Neuropsychiatric Symptoms in Dementia: Overview and Measurement Challenges

Introduction

Neuropsychiatric symptoms (NPS) are now known to occur almost universally over the course of dementia, Alzheimer’s in particular (1, 2). They also occur with higher-than-expected frequency in the dementia prodrome of mild cognitive impairment (3). Further, NPS in the form of “mild behavioral impairment,” in the absence of cognitive impairment, may constitute a dementia prodrome (4). NPS are associated with a number of adverse outcomes including accelerated transition from prodrome to dementia (4), and faster progression from early dementia to severe dementia or death [Peters et al, Am J Psychiatry, in press], as well as serious adverse effects for patients and caregivers such as greater disability, worse quality of life, earlier institutionalization, increased burden, and higher health care costs (2). Given their central importance, NPS are increasingly a focus of study with an eye to the development of effective treatments.

The heterogeneity of NPS complicates treatment development; they are heterogeneous in both phenomenology and cause. A wide range of symptoms has been reported although they tend to aggregate into predictable groups, especially in milder disease (5). The most reproducible groupings have been: depression, apathy, psychosis, agitation, and sleep disturbances [6]. Recent treatment development has targeted presumptive or proposed syndromes in these areas (7, 8). From the point of view of cause, NPS likely result from several interacting factors. The brain neurodegenerative process, through disruption of brain circuits involved in affect, behavior, motivation, or perception, is central to their emergence either through direct damage to circuits, or through creating vulnerabilities acted on by the environment. Additionally, NPS may result because of difficulties individuals face adapting to the surroundings as they lose cognitive abilities. Further, NPS may arise because of unmet needs, acute comorbid illnesses causing confusional states akin to delirium, or environments and caregiving that are mismatched with the patient’s current capabilities and skills [Kales et al, BMJ, under review).

Treatment development for NPS has to grapple with a number of issues concurrently. Efforts have included the development of nonpharmacologic approaches (9), as well as pharmacologic approaches. The latter initially involved importation of psychopharmaca developed for the treatment of psychiatric disorders into this setting with mixed results (2, 6). Of particular concern has been unexpected serious risks associated with these medications, antipsychotics in particular (6).

Central to treatment development for NPS is clinical measurement. As there are no specific “direct” measures of NPS, as with measures of cognitive functioning, measurement relies on reporting of observable behaviors and mental states by patients and others. Measurement is affected by a number of variables. These include aspects of the cognitive disorder that limit patient ability to reveal mental state, or lead to forgetting prior experiences and behaviors. As a result, measurement depends on input from caregivers who themselves are “filters,” leading at times to biased reporting in particular when caregivers suffer from psychological disturbances—a common finding in dementia caregivers (10). Further, because NPS are episodic, relapsing and remitting frequently, often “real-time” in response to environmental situations, the quantification of NPS frequency and severity over longer time frames is difficult.

In the last several decades a number of measures have been developed for the study of NPS. The most widely used are discussed in the most recent Handbook of Psychiatric Measures (11). They include “broad spectrum” measures to quantify the occurrence, frequency, severity, or other attributes of NPS (e.g., associated caregiver distress), without going into depth for individual disturbances. Other measures are “narrow spectrum” focused on individual disturbances such as depression, apathy, or agitation.

Some measures acquire information about the NPS entirely from structured caregiver interviews. Other measures use structured patient interviews, and yet others combinations of these approaches. The introduction of rating judgments from experienced clinicians, following the LED standard (longitudinal, expert, all data) is the optimal way of quantifying NPS; it is also the most complicated, costly, and time-consuming (12). Development of the Neuropsychiatric Inventory-clinician rating (NPI-C) (13), has produced a versatile measure that can be used as a broad spectrum measure based entirely on caregiver report or as a narrow spectrum measure of a particular domain, such as agitation. In the latter case NPI-C incorporates information from patient, caregiver and other sources and applies clinician judgment in making final ratings. NPI-C has better inter-rater reliability then the conventional NPI (13) perhaps because the latter is highly structured, based only on caregiver report, and not very sensitive to change (14). Rating judgments from well-trained clinicians are least affected by biases related to caregiver report.

In summary, NPS are a central focus of study and treatment development in the context of Alzheimer’s and other neurodegenerative brain diseases. It should be no surprise that such symptoms are universal in diseases that affect key brain areas regulating behavior, or disrupt multiple brain areas over time. Despite heterogeneity in phenomenology and etiology, treatment development for NPS has accelerated in the last few years (15) with the promise of more effective novel treatments on the immediate horizon.