Neuropsychiatric Inventory–Questionnaire

Abstract

The Neuropsychiatric Inventory–Questionnaire (NPI-Q) is a validated, brief, informant-rated clinical assessment tool designed to assess behavioral and psychological symptoms of dementia (BPSD) and the associated emotional distress experienced by primary caregivers. Developed by Daniel I. Kaufer, Jeffrey L. Cummings, and colleagues in 2000, the NPI-Q represents a streamlined adaptation of the standard, semi-structured clinical interview known as the Neuropsychiatric Inventory (NPI; Cummings et al., 1994). The instrument evaluates 12 discrete neuropsychiatric domains: Delusions, Hallucinations, Agitation/Aggression, Depression/Dysphoria, Anxiety, Elation/Euphoria, Apathy/Indifference, Disinhibition, Irritability/Lability, Motor Disturbance, Night-time Behaviors, and Appetite/Eating changes. For each domain, informants indicate symptom presence over the preceding month via a binary (Yes/No) screening gate. When present, informants rate symptom severity on a 3-point scale (Mild, Moderate, Severe) and caregiver distress on a 6-point scale ranging from 0 (Not distressing at all) to 5 (Extreme or Very Severe). Psychometric investigations across diverse clinical memory cohorts demonstrate that the NPI-Q possesses robust test-retest reliability (Pearson correlations ranging from r = 0.80 to 0.94 for severity and distress), high convergent validity when compared against the comprehensive NPI interview (r = 0.89 to 0.92 for total severity; r = 0.84 to 0.90 for total caregiver distress), and robust discriminative ability across neurodegenerative etiologies including Alzheimer’s disease, frontotemporal lobar degeneration, and dementia with Lewy bodies. Factor analytic models indicate multi-syndromic clusters encompassing hyperactivity/disinhibition, affective/mood alterations, psychotic features, and vegetative-homeostatic disturbances. The instrument’s brevity (requiring approximately 5 minutes for completion) and strong psychometric profile make it an indispensable standard in global clinical dementia research and routine ambulatory neurological practice.

Keywords

Neuropsychiatric Inventory-Questionnaire, NPI-Q, Behavioral and Psychological Symptoms of Dementia, BPSD, Alzheimer’s disease, caregiver distress, informant-rated assessment, neuropsychiatric symptoms, neurodegenerative disorders, psychometric validation

Authors

The Neuropsychiatric Inventory–Questionnaire was derived and validated by a multidisciplinary team of behavioral neurologists, neuropsychiatrists, and biostatisticians led by:

• Daniel I. Kaufer, MD (1961–2020): Late Professor of Neurology and Psychiatry, Founding Director of the Memory and Cognitive Disorders Program, and Chief of the Division of Cognitive and Behavioral Neurology at the University of North Carolina at Chapel Hill School of Medicine. Dr. Kaufer was a preeminent authority on frontotemporal dementia, dementia with Lewy bodies, and the measurement of neuropsychiatric manifestations in neurocognitive disorders.
• Jeffrey L. Cummings, MD, ScD: Research Professor in the Department of Brain Health at the School of Integrated Health Sciences, University of Nevada, Las Vegas (UNLV); Founding Director of the Cleveland Clinic Lou Ruvo Center for Brain Health; former Director of the Mary S. Easton Center for Alzheimer’s Disease Research at the University of California, Los Angeles (UCLA) David Geffen School of Medicine. Dr. Cummings is the primary architect of the original Neuropsychiatric Inventory (1994) and an internationally recognized pioneer in neurotherapeutics and clinical trial design in neurodegenerative disease.
• Patrick A. Ketchel, BS: Research Associate and clinical data coordinator within the Department of Neurology at the David Geffen School of Medicine at UCLA, contributing extensively to the standardization and validation protocols of the brief NPI forms.
• Collaborating Investigators: Additional contributors to the original multi-site validation included clinical researchers from the UCLA Alzheimer’s Disease Center and affiliated memory disorder clinics across North America.

Primary Contact / Inquiries: Correspondence historically directed to Jeffrey L. Cummings, MD, ScD via University of California, Los Angeles / Cleveland Clinic Lou Ruvo Center for Brain Health, or through the official instrument clearinghouse at http://npitest.net/ (archival contact: [email protected]).

Purpose

The Neuropsychiatric Inventory–Questionnaire (NPI-Q) was engineered to address a pervasive diagnostic and clinical management challenge: the systematic, rapid, and psychometrically sound assessment of behavioral and psychological symptoms of dementia (BPSD) within routine clinical practice, large-scale epidemiology, and clinical trial screening. While cognitive deficits constitute the core diagnostic criteria for neurodegenerative conditions, non-cognitive neuropsychiatric manifestations—such as agitation, visual hallucinations, persecutory delusions, profound apathy, and affective lability—occur in upwards of 90% of individuals over the course of dementia progression. These symptoms drive institutionalization, accelerate cognitive and functional decline, elevate health economic costs, and precipitate significant morbidity and psychological distress among family caregivers.

Prior to the introduction of the NPI-Q, the primary reference standard for capturing BPSD was the original Neuropsychiatric Inventory (Cummings et al., 1994). The parent NPI is a comprehensive clinician-administered, semi-structured interview requiring 20 to 30 minutes of specialized professional administration. In tertiary academic centers and dedicated pharmaceutical clinical trials, the full NPI remains exceptional; however, its substantial administration time renders it impractical for standard outpatient neurological consultations, primary care evaluations, memory screening clinics, and high-throughput registry enrollments. The fundamental purpose of the NPI-Q was to translate the 12 core domains of the validated NPI into a brief, self-administered informant questionnaire that can be completed independently by a caregiver in an ambulatory waiting room or clinical triage environment in approximately five minutes.

In addition to serving as a diagnostic screening instrument, the NPI-Q fulfills three vital clinical and research mandates:

• Dual-Construct Profiling: The instrument concurrently decouples and measures two interrelated dimensions: (a) the objective severity of the patient’s neuropsychiatric phenomenology, and (b) the subjective psychological distress imposed upon the caregiver by that specific symptom. This separation allows clinicians to identify situations where low-severity behaviors induce disproportionate caregiver burnout, guiding targeted psychoeducational and supportive interventions.
• Longitudinal Treatment Monitoring: The NPI-Q provides a quantifiable metric to evaluate therapeutic responses to non-pharmacological behavioral modifications and psychopharmacological interventions (such as acetylcholinesterase inhibitors, memantine, selective serotonin reuptake inhibitors, or atypical antipsychotics), tracking symptom regression or emergence over time.
• Health Services and Translational Research: The tool serves as an essential covariate in observational cohorts and pragmatic clinical trials, facilitating phenotypic subgrouping, tracking disease trajectories, and clarifying the neurobiological underpinnings of distinct behavioral variants of neurodegenerative diseases.

Psychological Construct

The NPI-Q operationalizes non-cognitive psychopathology in dementia across 12 discrete, clinically verified neuropsychiatric symptom domains, accompanied by an independent caregiver distress construct. Each domain reflects specific functional disruptions within cortical and subcortical networks:

• Delusions: Evaluates fixed, false idiosyncratic beliefs unsupported by external reality and resistant to logical counter-argumentation. Typical manifestations in neurodegenerative cohorts include persecutory ideation (beliefs that caregivers, neighbors, or staff are stealing possessions), delusions of infidelity, abandonment beliefs, or phantom boarder syndromes (the belief that uninvited persons are residing within the patient’s home).
• Hallucinations: Evaluates sensory perceptions occurring in the absence of external stimuli, most commonly visual (seeing deceased relatives, animals, or shadowy figures, highly characteristic of Lewy body pathology) or auditory (hearing nonexistent voices, music, or mechanical sounds).
• Agitation/Aggression: Captures behavioral resistance, overt defiance, physical combativeness, or verbal hostility, frequently manifested during hands-on caregiving routines such as bathing, dressing, or medication administration. It reflects dysregulated distress, threat hypersensitivity, and impaired frontolimbic inhibition.
• Depression/Dysphoria: Focuses on persistent sadness, weeping, expressions of worthlessness, self-deprecating remarks, passive or active suicidal ideation, and pervasive anhedonia beyond transient situational discouragement.
• Anxiety: Assesses excessive worry, autonomic signs of nervousness, panic phenomena, somatic preoccupations, and severe separation anxiety (e.g., severe clinging or “shadowing” behavior, wherein the patient follows the primary caregiver incessantly due to terror of abandonment).
• Elation/Euphoria: Evaluates inappropriately elevated mood, exaggerated cheerfulness, grandiose assertions, or juvenile hilarity incongruent with situational context, a presentation frequently identified in the behavioral variant of frontotemporal dementia (bvFTD).
• Apathy/Indifference: Quantifies profound loss of motivation, emotional blunting, spontaneous avolition, social withdrawal, and lack of goal-directed engagement, distinguishing primary motivational depletion from depressive sadness.
• Disinhibition: Captures failure of impulse control, loss of social tact, inappropriate approach to strangers, tactless verbal utterances, socially inappropriate sexualized commentary or actions, and disregard for conventional behavioral norms.
• Irritability/Lability: Measures rapid shifts in affective state, severe impatience, petulance, low frustration tolerance, and episodic emotional outbursts triggered by minor environmental or schedule disruptions.
• Motor Disturbance (Aberrant Motor Behavior): Evaluates purposeless, repetitive, stereotypical motor patterns, including persistent pacing, aimless wandering, rummaging through drawers, repetitive buttoning/unbuttoning, or compulsive manipulation of objects.
• Night-time Behaviors: Assesses disruptions of the normal sleep-wake architecture, including nocturnal awakenings, night-time wandering, excessive early-morning rising, confusion upon waking, and frequent daytime hypersomnolence.
• Appetite/Eating: Evaluates alterations in nutritional intake, profound changes in food preferences (such as selective craving for refined sweets and carbohydrates, frequently observed in frontotemporal degenerations), ravenous hyperphagia, or marked hypophagia culminating in clinical weight loss.

In addition to patient-centered symptom severity, the NPI-Q assesses Caregiver Distress. Rooted in psychological theories of stress appraisal, this construct reflects the emotional burden, subjective strain, anxiety, and perceived loss of coping efficacy directly induced in the caregiver by each endorsed neuropsychiatric behavior.

Theoretical Framework

The theoretical architecture of the NPI-Q rests upon the intersection of modern behavioral neurology, neuropsychiatry, and transactional models of psychological stress and caregiver appraisal. The instrument rejects the historically pervasive view that non-cognitive symptoms in dementia are merely nonspecific epiphenomena or reactive psychological frustrations resulting from memory loss. Instead, the framework posits that BPSD manifestations are direct phenotypic expressions of site-specific neuropathological disruptions within dedicated neural circuits, compounded by environmental mismatch and neurochemical depletion.

From a neurobiological standpoint, Cummings and colleagues conceptualized neuropsychiatric symptoms as signatures of specific frontosubcortical and limbic network disruptions. Agitation, disinhibition, and apathy arise from selective structural and functional disconnections within the orbitofrontal cortex, anterior cingulate cortex, and dorsolateral prefrontal-striato-pallido-thalamic circuits. Psychotic symptoms (delusions and visual hallucinations) emerge from imbalances in cholinergic-dopaminergic neurotransmission, paired with neuropathological burden across temporal-parietal association cortices and visual processing streams (the ventral “what” pathway). Vegetative disturbances reflect hypothalamic dysregulation and monoaminergic brainstem nucleus degradation (locus coeruleus, raphe nuclei). By establishing domain-specific questions, the NPI-Q operationalizes this neuroanatomical localization model for bedside and ambulatory application.

Simultaneously, the NPI-Q integrates Lazarus and Folkman’s transactional model of stress and coping and Pearlin’s stress process model of caregiving. These models postulate that objective patient behaviors represent primary stressors that undergo subjective cognitive appraisal by the caregiver. Caregiver distress is not an automatic, one-to-one linear consequence of symptom frequency or intensity; rather, it is mediated by the perceived threat, coping resources, relationship history, and contextual disruption caused by the symptom. For instance, whereas severe apathy may cause limited physical disruption in the home, it frequently causes profound existential distress for a spouse grieving the loss of reciprocal emotional intimacy. Conversely, severe aberrant motor behavior (e.g., repetitive folding) may be rated low in distress if the caregiver possesses adequate coping mechanisms. By measuring symptom severity and caregiver distress simultaneously across the identical behavioral domain, the NPI-Q structurally reflects this dual-layered theoretical model.

Validity

The psychometric validity of the NPI-Q has been corroborated through rigorous clinical investigations across international cohorts, encompassing Alzheimer’s disease, Lewy body dementia, frontotemporal lobar degeneration, and vascular cognitive impairment.

Convergent and Criterion Validity

The seminal validation study by Kaufer et al. (2000) evaluated the NPI-Q against the reference-standard, clinician-administered Neuropsychiatric Inventory (NPI) in a cohort of outpatients with cognitive disorders. The concurrent validity coefficients demonstrated exceptionally high agreement:

• Total Severity Score: The correlation between the NPI-Q total severity score and the standard NPI total score (Frequency × Severity composite) was statistically robust, with Pearson correlation coefficients ranging from r = 0.89 to 0.92 (p < 0.001).
• Total Caregiver Distress Score: The correlation between the NPI-Q total caregiver distress score and the standard NPI total caregiver distress score yielded Pearson coefficients of r = 0.84 to 0.90 (p < 0.001).
• Individual Domain Agreement: Across the individual 12 domains, Cohen’s kappa (κ) coefficients of concordance for symptom presence between the self-administered questionnaire and the clinical interview ranged from 0.60 to 0.86, signifying moderate to near-perfect diagnostic convergence. High concordance was notably observed in the domains of Delusions (κ = 0.76), Agitation (κ = 0.78), and Apathy (κ = 0.80).

Discriminant and Known-Groups Validity

The NPI-Q successfully differentiates distinct neurodegenerative syndromes based on their canonical behavioral profiles. Studies comparing patients with frontotemporal dementia (bvFTD) and Alzheimer’s disease demonstrate that the NPI-Q captures statistically significant elevations in Disinhibition, Apathy, and Appetite/Eating alterations in bvFTD groups, consistent with fronto-insular and anterior temporal atrophy. In contrast, patients with dementia with Lewy bodies (DLB) display significantly higher scores in the Hallucinations and Night-time Behaviors domains compared to matched Alzheimer’s disease cohorts. Furthermore, the NPI-Q sensitively discriminates between stages of cognitive decline, exhibiting significant linear increases in total severity and distress scores across stages defined by the Clinical Dementia Rating (CDR) scale (CDR 0.5, CDR 1, CDR 2, and CDR 3).

Construct and Predictive Validity

Construct validity is evidenced by strong correlations between NPI-Q caregiver distress indices and established caregiver burden instruments, such as the Zarit Burden Interview (ZBI; correlations consistently exceeding r = 0.65). In longitudinal investigations, baseline NPI-Q total severity scores and specific sub-domain elevations (particularly Agitation and Psychosis) strongly predict premature nursing home placement, accelerated cognitive decline, elevated emergency department utilization, and higher all-cause mortality, independent of baseline cognitive function measured by the Mini-Mental State Examination (MMSE).

Reliability

The psychometric reliability of the NPI-Q has been evaluated via test-retest assessments, inter-rater reliability paradigms, and internal consistency evaluations across numerous linguistic adaptations (e.g., German, Spanish, Japanese, Chinese, French, and Italian).

Test-Retest Reliability

In the primary validation study by Kaufer et al. (2000), test-retest reliability was evaluated by administering the NPI-Q to primary caregivers on two separate occasions within a mean interval of several days to two weeks, ensuring the underlying clinical state remained stable. Test-retest reliability for the identification of symptom presence across individual domains yielded Cohen’s kappa values ranging from 0.70 to 0.95. For continuous metric totals, the test-retest correlation coefficients reached:

• Total Symptom Severity: Intraclass correlation coefficient (ICC) = 0.89 to 0.94.
• Total Caregiver Distress: ICC = 0.80 to 0.89.

Internal Consistency

Because the NPI-Q measures a multidimensional array of behavioral symptoms that do not necessarily co-occur (for example, severe apathy can manifest in the complete absence of visual hallucinations or euphoria), conventional internal consistency estimates (Cronbach’s alpha) must be interpreted with caution. However, across comprehensive clinical cohorts, the total severity scale exhibits acceptable to good internal consistency, with Cronbach’s alpha coefficients typically ranging from α = 0.74 to α = 0.82. The caregiver distress scale yields higher internal consistency estimates, frequently between α = 0.81 and α = 0.88, reflecting the cumulative, systemic psychological burden experienced across multiple behavioral domains.

Inter-Informant and Inter-Rater Reliability

Studies examining concordance between different primary caregivers residing within the same household (e.g., primary spouse caregiver vs. adult child secondary caregiver) have shown moderate to substantial agreement, with weighted kappa coefficients averaging between 0.62 and 0.75 across domains. When administered by trained clinical staff assisting an informant versus complete self-administration by an informant, intra-class agreement exceeds 0.92, verifying the robust clarity and operational simplicity of the questionnaire’s standardized prompts.

Factor Analysis

Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across diverse international geriatric populations indicate that the 12 domains of the NPI-Q do not operate as isolated entities; rather, they coalesce into clinically coherent, neurobiologically grounded syndromic clusters.

Across multiple large-scale investigations (such as those conducted using the National Alzheimer’s Coordinating Center [NACC] database and European multicenter cohorts), factor analyses have consistently extracted three-factor and four-factor structural solutions, explaining between 48% and 62% of the total variance in neuropsychiatric symptoms:

• Factor 1: Hyperactivity / Agitation / Impulsivity Sub-syndrome: Consistently includes high factor loadings for Agitation/Aggression (loadings typically > 0.65), Irritability/Lability (loadings > 0.68), Disinhibition (loadings > 0.58), and Aberrant Motor Disturbance (loadings > 0.52). In some 4-factor extractions, Elation/Euphoria cross-loads here or forms a distinct frontal-affective component.
• Factor 2: Affective / Mood Sub-syndrome: Characterized by strong primary loadings for Depression/Dysphoria (loadings > 0.75) and Anxiety (loadings > 0.70), frequently accompanied by moderate loadings from Apathy/Indifference and Irritability.
• Factor 3: Psychosis Sub-syndrome: Defined by dominant, highly specific loadings for Delusions (loadings > 0.78) and Hallucinations (loadings > 0.81). This factor demonstrates minimal cross-loadings with affective items, underscoring its distinct underlying neuropathology (often involving Lewy pathology or advanced temporoparietal neurofibrillary tangle burden).
• Factor 4: Vegetative / Homeostatic Dysregulation Sub-syndrome: In four-factor solutions, Night-time Behaviors (sleep disruptions; loadings > 0.65) and Appetite/Eating changes (loadings > 0.62) isolate together, reflecting hypothalamic and circadian pacemaker dysfunction.

Confirmatory factor analytic investigations have verified that these syndromic models exhibit acceptable to excellent goodness-of-fit indices across clinical samples. Standard structural equation modeling typically yields a Comparative Fit Index (CFI) > 0.92, a Tucker-Lewis Index (TLI) > 0.90, and a Root Mean Square Error of Approximation (RMSEA) < 0.055 (90% CI: 0.042–0.068), affirming the multidimensional structural stability of the instrument across varying stages of cognitive impairment.

Instrument / Measurement Tool

The Neuropsychiatric Inventory–Questionnaire is an informant-completed, paper-and-pencil or digital screening instrument. It comprises 12 domain-specific screening modules followed by conditional branching ratings. Administration requires approximately 5 minutes.

Structure and Format

• Assessment Period: The informant evaluates behaviors observed strictly over the preceding one month (last 30 days).
• Screening Phase (Binary Gating): Each of the 12 domains begins with a clear, lay-worded screening question outlining cardinal symptoms. Informants circle “Yes” if the symptom has been present during the past month, or “No” if it was absent.
• Conditional Severity Rating (Patient Impact): If the informant marks “Yes”, they must rate the overall clinical severity of that symptom on a 3-point ordinal scale:
  • 1 = Mild (noticeable, but not a significant change)
  • 2 = Moderate (significant, but not a dramatic change)
  • 3 = Severe (very marked or prominent, a dramatic change)

  (If “No” was marked, the severity score for that domain is automatically scored as 0.)

• Conditional Caregiver Distress Rating (Informant Impact): For every domain marked “Yes”, the informant independently rates how emotionally distressing or disruptive that symptom was for them, using a 6-point ordinal scale:
  • 0 = Not distressing at all (points: 0)
  • 1 = Minimal (slightly distressing, not a problem to cope with; points: 1)
  • 2 = Mild (not very distressing, generally easy to cope with; points: 2)
  • 3 = Moderate (fairly distressing, not always easy to cope with; points: 3)
  • 4 = Severe (very distressing, difficult to cope with; points: 4)
  • 5 = Extreme or Very Severe (extremely distressing, unable to cope with; points: 5)

  (If “No” was marked, the caregiver distress score for that domain is automatically scored as 0.)

Scoring Metrics

• Total NPI-Q Severity Score: Calculated as the arithmetic sum of the severity scores across all 12 domains. Range: 0 to 36. Higher scores denote greater global neuropsychiatric symptom burden in the patient.
• Total NPI-Q Caregiver Distress Score: Calculated as the arithmetic sum of the distress scores across all 12 domains. Range: 0 to 60. Higher scores indicate greater psychological burden, caregiver strain, and potential risk for burnout.
• Domain-Specific Profiling: Individual domain scores are tracked independently to monitor targeted pharmacological or behavioral interventions (e.g., assessing if a reduction in Agitation severity from 3 to 1 correlates with a decline in Caregiver Distress from 5 to 1).

Permissions & Fee and Test Year

The Neuropsychiatric Inventory–Questionnaire was formally validated and published in 2000 by Daniel I. Kaufer, Jeffrey L. Cummings, and colleagues, building on the original NPI introduced in 1994 (10-item form) and expanded in 1997 (12-item form including sleep and appetite).

Copyright and Licensing: The copyright for the NPI, NPI-Q, and their authorized derivatives is held by Jeffrey L. Cummings, MD, ScD. The instrument is protected by international copyright laws. The terms of use are structured as follows:

• Non-Commercial Clinical and Academic Research Use: The NPI-Q is made accessible free of charge for individual clinical practice, non-funded academic research, and non-commercial educational activities. Clinicians and investigator-initiated researchers may access the instrument through academic repositories or via direct registration at the official clearinghouse (http://npitest.net/).
• Commercial and Funded Clinical Trials Use: For industry-sponsored studies, pharmaceutical clinical trials, commercial registries, or integration into proprietary electronic health record (EHR) platforms or mobile digital health technologies, formal licensing agreements, user fees, and permissions must be secured prior to administration. Licensing management is handled through designated commercial copyright administration entities acting on behalf of Dr. Cummings.

References

The scientific literature and clinical validation studies establishing the psychometric foundations of the NPI-Q include:

• Aalten, P., Verhey, F. R., Boziki, M., Bullock, R., Byrne, E. J., Camus, V., Caputo, M., Collins, D., De Deyn, P. P., Elahi, F., Frisoni, G., Gauthier, S., Holmes, C., Hurt, C., Karuza, J., Kaufer, D., Kerry, A., Koch, H., Lyketsos, C. G., … Robert, P. H. (2007). Neuropsychiatric syndromes in dementia. Results from the European Alzheimer Disease Consortium: Part I. Dementia and Geriatric Cognitive Disorders, 24(6), 457–463. https://doi.org/10.1159/000110738
• Cummings, J. L. (1997). The Neuropsychiatric Inventory: Assessing psychopathology in dementia patients. Neurology, 48(5 Suppl 6), S10–S16. https://doi.org/10.1212/wnl.48.5_suppl_6.10s
• Cummings, J. L., Mega, M., Gray, K., Rosenberg-Thompson, S., Carusi, D. A., & Gornbein, J. (1994). The Neuropsychiatric Inventory: Comprehensive assessment of psychopathology in dementia. Neurology, 44(12), 2308–2314. https://doi.org/10.1212/wnl.44.12.2308
• Hollingworth, P., Hamshere, M. L., Moskvina, V., Swanson, C., Harold, D., Abraham, R., Lovestone, S., Powell, J., Holmes, C., Williams, J., & Owen, M. J. (2006). Four-factor structure of the Neuropsychiatric Inventory in an Alzheimer’s disease sample. The Journal of Neuropsychiatry and Clinical Neurosciences, 18(4), 513–520. https://doi.org/10.1176/jnp.2006.18.4.513
• Kaufer, D. I., Cummings, J. L., Ketchel, P., Smith, V., MacMillan, A., Shelley, T., Lopez, O. L., & DeKosky, S. T. (2000). Validation of the NPI-Q, a brief clinical form of the Neuropsychiatric Inventory. The Journal of Neuropsychiatry and Clinical Neurosciences, 12(2), 233–239. https://doi.org/10.1176/jnp.12.2.233
• Lyketsos, C. G., Lopez, O., Jones, B., Fitzpatrick, A. L., Breitner, J., & DeKosky, S. (2002). Prevalence of neuropsychiatric symptoms in dementia and mild cognitive impairment: Results from the Cardiovascular Health Study. JAMA, 288(12), 1475–1483. https://doi.org/10.1001/jama.288.12.1475
• Trzepacz, P. T., Hochstetler, H., Wang, S., Walker, B., & Saykin, A. J. (2013). Relationship between the Neuropsychiatric Inventory-Questionnaire and other clinical measures in Alzheimer’s Disease Neuroimaging Initiative. Alzheimer’s & Dementia, 9(4), P481–P482. https://doi.org/10.1016/j.jalz.2013.04.225

13. Items of the Scale (Questionnaire)