trial

Appendix Table H.17Summary of strength of evidence: sertraline versus placebo

Antidepressant vs. Placebo Outcome AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
General behavior Probable or possible AD	39 weeks	1 RCT (n=219)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient
Depression Probable or possible AD	39 weeks	1 RCT (n=219)	No difference	Medium	Unknown	Direct	Imprecise	Insufficient

: Abbreviations: AD=Alzheimer’s Disease; RCT=Randomized Controlled Trial

Donepezil Versus Placebo

Appendix Table H.18Risk of bias ratings: donepezil versus placebo

Study

Time

Selection Bias

Attrition Bias

Performance Bias

Detection Bias

Reporting Bias

Overall Rating

Howard 2007²⁸⁷

12 weeks

Low

Medium (CMAI)

High (NPI; NPI Caregiver Distress)

Low

Medium (CMAI)

High (NPI; NPI Caregiver Distress)

: Abbreviations: CMAI=Cohen-Mansfield Agitation Inventory; NPI=Neuropsychiatric Inventory

Appendix Table H.19Characteristics of eligible studies: donepezil versus placebo

Study Characteristics:

Author/Year

Design

Country

Risk of Bias

Population Characteristics

Intervention: Intervention mode n/group Maximum dose Frequency Duration

Comparison: Comparison mode n/group Dose Frequency Duration

Outcome Timing

Outcome Domain [Instrument]

Howard 2007²⁸⁷

RCT

259

AD Severity NR

Mean Age 84.7

85% Female

97% White

Education 10.5 years

Baseline Cognition:

MMSE 8.2

Baseline BPSD:

NPI 23.7

CMAI 61.6

Eligible participants had no response to a prior psychosocial program

Donepezil, 10mg/day, titrated in 5mg/day increments, for 12 weeks

No concomitant psychosocial intervention reported

Placebo

No concomitant psychosocial intervention reported

12 weeks

Agitation

Cohen-Mansfield Agitation

Inventory

Adverse Events

Falls

Stroke

Death

: Abbreviations: AD=Alzheimer’s Disease

Appendix Table H.20Primary outcomes summary low and medium risk of bias studies: donepezil versus placebo

Drug Comparison

AD Severity

Study Characteristics:

Author/Year

Followup

Risk of Bias

Agitation

Aggression

Hypersexuality

Harms

Donepezil vs. Placebo

Howard, 2007²⁸⁷

12 weeks

Medium

Proportion making 30% reduction in baseline CMAI score

I: 19.5% C: 20.4%; RR=0.96; 95% CI=0.56 to 1.62

CMAI, change in total score, difference in mean change (adjusting for baseline CMAI score and stratification variables)

(95% CI): -0.18 (-4.59 to 4.22)

p = 0.94

CMAI, change in total score, standardized mean difference in (adjusting for baseline CMAI score and stratification variables)

(95% CI): 0.00 (-0.24, 0.25)

N/A

Falls

I: 2

C: 2

p-value not reported

Stroke

I: 1

C: 0

p-value not reported

Death

I: 3

C: 4

p-value not reported

: Abbreviations: AD=Alzheimer’s Disease

Appendix Table H.21Summary of strength of evidence: donepezil versus placebo

Outcome

AD Severity

Timing

# Studies/Design (n analyzed)

Finding or Summary Statistic

Study Limitations

Consistency

Directness

Precision

Overall Grade/Conclusion

Cohen Mansfield Agitation Inventory

12 weeks

1 RCT

Proportion making 30% reduction in baseline CMAI score: I: 19.5%

C: 20.4%; RR=0.96; 95% CI=0.56 to 1.62

CMAI, total score, difference in mean change (adjusting for baseline CMAI score and stratification variables): (95% CI): -0.18 (-4.59 to 4.22)

p = 0.94

CMAI, change in total score, standardized mean difference in (adjusting for baseline CMAI score and stratification variables) (95% CI):

0.00 (-0.24, 0.25)

Medium

Unknown

Direct

Imprecise

Low

Anticonvulsants Versus Placebo

Appendix Table H.22Risk of bias ratings: anticonvulsants versus placebo for treatment of BPSD

Study	Time		Selection Bias	Attrition Bias	Performance Bias	Detection Bias	Reporting Bias	Overall Rating
Tariot 2005²⁸⁸	6 weeks		Low	Medium	Medium	Medium	Low	Medium

: BPSD=behavioral and psychological symptoms of dementia

Appendix Table H.23Characteristics of eligible studies: anticonvulsants versus placebo for treatment of BPSD

Study Characteristics:

Author/Year

Design

Country

Risk of Bias

Population Characteristics

Intervention: Intervention mode Components Frequency Duration

Comparison: Comparison mode Components Frequency Duration

Outcome Timing

Outcome

Domain [Instrument]

Tariot 2005²⁸⁸

RCT

153

AD Severity NR

Mean Age 86.0

69% Female

92% White

Education NR

Baseline Cognition:

MMSE 10.7

Baseline BPSD:

BPRS, total score 33.7

BPRS agitation score 8.3

CMAI, 36.5

Divalproex Sodium, target dose of 750mg/day, titrated in 250mg/day increments every 3 days, for 12 weeks (6 weeks double blind; 6 weeks open label)

No concomitant psychosocial intervention reported

Placebo

No concomitant psychosocial intervention reported

6 weeks

Agitation

Brief Psychiatric Rating Scale, agitation factor

Cohen-Mansfield Agitation

Inventory

Adverse Events

Falls

Psychiatric disorders^*

*: Composite outcome that included somnolence, but also agitation and aggression.
: AD=Alzheimer’s Disease; BPRS=Brief Psychiatric Rating Scale; BPSD=behavioral and psychological symptoms of dementia; MMSE=Mini-Mental State Exam; NPI=Neuropsychiatric Inventory; CMAI=Cohen-Mansfield Agitation Inventory

Appendix Table H.24Primary outcomes summary low and medium risk of bias studies: anticonvulsants versus placebo for treatment of BPSD

Drug Comparison

AD Severity

Study

Followup

RoB

Agitation

Aggression

Hypersexuality

Harms

Divaloprex Sodium vs. Placebo

Tariot 2005²⁸⁸

4 weeks

Medium

BPRS, Agitation Factor, 6-Week

Change (SD)

From Baseline

I: -2.08 (3.1)

C: -1.72 (3.1)

95% CI for Difference: -1.4 to 0.6

Standard Mean Difference (95% CI)

-0.12 (-0.44 to 0.21)

CMAI

I: -3.5 (14.5)

C: -6.7 (15.6)

95% CI for Difference: -8.0 to 1.4

Standard Mean Difference (95% CI)

-0.21 (-0.54 to 0.11)

N/A

Falls

I: 21% divalproex vs. C: 17% placebo

p=0.54

“Psychiatric disorders”*:

I: 23% divalproex

C: 15% placebo

p=0.30

: AD=Alzheimer’s Disease; BPRS=Brief Psychiatric Rating Scale; BPSD=behavioral and psychological symptoms of dementia; MMSE=Mini-Mental State Exam; NPI=Neuropsychiatric Inventory; CMAI=Cohen-Mansfield Agitation Inventory

Appendix Table H.25Summary of strength of evidence: anticonvulsants versus placebo for treatment of BPSD

Outcome	AD Severity	Timing	# Studies/Design (n analyzed)	Finding or Summary Statistic	Study Limitations	Consistency	Directness	Precision	Overall Grade/Conclusion
Agitation (BPRS Agitation; CMAI)		6 weeks	1 RCT	No significant difference between groups on either of two tests (BPRS, CMAI)	Medium	Unknown	Direct	Imprecise	Insufficient

: BPSD=behavioral and psychological symptoms of dementia

Memantine Versus Placebo

Appendix Table H.26Risk of bias ratings: memantine versus placebo for treatment of BPSD

Study	Time	Selection Bias	Attrition Bias	Performance Bias	Detection Bias	Reporting Bias	Overall Rating
Fox 2012²⁸⁹	6 weeks	Low	High	Low	Low	Low	High
Hermann 2013¹⁸³	24 weeks	Medium	Medium	High	Low	High	High

Appendix Table H.27Characteristics of eligible studies: memantine versus placebo for treatment of BPSD

Study Characteristics:

Author/Year

Design

Country

Risk of Bias

Population Characteristics

Intervention: Intervention mode Components Frequency Duration

Comparison: Comparison mode Components Frequency Duration

Outcome Timing

Outcome

Domain [Instrument]

Fox 2012²⁸⁹

RCT

United Kingdom

High

153

Moderate to Severe AD

Mean Age 85

74% Female

98% White

Education NR

Baseline Cognition:

MMSE 7.3

Baseline BPSD:

CMAI 68.3

NPI 36.6

Memantine, 20mg/day in 2 doses, titrated in 10mg/day increments, for 12 weeks

Placebo

6 and 12 weeks

Agitation

CMAI

General BPSD

NPI

Adverse Events

Death

Hermann 2013¹⁸³

RCT

Canada

369

AD Severity NR

Mean Age 75

58% Female

Race NR

Education NR

Baseline Cognition:

MMSE 11.9

Baseline BPSD:

CMAI (physical): 16

NPI: 30

Memantine, 20mg/day in 1 dose, titrated in 5 mg/day increments, for 24 weeks

Placebo

24 weeks

Agitation

CMAI

General BPSD

NPI

Adverse Events

Falls

: NR=not reported

Estrogen Versus Placebo

Appendix Table H.28Risk of bias ratings: estrogen versus placebo

Study	Time	Selection Bias	Attrition Bias	Performance Bias	Detection Bias	Reporting Bias	Overall Rating
Hall 2005²⁹⁰	8 weeks	Low	High	Low	Low	Medium	High
Kyomen 1999²⁹¹	4 weeks	Medium	Low	High	High	High	High

Appendix Table H.29Characteristics of eligible studies: estrogen versus placebo

Study Characteristics:

Author/Year

Design

Country

Risk of Bias

Population Characteristics

Intervention: Intervention mode Components Frequency Duration

Comparison: Comparison mode Components Frequency Duration

Outcome Timing

Outcome

Domain [Instrument]

Hall 2005²⁹⁰

RCT

Australia

AD Severity NR

Mean Age 78.5

0% Female

Race NR

Education NR

Baseline Cognition:

MMSE 17.2

Baseline BPSD:

Rating Scale for Aggressive Behavior in Elderly (RAGE) 22.5

Estrogen patch, 10mcg/day, titrated in 5mcg/day increments

Placebo

8 weeks

Aggression

Rating Scale for Aggressive

Behavior in the Elderly (RAGE)

Adverse Events

The study did not report data for any of the harms of interest specified in this systematic review protocol.

Kyomen 1999²⁹¹

RCT

Moderate to Severe AD

Mean Age 83.7

85.7% Female

Race NR

Education NR

Baseline Cognition:

MMSE 4.72

Baseline BPSD NR

Estrogen pill, 2.5 mg/day, titrated in 0.625mg doses

Placebo

4 weeks

Aggression

Overt Aggression Scale, modified: physical and verbal aggression domains

Disinhibited Sexual Behavior

Overt Aggression Scale, modified: sexually aggressive domain

Adverse Events

The study did not report data for any of the harms of interest specified in this systematic review protocol.

Bookshelf ID: NBK556563

Contents

< Prev Next >

PubReader
Print View
Cite this Page
Fink HA, Hemmy LS, Linskens EJ, et al. Diagnosis and Treatment of Clinical Alzheimer’s-Type Dementia: A Systematic Review [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2020 Apr. (Comparative Effectiveness Review, No. 223.) Appendix H, Key Question 6: Efficacy and Harms of Prescription Drug Treatment Versus Placebo for Behavioral and Psychological Symptoms of Dementia.
PDF version of this title (14M)
Disable Glossary Links

Key Question 6: Efficacy and Harms of Prescription Drug Treatment Versus Placebo...
Key Question 6: Efficacy and Harms of Prescription Drug Treatment Versus Placebo for Behavioral and Psychological Symptoms of Dementia - Diagnosis and Treatment of Clinical Alzheimer’s-Type Dementia: A Systematic Review

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

Bookshelf

Diagnosis and Treatment of Clinical Alzheimer’s-Type Dementia: A Systematic Review [Internet].

Appendix HKey Question 6: Efficacy and Harms of Prescription Drug Treatment Versus Placebo for Behavioral and Psychological Symptoms of Dementia

Antipsychotics Versus Placebo

Appendix Table H.1Characteristics of eligible studies: antipsychotics versus placebo

Appendix Table H.2Risk of bias ratings: antipsychotics versus placebo

Appendix Table H.3Primary outcomes summary low and medium risk of bias studies: antipsychotics versus placebo

Appendix Table H.4Primary outcomes summary low and medium risk of bias studies: antipsychotics dose response

Appendix Table H.5Summary of strength of evidence: aripiprazole versus placebo

Appendix Table H.6Summary of strength of evidence: haloperidol versus placebo

Appendix Table H.7Summary of strength of evidence: pimavanserin versus placebo

Appendix Table H.8Summary of strength of evidence: quetiapine versus placebo

Appendix Table H.9Summary of strength of evidence: haloperidol standard dose (2-3 mg) versus low dose (0.5-0.75 mg)

Antidepressants Versus Placebo

Appendix Table H.10Characteristics of eligible studies: antidepressants versus placebo

Appendix Table H.11Risk of bias ratings: antidepressant versus placebo

Appendix Table H.12Primary outcomes summary low and medium risk of bias studies: antidepressant versus placebo

Appendix Table H.13Secondary outcomes summary low and medium risk of bias studies: antidepressant versus placebo

Appendix Table H.14Summary of strength of evidence: citalopram versus placebo

Appendix Table H.15Summary of strength of evidence: donepezil and sertraline versus donepezil and placebo

Appendix Table H.16Summary of strength of evidence: mirtazapine versus placebo

Appendix Table H.17Summary of strength of evidence: sertraline versus placebo

Donepezil Versus Placebo

Appendix Table H.18Risk of bias ratings: donepezil versus placebo

Appendix Table H.19Characteristics of eligible studies: donepezil versus placebo

Appendix Table H.20Primary outcomes summary low and medium risk of bias studies: donepezil versus placebo

Appendix Table H.21Summary of strength of evidence: donepezil versus placebo

Anticonvulsants Versus Placebo

Appendix Table H.22Risk of bias ratings: anticonvulsants versus placebo for treatment of BPSD

Appendix Table H.23Characteristics of eligible studies: anticonvulsants versus placebo for treatment of BPSD

Appendix Table H.24Primary outcomes summary low and medium risk of bias studies: anticonvulsants versus placebo for treatment of BPSD

Appendix Table H.25Summary of strength of evidence: anticonvulsants versus placebo for treatment of BPSD

Memantine Versus Placebo

Appendix Table H.26Risk of bias ratings: memantine versus placebo for treatment of BPSD

Appendix Table H.27Characteristics of eligible studies: memantine versus placebo for treatment of BPSD

Estrogen Versus Placebo

Appendix Table H.28Risk of bias ratings: estrogen versus placebo

Appendix Table H.29Characteristics of eligible studies: estrogen versus placebo

Views

In this Page

Other titles in this collection

Recent Activity