The Canadian Coalition for Seniors’ Mental Health Canadian Clinical Practice Guidelines for Assessing and Managing Behavioural and Psychological Symptoms of Dementia (BPSD)

Abstract

In Canada, approximately 730,000 people are currently living with dementia. Over 75% will experience behavioural and psychological symptoms of dementia (BPSD). There is a lack of consensus on best practices for the assessment and management of BPSD. In 2024, the Canadian Coalition for Seniors Mental Health (CCSMH) developed a Clinical Practice Guideline (CPG) for assessing and managing BPSD, specifically for agitation, depression, anxiety, psychosis, and sexual expressions of potential risk, and deprescribing antipsychotics and psychotropic medications. Development of the BPSD CPG followed the Guideline International Network (GIN)—McMaster Guideline Development checklist. The guideline is intended for people living with dementia, caregivers of people living with dementia, and health-care providers in community, outpatient, inpatient, long-term care, and other residential care settings. Recommendations were informed by a Canada-wide prioritization exercise to identify CPG topics and preferred terms for describing BPSD. A systematic review of existing dementia CPGs, an overview of systematic reviews on assessing and managing BPSD, and systematic reviews of tools for measuring psychosis, anxiety, and depressive symptoms in people living with dementia was undertaken, along with a rapid review of studies of pharmacologic and nonpharmacologic interventions for reducing sexual expressions of potential risk in people living with dementia. Guideline panel members voted on recommendation strength and quality of evidence, per the Grading of Recommendations, Assessment, Development, and Evaluations approach. This CPG resulted in 11 good practice statements and 63 guideline recommendations that will inform BPSD best practices in a Canadian health-care context.

INTRODUCTION

Over 55 million people are currently living with dementia worldwide.⁽¹⁾ In Canada, 733,000 people live with dementia, and approximately 1.7 million people are expected to be living with dementia by 2050.⁽²⁾ Alzheimer disease (AD) is the most frequently diagnosed cause of dementia and accounts for 50% to 70% of cases.⁽²⁾ Other causes of dementia include vascular, mixed, Lewy body, frontotemporal, and Parkinson disease.^(1,3) The annual economic impact of dementia care in Canada is substantial. The costs associated with dementia care include direct medical expenses, such as hospital stays and emergency department visits, and were estimated at $40 billion in 2022.⁽⁴⁾ The indirect costs associated with dementia care are frequently driven by lost income for caregivers through workplace absenteeism, reduced productivity in the workplace, and the early retirement of caregivers of people living with dementia.⁽⁴⁾ In 2018, an estimated 350,000 people provided care to people living with dementia, spending an average of 26 hours per week.⁽⁵⁾ In 2020, the Canadian Centre for Economic Analysis calculated that, at federal minimum wage dollars of $15.50, caregivers had provided the equivalent of more than $7.3 billion of care in that year.^(4,6)

In addition to the cognitive symptoms of dementia, more than 75% of community-dwelling people living with dementia, and more than 80% of people living with dementia in long-term care (LTC) homes, will experience changes to mood, perception, and behaviours, often referred to as the behavioural and psychological symptoms of dementia (BPSD).^(1,7) BPSD include a variety of symptoms such as agitation, psychosis, anxiety, and depression, and occur in all dementia types.^(7,8) BPSD are associated with adverse mental health outcomes for people living with dementia and their caregivers, including decreased quality of life and increased caregiver burden.^(9–12) BPSD are also associated with more rapid cognitive and physical decline, and earlier admissions to LTC.^(10,13) The type of BPSD experienced by individuals can vary depending upon the underlying cause of dementia and stage or severity of dementia.^(14,15)

There is a lack of current clinical guidance on best practices for the assessment and management of BPSD. In 2023, our guideline group published a systematic review of clinical practice guidelines (CPG) on dementia care making at least one recommendation on BPSD assessment or management.⁽¹¹⁾ No current CPGs focused specifically on the assessment and management of BPSD, although some existing CPGs included recommendations on BPSD.⁽¹¹⁾ Therefore, the goal of our CPG was to develop recommendations for assessing and managing common BPSD in AD and related dementias, specifically agitation, anxiety, psychosis (hallucinations and delusions), depression and depressive symptoms, sexual expressions of potential risk, as well as for deprescribing of antipsychotics and other psychotropic medications. We intend for the CCSMH CPG on assessing and managing BPSD to inform shared decision-making among people living with dementia, caregivers of people living with dementia, and health-care providers across community, outpatient, inpatient, LTC home, and other residential care settings. We also hope that this CPG will support health-care leaders, policymakers, and researchers to identify potential areas for the future development of health services and interventions to prevent and reduce BPSD.

METHODS

The development of this CPG followed the Guideline International Network (GIN)—McMaster Guideline Development Checklist.^(16,17) The guideline panel consisted of 15 members (Appendix A), including the two guideline co-chairs (DPS and JAW). Guideline panel membership was based on members’ area of professional expertise in the care of people living with dementia, as well as multidisciplinary and geographic representation from across Canada. Guideline panel members included expertise in geriatric psychiatry, geriatric medicine, family medicine, pharmacy, psychology, neurology, nursing, and pharmacology. All panel members completed INGUIDE Level 1: Guideline Group or Panel Member training for participants of guideline panels.⁽¹⁶⁾ Guideline working group members were offered an honorarium for their participation in the CPG development process. Guideline panel members met monthly by videoconference throughout the guideline development process. Conflict of interest statements were completed at two points: the beginning of the development of the CPG; and the beginning of the writing of the CPG.⁽¹⁸⁾ Terms of reference included the disclosure of all perceived and actual conflicts of interest for each panel member. Panel members who identified conflicts of interest participated in discussions and voting (Appendix A).

A Canada-wide guideline topic prioritization exercise survey was completed, and the results informed the development of the guidelines. The survey reached a diverse group of 254 respondents, including people living with dementia, family and friend caregivers, and health-care providers. The distribution of the top groups of respondents was as follows: nurses (83%); family and friend caregivers (38%); psychiatrists or geriatric psychiatrists (37%); allied health-care providers (31%); and organizational managers (31%). The survey identified preferred BPSD terminology and prioritized CPG topics. Prioritized topics included diagnostic criteria for individual BPSD; detecting and measuring BPSD; and pharmacologic and nonpharmacologic interventions. Outcomes, such as individual BPSD symptoms, that were voted as critical by guideline panel members were included in this CPG. Critical BPSD topics selected for the guideline included agitation, psychosis, depression and depressive symptoms, anxiety, and sexual expressions of potential risk. Deprescribing was identified as a separate priority topic area. Due to resource constraints related to the funding available for developing this CPG, we present recommendations related to AD and related dementias of vascular and mixed type. Other types of dementia (for example, dementia with Lewy bodies and frontotemporal dementia) were not included in the CPG.

CPG recommendations were informed by a systematic review of existing dementia CPGs to identify recommendations related to assessing and managing BPSD. A rapid overview of systematic reviews on assessing and managing BPSD was undertaken, as well as a rapid review of studies describing the efficacy or safety of pharmacological, non-pharmacological, and psychosocial interventions for sexual expressions of potential risk. A systematic review describing the diagnostic test accuracy of tools for assessing psychosis, as well as updated systematic reviews describing the diagnostic test accuracy of tools for assessing anxiety and depressive symptoms, were also completed.^(11,19)

The draft recommendations were developed through topic working groups. Each CPG topic working group revised the draft recommendations which were presented to the full guideline panel. The presentation of each recommendation was comprised of an evidence-to-decision table of the strength and quality of evidence, a review and discussion of the recommendation, and voting. Recommendations were revised by the CPG co-chairs based on discussions with guideline panel members, which provided an opportunity to modify recommendations based on group discussion prior to voting. Recommendations were approved if at least 80% of guideline panel members voted in favour of the recommendation.

Recommendations were rated as either strong or conditional, and reflected the confidence of the guideline panel that the benefits of each recommendation outweighed the risks (Appendix B). The quality of evidence was rated per the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) methodology as very low, low, moderate, or high.⁽²⁰⁾ Recommendation strength also considered the feasibility of implementing recommendations, recommendation cost, equity, and the values and preferences of people living with dementia and their caregivers. Resource implications were specific to providing care in Canadian environments. Under the GRADE framework, actionable good practice statements can also be included, with clear rationale for their consideration.⁽²⁾ Therefore, the guideline panel created good practice statements regarding dementia care.

Recommendations were accepted at 80% consensus among the guideline panel members, with most recommendations receiving unanimous agreement. The guideline panel acknowledged that the majority of pharmacologic recommendations for treatments of BPSD are not Health Canada-approved at this time, unless otherwise stated in the CPG.

Recommendations for nonpharmacologic interventions are summarized as categories of interventions, with specific recommendations noted for agitation, psychosis, depressive symptoms and depression, anxiety, and sexual expressions of potential risk.⁽²¹⁾ The guideline panel made recommendations for primary care, tertiary or specialty care, and LTC homes, where such information was available.

We anticipate updating the CPG recommendations within three to four years. The CCSMH will re-survey people living with dementia, caregivers of people living with dementia, and health-care providers to identify priorities for future CPG recommendation development. Key informants and experts in the area of providing care for people living with dementia will be asked to identify new evidence developments. The guideline panel noted that, although apathy and sleep disturbances were deemed less critical than other topics during the prioritization exercise, stakeholders and some guideline panel members later determined these to be important topics to be included in the next iteration of the CPG. Resource constraints prevented the development of recommendations on these topics, and the feasibility of including such BPSD in future CPG updates will be considered. Future iterations will also include caregiver support recommendations.

RESULTS

This CPG made 11 good practice statements and 63 clinical practice recommendations for assessing and managing BPSD. The full BPSD CPG guidelines are available on the CCSMH website (https://ccsmh.ca/areas-of-focus/dementia/).

General Principles of Assessing and Managing BPSD

The guideline panel made 11 good practice statements for assessing and managing BPSD (Table 1). These statements emphasized respecting autonomy and collaboration in the decision-making process; communicating with person-centred language; and incorporating the values, preferences, and prior expressed wishes of people living with dementia (Table 1). Implementing a tailored interdisciplinary approach to the assessment and management of BPSD that integrates the biopsychosocial contributors to BPSD was also emphasized⁽²²⁾ (Table 1).

TABLE 1.

General principles of assessing and managing BPSD

Good Practice Statement 1: Provide health-care providers and caregivers with the education and organizational system of support needed to implement a structured approach for assessing and managing BPSD.

Good Practice Statement 2: Obtain informed consent for the assessment and management of BPSD.

Good Practice Statement 3: Incorporate information about the values, goals of care, and advance wishes of the person living with dementia in assessing and managing BPSD.

Good Practice Statement 4: Review the underlying etiology of dementia, stage of dementia (mild, moderate, advanced), and the specific BPSD of concern, including the frequency, duration, severity and any associated risks when assessing BPSD.

Good Practice Statement 5: Conduct a thorough evaluation of potential biological contributors to BPSD, including an assessment for delirium, a general medical and mental health history, medication review, substance use, hearing and vision assessment, pain, and other contributors.

Good Practice Statement 6: Conduct a thorough review of the personhood of the person living with dementia, including sex, gender, sexual orientation, language, race, ethnicity, cultural background, trauma history, religious or spiritual beliefs, and other factors when attempting to understand contributors to BPSD, and to inform assessment and management.

Good Practice Statement 7: Conduct a thorough review of psychosocial and environmental contributors to BPSD.

Good Practice Statement 8: Use person-centred language and incorporate specific descriptions of the BPSD using language appropriate for the intended audiences when communicating with people living with dementia, caregivers, or health-care providers.

Good Practice Statement 9: Psychosocial interventions are recommended for all BPSD, either alone or in combination with pharmacological treatments.

Good Practice Statement 10: Select and tailor interventions that are likely to be safe and effective for the specific BPSD and avoid treatments that are neither safe nor effective.

Good Practice Statement 11: Routinely assess the effectiveness of the BPSD management plan and evaluate the plan to consider adjusting, changing or discontinuing strategies as appropriate

Recommendations for Assessing and Managing BPSD

The guideline panel made 63 recommendation statements for assessing and managing agitation, psychosis, depression and depressive symptoms, anxiety, and sexual expressions of risk. A high-level summary of recommendations for each BPSD topic area is presented in the following section.

Agitation

The guideline panel made 30 recommendation statements on assessing and managing agitation in people living with dementia. The panel recommended the International Psychogeriatric Association (IPA) consensus criteria for agitation in cognitive disorders for the diagnosis of agitation in dementia⁽²³⁾ (Table 2), with one dissenting opinion (Appendix C). Additional recommendations were provided for evidence-based tools to assess agitation in dementia (Table 2). Nonpharmacologic recommendations included interdisciplinary approaches to care such as education about BPSD, physical exercise, home-based problem-solving behaviour therapy, and structured approaches to individualized care plans (Table 3). Recommendations for individualized care plans included psychosocial interventions with an emphasis on meaningful activities, pleasurable activities, and music therapy with the preferred music of the person living with dementia (Table 3).

TABLE 2.

Diagnostic and detection and assessment tools (strength, quality of evidence)

	Agitation	Psychosis	Depression and Depressive Symptoms	Anxiety	Sexual Expression of Potential Risk
Diagnosis	International Psychogeriatric Associationa consensus criteria for agitation in cognitive (S, M)	International Psychogeriatrics Association criteria for psychosis in major neurocognitive disorders (S, M)	National Institutes of Mental Health – depression in Alzheimer’s disease criteria for the diagnosis of depression in dementia (S, L)	Diagnostic and Statistical Manual of Mental Disorders-5-TR (C, VL)	Defined as “disruptive verbal or physical act of an explicit or perceived sexual nature, which is either intrusive or engaged in without the consent of those around the person living with dementia” (C, VL)
Detection and Assessment Tool	NPI-Agitation subscale, NRS, EBRS, PAS, Spanish NPI-Agitation, F-RAGE, Speciality clinics: (C, L) CMAI Speciality clinics (C, VL) RAGE, PAS, NRS, EBRS, Spanish NPI-Agitation, F-RAGE, CMAI Primary care and LTC: (C, VL)	NPI psychosis subscale (C, VL)	CSDD Specialty clinics (C, M) Primary care and LTC: (C, L)	RAID Speciality Clinics: (S, M) Primary care and LTC: (C, L)	St. Andrew’s Sexual Behaviour Assessment Scale (C, VL)

^aInternational Psychogeriatric Association (IPA) consensus criteria for agitation in cognitive disorders.

NPI-Agitation = Neuropsychiatric Inventory subscale; NRS = Neurobehavioral Rating Scale; EBRS = Empirical Behavioural Rating Scale; PAS = Psychogeriatric Assessment Scale; F-RAGE = Rating Scale for Aggressive Behavior in the Elderly – French version; CMAI = Cohen Mansfield Agitation Inventory; RAGE = Rating Scale for Aggressive Behaviour in the Elderly; NPI = Neuropsychiatric Inventory; CSDD = Cornell Scale for Depression in Dementia; RAID = Rating Anxiety in Dementia; (S) = strong recommendation; (C) = conditional recommendation; (M) = moderate quality evidence; (L) = low quality evidence; (VL) = very low quality evidence.

TABLE 3.

Psychosocial and non-pharmacological interventions: BPSD syndrome (strength, quality of recommendation)

	Agitation	Psychosis	Depression; Depressive Symptoms	Anxiety	Sexual Expressions of Potential Risk
Interdisciplinary Approaches	Health-care provider education structured approaches to assessment, individualized care plans, personalized meaningful activities (S, M)	Health-care provider training, communication strategies (C, VL)	Health-care provider education, structured approaches to assessment, individualized care plans, personalized meaningful activities (S, M)	Education and training programs for caregivers (C, L)	Patient and caregiver education, removal of environmental triggers, strategies to engage in other activities, changes to environment (C, VL)
Behavioural Therapies	Physical exercise (C, VL)	Physical exercise (C, VL)	Cognitive stimulation therapy (S, M) Reminiscence therapy: LTC (S, M) Community settings: (C, L) Occupational therapy (C, L) Home-based problem-solving behaviour therapy (C, L) Physical exercise (S, M)	Cognitive behavioral therapy, adapted, in mild to moderate dementia (C, L)	—
Personalized Pleasant Activities	Music therapy with preferred music (S, M) Animal assisted therapy (C, VL) Robotic pets (C, M)	Music therapy with preferred music, art-based activities (C, VL)	Animal assisted therapy (C, VL-L) Robotic pets (C, M)	Music therapy with preferred music (S, M)	—
Sensory Approaches	Massage (C, M) Aromatherapy (C, L)	—	Massage and touch therapy: mild to moderate dementia (S, M); severe dementia (C, L)	—	—

(S) = strong recommendation; (C) = conditional recommendation; (M) = moderate quality evidence; (L) = low quality evidence; (VL) = very low quality evidence; — = no recommendation.

The guideline panel made 21 pharmacologic recommendations for managing agitation in people living with dementia (Table 4). Citalopram was recommended for first-line treatment of moderate agitation, and aripiprazole, brexpiprazole, and risperidone were recommended as first-line treatments for severe agitation. Quetiapine and low doses of typical antipsychotics were recommended as potential options for severe agitation in dementia where there was poor tolerability or response to other first-line treatments. Nabilone and carbamazepine were recommended where no response was noted to any of the aforementioned medications. Recommendations were made against the use of medications that are ineffective or have high propensity for side-effects. The guideline panel recommended against the use of cognitive enhancers where agitation is the primary focus of treatment, most antidepressants other than citalopram, and against the use of olanzapine except in the emergency treatment of acute agitation in dementia. Switching treatments was recommended if pharmacologic interventions are ineffective after eight weeks. The guideline panel recommended against polypharmacy in treating agitation in dementia.

TABLE 4.

Pharmacological interventions: BPSD syndrome (strength, quality of recommendation)

	Agitation	Psychosis	Depression; Depressive Symptoms	Anxiety	Sexual Expressions of Risk
Recommendations For Use
Antidepressants	Citalopram for moderate agitation (S, L) Citalopram, severe agitation where risks and benefits preclude alterative medications (C, VL)	Citaloprama for moderate psychosis (C, L)	Antidepressants for moderate to severe depression that has not responded to psychosocial approaches (CCSMH depression guidelines) (C, L)	Citalopram for moderate to severe anxiety (CCSMH anxiety guidelines) (C, L)	Neither for nor against the use of any pharmacological intervention for reducing sexual expressions of potential risk
Antipsychotics	Aripiprazole,b brexpiprazolec or risperidone for severe agitation (C, M) Quetiapine for severe agitation if symptoms are refractory to other pharmacological treatments (C, L) Typical antipsychotics, if symptoms are refractory to other pharmacological treatments (C, VL)	Aripiprazole or risperidone for severe psychosis (C, L)	—	—
Cannabinoids	Synthetic cannabinoids for severe agitation if symptoms are refractory to other pharmacological treatments (C, L)	—	—	—
Anticonvulsants	Carbamazepine for severe agitation if symptoms are refractory to other pharmacological treatments (C, VL)	—	—	—
Short-Term Emergency Treatment	Short-acting antipsychotics available PO/IM for severe agitation (C, VL) Short-acting benzodiazepines available PO/IM for severe agitation (C, VL)	—	—	—
Recommendations Against Use
Cholinesterase Inhibitor/ Memantine	Recommend against starting cholinesterase inhibitors for moderate-to-severe agitation (C, L) Recommend against starting memantine for moderate-to-severe agitation (C, M)	—	—	—	—
Antidepressants	Recommend against trazodone, sertraline, mirtazapine, fluoxetine (C, L) Recommend against paroxetine, fluvoxamine, tricyclic antidepressants (C, VL)	—	Recommend against use of antidepressants for depressive symptoms in dementia without concurrent diagnosis of depression (C, L)	—	—
Antipsychotics	Recommend against olanzapine (S, L) Recommend against long-acting injectable antipsychotics (S, L)	—	—	—	—
Alpha-blockers	Recommend neither for nor against use of prazosin (C, L)	—	—	—	—
Anticonvulsants	Recommend against valproic acid or sodium divalproex (S, M)	—	—	—	—
Polypharmacy	Recommend against polypharmacy (S, L)	—	—	—	—

^aCould be considered for severe agitation in some circumstances.

^bIf non-response or not tolerated with either, switch to risperidone or another medication if risperidone previously trialed.

^cExcept as short-term emergency treatment of acute agitation.

(S) = strong recommendation; (C) = conditional recommendation; (M) = moderate quality evidence; (L) = low quality evidence; (VL) = very low quality evidence; — = no recommendation.

The guideline panel recommended against using restraints or seclusion for individuals hospitalized with agitation in dementia.^(24,25) Preliminary evidence from observational studies indicated that electroconvulsive therapy (ECT) may be effective in agitation refractory to pharmacologic treatment.^(26,27) However, based on the current lack of randomized controlled studies, the potential for adverse effects, and resource implications, the guideline panel recommended neither for nor against the use of ECT at this time.

Psychosis

Five recommendations were made for assessing and managing psychosis in people living with dementia. The IPA criteria for psychosis in major neurocognitive disorders were recommended for the diagnosis of psychosis in dementia⁽²³⁾ (Table 2). The Neuropsychiatric Inventory was recommended for the detection of symptoms in psychosis⁽²⁸⁾ (Table 2). Nonpharmacologic recommendations included education for caregivers and health-care providers, while recommended psychosocial interventions included pleasant activities such as art- and music-based therapies (Table 3). Pharmacologic recommendations for the treatment of moderate to severe symptoms in psychosis in Alzheimer’s disease and related dementia were similar to those made for agitation in dementia and included a recommendation for the use of citalopram in moderate symptoms of psychosis in dementia, and aripiprazole or risperidone for severe symptoms of psychosis in dementia (Table 4). These recommendations related to antipsychotic medications do not apply to other types of dementia, such as dementia with Lewy bodies or Parkinson’s disease dementia, which were not the focus of this guideline.

Depression and Depressive Symptoms

Recommendations for depressive symptoms in people living with dementia were considered separately from a diagnosis of major depressive disorder in people living with dementia. Four recommendations were made for assessing and managing depressive disorder in dementia, and nine recommendations were made for assessing and managing depressive symptoms in dementia. The National Institutes of Mental Health Depression in Alzheimer’s disease criteria was recommended for the diagnosis of depressive disorder in dementia.⁽²⁹⁾ The Cornell Scale for Depression in Dementia was recommended for the detection and assessment of depressive symptoms in dementia.⁽³⁰⁾

Recommended nonpharmacologic interventions for the management of depressive symptoms in people living with dementia were interdisciplinary approaches to care including educational interventions, behavioural therapies, physical exercise, and cognitive therapy adapted for people living with mild-to-moderate dementia (Table 3). Psychosocial recommendations included personalized pleasant activities, and sensory approaches such as animal-assisted therapies and massage therapy (Table 3).

The guideline panel recommended against the use of pharmacologic treatments for depressive symptoms where the person living with dementia does not meet the threshold for a diagnosis of depression in dementia (Table 4). Where a person living with dementia has been diagnosed with depression, the guideline panel recommended antidepressants for moderate to severe depression that did not respond to nonpharmacologic approaches. The specific antidepressants recommended for the treatment of depression in dementia, such as sertraline or duloxetine, are described in the CCSMH Canadian Guidelines on Prevention, Assessment and Treatment of Depression in Older Adults (https://ccsmh.ca/areas-of-focus/depression/).

Anxiety

The guideline panel made six recommendations for assessing and managing symptoms of anxiety in people living with dementia. With no existing standardized criteria for the diagnosis of anxiety in dementia, the guideline panel recommended the Diagnostic Statistical Manual-5 (DSM-5) criteria for the diagnosis of generalized anxiety disorder.⁽³¹⁾ The Rating Anxiety in Dementia (RAID) scale was recommended for the detection of anxiety symptoms in dementia.⁽³²⁾ Recommendations for nonpharmacologic interventions to manage anxiety in dementia included education and training programs for family caregivers, and adapted psychotherapies such as cognitive behavioural therapy incorporating family caregivers (Table 3). Music therapy with the preferred music of the person living with dementia was recommended as an effective psychosocial intervention (Table 3). Where no response to nonpharmacologic or psychosocial interventions has been observed, citalopram was recommended as a pharmacologic treatment for anxiety in dementia (Table 4). Further pharmacologic recommendations may be found in the CCSMH Canadian Guidelines for the Assessment and Treatment of Anxiety in Older Adults (https://ccsmh.ca/areas-of-focus/anxiety/).

Sexual Expressions of Potential Risk

Four recommendation statements were made on assessing and managing sexual expressions of potential risk in people living with dementia. As there is no standard definition of sexual expressions of potential risk in dementia, the guideline panel recommended a definition (Table 2). The guideline panel made a conditional recommendation that sexual behaviours of potential risk in dementia be defined as disruptive vocal or physical acts of an explicit or sexual nature that would be considered intrusive or are engaged in by the person living with dementia without the consent of those present (Table 2). The St. Andrew’s Sexual Behaviour Assessment Scale was recommended for the detection and assessment of sexual expressions of potential risk.⁽³³⁾ While limited evidence was noted for this topic, evidence was found in support of nonpharmacologic interventions such as education for the person living with dementia and their caregiver (Table 3). Recommendations for psychosocial interventions included removing environmental triggers and engaging the person living with dementia in other activities (Table 3). Due to the limited evidence base concerning the effectiveness of pharmacologic interventions, the guideline panel recommended neither for nor against the use of pharmacologic treatments for sexual expressions of potential risk (Table 4).

Deprescribing of Antipsychotics and Other Psychotropics

The guideline panel recommended deprescribing of antipsychotic medications for the person living with dementia where there is no history of severe agitation, psychosis, depression or another potentially appropriate indication for antipsychotic treatment (Table 5). Where the person living with dementia previously exhibited severe symptoms and such symptoms are reduced or no longer reported, the guideline panel recommended an individualized approach for deprescribing that considers both the current symptoms exhibited, as well as the risks and benefits of deprescribing (Table 5). The guideline panel further recommended that deprescribing be undertaken by reducing psychotropic dosing by 25–50% every one to two weeks until discontinued, and to stop dose reduction at the lowest effective dose if symptoms recur. Interdisciplinary education programs regarding deprescribing were recommended for primary care providers in LTC and residential care settings, as well as pharmacist-led medication reviews (Table 5). A recommendation was made by the guideline panel that routine reviews of psychotropic medications, including antidepressants and benzodiazepines, be conducted for potential discontinuation in people living with dementia (Table 5).

TABLE 5.

Deprescribing: recommendation (strength, quality of recommendation)

Antipsychotics	We recommend deprescribing antipsychotics in people living with dementia who do not have a history of severe agitation or psychosis or another potentially appropriate indication for antipsychotics such as a history of serious mental illness (S, L) We suggest deprescribing antipsychotics in people living with dementia who initially had severe agitation or psychosis, after considering their current symptoms, the total duration of antipsychotic treatment, dosage of medication required to stabilize BPSD, and initial severity of symptoms (C, L) We suggest deprescribing antipsychotics by decreasing the dose by 25–50% every 1–2 weeks until discontinued, and that dosage reduction be stopped at the lowest effective dose if BPSD worsen (C, L)
Other psychotropic medications	We suggest that other psychotropic medications be reviewed routinely for potential discontinuation in people with dementia including benzodiazepines and antidepressants (C, VL)
Education	We suggest interdisciplinary education interventions, interdisciplinary medication reviews, educational interventions for family physicians, and pharmacist-led medication reviews to facilitate antipsychotic deprescribing in people with dementia at the organizational level in long-term care and other residential care settings (C, L)

BPSD = behavioural and psychological symptoms of dementia; (S) = strong recommendation; (C) = conditional recommendation; (M) = moderate quality evidence; (L) = low quality evidence; (VL) = very low quality evidence.

CONCLUSIONS

The CCSMH CPG for assessing and managing BPSD will inform best practices for BPSD care within a Canadian health-care context. This CPG provides recommendations for assessing and managing agitation, psychosis, depression and depressive symptoms, anxiety, and sexual expressions of potential risk in people living with dementia. Specific recommendations were made based on BPSD severity. Suggestions for tailored management plans depended upon the context and resources available. Recommendations regarding the deprescribing of antipsychotics and other medications frequently used in the management of BPSD were also provided.

This CPG was rigorously developed by a Canada-wide panel of professional clinical experts; people with lived experience of dementia provided external feedback on CPG recommendations. We present the only current CPG for assessing and managing BPSD in dementia within a Canadian context and have addressed a significant gap in Canadian CPGs, research, and recommendations. Some limitations to the CPG development process include a lack of high-quality studies to inform recommendations, most notably for nonpharmacologic and psychosocial interventions for BPSD other than agitation and depression. In addition, relatively few high-quality studies have examined pharmacological interventions for BPSD aside from agitation. We note a pressing need for research specific to these areas. Resource constraints did not permit addressing other critical BPSD in the CPG such as sleep changes and apathy. We recommend future research in the development of clinical practice recommendations for BPSD in people with Lewy body dementia, Parkinson’s disease dementia, frontotemporal dementia, and other dementias.

Finally, we note that many nonpharmacologic interventions were effective across multiple BPSD. For example, education and training programs for caregivers and health-care workers showed effectiveness in the management of all BPSD. Further, a common approach across several diagnoses for the management of multiple symptoms may show promise in terms of implementing nonpharmacologic and psychosocial approaches to BPSD more widely. We look to the effective utilization of psychotherapies such as cognitive behavioural therapy across multiple mental health disorders including anxiety,⁽³⁴⁾ depression,⁽³⁵⁾ for older adults with both anxiety and depression in residential care⁽³⁶⁾ and psychosis^(37,38) as a potential model. A recommendation for future research would be an examination of potential transdiagnostic psychosocial approaches for multiple BPSD symptoms.

Implementation and evaluation of this CPG will be an ongoing process. The CCSMH has developed a knowledge mobilization resource to collate evidence-based resources related to BPSD at the Behaviours in Dementia Toolkit website (www.behavioursindementia.ca) to help people living with dementia, caregivers, and health-care providers identify resources that may assist in the practical implementation of CPG recommendations. The CCSMH hopes that this CPG will stimulate improvements in clinical care. The guideline panel hopes that health-care providers, organizations, and researchers will implement and evaluate the impact of these recommendations in their current and future work.

APPENDIX A. Guideline Panel Members and Conflict of Interest Disclosures

We have read and understood the Canadian Geriatrics Journal’s policy on disclosing conflicts of interest and declare the following interests:

Co-Author	Affiliation	Conflict Of Interest Disclosure
Stacey Hatch, PhD, RP (first author)	University of Calgary, Calgary, ON, Canada	No conflict of interest disclosed
Dallas P. Seitz MD PhD	University of Calgary, Calgary, ON, Canada	No conflict of interest disclosed
Marie-Andree Bruneau, MD, MSc	University of Montreal, Montreal, QC, Canada	Dr. Bruneau has received funding from Lifespeaks Inc. (conference honorarium), Janssen (conference honorarium) and Otsuka/Lundbeck (advisory board).
Vivian Ewa, MD	University of Calgary, Calgary, ON, Canada	No conflict of interest disclosed
Sid Feldman, MD	Baycrest Health Sciences, Toronto, ON, Canada, Temerty Faculty of Medicine, University of Toronto, Toronto, ON, Canada	No conflict of interest disclosed
Yael Goldberg PhD. C.Psych	Baycrest Health Sciences, Toronto, ON, Canada	No conflict of interest disclosed
Zahra Goodarzi MD MSc	University of Calgary, Calgary, ON, Canada	No conflict of interest disclosed
Nathan Herrmann MD	Temerty Faculty of Medicine, University of Toronto, Toronto, ON, Canada, Sunnybrook Health Sciences Centre, Toronto, ON, Canada	No conflict of interest disclosed
Debbie Hewitt Colborne RN, MScN	Behavioural Supports Ontario Provincial Coordinating Office, North Bay, ON, Canada	No conflict of interest disclosed
Alexandre Henri-Bhargava, MDCM, MScCH	Neil and Susan Manning Cognitive Health Initiative, University of British Columbia, Victoria, BC, Canada	Dr. Henri-Bhargava has received consulting fees from Roche, and sits on advisory boards for Roche, Integra and Biogen, and receives funding from Intelgenx, NovoNordisk, Cerevel and Anavex for clinical trials in which he is the principle investigator.
Zahinoor Ismail, MD	University of Calgary, Calgary, ON, Canada	Dr. Ismail has received consulting fees from Acadia, Otsuka/Lundbeck and Roche, and receives funding from Avanir, Biogen, and Roche for studies in which he is the site principle investigator and site investigator.
Julia Kirkham MD., MSc	University of Calgary, Calgary, ON, Canada	No conflict of interest disclosed
Sanjeev Kumar, MD	Temerty Faculty of Medicine, University of Toronto, Toronto, ON, Canada, Centre for Addiction and Mental Health, Toronto, ON, Canada	Dr. Kumar’s institution receives funding from Soterix Medical for a study in which he is the principle investigator.
Krista L. Lanctôt, PhD	Temerty Faculty of Medicine, University of Toronto, Toronto, ON, Canada, Sunnybrook Health Sciences Centre, Toronto, ON, Canada	Dr. Lanctôt has received fees as an advisory board member for Bright Minds, Cerevel Therapeutics, Eisai Co. Ltd., Exciva, ICG Pharma, Kondor Pharma, H. Lundbeck A/S, Merck Sharp Dohme, Novo Nordisk, Otsuka, Praxis Therapeutics, and receives funding for studies from Cerevel Therapeutics in which she is an investigator. She is also supported by the Bernick Chair in Geriatric Psychopharmacology.
Wade Thompson PharmD PhD	University of British Columbia, Vancouver, BC, Canada	No conflict of interest disclosed
Jennifer Porter, MPH	University of Calgary, Calgary, ON, Canada	No conflict of interest disclosed
Jennifer A. Watt MD PhD	University of Toronto, Toronto, ON, Canada, Unity Health Toronto, Toronto, ON, Canada	No conflict of interest disclosed

APPENDIX B. Strength of recommendations and quality of evidence

Strength of Recommendations	GRADE: The Extent to Which Confidence in the Evidence Supports a Recommendation(26)
Strong	Most patients would want the proposed care approach Most clinicians would advise the proposed care approach Policymakers should implement the recommendations (with adaptions as appropriate)
Conditional	The majority (but not all) patients would want the proposed care approach The majority (but not all) clinicians would advise the proposed care approach Policymaking will require considerable debate to implement the recommendation
No recommendation	Insufficient evidence to judge risks and benefits associated with a recommendation statement
Strength of Evidence	Definition
High	Any future research would be unlikely to establish further confidence in treatment effect
Moderate	Future research could impact confidence in the treatment effect
Low	Future research will very likely impact the confidence of a treatment effect
Very low	Evidence for a treatment effect is uncertain(24)

APPENDIX C. Dissenting Opinion Related to Recommendation 1: We recommend the International Psychogeriatrics Association (IPA) consensus criteria for agitation in cognitive disorders to diagnose agitation in dementia. (Strong recommendation, moderate quality)

I respectfully disagree with the majority decision provided by the guideline panel for this recommendation and generally disagree with diagnosing behaviours as ‘agitation’. Although I appreciate the benefits to standardizing language, there are significant risks in adopting such language for behaviours that don’t fit the IPA definition and for those that do. Many behaviours expressed by individuals living with dementia are distressing for them and those around them, but do not meet the IPA criteria (e.g. behaviour that has occurred for less than two weeks); and thus, fall outside of the recommendations of this guideline. This gap leaves care providers without guidance regarding how best to respond.

More concerning are behaviours that meet the IPA criteria leading to people living with dementia being diagnosed/labelled with ‘agitation’. This language reflects a biomedical lens that pathologizes the behavioural responses expressed by people living with dementia. Instead of viewing behaviours as expressions of unmet needs or normal human reactions to situations (e.g. a stranger attempting to undress you and washing your genitals without you asking them to do so), the person gains a new diagnosis that needs to be ‘treated’. Alternatively, an unmet need lens puts the responsibility on the care team to consider context and possible contributing factors to the specific behaviours, and to find individualized approaches to meet the need.

Debbie Hewitt Colborne RN, MScN . Registered Nurse. Project Advisor, Behavioral Supports Ontario Provincial Coordinating Office, North Bay Regional Health Centre