The clinical problem of neuropsychiatric signs and symptoms in dementia

Predicting neuropsychiatric symptoms of persons with dementia in a day care center using a facial expression recognition system

BACKGROUND

Behavioral and psychological symptoms of dementia (BPSD) affect 90% of persons with dementia (PwD), resulting in various adverse outcomes and aggravating care burdens among their caretakers. This study aimed to explore the potential of artificial intelligence-based facial expression recognition systems (FERS) in predicting BPSDs among PwD.

METHODS

A hybrid of human labeling and a preconstructed deep learning model was used to differentiate basic facial expressions of individuals to predict the results of Neuropsychiatric Inventory (NPI) assessments by stepwise linear regression (LR), random forest (RF) with importance ranking, and ensemble method (EM) of equal importance, while the accuracy was determined by mean absolute error (MAE) and root-mean-square error (RMSE) methods.

RESULTS

Twenty-three PwD from an adult day care center were enrolled with ≥ 11,500 FERS data series and 38 comparative NPI scores. The overall accuracy was 86% on facial expression recognition. Negative facial expressions and variance in emotional switches were important features of BPSDs. A strong positive correlation was identified in each model (EM: r = 0.834, LR: r = 0.821, RF: r = 0.798 by the patientwise method; EM: r = 0.891, LR: r = 0.870, RF: r = 0.886 by the MinimPy method), and EM exhibited the lowest MAE and RMSE.

CONCLUSIONS

FERS successfully predicted the BPSD of PwD by negative emotions and the variance in emotional switches. This finding enables early detection and management of BPSDs, thus improving the quality of dementia care.

Prediction of Relapse After Discontinuation of Antipsychotic Treatment in Alzheimer's Disease: The Role of Hallucinations

OBJECTIVE

In Alzheimer's disease, antipsychotic medications are often used for a period, with relief of symptoms, and then discontinued, after which relapse may occur. The authors sought to determine which neuropsychiatric symptoms predict relapse.

METHOD

In the Antipsychotic Discontinuation in Alzheimer's Disease trial, 180 patients with Alzheimer's disease and symptoms of agitation or psychosis were treated with risperidone for 16 weeks, after which patients who responded (N=110) were randomly assigned to continue risperidone for 32 weeks, to continue risperidone for 16 weeks followed by switch to placebo for 16 weeks, or to receive placebo for 32 weeks. As reported previously, discontinuation of risperidone was associated with a two- to fourfold increased risk of relapse over 16-32 weeks. In planned post hoc analyses, the authors examined associations between the 12 symptom domains in the Neuropsychiatric Inventory (NPI) and relapse in the first 16-week phase after randomization.

RESULTS

Compared with patients with mild hallucinations or no hallucinations, patients with severe hallucinations as a presenting symptom at baseline had a higher likelihood of relapse (hazard ratio=2.96, 95% CI=1.52, 5.76). This effect was present for the subgroup with auditory hallucinations, but not the subgroup with visual hallucinations. Among patients with baseline hallucinations, 13 of 17 (76.5%) who discontinued risperidone relapsed, compared with 10 of 26 (38.5%) who continued risperidone (p<0.02). This group difference remained significant for severe (77.8%) compared with mild (36%) hallucinations. NPI domain scores after the initial open-treatment phase were not associated with relapse.

CONCLUSIONS

Patients with severe baseline hallucinations were more likely to relapse after randomization, and the presence of baseline hallucinations was associated with a higher risk of relapse after discontinuation of risperidone compared with continued risperidone treatment. For patients with hallucinations, particularly auditory hallucinations, antipsychotic discontinuation should be approached cautiously because of high relapse risk.

Determinants of Capacity to Consent to Research on Alzheimer's disease

OBJECTIVE

Investigators conducting Alzheimer's disease (AD) research need to consider participants' capacity to consent. Cognitive functioning is a significant predictor of decisional capacity, but there is a dearth of information on the influence of neuropsychiatric symptoms in AD on decisional capacity. We examined the rates of decisional capacity associated with two types of research protocols, and the association of capacity with neuropsychiatric symptoms and other participant characteristics.

METHODS

We comprehensively evaluated decisional capacity among 64 patients with mild-to-moderate AD and 70 healthy comparison (HC) subjects randomized to consider either a medium risk or higher risk hypothetical research protocol. Additional measures included sociodemographics, cognitive deficits, and neuropsychiatric symptoms.

RESULTS

Twenty AD patients (31.3%) and 67 HCs (95.7%) were deemed capable; 44 AD patients (68.8%) and 3 HCs (4.3%) incapable of consent. Age, education, and severity of cognitive deficits were associated with incapable status; there were no significant associations with severity of neuropsychiatric symptoms or protocol risk level.

CONCLUSIONS

Findings highlight the importance of understanding of capacity and its assessment among people with AD, rather than treating AD diagnosis as synonymous with impaired capacity. As novel treatments move from bench to bedside, methods of assessing and addressing capacity impairment must similarly advance.

CLINICAL IMPLICATIONS

In assessing research consent capacity, use structured assessments with population specific cut scores interpreted in the context of the person's background including education, culture, and language. Individuals should be encouraged to execute research proxy documents when able.