Residency · Residency · Geriatrics

Behavioral and Psychological Symptoms of Dementia

Introduction

Behavioral and psychological symptoms of dementia (BPSD) affect up to 90 percent of individuals with dementia at some point during the disease course and represent one of the most challenging aspects of dementia care for clinicians, families, and care systems alike. The term encompasses a broad spectrum of non-cognitive manifestations including agitation, aggression, psychosis (delusions and hallucinations), depression, anxiety, apathy, disinhibition, aberrant motor behavior, sleep disturbances, and appetite changes. These symptoms are the primary driver of caregiver burden, the leading precipitant of institutionalization, and the most common reason for the use of physical and chemical restraints in care facilities. The management of BPSD demands a systematic approach in which non-pharmacological strategies are always first-line and pharmacotherapy is reserved for situations in which behavioral interventions have proven insufficient and the symptoms pose significant danger or distress.

Classification and Epidemiology

Behavioral Symptoms

Agitation and aggression, which may be verbal (shouting, cursing, repetitive vocalization) or physical (hitting, pushing, resisting care), affect 30 to 50 percent of individuals with dementia and are among the most distressing symptoms for caregivers and care staff. Aberrant motor behavior, including wandering, pacing, and repetitive purposeless movements, occurs in 15 to 65 percent of patients and poses significant safety concerns, particularly the risk of elopement. Disinhibition manifests as socially inappropriate behavior, impulsivity, and loss of adherence to social norms, affecting 10 to 30 percent of patients. Appetite and eating disturbances, encompassing hyperphagia, hyporexia, changes in food preferences, and pica (eating non-food objects), occur in 20 to 30 percent of patients. Sleep disturbances, including insomnia, sundowning (increased confusion and behavioral symptoms in the late afternoon and evening), and day-night reversal, affect 25 to 50 percent of patients and contribute substantially to caregiver exhaustion.

Psychological Symptoms

Apathy is the most common neuropsychiatric symptom in dementia, affecting 50 to 70 percent of patients and manifesting as reduced motivation, diminished initiative, and blunted emotional responsiveness. Although frequently mistaken for depression, apathy is a distinct syndrome that may coexist with depression but requires different management approaches. Depression itself has a prevalence of 20 to 50 percent in dementia populations and overlaps significantly with apathy in its clinical presentation. Anxiety, including separation anxiety, fear of being alone, and generalized worry, affects 20 to 50 percent of patients.

Psychotic symptoms are common and clinically significant. Delusions occur in 25 to 50 percent of patients with dementia and are typically paranoid in nature, with common themes including theft (believing possessions are being stolen), infidelity, and the phantom boarder delusion (believing uninvited strangers are living in the home). Hallucinations, predominantly visual, occur in 15 to 30 percent of patients. Misidentification syndromes represent a distinct category of psychotic phenomena: Capgras syndrome involves the belief that a familiar person has been replaced by an impostor, the TV sign describes the inability to distinguish televised events from reality, and the mirror sign involves failure to recognize one's own reflection.

Assessment Tools

The Neuropsychiatric Inventory (NPI) is the gold standard assessment instrument for BPSD, evaluating twelve behavioral domains by frequency and severity scores, yielding a total score from 0 to 144, with an additional measure of caregiver distress associated with each symptom. The Cohen-Mansfield Agitation Inventory (CMAI) rates the frequency of 29 specific agitation behaviors and is particularly useful in long-term care settings. The Cornell Scale for Depression in Dementia (CSDD) is an observer-rated instrument designed specifically for assessing depression in individuals with cognitive impairment, with a score of 8 or greater suggesting depression. The Pittsburgh Agitation Scale provides a brief clinician-rated assessment of acute agitation episodes.

Etiology and the DICE Framework

Understanding the Behavior

A foundational principle of BPSD management is that behavioral symptoms are communications of unmet needs. The behavior itself is not the problem; it is a signal pointing toward an underlying cause that must be identified and addressed. The clinician's first response to any behavioral symptom should be the question: "What is driving this behavior?"

DICE Approach (Kales et al., 2014)

The DICE framework provides a structured, evidence-based methodology for evaluating and managing BPSD.

The first step is to Describe the behavior with precision: what exactly is the behavior, when does it occur, how frequently, how severe is it, what are the identifiable triggers, and what are the consequences for the patient and others? Vague characterizations such as "the patient is agitated" are insufficient; the behavior must be operationalized in concrete, observable terms.

The second step is to Investigate underlying causes through a systematic evaluation of multiple domains. Medical causes must be considered first, as pain accounts for up to 50 percent of agitation in dementia and is frequently underrecognized due to communication barriers. Infection, constipation, urinary retention, medication side effects, and delirium superimposed on dementia are other common medical drivers. Psychiatric causes include depression, anxiety, and psychosis. Environmental factors encompass overstimulation, understimulation, disruptions in routine, and unfamiliar caregivers. Caregiver factors include communication approach (speaking too quickly, providing complex instructions), rushing the patient, and unfamiliarity with the individual's preferences and history. Unmet basic needs, including hunger, thirst, boredom, loneliness, need to toilet, and thermal discomfort, must always be considered.

The third step is to Create a plan that targets the identified causes with specific interventions tailored to the individual patient and their circumstances.

The fourth step is to Evaluate the effectiveness of the plan within one to two weeks, modifying the approach based on response.

<image>A clinical decision-making flowchart using the DICE framework for BPSD management. Start with "BPSD Identified" and move through four sequential boxes: (1) DESCRIBE — show a checklist including behavior type, frequency, timing, triggers, and safety risk assessment; (2) INVESTIGATE — show branching causes with icons: medical (stethoscope — pain, infection, constipation, delirium, medications), psychiatric (brain — depression, anxiety, psychosis), environmental (house — noise, lighting, routine changes), caregiver (people — communication style, approach), unmet needs (list — hunger, thirst, toileting, boredom); (3) CREATE — show parallel tracks for non-pharmacological interventions (first-line, always) and pharmacological interventions (second-line, if severe); (4) EVALUATE — reassess at 1-2 weeks with modification loop. Include a prominent side banner stating "Treat the cause, not just the behavior."</image>

Non-Pharmacological Management (First-Line, Always)

Person-Centered Care

Person-centered care is the philosophical foundation of BPSD management and requires knowing the individual as a person rather than merely as a patient with dementia. This means understanding their life history, preferences, daily routines, and deeply held values through structured life story work. Consistent caregivers and maintenance of predictable daily routines provide the environmental stability that reduces confusion and distress. Communication should be simplified to short sentences with one instruction at a time, allowing adequate processing time before expecting a response. Argumentation, correction, and reality orientation should be avoided in favor of validation of the patient's emotional experience and gentle redirection. Throughout all interactions, dignity must be maintained by offering choices wherever possible and preserving autonomy in whatever abilities remain.

Specific Interventions by Evidence Level

Strong Evidence

Music therapy, particularly personalized music programs such as the MUSIC and MEMORY program, has the strongest evidence base among non-pharmacological interventions for BPSD. Research demonstrates reductions in agitation of 50 to 67 percent during music sessions, with the most robust effects achieved when using music from the patient's young adulthood, typically between ages 18 and 25, which corresponds to the period of peak autobiographical memory encoding.

Exercise and physical activity programs, including walking programs and chair-based exercise, reduce agitation and improve sleep quality. Caregiver training programs represent another strongly evidence-supported intervention. The REACH II (Resources for Enhancing Alzheimer's Caregiver Health) multicomponent caregiver intervention has demonstrated reductions in caregiver burden, depression, and BPSD simultaneously, recognizing the bidirectional relationship between caregiver well-being and patient behavioral symptoms. Behavioral management techniques using ABC analysis (Antecedent-Behavior-Consequence) and positive reinforcement provide systematic frameworks for understanding and modifying specific behaviors.

Moderate Evidence

Aromatherapy, particularly lavender oil, has shown promise for reducing agitation in small studies, though the overall evidence quality remains low. Light therapy using bright light of 2,500 to 10,000 lux delivered in the morning can stabilize circadian rhythms and reduce sundowning. Montessori-based activities, which are structured, person-centered activities carefully matched to the individual's current ability level, engage preserved cognitive and motor capacities. Sensory stimulation approaches include Snoezelen (multisensory environments), tactile stimulation, and doll therapy. Pet therapy and animal-assisted interventions reduce agitation and social withdrawal in structured programs.

For Specific Symptoms

Wandering should be managed through environmental modifications that prioritize safety without imposing physical restraint: safe walking areas, door alarms, GPS tracking devices, and environmental design features such as disguised exits and stop signs placed on doors. Physical restraints must be avoided. Sundowning benefits from increased afternoon light exposure, structured evening routines, minimization of evening stimulation, and systematic exclusion of pain and constipation as contributing factors. Repetitive vocalizations should prompt assessment for unmet needs, particularly pain, hunger, and boredom, with provision of music or comfort objects. Resistance to care during activities such as bathing responds to a slow, gentle approach, a warm environment, music during the activity, distraction techniques, and assignment of consistent caregivers who know the patient.

Pharmacological Management

General Principles

Pharmacotherapy for BPSD should be reserved for behaviors that are severe, dangerous to the patient or others, or causing significant distress despite adequate non-pharmacological interventions. Treatment should target specific symptom clusters (psychosis, depression, agitation) rather than aiming for generic sedation. The guiding principle is to start at low doses, titrate slowly, use the lowest effective dose for the shortest duration necessary, and reassess the continued need for treatment every three to six months with an attempted taper. It is critical to recognize that no medication is specifically FDA-approved for BPSD, with the exception of brexpiprazole for agitation associated with Alzheimer disease dementia.

Antipsychotics

All antipsychotics carry an FDA black box warning for increased mortality when used in elderly patients with dementia, with a relative risk of 1.6 to 1.7 and a number needed to harm of 53 to 100 over 10 to 12 weeks. The causes of excess mortality include cardiovascular events, infections (particularly pneumonia), and cerebrovascular events. The CATIE-AD trial conducted by Schneider and colleagues in 2006 demonstrated that olanzapine, quetiapine, and risperidone produced no significant benefit over placebo on the Clinical Global Impression of Change, while causing significant adverse effects including sedation, extrapyramidal symptoms, metabolic disturbances, and weight gain.

Despite these concerning findings, antipsychotics remain the most commonly used pharmacological agents for severe BPSD when non-pharmacological measures fail. Risperidone has the best evidence of any antipsychotic for BPSD, with a number needed to treat of 5 to 8 for aggression at doses of 0.5 to 1 mg per day. It is approved in the United Kingdom for short-term treatment (up to six weeks) of persistent aggression in moderate-to-severe Alzheimer disease. Olanzapine at 2.5 to 5 mg per day is sedating and carries metabolic and weight gain risks. Quetiapine at 12.5 to 50 mg twice daily has the weakest evidence for BPSD among the atypical antipsychotics but is the safest option in dementia with Lewy bodies and Parkinson disease dementia. Aripiprazole at 2 to 5 mg per day, as a partial dopamine D2 agonist, produces less sedation and fewer metabolic effects.

Brexpiprazole (Rexulti) received FDA approval in 2023 for the treatment of agitation associated with Alzheimer disease dementia at doses of 2 to 3 mg daily, making it the first and only medication with this specific indication. Phase 3 trials demonstrated a modest benefit with CMAI score reductions of approximately 5 points versus placebo and a number needed to treat of approximately 8. Side effects include headache, dizziness, somnolence, and urinary tract infection.

Haloperidol should be avoided in dementia with Lewy bodies and has limited use in the general dementia population due to significant extrapyramidal risk. Pimavanserin, a 5-HT2A inverse agonist with no D2 receptor activity, is useful for psychosis in dementia with Lewy bodies and Parkinson disease dementia.

Antidepressants

The CitAD trial conducted by Porsteinsson and colleagues in 2014 demonstrated that citalopram at 30 mg per day significantly reduced agitation in Alzheimer disease with a number needed to treat of 7. However, citalopram at 30 mg produces clinically significant QTc prolongation (mean increase of 18 milliseconds), and the FDA-recommended maximum dose in elderly patients is 20 mg. This creates a clinical dilemma in which the effective dose for agitation exceeds the recommended maximum. At 20 mg, some benefit may still be present; a baseline electrocardiogram should be obtained and QTc monitored.

Trazodone at 25 to 100 mg is useful for agitation, sundowning, and insomnia with minimal anticholinergic effects. SSRIs including sertraline and escitalopram are first-line agents for depression in dementia and are generally well tolerated. Mirtazapine at 7.5 to 15 mg at bedtime is particularly useful when insomnia and poor appetite coexist with depression.

Anticonvulsants/Mood Stabilizers

Carbamazepine has some evidence supporting its use for agitation at doses of 100 to 300 mg twice daily, but significant drug interactions and the risk of bone marrow suppression limit its utility. Valproic acid does not have evidence supporting its use for BPSD; the VALID trial was negative, and valproic acid has been associated with accelerated brain atrophy in dementia patients. It is not recommended. Gabapentin and pregabalin have limited evidence but may help with anxiety-driven behaviors.

Other Agents

Benzodiazepines should generally be avoided in dementia patients due to increased falls, sedation, paradoxical agitation, and cognitive worsening, with the exception of acute severe distress requiring emergency management. Cholinesterase inhibitors may modestly reduce apathy and psychosis as part of overall Alzheimer disease treatment but are not indicated specifically for acute BPSD. Memantine may reduce agitation in moderate-to-severe Alzheimer disease as reflected in NPI improvements and should be considered as part of the baseline dementia treatment regimen. Prazosin, an alpha-1 blocker, has emerging evidence for BPSD-related agitation and nightmares at doses of 1 to 2 mg at bedtime. Dextromethorphan/quinidine (Nuedexta) showed positive results in a Phase 2 trial for agitation in Alzheimer disease and is used off-label. Cannabinoids, including nabilone, have shown agitation reduction in a small randomized controlled trial, and dronabinol may improve appetite, but the evidence is limited and routine use is not recommended.

<image>A medication management table for BPSD organized by target symptom. Create a grid with columns for: Symptom Cluster, First-Line Medication, Dose Range, Key Evidence, and Critical Warnings. Rows should include: (1) Psychosis/Hallucinations — risperidone 0.5-1mg, pimavanserin 34mg for DLB; (2) Agitation/Aggression — brexpiprazole 2-3mg (FDA-approved), citalopram ≤20mg (QTc), trazodone 25-100mg; (3) Depression — sertraline 25-100mg, escitalopram 5-10mg; (4) Apathy — cholinesterase inhibitors, methylphenidate (limited evidence); (5) Sleep Disturbance — trazodone 25-50mg, melatonin 3-10mg; (6) Anxiety — SSRI, buspirone 5-15mg BID. Include a red warning banner at the top with the FDA black box warning about antipsychotic mortality in elderly dementia patients. Add a green banner below stating "Non-pharmacological interventions are ALWAYS first-line."</image>

Special Considerations

Emergency/Acute Severe Agitation

When a patient is at imminent risk of harming themselves or others, verbal de-escalation should always be attempted first. If pharmacotherapy becomes necessary, options include olanzapine 2.5 to 5 mg given intramuscularly or orally (with the caveat that concurrent benzodiazepine use must be avoided due to the risk of respiratory depression), haloperidol 0.5 to 1 mg intramuscularly (which must not be used in dementia with Lewy bodies and should not be given intravenously due to QTc risk), and lorazepam 0.5 mg intramuscularly or orally as a last resort for short-term use only. Physical restraints represent the absolute last resort, as they increase rates of delirium, injuries, and mortality; when used, they require a physician order and frequent reassessment.

Antipsychotic Deprescribing

Deprescribing of antipsychotics should be attempted every three to six months unless the original behavior was severe or life-threatening. The DART-AD trial conducted by Ballard and colleagues in 2009 demonstrated that the majority of nursing home residents did not experience behavioral worsening upon antipsychotic discontinuation, while long-term antipsychotic use was associated with a two-fold increase in mortality. Tapering should be gradual, with dose reductions of approximately 25 percent every two weeks over a period of one to two months. Close monitoring during and after the taper is essential, as relapse occurs in approximately 30 percent of cases. Centers for Medicare and Medicaid Services nursing home regulations require documented antipsychotic dose reduction attempts.

Advanced Dementia

In advanced dementia, the character of behavioral symptoms may shift from overt agitation and psychosis to more subtle forms of distress including resistive behaviors during care, vocalizations, and nonverbal distress cues. Pain is the most common treatable cause of distress in advanced dementia and should be assessed using observational tools such as the PAINAD (Pain Assessment in Advanced Dementia) scale. Management in this stage should adopt a comfort-focused approach that avoids restraints, minimizes invasive procedures, and prioritizes quality of life above all other considerations.

Key Clinical Pearls

• BPSD are communications of unmet needs — always investigate underlying causes (pain, infection, medication, environment) before reaching for pharmacotherapy
• Non-pharmacological interventions (person-centered care, music therapy, caregiver training) are first-line and should never be skipped
• All antipsychotics carry a black box warning for increased mortality in elderly dementia — NNH 53-100; use only when non-pharmacological measures fail and behaviors are dangerous
• Brexpiprazole (Rexulti) is the only FDA-approved medication for agitation in Alzheimer dementia (2023) — benefit is modest (NNT ~8)
• Citalopram for agitation is effective at 30 mg but exceeds the FDA-recommended maximum of 20 mg in elderly due to QTc prolongation — always check ECG
• Attempt antipsychotic deprescribing every 3-6 months — DART-AD showed long-term use doubles mortality
• Pain accounts for up to 50% of agitation in dementia — always assess and treat pain before attributing behaviors to "dementia"

References

1. Kales HC, Gitlin LN, Lyketsos CG. Assessment and management of behavioral and psychological symptoms of dementia. BMJ. 2015;350:h369.
2. Schneider LS, Tariot PN, Dagerman KS, et al. Effectiveness of atypical antipsychotic drugs in patients with Alzheimer's disease. N Engl J Med. 2006;355(15):1525-1538.
3. Porsteinsson AP, Drye LT, Pollock BG, et al. Effect of citalopram on agitation in Alzheimer disease: the CitAD randomized clinical trial. JAMA. 2014;311(7):682-691.
4. Ballard C, Hanney ML, Theodoulou M, et al. The dementia antipsychotic withdrawal trial (DART-AD): long-term follow-up of a randomised placebo-controlled trial. Lancet Neurol. 2009;8(2):151-157.
5. Livingston G, Kelly L, Lewis-Holmes E, et al. Non-pharmacological interventions for agitation in dementia: systematic review of randomised controlled trials. Br J Psychiatry. 2014;205(6):436-442.

Read this lecture as Markdown