• Systems Integration
• Education & Leadership
• Assessment
• Non-Pharmacological Approaches

Assessment of Behavioral and Psychological Symptoms of Dementia (BPSD)

One of the first steps in responding to the Behavioral and Psychological Symptoms of Dementia (BPSD) through pharmacological treatments or non-pharmacological approaches is to detect their occurrence.1 There is a growing consensus that early and ongoing assessment of behaviors can lead to a more effective response. For example, the AMA Performance Dementia Measurement Set recommends that health providers systematically screen for behaviors at a minimum once a year using a standardized assessment tool. It is unclear, however, which assessment tool to use and whether sufficient standardized approaches to ongoing behavioral detection are available from which effective plans for response can be derived. Nevertheless, behavior should be assessed on a frequent, ongoing, and systematic basis so that patterns can be identified that will help inform the design of NPA.

To evaluate the state-of-science concerning behavioral assessments, a comprehensive review of available published measures was conducted and their psychometric properties summarized. A final set of 44 measures were identified, with most having adequate psychometric properties. The majority of measures (n=15) covered a broad range of behaviors and were developed for use in dementia care or in nursing homes. Measures were also found for specific neuropsychiatric symptoms, including agitation, apathy, aggression, depression, anxiety, sleep, and wandering, though these were less likely to be specific for a dementia population. No specific measures were identified separate from the general measures, which addressed behaviors such as euphoria, hallucinations, irritability apart from aggression or anxiety, and/or motor and verbal disturbances. Table 1 lists all 44 measures and their essential properties. Table 2 summarizes the number of measures for each behavioral category considered. Table 3 compares the 44 measures in terms of settings, number of items, and behavioral categories included.

¹Note: For any resident who exhibits an acute and fluctuating change in behavior, an initial assessment for delirium should be made to rule out the presence of acute medical conditions. Two recent systematic reviews of instruments to detect delirium may be useful to staff trained in their administration:

• Morandi, A., McCurley, J., Vasilevskis, E., et al. Tools to detect delirium superimposed on dementia: A systematic review. Journal of the American Geriatrics Society. 2012;60:2005-2012.


• Wong, C., Holroyd-Leduc, J., Simel, D., et al. Does this patient have delirium? Value of bedside instruments.Journal of the American Medical Association. 2010;304:779-786

• Table 1. A Systematic Review of Measures of Behaviors for Use with Dementia Patients
• Table 2. Summary of Results from Systematic Review of Behavioral Measures for Use with Dementia Patients (N=44)
• Table 3. Comparison of 44 Behavioral Symptom Scales along Settings, Number of Items and Behavioral Categories Included

Table 1. A Systematic Review of Measures of Behaviors for Use with Dementia Patients

Table 1. Notes

Brief Psychiatric Rating Scale not included since generally used to assess psychosis in patients with schizophrenia.

Scale for the Assessment of Negative Symptoms (SANS) and Positive and Negative Symptom Scale (PANSS) are not included since primarily used in patients with schizophrenia.

Unified Parkinson's Disease rating scale not included as it is a single item for apathy that does not have any reliability measure.

Hamilton Depression Rating Scale not included because it was first published in 1960 and is no longer considered the gold standard. While it has adequate internal reliability, many of the items do not contribute to depression severity, response options are not optimal ,and the retest reliability is poor. Content validity has also been found to be poor (Bagby, Ryder, Schuller & Marshall, 2004).

The Mini Nutritional Assessment was not included because it does not address behavior. The Nutritional Risk Index and DETERMINE Your Nutritional Health Checklist were not included because they do not address behavior and because validity has not been established.

GDS was not included as at least two studies have found it not to be as reliable and valid in a dementia population as in the general geriatric population (Burke, 1989; Korner, 2006).

CES-D was not included as there was no evidence of reliability and validity in a dementia population.

State/Trait anxiety scale: originally published prior to 1980.

Geriatric Evaluation by Relative's Rating Instrument (GERRI) Schwartz, G. (1983). Development and validation of the Geriatric Evaluation by Relative's Rating Instrument (GERRI). Psychological Reports. 53:479-88-not included because there is no subscale for neuropsychiatric symptoms.

Clinical Assessment of Psychopathology among Elderly Residents (CAPER): Reichenfeld (1992) not included because it does not address specific behaviors of dementia-it is for diagnosis of psychotic disorders. It is also based on psychiatrist interview.

Table 2. Summary of Results from Systematic Review of Behavioral Measures for Use with Dementia Patients (N=44)

Table 3. Comparison of 44 Behavioral Symptom Scales along Settings, Number of Items and Behavioral Categories Included

Table 3. Notes

References