The Neuropsychiatric Inventory-Diary Rating Scale (NPI-Diary): A Method for Improving Stability in Assessing Neuropsychiatric Symptoms in Dementia

Identification of Dementia & Mild Cognitive Impairment in Chinese Elderly Using Machine Learning

OBJECTIVE

To assess the role of Machine Learning (ML) in identification critical factors of dementia and mild cognitive impairment.

METHODS

371 elderly individuals were ultimately included in the ML analysis. Demographic information (including gender, age, parity, visual acuity, auditory function, mobility, and medication history) and 35 features from 10 assessment scales were used for modeling. Five machine learning classifiers were used for evaluation, employing a procedure involving feature extraction, selection, model training, and performance assessment to identify key indicative factors.

RESULTS

The Random Forest model, after data preprocessing, Information Gain, and Meta-analysis, utilized three training features and four meta-features, achieving an area under the curve of 0.961 and a accuracy of 0.894, showcasing exceptional accuracy for the identification of dementia and mild cognitive impairment.

CONCLUSIONS

ML serves as a identification tool for dementia and mild cognitive impairment. Using Information Gain and Meta-feature analysis, Clinical Dementia Rating (CDR) and Neuropsychiatric Inventory (NPI) scale information emerged as crucial for training the Random Forest model.

Assessing behavioral and psychological symptoms of dementia: a comprehensive review of current options and future perspectives

The behavioral and psychological symptoms of dementia (BPSD) are a heterogeneous set of challenging disturbances of behavior, mood, perception, and thought that occur in almost all patients with dementia. A huge number of instruments have been developed to assess BPSD in different populations and settings. Although some of these tools are more widely used than others, no single instrument can be considered completely satisfactory, and each of these tools has its advantages and disadvantages. In this narrative review, we have provided a comprehensive overview of the characteristics of a large number of such instruments, addressing their applicability, strengths, and limitations. These depend on the setting, the expertise required, and the people involved, and all these factors need to be taken into account when choosing the most suitable scale or tool. We have also briefly discussed the use of objective biomarkers of BPSD. Finally, we have attempted to provide indications for future research in the field and suggest the ideal characteristics of a possible new tool, which should be short, easy to understand and use, and treatment oriented, providing clinicians with data such as frequency, severity, and triggers of behaviors and enabling them to find appropriate strategies to effectively tackle BPSD.

Machine learning-based predictive models for the occurrence of behavioral and psychological symptoms of dementia: model development and validation

The behavioral and psychological symptoms of dementia (BPSD) are challenging aspects of dementia care. This study used machine learning models to predict the occurrence of BPSD among community-dwelling older adults with dementia. We included 187 older adults with dementia for model training and 35 older adults with dementia for external validation. Demographic and health data and premorbid personality traits were examined at the baseline, and actigraphy was utilized to monitor sleep and activity levels. A symptom diary tracked caregiver-perceived symptom triggers and the daily occurrence of 12 BPSD classified into seven subsyndromes. Several prediction models were also employed, including logistic regression, random forest, gradient boosting machine, and support vector machine. The random forest models revealed the highest area under the receiver operating characteristic curve (AUC) values for hyperactivity, euphoria/elation, and appetite and eating disorders; the gradient boosting machine models for psychotic and affective symptoms; and the support vector machine model showed the highest AUC. The gradient boosting machine model achieved the best performance in terms of average AUC scores across the seven subsyndromes. Caregiver-perceived triggers demonstrated higher feature importance values across the seven subsyndromes than other features. Our findings demonstrate the possibility of predicting BPSD using a machine learning approach.

Factors Associated With Behavioral and Psychological Symptoms of Dementia: Prospective Observational Study Using Actigraphy

BACKGROUND

Although disclosing the predictors of different behavioral and psychological symptoms of dementia (BPSD) is the first step in developing person-centered interventions, current understanding is limited, as it considers BPSD as a homogenous construct. This fails to account for their heterogeneity and hinders development of interventions that address the underlying causes of the target BPSD subsyndromes. Moreover, understanding the influence of proximal factors-circadian rhythm-related factors (ie, sleep and activity levels) and physical and psychosocial unmet needs states-on BPSD subsyndromes is limited, due to the challenges of obtaining objective and/or continuous time-varying measures.

OBJECTIVE

The aim of this study was to explore factors associated with BPSD subsyndromes among community-dwelling older adults with dementia, considering sets of background and proximal factors (ie, actigraphy-measured sleep and physical activity levels and diary-based caregiver-perceived symptom triggers), guided by the need-driven dementia-compromised behavior model.

METHODS

A prospective observational study design was employed. Study participants included 145 older adults with dementia living at home. The mean age at baseline was 81.2 (SD 6.01) years and the sample consisted of 86 (59.3%) women. BPSD were measured with a BPSD diary kept by caregivers and were categorized into seven subsyndromes. Independent variables consisted of background characteristics and proximal factors (ie, sleep and physical activity levels measured using actigraphy and caregiver-reported contributing factors assessed using a BPSD diary). Generalized linear mixed models (GLMMs) were used to examine the factors that predicted the occurrence of BPSD subsyndromes. We compared the models based on the Akaike information criterion, the Bayesian information criterion, and likelihood ratio testing.

RESULTS

Compared to the GLMMs with only background factors, the addition of actigraphy and diary-based data improved model fit for every BPSD subsyndrome. The number of hours of nighttime sleep was a predictor of the next day's sleep and nighttime behaviors (odds ratio [OR] 0.9, 95% CI 0.8-1.0; P=.005), and the amount of energy expenditure was a predictor for euphoria or elation (OR 0.02, 95% CI 0.0-0.5; P=.02). All subsyndromes, except for euphoria or elation, were significantly associated with hunger or thirst and urination or bowel movements, and all BPSD subsyndromes showed an association with environmental change. Age, marital status, premorbid personality, and taking sedatives were predictors of specific BPSD subsyndromes.

CONCLUSIONS

BPSD are clinically heterogeneous, and their occurrence can be predicted by different contributing factors. Our results for various BPSD suggest a critical window for timely intervention and care planning. Findings from this study will help devise symptom-targeted and individualized interventions to prevent and manage BPSD and facilitate personalized dementia care.

Association between care burden, depression and personality traits in Alzheimer's caregiver: A pilot study

Introduction

In this study correlations between care burden, depression, and personality at intrapsychic level in caregivers of Alzheimer’s disease patients were evaluated.

Materials and methods

Caregivers: n.40. Tests: Social-schedule; CBI; BDI; SASB-Structural-Analysis of Social Behaviours- Form-A- intrapsychic behaviours (8 Cluster); ECOG. Patients:MMSE. Statistical analysis: Chi-squared test; Anova one way F test; Pearson’s R coefficient.

Results

Correlations: CBI-total and NPI(p < .001); CBI-total—ECOG (p = .042); CBI-total—BDI(p< .001); CBI- total-SASB-Cl7(p = .014); SASB-CL8(p<0.000); BDI and SASB-Cl 2 (p = .018), SASB-Cl 3 (p = .004), SASB-Cl7(p < .000), SASB-CL8 (p < .000). High CBI is correlated with high depression, neuropsychiatry symptoms, low cognitive patient’s functions. Caregivers have the following intrapsychic behaviors: poor self-care, poor ability to take care of themselves; they exercise control over themselves and do not consider and/or ignore their basic needs at emotional and physical levels. These intrapsychic behaviours are indicators of depression (SASB Model) and are correlated with high care burden–CBI and high depression-CDQ.

Discussion

Care burden is closely related to the depression and individual personality (intrapsychic experience) of the caregiver. This may reveal a source of strength and may suggest areas of multidimensional and psychotherapeutic interventions.

Biodanza as a Nonpharmacological Dance Movement-Based Treatment in Older People With Alzheimer's Disease: An Italian Pilot Study in 2 Tuscan Nursing Homes

The aim of this pilot, feasibility study was to assess health improvements in 16 institutionalized older people with Alzheimer's disease, after the Biodanza intervention, a nonpharmacological dance movement-based treatment. Biodanza significantly decreased agitated and neuropsychiatric behaviors. Effectiveness studies will be performed in order to assess the implication of such interventions.

[Neuropsychiatric symptoms in patients with mild cognitive impairment]

OBJECTIVE

To study the prevalence of neuropsychiatric symptoms (NPS) in patients with mild cognitive impairment (MCI) who asked for help in memory clinics.

MATERIAL AND METHODS

We analysed data on 729 patients with MCI (average age 76.6 years, average MMSE score 25.3), who underwent a course of cognitive neurorehabilitation in a specialized department - a memory clinic. A Russian version of Neuropsychiatric Inventory (NPI) was used. We compared the indicators for the main psychometric scales for the diagnosis of MCI in comparison with the dynamics of NPS.

RESULTS

The prevalence rates for NPS differed in part from those reported in the literature. The most common symptom was anxiety (54.7%) and irritability (56.5%), while euphoria, as well as delusions and hallucinations were not detected. All disorders significantly reduced at the end of the rehabilitation program.

CONCLUSION

MCI influences the level of functioning and social interaction in older patients and mediates the quality of life. Thus, given the increase in life expectancy, it is necessary to introduce new methods of examination applicable in the practice of psychiatrists to diagnose and rehabilitate such patients. NPS turned out to be widespread in MCI, but may reduce during the course of complex neurorehabilitation.

A biopsychosocial interpretation of the Neuropsychiatric Inventory - Nursing Home assessment: reconceptualising psychiatric symptom attributions

BACKGROUND

The Neuropsychiatric Inventory (NPI) is predicated on the assumption that psychiatric symptoms are manifestations of disease. Biopsychosocial theories suggest behavioural changes viewed as psychiatric may also arise as a result of external behavioural triggers. Knowing the causes of psychiatric symptoms is important since the treatment and management of symptoms relies on this understanding.

AIMS

This study sought to understand the causes of psychiatric symptoms recorded in care home settings by investigating qualitatively described symptoms in Neuropsychiatric Inventory-Nursing Home (NPI-NH) interviews.

METHOD

The current study examined the NPI-NH interviews of 725 participants across 50 care homes. The qualitatively described symptoms from each of the 12 subscales of the NPI were extracted: 347 interviews included at least one qualitatively described symptom (n = 651 descriptions). A biopsychosocial algorithm developed following a process of independent researcher coding (n = 3) was applied to the symptom descriptions. This determined whether the description had predominantly psychiatric features, or features that were cognitive or attributable to other causes (i.e. issues with orientation and memory; expressions of need; poor care and communication; or understandable reactions).

RESULTS

Our findings suggest that the majority (over 80%) of descriptions described symptoms with features that could be attributable to cognitive changes and external triggers (such as poor care and communication).

CONCLUSIONS

The finding suggest that in its current form the NPI-NH may over attribute the incidence of psychiatric symptoms in care homes by overlooking triggers for behavioural changes. Measures of psychiatric symptoms should determine the causes of behavioural changes in order to guide treatments more effectively.

Psychogeriatric Inventory of Disconcerting Symptoms and Syndromes (PGI-DSS): validity and reliability of a new brief scale compared to the Neuropsychiatric Inventory for Nursing Homes (NPI-NH)

OBJECTIVES

To validate the Psychogeriatric Inventory of Disconcerting Symptoms and Syndromes (PGI-DSS), a single scale in A4 format comprising four disconcerting syndromes (violence, refusal, words, and acts). The scale enables an immediate conversion of a qualitative assessment to a quantitative assessment. The PGI-DSS was compared with the Neuro Psychiatric Inventory for Nursing Homes (NPI-NH).

DESIGN

Cross-sectional descriptive and correlational studies.

SETTING

Thirty geriatric care units and nursing homes.

PARTICIPANTS

Raters interviewed nurses and nursing assistants in charge of older adults hospitalized in geriatric care units or living in nursing homes (N = 226).

MEASUREMENTS

The French version of the PGI-DSS and the French version of the NPI-NH.

RESULTS

The correlation coefficient between the PGI-DSS and the NPI-NH was 0.70 (p < 0.0001). The PGI-DSS threshold score corresponding to the NPI threshold score was 17 (specificity: 87%, sensitivity: 63%). Four statistical factors, corresponding to the four clinical syndromes, explained 53.4% of the total variance. The internal consistency of the PGI-DSS (Cronbach's alpha = 0.695) was higher than that of the NPI-NH (Cronbach's alpha = 0.474). Test-retest reliability was better for the PGI-DSS than for the NPI-NH. The intraclass correlations were 0.80 [0.73; 0.86] and 0.75 [0.67; 0.83], respectively. Interrater reliability was better for the PGI-DSS than for the NPI-NH. The intraclass correlations were 0.65 [0.55-0.76] and 0.55 [0.43-0.68], respectively.

CONCLUSION

The PGI-DSS was developed to overcome the limitations of the NPI-NH. New, brief, easy to administer in less than 4 minutes, foldable in four parts, pocket-sized, easy-to-read in the palm of the hand, PGI-DSS could have similar or better statistical properties than the NPI-NH. Whereas the 10 domains in the NPI-NH have clinical utility for clinicians, the four easily understandable syndromes in the PGI-DSS can help avoid inappropriate attitudes and can guide psychosocial interventions. It could likewise improve dialogue between caregivers and clinicians.

Steps Towards Developing Effective Treatments for Neuropsychiatric Disturbances in Alzheimer's Disease: Insights From Preclinical Models, Clinical Data, and Future Directions

Alzheimer's disease (AD) is the most common form of dementia worldwide. It is mostly known for its devastating effect on memory and learning but behavioral alterations commonly known as neuropsychiatric disturbances (NPDs) are also characteristics of the disease. These include apathy, depression-like behavior, and sleep disturbances, and they all contribute to an increased caregiver burden and earlier institutionalization. The interaction between NPDs and AD pathology is not well understood, but the consensus is that they contribute to disease progression and faster decline. Consequently, recognizing and treating NPDs might improve AD pathology and increase the quality of life for both patients and caregivers. In this review article, we examine previous and current literature on apathy, depressive symptoms, and sleep disturbances in AD patients and preclinical AD mechanistic models. We hypothesize that tau accumulation, beta-amyloid (Aβ) aggregation, neuroinflammation, mitochondrial damage, and loss of the locus coeruleus (LC)-norepinephrine (NE) system all collectively impact the development of NPDs and contribute synergistically to AD pathology. Targeting more than one of these processes might provide the most optimal strategy for treating NPDs and AD. The development of such clinical approaches would be preceded by preclinical studies, for which robust and reliable mechanistic models of NPD-like behavior are needed. Thus, developing effective preclinical research models represents an important step towards a better understanding of NPDs in AD.

Cirrhosis and frailty assessment in elderly patients: A paradoxical result

The frailty represents a key determinant of elderly clinical assessment, especially because it allows the identification of risk factors potentially modifiable by clinical and therapeutic interventions. The frailty assessment in elderly patients usually is made by using of Fried criteria. However, to assess the frailty in cirrhotic patients, multiple but different tools are used by researchers. Thus, we aimed to compare frailty prevalence in elderly patients with well-compensated liver cirrhosis and without cirrhosis, according to Fried criteria.Among 205 elderly patients screened, a total of 148 patients were enrolled. The patients were divided into 2 groups according to the presence/absence of well-compensated liver cirrhosis.After clinical examination with conventional scores of cirrhosis, all patients underwent anthropometric measurements, nutritional, biochemical, comorbidity, and cognitive performances. Frailty assessment was evaluated according to Fried frailty criteria.Unexpectedly, according to the Fried criteria, non-cirrhotic patients were frailer (14.2%) than well-compensated liver cirrhotic patients (7.5%). The most represented Fried criterion was the unintentional weight loss in non-cirrhotic patients (10.1%) compared to well-compensated liver cirrhotic patients (1.4%). Moreover, cumulative illness rating scale -G severity score was significantly and positively associated with frailty status (r = 0.234, P < .004). In a multivariate linear regression model, only female gender, body mass index and mini nutritional assessment resulted associated with frailty status, independently of other confounding variables.Despite the fact that elderly cirrhotic patients are considered to be frailer than the non-cirrhotic elderly patient, relying solely on "mere visual appearance," our data show that paradoxically non-cirrhotic elderly patients are frailer than elderly well-compensated liver cirrhotic patients. Thus, clinical implication of this finding is that frailty assessment performed in the well-compensated liver cirrhotic patient can identify those cirrhotic patients who may benefit from tailored interventions similarly to non-cirrhotic elderly patients.