A multicentric exploratory study of behavioral and psychological symptom characteristics of dementia.

Background: Behavioral and psychological symptoms (BPS) are usually the expected consequences of dementia. BPS increases morbidity and burden, affects the quality of life, and impacts care costs. However, the symptom characteristics, clinical correlations, and symptom-specific clusters aiding the diagnosis are less well studied, especially in the Indian population. Materials and
Methods: The present study examined the BPS clusters based on various cognitive and neuropsychiatric profiles in patients with dementia under a multicentric study in India. We did a cross-sectional assessment using the Neuropsychiatric Inventory Questionnaire (NPI-Q) and cognitive functions by Montreal Cognitive Assessment (MoCA), and the severity of dementia using the Clinical Dementia Rating (CDR) scale. In addition, all of the participants were evaluated on a structured Clinical Interview for DSM-5 Research Version for past or current psychiatric disorder(s).
Results: We describe the various BPS clusters uniquely associated with the severity of dementia. Further, on linear regression analysis, we predicted three symptom clusters (anxiety, irritability, aberrant motor) in mild, two symptom clusters (disinhibition, agitation/aggression) in moderate and three symptom clusters (delusion, euphoria/elation, disinhibition) in severe dementia.
Conclusion: The study provides insights into the various symptom characteristics and inter-relationship of BPS, which may benefit the clinician while assessing patients with dementia. Copyright:

INTRODUCTION

Dementia is a disease of the elderly, characterized by progressive loss of memory and impairment in various domains of cognitive functions.[1] It was earlier estimated that around 24.3 million people live with dementia, with 4.6 million new cases of dementia added every year.[2] In 2018, the Global Burden of Disease Study (GBD) showed an alarming global trend of 43.8 million people living with dementia.[3] This increasing trend is reflected more prominently in developing countries like India. India has a unique state of affairs due to a rapid epidemiological transition leading to an increasingly ageing population, growing risk factors, and a rising prevalence of dementia.[4] Hence, dementia in India is more understood as an epidemic than merely a neuropsychiatric disease.[4] Furthermore, the current understanding of dementia is inadequate in diagnosis, management and prevention, which is perhaps responsible for the burden caused by the disease.

Behavioral and psychological symptoms (BPS) are common consequences of dementia.[5] Although cognition and behavior are independent dimensions involved in dementia, they tend to influence each other and cause greater impairment in activities of daily living and affect the quality of life substantially.[6] The symptoms affect the patients and are a significant source of distress for the primary caregivers.[7] Petrovic et al.[8] found that 96% of patients with dementia manifest with at least one BPS. Apathy and depression are relatively common, followed by anxiety and agitation. Some of the psychological factors that cause predominant impairment include psychosis, mood lability, aberrant motor behavior and altered biological functions.[58] Data from the Indian subcontinent are limited[14910111213] and reveal around 99% of people with dementia have at least one behavioral and psychological symptom, and 71% have ≥4 symptoms. Most frequent symptoms include apathy and agitation followed by irritability, altered biological functions and emotional lability. Psychosis, although noted, occur in a relatively lesser proportion of the patients.[14] Also, the pattern of BPS differs depending on the subtypes of dementia; for example, Alzheimer’s disease patients have significantly more delusions, hallucinations and anxiety symptoms than vascular dementia.[11] Abnormal motor behavior, disinhibited behavior, and aberrant feeding behavior are more prominent in frontotemporal dementia.[9]

In developing countries like India, there is a desperate need to improve the recognition of symptoms under various BPS clusters. Prompt identification of various symptoms would help in better clinical treatment decisions. Also, due to Indias multi-ethnic, multicultural, and environmental differences, dementia prevalence from different regions differs, and so would BPS. The present study aimed to identify BPS clusters in dementia. We also intended to ascertain the association of severity of dementia to various behavioral and psychological symptoms.

MATERIALS AND METHODS

This multicentric study was conducted under the egis of the Research, Education, and Training Foundation subcommittee of the Indian Psychiatric Society (IPS) after obtaining necessary approval from the Ethics Committee of the Indian Psychiatry Society for Research. This was a cross-sectional multicentric study conducted at six centers spanning India’s northern, southern and eastern parts. All the participating centers followed a uniform research protocol and institutional ethics committees at all participating centers approved the study. Psychiatrists experienced in diagnosing and treating dementia and associated behavioral and psychological symptoms supervised data collection at each center. We used the purposive sampling technique and included patients who were more than or equal to 60 years with a primary diagnosis of dementia according to ICD-10 DCR. Patients with a history of psychosis, mood, anxiety, obsessive-compulsive, substance use (except tobacco and caffeine) or personality disorder antedating the onset of dementia were excluded from the study. Written informed consent was taken from all participants. Sample size estimation was done based on infinite sample calculation keeping precision at 5% and confidence interval (CI) at 99%, with the estimated prevalence of dementia as 5%–8%.[1] The sample size of 240 was estimated with apparent precision of 0.1.

Tools

The data was collected on a comprehensive semi-structured proforma to record sociodemographic and clinical characteristics.

Structured Clinical Interview for DSM-5 Research Version (SCID-5-RV) is a semi-structured interview to help in the diagnosis of psychiatric disorders as per the DSM-5. It is used by clinicians or other trained mental health professionals. The Research Version (SCID-5-RV) is a comprehensive tool that generates current and lifetime disorders.[12]

Montreal Cognitive Assessment (MoCA) is a cognitive screening single-page test administered in about 10 minutes and scores 30 points. The MoCA contain 12 subsets that tap on 7 cognitive functions, i.e., visuospatial/executive function (trail-making test - 1, cube copy - 1, clock drawing - 3); attention and concentration (digit span test - 2, serial subtraction - 3, tapping - 1); language (naming - 3, repetition - 2, fluency - 1); memory (delayed recall - 5); abstraction – 2; and orientation - 6.[13] One score is added to MOCA’s total score, whose formal education was less than 12 years.

Neuropsychiatric Inventory Questionnaire (NPI-Q) is used to measure neuropsychiatric disturbances in dementia. However, it was originally devised for patients with Alzheimer’s disease. It is considered equally applicable to all types of dementia and examines 12 different symptoms. It is performed with the caregiver and thus reflects the caregivers experience of caring for the patient. Symptoms measured include delusions, hallucinations (visual, auditory, gustatory, or somatic), agitation, depression, anxiety, elation, apathy, disinhibition, irritability, aberrant motor behavior (e.g., pacing, rummaging, repetitive movements), sleep disturbances, and appetite or eating disorders.[14]

The Clinical Dementia Rating (CDR) is a global clinical scale with established diagnostic and severity-ranking utility. The scale evaluates the cognitive performance of the individual’s performance of everyday tasks, cognitive demands in one’s natural environment within his own cultural and ethnic context. It demonstrates good face validity and is generally more robust against the influence of cultural bias.[15] The severity of dementia is classified as mild (CDR 1), moderate (CDR 2), severe (CDR 3) and questionable (CDR 0.5).

Procedure

Patients were recruited after fulfilling the inclusion and exclusion criteria. The information and purpose of the study were explained in detail. Written, informed consent was taken from the patients or their nominated representatives. A psychiatrist conducted a detailed interview for diagnosing and assessing various behavioral and psychological symptoms. Subsequently, various scales were applied to evaluate the cognitive and neuropsychiatric symptoms. The data was collected from October 2019 to June 2020. However, due to the ongoing COVID-19 pandemic and reduced patient visits to the hospital, a few centers restricted the data collection up to March 2020.

Statistical analyses

Categorical variables are expressed as frequency and percentage. Continuous variables are expressed as mean ± standard deviation. In addition, Pearson correlation coefficient and linear regression analysis were performed to assess the strength of association. Statistical analysis was done using Statistical Package for the Social Sciences (SPSS) 24.0 for Windows.

RESULTS

Sample characteristics (sociodemographic and clinical)

A total of 292 participants were recruited for the study. The participants were predominantly recruited from government-run teaching institutions. The samples from respective centers were as follows: Ranchi (Central Institute of Psychiatry, Ranchi and Rajendra Institute of Medical Sciences; n = 82); Chandigarh (PGIMER; n = 75), Murshidabad (Murshidabad Medical College and Hospital; n = 74), Puducherry (JIPMER; n = 31), Bhubaneswar (Sum Hospital; n = 30). The sample was predominantly male (55.1%), married (69.9%), of Hindu religion (75%), in urban residence (43.8%), extended/joint family type (55.5%), and unemployed (32.5%). The mean age was 69.97 ± 6.84 years, and the mean years of education were 7.99 ± 5.43 years [Table 1]. The mean age of onset of dementia was 65.90 ± 6.67 years, and the duration of illness was 20.19 ± 11.74 months. 47.3% of the participants were provisionally diagnosed with Alzheimer’s dementia, 24.7% with vascular dementia and 23.3% with mixed dementia. 20.9% of patients were previously on cognitive enhancers and antipsychotics, but 22.94% were either drug-naïve or drug-free. Children of patients were the predominant caregivers (45.9%), followed by the spouse (37.7%). Awareness of dementia was good in 34.2%, and 21.9% of caregivers were unaware of dementia/BPS [Table 2].

Table 1

Sociodemographic characteristics of patients with the diagnosis of dementia

VARIABLES (N=292)	SUMMARY STATISTICS
Age in years (Mean±SD)	69.97±6.84
Gender n (n%)
Males	161 (55.1%)
Females	131 (44.9%)
Religion n (n%)
Hindu	219 (75%)
Muslim	35 (12%)
Sikh	25 (8.6%)
Christian	11 (3.8%)
Others	02 (0.7%)
Residence n (n%)
Rural	128 (43.8%)
Urban	117 (40.1%)
Sub urban	47 (16.1%)
Marital status n (n%)
Single	06 (2.1%)
Married	204 (69.9%)
Widow (er)	66 (22.6%)
Divorced	04 (1.4%)
Others	12 (4.1%)
Years of education (Mean±SD) in years	7.99±5.43
Monthly income (INR)
<2390	51 (17.5%)
2391-7101	48 (16.4%)
7102-11836	53 (18.2%)
11837-17755	34 (11.6%)
17756-23673	24 (8.2%)
23674-47347	44 (15.1%)
47348 and above	38 (13.0%)
Family type
Nuclear	115 (39.4%)
Extended/Joint	162 (55.5%)
Living alone	11 (3.8%)
Old age home	4 (1.4%)
Previous Profession
Professional	20 (6.8)
Semi-professional	26 (8.9%)
Clerical/Shop Owner/Farmer	44 (15.1%)
Skilled worker	26 (8.9%)
Semi-skilled worker	30 (10.3%)
Unskilled worker	51 (17.5%)
Unemployed	95 (32.5%)

Table 2

Clinical characteristics of patients with the diagnosis of dementia

VARIABLES (N=292)	SUMMARY STATISTICS
Age of onset of illness (Mean±SD) in years	65.90±6.67
Duration of illness (Mean±SD) in months (Range)	27.04±29.34 (3-156)
Duration of treatment (Mean±SD) in months	20.19±11.74
Diagnosis
Dementia in Alzheimer’s disease	138 (47.3%)
Vascular dementia	72 (24.7%)
Mixed dementia	68 (23.3%)
Other dementias	08 (2.73%)
Unspecified dementia	06 (2.1%)
Treatment
No treatment	67 (22.94%)
Cognitive enhancers (CE)	53 (18.2%)
Antipsychotics (AP)	30 (10.3%)
CE+AP	61 (20.9%)
Others	04 (1.4%)
Treatment Unknown	77 (26.4%)
Primary caregiver
Spouse	110 (37.7%)
Children	134 (45.9%)
Other informal caregivers	47 (16.1%)
None	01 (0.3%)
Awareness of dementia in the family
Good	100 (34.2%)
Satisfactory	114 (39.0%)
Poor	14 (4.8)
No response	64 (21.9%)

Behavioral and psychological symptoms and cognitive profile of dementia

The neuropsychiatric profile was performed based on the severity of the BPS. 43.5% of patients had mild symptoms of appetite/eating, 29.8% nighttime behavior, and 28.1% anxiety symptoms. 35.6% were having moderate levels of irritability/lability, 34.9% agitation/aggression, 29.8% dysphoria/depression and 28.8% apathy/indifference. Severe BPS were predominantly irritability/lability (23.6%), nighttime behavior (23.6%), appetite/eating (20.9%) and aberrant motor behavior (20.5%) [Table 3]. The mean MoCA score was 11.39 ± 6.08.

Table 3

Behavioral and psychological symptom severity profile in dementia patients (N=292)

Variables (n=292)	Absent n (n%)	Mild n (n%)	Moderate n (n%)	Severe n (n%)
Delusions	114 (49.3%)	45 (15.4%)	63 (21.6%)	40 (13.2%)
Hallucinations	162 (55.5%)	42 (14.4%)	47 (16.1%)	41 (14%)
Agitation/Aggression	86 (29.5%)	41 (14%)	102 (34.9%)	63 (21.6%)
Dysphoria/Depression	116 (39.7%)	54 (18.5%)	87 (29.8%)	35 (12%)
Anxiety	120 (41.4%)	82 (28.1%)	76 (26%)	14 (4.8%)
Euphoria/Elation	189 (64.7%)	26 (8.9%)	50 (17.1%)	27 (9.7%)
Apathy/Indifference	118 (40.4%)	29 (9.9%)	84 (28.8%)	61 (20.9%)
Disinhibition	156 (53.4%)	44 (15.1%)	62 (21.2%)	30 (10.3%)
Irritability/Lability	77 (26.4%)	42 (14.4%)	104 (35.6%)	69 (23.6%)
Aberrant Motor Behavior	123 (42.1%)	33 (11.3%)	76 (26%)	60 (20.5%)
Night-time Behaviour	87 (29.8%)	40 (29.8%)	96 (32.9%)	69 (23.6%)
Appetite/Eating	127 (43.5%)	127 (43.5%)	49 (16.8%)	61 (20.9%)

Bivariate correlation of severity of behavioral and psychological symptoms with the severity of dementia measured by CDR

Pearson correlation coefficient was applied to understand the relationship between BPS and dementia severity. Poor memory functions were positively correlated with delusion severity, dysphoria/depression, anxiety and appetite/eating. However, good memory functions were negatively correlated with the severity of irritability/lability. Poor orientation was positively correlated with the severity of delusions, dysphoria/depression, and euphoria/elation. Poor judgement and problem-solving were positively correlated with the severity of delusions, hallucinations, and euphoria/elation. Poor outdoor/social activities were positively correlated to the severity of delusions, hallucinations, dysphoria/depression, and agitation/aggression. A notably similar trend was noted in household hobbies and personal care. Overall, with the increase in the severity of dementia, there is an increment in the behavioral and psychological symptoms [Table 4].

Table 4

Bivariate correlation of severity of behavioral and psychological symptoms with the severity of dementia measured by CDR

	Del	Hal	Ag	Dys	Anx	Eup	Apa	Dis	Irr	MB	NB	App
Memory	0.251**	0.253**	0.079	0.153**	0.132*	0.128	-0.005	0.003	-0.126*	-0.044	0.006	0.116*
Orientation	0.146*	0.105	0.000	0.127*	-0.034	0.191**	0.038	-0.031	-0.089	-0.055	0.099	-0.079
Judgement and Problem solving	0.284**	0.164**	0.097	0.150*	0.137*	0.374**	-0.026	0.016	-0.069	-0.081	0.040	-0.083
Outer activities	0.216**	0.130*	0.128*	0.148*	0.101	0.026	0.068	0.060	-0.111	0.066	0.037	-0.046
Housework Hobbies	0.247**	0.212**	0.026	0.165**	0.053	-0.032	0.032	-0.054	0.196**	-0.048	-0.018	-0.086
Personal care	0.134*	0.099	0.147*	0.005	0.005	0.036	0.205**	0.175**	0.032	0.198**	0.136*	0.093

Significance level *P<0.05, **P<0.01; Del - Delusions, Hal - Hallucinations, Ag - Agitation/Aggression, Dys - Dysphoria/Depression, Anx - Anxiety, Eup - Euphoria/ Elation, Apa - Apathy/Indifference, Dis - Disinhibition, Irr-Irritability/Lability, MB-Aberrant Motor Behaviour, NB-Night-time Behaviour, App-Appetite/Eating

Multiple regression of behavioral and psychological symptoms (dependent factors) with cognitive scores as a predictor

A linear regression analysis was carried out to assess the strength of association between the dependent factors, i.e., behavioral and psychological symptoms and severity of dementia. As a result, a significant association was found between mild dementia and anxiety, irritability/lability and aberrant motor behavior, a trend level association between moderate dementia and disinhibition; agitation/aggression and severe dementia and delusions; euphoria/elation (trend level) and disinhibition [Table 5].

Table 5

Predictor role of global cognitive scores on behavioral and psychological symptoms in dementia

Dependent Factors	Unstandardized Beta	Sth. Error	t	P
Mild Dementia
Anxiety	0.029	0.014	2.110	0.037*
Irritability/Lability	0.036	0.017	2.090	0.038*
Aberrant Motor	0.038	0.019	1.979	0.050
Moderate Dementia
Disinhibition	0.026	0.013	1.966	0.052
Agitation/Aggression	0.036	0.020	1.863	0.065
Severe Dementia
Delusions	0.109	0.046	2.353	0.025*
Euphoria/Elation	0.056	0.030	1.882	0.069
Disinhibition	0.097	0.047	2.065	0.047*

Predictor factor: Total MoCA score; Significance level *P<0.05

DISCUSSION

The present study was a multicentric study on BPS of dementia under the egis of the Research, Education, and Training Foundation subcommittee of the Indian Psychiatric Society (IPS). The study recruited 292 participants from Jharkhand, Chandigarh, West Bengal, Puducherry and Odisha. The study reports the sociodemographic, clinical characteristics, cognitive profile, and BPS clusters based on various clinical diagnoses on hospital-based samples. Previous studies from India have reported considering either hospital-based samples or community samples.[51116] In a community-based study by Shaji et al.[17] the mean age of the sample who presented with Alzheimer’s dementia was 76.5 ± 9.1 and 81.7 ± 8.6 in vascular dementia. In another study by Pinto and Seethalakshmi,[16] the mean age of samples was 65.1 ± 12.25 years. The present study found the mean age to be 69.97 ± 6.84 years, similar to previous literature. The samples were predominantly male (55.1%), hailing from the rural background (43.8%) and belonged to joint/extended family (55.5%). First-degree relatives were mostly the informal caregivers (45.9%), similar to previous studies conducted in India.[18] These figures may reflect the sociodemographic background of the patients reaching the tertiary geriatric care settings across India.

Clinical characteristics of people with dementia

We found that the mean age of onset of dementia was 65.90 ± 6.67 years. The most common diagnosis was Alzheimer’s dementia (47.3%), followed by vascular (24.7%) and mixed dementia (23.3%). Previous studies from India have reported Alzheimer’s disease as the most common type (54%), followed by vascular dementia (39%).[19] The prevalence is also comparable to the global data. Data from developing countries suggest that age-adjusted dementia have prevalence estimates of 1%–3% in India and sub-Saharan Africa, and Alzheimer’s disease accounts for approximately 60%, and vascular dementia accounts for approximately 30% of the overall prevalence.[20] The study found a mean duration of dementia of around 20.19 ± 11.74 months. The patients visiting hospital settings for treatment may take a longer time since the onset, which may be due to the insidious onset of symptoms, slow progression of illness or new onset of BPS, which may indicate the caregivers for taking the required medical attention.

Behavioral and psychological symptoms of dementia

It is well known that BPS has a myriad of manifestations with varying frequency and severity. There were earlier attempts to understand and conceptualize the symptom clusters in a meaningful way.[21] Some of the clusters comprised the psychotic syndrome and affective syndrome and areas of abnormal motor behaviors, social interactions, speech, personality changes and somatic symptoms.[821] The present study found BPS predominantly in patients regardless of the diagnosis. The most common symptoms represented from our study include agitation/aggression (69.5%), nighttime behavior (69.2%) and dysphoria/depression (59.3%). Some form of psychosis was noted in at least 50% of patients visiting the hospital. Earlier, Shaji found a higher prevalence of BPS in patients with dementia, more commonly in Alzheimer’s dementia. In that study, 96.6% had at least one BPS, delusions being more prevalent, which accounted for 65.5%, activity disturbance in 65.5%, aggressive behaviors in 51.7% and hallucinations in 41.3%.

Moreover, delusions and hallucinations have been correlated with cognitive and functional decline in dementia patients.[22] The present study found some form of delusions in 51.7% and hallucinations in 44.5%, which is similar to previous literature.[5] Prior studies have also associated higher frequency and severity of sleep disturbances with depression, disinhibition, or aberrant motor behaviors.[23] We found nighttime behaviour in 69.2% and aberrant motor activity in 57.9%. Objective measures of nighttime behaviors showed that more prominent sleep disturbance was associated with impaired functioning in dementia.

Correlational analysis suggests that BPS has a close link to the severity of dementia in the cognitive domain and domains like orientation, judgment, problem-solving abilities, outdoor/social activities and personal care or, broadly, the individual’s functional status.[24] Furthermore, specific neuropsychiatric symptoms uniquely contribute to functional status, independent of age and specific measures of memory and executive function.[22] We found a significant positive correlation between dementia severity and BPS across daily living activities and functioning domains [Table 4]. Most importantly, impaired memory is positively correlated with depression/dysphoria, anxiety, delusions and hallucinations. This relationship is seemingly bidirectional and results in poorer outcomes in dementia.[22]

Predictor model of global cognitive scores on BPS in dementia

Results of the regression analysis suggest that some behaviors measured on the NPI-Q were significantly associated with the level of cognitive functioning measured on MoCA. Specifically, patients with mild dementia, having the most remarkable cognitive status, were most likely to exhibit a variety of BPS, including anxiety, aberrant motor activity and irritability/lability. The association between excessive motor activity (wandering and restlessness) and the severity of cognitive impairment has been demonstrated.[25] The present study suggests that aberrant motor activity is also seen in milder stages of dementia. Still, the severity of symptoms may vary due to many factors that may interplay in the symptom presentation. We found that disinhibition and agitation/aggression significantly associated with cognitive dysfunction in moderate dementia. Disinhibition[26] and aggression[27] are commonly associated with a grossly dysfunctional brain with the predominant influence of the response inhibition that may be playing a role in dementia.[28]

Regarding severe dementia, delusions, euphoria/elation and disinhibition were significantly associated with poorer cognitive performance in severe dementia. These “psychotic clusters” have been previously reported.[29] Psychotic symptoms may occur throughout the course, most commonly at the moderate-to-severe level of dementia (CDR 2 and 3).[30] However, the presence of psychosis is alarming and would lead to significant cognitive decline in the disease.[21]

CONCLUSION, LIMITATIONS AND FUTURE DIRECTIONS

The present study provides insights into various behavioral and psychological symptom clusters in dementia, considering the severity of dementia and cognitive dysfunction. Some BPS were commonly hinted at as a common cluster, whereas others were unique. However, the factors could not be pinpointed to the severity of the condition, largely due to inter-individual variability. Therefore, it may be essential to decide if symptoms should be studied in groups or individually.[21] Importantly, clinicians need to consider each symptom meticulously during the clinical work-up and be aware of the interrelations between them when assessing patients. There were some limitations to the study. We did not use recent criteria for diagnosing different types of dementias. Being cross-sectional, the study was unable to provide the duration of various BPS longitudinally and how these symptoms had a bearing on the individual’s cognitive status. Future studies may emphasize the longitudinal study design to comprehensively understand the BPS and assess their effect on daily life, social functions, cognitive functions, and overall activities. Also, future studies may emphasize the impact on caregivers due to the disease and BPS arising out of the same.

Financial support and sponsorship

This study was supported by the Indian Psychiatric Society. However, the sponsor had no role in the study designing and reporting of the findings.

Conflicts of interest

There are no conflicts of interest.