Neuropsychiatric Symptoms in Patients with Neurocognitive Disorder and Their Performance Between Mild and Major Stages

Abstract

Background:

The neuropsychiatric symptoms (NPS) in patients with neurocognitive disorders (NCD) increases the risk of exhibiting significant cognitive and functional decline. However, to the best of our knowledge, few studies have evaluated to what extent the presence of chronic and early NPS impacts cognition and functionality in patients with minor or major stages of NCD.

Objective:

We aimed to assess the interplay between early and chronic NPS and cognitive and functional presentation of patients with mild and major forms of NCD.

Methods:

We used two NPS tools tracking early and late NPS and assessed to what extent they determine cognitive and functional outcomes in patients with mild and major forms of NCD.

Results:

We found an inverse relationship between the presence of NPS, as measured by the Neuropsychiatric Inventory and Mild Behavioral Impairment Checklist (MBI-C), and cognitive and functional variables in major forms of NCD. In contrast, the minor stage of NCD was associated with increased MBI-C scores.

Conclusion:

Our results revealed that NPS are associated with cognitive and functional outcomes in mild and chronic forms of NCD. Crucially our results suggest that NPS could be considered as a pathological marker of the clinical course of dementia. Additionally, our study calls to study early and late forms of NPS as both impact cognition and functionality of NCD.

Keywords

Assessment of cognitive disorders/dementia behavioral disturbances neurocognitive disorder neurodegeneration neuropsychiatric symptoms

INTRODUCTION

Neurocognitive disorders (NCD) comprise a group of intractable diseases, including Alzheimer’s disease (AD), frontotemporal degeneration (FTD), Lewy body dementia, and NCD due to vascular disease, among others. NCD are considered progressive diseases characterized by multiple impairments in cognitive domains, physical functionality, social interactions, and occupational autonomy [1, 2]. Crucially, AD patients exhibit cognitive impairments affecting memory, language skills, executive functions, visual-constructional capacity, and instrumental capacities [3]. In contrast, FTD patients exhibit a group of behavioral disturbances, impairments in executive functioning, motor alterations, and interoceptive deficits [4, 5]. Patients with other types of NCD, including patients with vascular disease, Lewy body dementia, or Parkinson’s disease dementia, tend to exhibit a combination of cognitive, affective, behavioral, as well as motor alterations [6 –8].

Additionally, patients with major NCD, including the abovementioned etiologies, exhibit a group of neuropsychiatric symptoms (NPS), which are defined as symptoms affecting mood, thought, and behavior dementia [9 –11]. The NPS may predict a cognitive and functional decline [12 –14]. Besides, they are associated with higher rates of institutionalization, poorer quality of life, fast disease progression, higher mortality risk, and a more significant burden for caregivers [15 –18].

The patients with mild forms of NCD exhibit early clinical stages and course with cognitive decline that does not affect dramatically the daily life functionality and they are not even apparent in routine appointments [19, 20]. Previous studies suggested that NPS are highly prevalent even in mild forms of NCD and indicate greater cognitive and functional deterioration. The NPS could appear even before cognitive impairment [2, 21]. Therefore, an early detection of this symptomatology impacts the evolution of this disorder positively and favoring clinical interventions [15]. Although there is consistent evidence of the impact of NPS in patients with major forms of NCD, the study chronic and early NPS and its relationship with NCD at different stages [17 , 22–24] is still lacking.

Here we conducted an analytical cross-sectional study of a 552-patient cohort (408 major NCD and 144 mild NCD) recruited across a year (2018) in a center of memory and cognition in Bogota, Colombia. We aimed to evaluate the association between chronic and early forms of NPS with cognition performance and functionality in patients with different stages of neurocognitive disorder (major and mild NCD). To this end, we used two different instruments used to track chronic and early NPS including the Neuropsychiatric Inventory Questionnaire (NPI-Q) which tracks majorly chronic behavioral changes [25] and the Mild Behavioral Impairment Checklist (MBI-C), developed to track early and early forms of NPS in patients with NCDs [18].

Considering previous studies of NPS in major NCD [2, 26], we anticipated worse cognitive function and deteriorated instrumental functionality in patients with major forms of NCD that exhibit larger scores of chronic NPS. Moreover, considering that the behavioral impairments would start since mild stages of NCD [10, 11], we anticipate the presence of higher scores of early NPS compared than chronic NPS in patients with minor stages of NCD. Considering mild NCD patients did not course with greater functionality impairments, we did not expect a significant association between NPS and functionality measures.

METHODS

Population and clinical assessment

Our population consisted of patients who attended the Intellectus Memory and Cognition Center in Hospital Universitario San Ignacio, Bogotá, Colombia, between January 1 and December 31, 2018. In the analysis, we included those patients who meet the condition of major NCD or mild NCD (552 subjects). Patients were evaluated using a standardized interdisciplinary clinical assessment that includes geriatricians, psychiatrists, neurologists, and neuropsychologists. In geriatrics, the baseline situation of the patient was evaluated, including multimorbidity, functionality, and social status. The psychiatrist applied a semi-structured interview to determine the presence of NPS and ruled out their existence due to primary psychiatrist illness. However, we ruled out behavioral alterations attributable to another mental disorder such as depressive or anxiety disorders. The neurologist assessed comorbidities and risk factors of other neuropsychiatric conditions. Neuropsychologists measured cognitive domains such as memory, executive functions, attention, visuo-constructional function, language, and social cognition. Furthermore, an interdisciplinary meeting was held to reach a consensus diagnosis and come up with appropriate management suggestions [27].

Materials

The MBI-C assesses mild cognitive impairment and predicts dementia risk [18]. This scale arose from behavioral symptoms manifested in the early stages of FTD behavioral variant, for which the construct mild behavioral impairment (MBI) was proposed [17]. The MBI criteria and states that the symptoms must be present for at least six months. Studies demonstrate it is a predictor of other types of dementia and helps detect cognitive, neuropsychiatric, and functional symptoms. It comprises five domains (interest, affective symptoms, disinhibition, social norms, and delusions/hallucinations). These domains evaluate the presence and severity of symptoms (1 = mild, 2 = moderate, 3 = severe) [13 , 28]. Other studies use this scale on individuals without cognitive impairments or mild cognitive impairments [17, 18].

The NPI-Q was designated in 1994. It assesses chronic neuropsychiatric symptoms in people with dementia. It comprises twelve domains (delusions, hallucinations, aggressiveness, depression, anxiety, euphoria, apathy, disinhibition, irritability, motor disturbance, nocturnal behavior, and appetite), which evaluate the presence and severity of the symptoms (1 = mild, 2 = moderate, 3 = severe). The maximum score is 36 points, and it is validated in different languages [25]. This tool is precise and reliable for monitoring behavioral alterations in different neuropsychiatric and neurocognitive disorders [17 , 29–31].

The Barthel assesses physical activity [32]. It considers the self-care capacity of people with neuromuscular disorders. It demonstrates a strong focus on the leisure activity of the lower extremities. Today it is used to detect dependence on daily life basic activities. This easy application scale includes ten activities of everyday life. The maximum score is 100 points and is also helpful for patient longitudinal follow-up [32].

The Mini-Mental State Examination (MMSE) is used to assess cognitive domains [33]. It possesses good psychometric characteristics, with 88.3% for sensibility and 87% for specificity [34]. It evaluates cognitive domains, such as memory, language, executive function, and visuo-constructional capacity in people with neurocognitive disorder [33, 34].

The Montreal Cognitive Assessment (MoCA) is useful to detect cognitive impairment [35]. It has a cut-off point of 26 point, and consists of 19 questions, which assess eight different domains (executive skills, naming, memory, attention, language, abstraction, delayed memory, and orientation). It has a sensitivity greater than 87% and a specificity of 90% [36, 37].

The INECO Frontal Screening (IFS) is a neuropsychological test that allows the detection of executive impairment [38, 39]. It is composed of 7 domains: response inhibition and control over task change, the ability to respond appropriately to conflicting instructions, the ability to abstract, through the interpretation of proverbs, and working memory [40]. Other studies use this tool with patients diagnosed with different types of neurocognitive disorder [38, 39].

For severity measure, we used the Clinical Dementia Rating (CDR) that involves a semi-structured interview, which is conducted based on the patient, caregiver, and clinical judgment [41].

Statistical analysis

We conducted a descriptive analysis of each variable to explore the population distribution and the participants’ characteristics. The assessed variables were age and sex, Barthel score (physical functionality), MMSE, MoCA, and IFS (cognitive), CDR (severity), and NPI and MBI-C (neuropsychiatric). With this information, we carried out an exploratory analysis to identify missing data and outliers.

According to our objective, we carried out a stratified analysis based on the NCD severity established by clinical criteria (major and mild). To define the distribution of the variables, we evaluate normality using the Kolmogorov-Smirnov (Lilliefors’ correction) test because the sample size was greater than 50 [42]. Given that the variables did not have a normal distribution and presented an ordinal measurement scale (except for age), we used the Wilcoxon Sum of ranks and signs test (student’s t-test for age) to assess the mean differences. We found statistically significant differences between the two groups. Subsequently, we evaluate possible collinearity between the variables evaluated using Spearman’s correlation based on the distribution of the variables, and since it is less sensitive to outliers [43].

To evaluate the association between neuropsychiatric symptoms (with age as a control variable) and cognitive and functional measures in patients with a major and mild NCD. Then, we performed a multiple linear regression model for each clinical group consequent analysis of variance and effect size using ETA2 coefficient. We did not find a significant interaction between the variables. Possible confounding variables were sex and neurocognitive disorder etiology. For this reason, we formed distinct classifications based on such confounding variables. We then also created an independent model for men and another for women in both clinical groups.

Regarding etiology, we stratified in AD, FTD (behavioral and semantic variant), or others. The “other” category corresponded to vascular dementia, Parkinson’s disease, or dementia with Lewy bodies, which are less prevalent. In those cases, we only ran the model for the major NCD.

We made different additive risk models using logistic regression to estimate the association of neuropsychiatric, cognitive, and functional variables with major and mild NCD. In these models, the presence of major and mild NCD was considered dependent variables; as independent variables, we followed a hierarchical model in which different groups were included according to the established model, added to sex and age as control variables. All analysis and graphics were made using R studio [44].

RESULTS

We evaluated a total of 552 patients. They had a mean age of 71.6 years (SD: 13.21). Table 1 demonstrates the population’s descriptive parameters. Table 2 compares the parameters between the people with major and mild NCD. For neuropsychiatric, age, and severity variables, there was a higher median in the group with major cognitive disorder (p < 0.001). The cognitive and functional variables were higher in the mild population (p < 0.001).

Table 1

Population characteristics

	Median (IR)	25% ile	75% ile	90% ile
n	552
Demographics - Functionality
Age^*	71.6 (13.2)	66	80	85
Barthel	95 (15)	85	100	100
Cognitive
MMSE	24 (8)	19	27	29
MoCA	16 (10)	11	21	25
IFS	11 (12)	6	18	22.5
Severity
CDR	1 (1.5)	0,5	2	3
Neuropsychiatric
NPI	6 (6.75)	3	9.75	14
MBI-C	13 (19)	5	24	36.5

* mean (SD); CDR, Clinical Dementia Rating; MoCA, Montreal Cognitive Assessment; MMSE, Mini-Mental State Examination; IFS, INECO Frontal Screening.

Table 2

Population characteristics stratified by neurocognitive disorder stage

	Major NCD median (IR)	Mild NCD median (IR)	p
n	408	144
Demographics - Functionality
Age^*	76.91 (8.85)	70.51 (12.99)	< 0.001^**
Barthel	90 (25)	100 (5)	< 0.001
Cognitive
MMSE	20 (8)	25 (4)	< 0.001
MoCA	12 (8)	18 (6)	< 0.001
IFS	6.5 (9)	11 (13)	< 0.001
Severity
CDR	2 (1)	0.5 (0.5)	< 0.001
Neuropsychiatric
NPI	7 (7)	5 (4)	< 0.001
MBI-C	15 (21.25)	11 (12.25)	< 0.001

^*mean (SD); ^**student’s t test; NCD, neurocognitive disorder; CDR, Clinical Dementia Rating; MoCA, Montreal Cognitive Assessment; MMSE, Mini-Mental State Examination; IFS, INECO Frontal Screening.

The correlation between neuropsychiatric variables and the cognition and functionality variables demonstrated low correlation coefficients, which indicates a low risk of collinearity (Supplementary Table 1). There were no significant interactions between the variables when proceeding with the linear regression analysis in the population with major NCD. We found negative estimates in cognitive and functionality measurements, as neuropsychiatric variables increase in most cases (MoCA: R² = 0.045, p < 0.001; MMSE: R² = 0.077, p < 0.001; IFS: R² =0.082, p < 0.001; Barthel: R² = 0.07, p < 0.001). In the mild cognitive disorder population, models report similar trends (MoCA: R² = 0.035, p = 0.11; MMSE: R² = 0.043, p = 0.059; IFS: R² = 0.143, p < 0.001; Barthel: R² = 0.048, p = 0.041). In addition to the variability provided in each group, among those with major cognitive disorder, there was a negative association between MBI-C and NPI and the other four variables, while among those with mild cognitive disorder, there was a difference in the association between NPI and MBI-C and IFS (β_NPI=–0.286, p < 0.001; β_MBI - C=0.048, p = 0.309) (Table 3, Fig. 1). On the other hand, we found weak effect sizes, except for the NPI variable in the model that considered MMSE as a dependent variable and the model that considered Barthel as a dependent variable in the mild NCD population. Furthermore, we found a greater effect of the NPI variable among the cognitive and functional variables (Supplementary Table 2).

Table 3

Linear regression between neuropsychiatric symptoms and cognitive/functionality variables stratified by neurocognitive disorder stage

	Dependent variable
	MoCA		MMSE		IFS		Barthel
	R²	p	R²	p	R²	p	R²	p
Major NCD	0.045	< 0.001	0.077	< 0.001	0.082	< 0.001	0.069	< 0.001
Mild NCD	0.035	0.11	0.043	0.05	0.143	< 0.001	0.048	0.04

NCD, neurocognitive disorder; MoCA, Montreal Cognitive Assessment; MMSE, Mini-Mental State Examination; IFS, INECO Frontal Screening.

Fig. 1

Linear regression models between neuropsychiatric and cognitive/functionality variables. A) Linear regression models between neuropsychiatric and Barthel index. B) Linear regression models between neuropsychiatric and INECO Frontal Screening (IFS). C) Linear regression models between neuropsychiatric and Mini-Mental State Examination (MMSE). D) Linear regression models between neuropsychiatric and Montreal Cognitive Assessment (MoCA).

When controlling for sex, we found similar results. In some cases, we also found distinct dissociations in the β between NPI and MBI-C: in the men major cognitive disorder category: NPI-Q with MoCA and IFS, in the women major cognitive disorder category: NPI-Q with IFS, and in the women mild cognitive disorder category: MBI-C with MoCA, MMSE, and IFS. None of these dissociations were statistically significant. When we controlled for etiology, the inverse tendencies are preserved except for NPI-Q and MoCA, MMSE, IFS, and Barthel in AD; NPI-Q with IFS and Barthel, MBI-C with MoCA and MMSE in FTD. Neither of these results was statistically significant.

The results of the logit models also show the association between both major and mild NCD and the cognitive and functional variables. Within the combined models in the major cognitive disorder group, only the models that included both neuropsychiatric variables or just Barthel (together with sex and age as control variables) as independent variables were statistically significant except by sex. These models allow us to better describe the association between NPI-Q and MBI-C and major cognitive disorder. In the case of NPI-Q, it demonstrated a more significant association with this outcome because for each unit increase, this resulted in a 12% greater chance of presenting major cognitive disorder (OR:1.12, 95% CI: 1.09–1.16, p < 0.001). In the second model, which considered the MBI-C result, we can see a 4% greater probability of presenting major cognitive disorder for each unit increase (OR:1.04, 95% CI: 1.03–1.06, p < 0.001). Although the magnitude effect changes in the other models, both neuropsychiatric variables increase the probability of presenting major cognitive disorder (Table 4).

Table 4

Odds ratio to present major neurocognitive disorder depending on neuropsychiatric symptoms

	Major NCD				Mild NCD
	NPI-Q		MBI-C		NPI-Q		MBI-C
	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
Age	1.08	1.07–1.10^*	1.09	1.07–1.11^*	0.98	0.96–0.99^*	0.98	0.97–0.99^*
Sex	1.24	0.89–1.72	1.26	0.92–1.75	0.78	0.55–1.13	0.79	0.56–1.21
NPS	1.12	1.09–1.16^*	1.04	1.03–1.05^	0.90	0.87–0.94^	0.98	0.96–0.99^*

^*p < 0.05; NCD, neurocognitive disorder; NPS, neuropsychiatric symptoms; OR, odds ratio; CI, confidence interval.

Regarding the mild cognitive disorder group, we observe that an increase in both the MBI-C and NPI scores decreases the probability of presenting this outcome. For MBI-C there is a probability decrease of 2% (OR:0.98, 95% CI: 0.96–0.99, p < 0.001), while for NPI-Q it there is a decrease of 10% (OR:0.90, 95% CI: 0.87–0.94, p < 0.001). When comparing other models, in addition to changing the effect magnitude in both variables, the MBI-C scale presents a greater probability of presenting mild NCD in the models which calculated neuropsychiatric variables (95% CI: –0.01–0.02), neuropsychiatric and cognitive variables (95% CI: –0.01–0.03), and neuropsychiatric and functional variables (95% CI: –0.01–0.03).

DISCUSSION

In the present study, we evaluated the association of chronic and early NPS within different stages (major and mild) and etiologies of NCDs. We also explored the role NPS plays in determining cognitive and functional outcomes in patients. Our results indicated a particular pattern of association between NPS and major cognitive disorder versus NPS and mild cognitive disorder. Patients with higher NPS exhibited lower cognitive functioning and lower functionality in major NCD but also in patients with mild forms of NCD. These results remained significant when stratifying for sex and etiology. These results provide new evidence because they revealed the importance of evaluating NPS at an early stage. To the best of our knowledge, to date few studies have simultaneously studied the role of early and chronic NPS at different neurocognitive disorder stages, where they do not evaluate its association with functionality, or focus on studying the applicability of the scales [28, 45]. Additionally, our results allow clinical attention to focus more on NPS at different stages of neurocognitive disorders. This study could pave the way for new research on the evaluation of new treatment options.

The linear regression models demonstrated that patients with higher NPS presented lower functionality and cognitive performance in the major NCD group. The variability of the outcomes was minor or moderate without showing a significant difference between both scales. That pattern of results resembles previous findings [2, 46] where the NPS seems to affect the progression of cognitive and functional compromise in each type of neurocognitive disorder. These results add new evidence by revealing the progress of the neurodegenerative disease is affected by NPS and indicates its association with other variables depending on whether the symptoms studied were acute or chronic. The association also persists in studies with larger sample sizes, as in this article. When evaluating the association of cognitive, functional, and NPS variables with major cognitive disorder, we found that the logit models, which only included NPS (MBI-C and NPI-Q independently), were statistically significant. We found that the presence of chronic symptoms (evaluated with NPI-Q) contributed an additional 12% odds (OR: 1.12, 95% CI: 1.09–1.16, p < 0.001) of presenting major NCD compared to early symptoms (evaluated with MBI-C) that represented a 4% increase (OR: 1.04, 95% CI: 1.03–1.06, p < 0.001) of presenting this outcome. Similar results found in the literature show this association between cognitive and functional impairment with progression to major cognitive disorder [27], as well as the presence of depressive symptoms with impaired cognition [14, 47].

Regarding the models evaluating the population with mild cognitive disorder, we evidenced similar findings: each point increased in the NPS scale the cognition and functionality score decreased. This phenomenon also occurs in Crees et al.’s study [48] in which the patients with higher MBI-C scores have a greater cognitive impairment, particularly in working memory and attention, in patients who have not developed dementia. Furthermore, when categorizing by sex, only MBI-C is statistically significant for the association with mild NCD. When evaluating the association of the cognitive, functional, and NPS variables with the presence of mild NCD, a difference was evident with the scales by demonstrating a higher post-test probability when the MBI-C was positive concerning the NPI, since the risk of presenting the outcome decreased by 2% (OR: 0.98, 95% CI: 0.96–0.99, p < 0.001) versus 10% (OR: 0.90, 95% CI: 0.87–0.94, p < 0.001) respectively (taking into account that the major and mild cognitive disorders are exclusive outcomes). These results coincide with previous studies that shows a significant association between cognitive impairment and a higher score in the NPS tools, mainly with delusions, hallucinations, agitation, apathy, and alterations in motor behavior [49].

Concerning this, NPS has been identified as risk factors for cognitive alterations, recognizing that people with these alterations can present up to 4 times higher probability of major cognitive disorder, and more than half develop dementia (cumulative conversion of 10%, 29%, 45.5%, 50%, and 56.3% at one, two, three, four, and five years respectively) [15, 50]. This progression is higher in those patients who present MBI added or without MCI [51]. Regarding the tools used, we found a greater association of MBI-C than NPI-Q with mild cognitive disorder. These results indicate a possible greater capacity of MBI-C on NPI-Q to detect cognitive and functional alterations in the presence of NPS. This result exposes the MBI-C as a valid tool to identify behavioral alterations in mild stages [15, 17]. Also, it is in line with another study [52], which demonstrates that the MBI-C had high reliability and validity to assess cognitive performance in AD. However, more longitudinal studies should be carried out to support these findings. Based on this, with our present results, we add to evidence supporting the use of MBI-C because it had not been comparing previously with NPI-Q in patients with the cognitive disorder [18]. Even though the association of major cognitive disorder and NPS had been studied using the NPI-Q, Yesavage, and Cornell tools [2, 21]. This relationship had not been evaluated so far with tools that capture NPS from early or chronic stages. However, the MBI-C instrument is recent, and few studies assessed the relationship with the mild cognitive disorder [28, 51]. Also, the effectiveness of capturing both tools simultaneously had not been evaluated yet.

Our findings suggest 1) both MBI-C and NPI-Q are valuable tools for assessing NPS in major and mild neurocognitive disorders; 2) identifying NPS could increase the probability of detecting cognitive and functional changes in patients with mild stages of neurocognitive disorder; 3) the importance of evaluating NPS through different methodologies. In our study, MBI-C seemed to be more sensitive for detecting cognitive and functional changes in mild cognitive disorder when comparing OR. Overall, our results underscore the importance of studying NPS from the mild stages of neurocognitive disorder and reveals new opportunities for future research on these changes in people with early cognitive changes without the mild cognitive disorder. Furthermore, our results strengthen a path for the early detection of NPS, creating an opportunity for research on early NPS and the evaluation of behavioral changes. It could have an impact on the progression or appearance of neurocognitive disorder. This agrees with what is suggested by Taragano et al. [15], who reveals MBI-C as a valid tool for identifying behavioral alterations in mild stages of dementia.

This study presents strengths based on the evaluation of NPS at different stages of neurocognitive disorders. It also evaluates the association of cognitive and functionality measures, which are usually direct markers of neurocognitive disorders. Additionally, it is one of the few studies which has investigated these issues. This could, consequently, improve the prognosis of a population with access difficulties and allows them to receive specialized care. However, our study also demonstrates some limitations. For example, our current design does not allow us to infer causality nor temporality, indicating the need for future research with longitudinal designs to confirm our results. The study evaluates behavioral changes with measures used for regular and early follow-up. It is necessary to carry out new studies evaluating which symptoms are associated with greater functional and cognitive differences. Although our power analyzes demonstrate that the sample is adequate, when stratifications are made, the study’s power decreases.

Consequently, this does not allow us to identify possible differences between the groups. Future studies should balance the samples and include a greater sample size. Besides, tools evaluated have different domains or symptoms; however, the results obtained in this work show apparent differences. Future studies should assess the association between symptoms and discriminated domains and the outcomes studied. On the other hand, the evaluations carried out by MBI-C and NPI-Q may be subject to self-reporting bias from the patients’ companions who completed the scale and may have magnified some symptoms. Finally, this research did not consider some variables such as comorbidities, psychiatrist and not psychiatrist, nor pharmacological treatment, that could be effect modifying variables; then, following studies should include this clinical information to enhance results’ analysis.

Conclusion

The results of this work support the association described between NPS and NCD. It also highlights the importance of measuring NPS using different methodologies and at different stages of the disease. Our study suggests a possible pathological marker of great significance for the diagnosis, course, and management of dementias by encompassing early manifestations. It also indicates a potential synergy in the use of the NPI-Q and MBI-C tools, with the opportunity to study treatment for early NPS and evaluate whether control of these behavioral changes could impact the progression or appearance of neurocognitive disorder.

Footnotes

ACKNOWLEDGMENTS

We wish to thank the patients and families who kindly contributed to this study. Also, to the working team at the Intellectus Memory and Cognition Center.

This work was partially supported by grants from the Centro de Memoria y Cognition Intellectus, Hospital Universitario San Ignacio and Javeriana University.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Arvanitakis

, Shah

, Bennett

(2019) Diagnosis and managementof dementia: Review. JAMA 322, 1589–1599.

Santacruz Escudero

, Beltrán

, Palacios

, Chimbí

, Matallana

, Reyes

, Perez-Sola

, Santamaría-García

(2019) Neuropsychiatric symptoms as predictors of clinical coursein neurodegeneration. A longitudinal study. Front Aging Neurosci 11, 176.

Okazaki

, Kasai

, Meguro

, Yamaguchi

, Ishii

(2009) Disturbances in everyday life activities and sequence disabilitiesin tool use for Alzheimer disease and vascular dementia. Cogn Behav Neurol 22, 215–221.

Santamaría-García

, Reyes

, García

, Baéz

, Martinez

, Santacruz

, Slachevsky

, Sigman

, Matallana

, Ibañez

(2016) First symptoms and neurocognitive correlates ofbehavioral variant frontotemporal dementia . J Alzheimers Dis 54, 957–970.

Medscape, Frontotemporal dementia and frontotemporal lobar degeneration: Overview, etiology, genetic distribution and variation, https://emedicine.medscape.com/article/1135164-overview.

Schapira

AHV

, Chaudhuri

, Jenner

(2017) Non-motor features ofParkinson disease. Nat Rev Neurosci 18, 435–450.

Robinson

, Jorge

(2016) Post-stroke depression: A review. Am J Psychiatry 173, 221–231.

Bruun

, Rhodius-Meester

HFM

, Koikkalainen

, Baroni

, Gjerum

, Lemstra

, Barkhof

, Remes

, Urhemaa

, Tolonen

, Rueckert

, van Gils

, Frederiksen

, Waldemar

, Scheltens

, Mecocci

, Soininen

, Lötjönen

, Hasselbalch

, van der Flier

(2018) Evaluating combinations of diagnostic tests to discriminatedifferent dementia types. Alzheimers Dement 10, 509–518.

Duong

, Patel

, Chang

(2017) Dementia: What pharmacist need toknow. Can Pharm J 150, 118–129.

10.

Alzheimer’s Disease International,World Health Organization (2015) DEMENTIA. A Public Health Priority.

11.

International Psychogeriatric Association (2015) The IPA complete guides to behavioral and psychological symptoms of dementia.

12.

Taragano

, Allegri

, Krupitzki

, Sarasola

, Serrano

, Loñ

, Lyketsos

(2009) Mild behavioral impairment and risk ofdementia: A prospective cohort study of 358 patients. J ClinPsychiatry 70, 584–92.

13.

Forlenza

, Diniz

, Stella

, Teixeira

, Gattaz

(2013) Mildcognitive impairment (part 1): Clinical characteristics andpredictors of dementia. Braz J Psychiatry 35, 178–185.

14.

Chan

, Soldan

, Pettigrew

, Wang

M-C

, Wang

, Albert

, Rosenberg

, BIOCARD Research Team (2019) Depressive symptoms inrelation to clinical symptom onset of mild cognitive impairment. Int Psychogeriatr 31, 561–569.

15.

Taragano

, Allegri

, Heisecke

, Martelli

, Feldman

, Sánchez

, García

, Tufro

, Castro

, Leguizamón

, Guelar

, Ruotolo

, Zegarra

, Dillon

(2018) Risk ofconversion to dementia in a mild behavioral impairment groupcompared to a psychiatric group and to a mild cognitive impairmentgroup. J Alzheimers Dis 62, 227–238.

16.

Peters

, Rosenberg

, Steinberg

, Norton

, Welsh-Bohmer

, Hayden

, Breitner

, Tschanz

, Lyketsos

(2013) Neuropsychiatric symptoms as risk factors for progression from CINDto dementia: The Cache County Study . Am J Geriatr Psychiatry 21, 1116–1124.

17.

Ismail

, Smith

, Geda

, Sultzer

, Brodaty

, Smith

, Agüera-Ortiz

, Sweet

, Miller

, Lyketsos

; ISTAARTNeuropsychiatric Symptoms Professional Interest Area (2016) Neuropsychiatric symptoms as early manifestations of emergentdementia: Provisional diagnostic criteria for mild behavioralimpairment. Alzheimers Dement 12, 195–202.

18.

Ismail

, Agüera-Ortiz

, Brodaty

, Cieslak

, Cummings

, Fischer

, Gauthier

, Geda

, Herrmann

, Kanji

, Lanctôt

, Miller

, Mortby

, Onyike

, Rosenberg

, Smith

, Sultzer

, Lyketsos

, NPS Professional Interest Area of theInternational Society of to Advance Alzheimer’s Research andTreatment (NPS-PIA of ISTAART) (2017) The Mild Behavioral ImpairmentChecklist (MBI-C): A rating scale for neuropsychiatric symptoms inpre-dementia populations. J Alzheimers Dis 56, 929–938.

19.

Boise

, Camicioli

, Morgan

, Rose

, Congleton

(1999) Diagnosing dementia: Perspectives of primary care physicians. Gerontologist 39, 457–464.

20.

Shao

, Zeng

, Chen

, Shutes-David

, Thielke

, Tsuang

(2019) Detection of probable dementia cases in undiagnosed patientsusing structured and unstructured electronic health records. BMC Med Inform Decis Mak 19, 1–11.

21.

Lyketsos

, Lopez

, Jones

, Fitzpatrick

, Breitner

, Dekosky

(2002) Prevalence of neuropsychiatric symptoms in dementia and mildcognitive impairment: Results from the cardiovascular health study. JAMA 288, 1475–1483.

22.

Mortby

, Lyketsos

, Geda

, Ismail

(2018) Special Issue onmild behavioral impairment and non-cognitive prodromes to dementia. Int Psychogeriatr 30, 167–169.

23.

Canevelli

, Blasimme

, Vanacore

, Bruno

, Cesari

(2016) Mildbehavioral impairment: Ethical, methodological and clinicalreflections. Neurosci Biobehav Rev 69, 402–403.

24.

Elefante

, Lattanzi

, Ismail

, Medda

, Bacciardi

, Mainardi

, Perugi

(2019) Mild Behavioral Impairment: Presentation of thediagnostic criteria and the Italian version of the MBI-Checklist. Riv Psichiatr 54, 59–66.

25.

Cummings

, Mega

, Gray

, Rosenberg-Thompson

, Carusi

, Gornbein

(1994) The neuropsychiatric inventory: Comprehensiveassessment of psychopathology in dementia. Neurology 44, 2308–2314.

26.

Cerejeira

, Lagarto

, Mukaetova-Ladinska

(2012) Behavioral andpsychological symptoms of dementia. Front Neurol 3, 1–21.

27.

Chimbí-Arias

, Santacruz-Escudero

, Chavarro-Carvajal

, Samper-Ternent

, Santamaría-García

(2020) Alteracionesdel comportamiento de pacientes con diagnóstico de trastornoneurocognoscitivo en Bogotá (Colombia). Rev ColombPsiquiatr 49, 136–141.

28.

Mallo

, Ismail

, Pereiro

, Facal

, Lojo-Seoane

, Campos-Magdaleno

, Juncos-Rabadán

(2018) Assessing mildbehavioral impairment with the mild behavioral impairment-checklistin people with mild cognitive impairment . J Alzheimers Dis 66, 83–95.

29.

Ismail

, Emeremni

, Houck

, Mazumdar

, Rosen

, Rajji

, Pollock

, Mulsant

(2013) A comparison of the E-BEHAVE-AD, NBRS,and NPI in quantifying clinical improvement in the treatment of agitation and psychosis associated with dementia. Am J Geriatr Psychiatry 21, 78–87.

30.

Lai

CKY

(2014) The merits and problems of Neuropsychiatric Inventoryas an assessment tool in people with dementia and other neurologicaldisorders. Clin Interv Aging 9, 1051–1061.

31.

Nowrangi

, Lyketsos

, Rosenberg

(2015) Principles and management of neuropsychiatric symptoms in Alzheimer’s dementia. Alzheimers Res Ther 7, 12.

32.

Mahoney

, Barthel

(1965) Functional evaluation: The Barthelindex. Md State Med J 14, 61–5.

33.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental”. Apractical method for grading the cognitive state of patients for theclinician. J Psychiatr Res 12, 189–98.

34.

Creavin

, Wisniewski

, Noel-Storr

, Trevelyan

, Hampton

, Rayment

, Thom

, Nash

KJE

, Elhamoui

, Milligan

, Patel

, Tsivos

D V

, Wing

, Phillips

, Kellman

, Shackleton

, Singleton

, Neale

, Watton

, Cullum

(2016) Mini-Mental State Examination (MMSE) for the detection of dementia in clinically unevaluated people aged 65 and over in community and primary care populations. Cochrane Database Syst Rev 1, CD011145.

35.

MoCA - Cognitive assessment, https://www.mocatest.org/.

36.

Nasreddine

, Phillips

, Bédirian

, Charbonneau

, Whitehead

, Collin

, Cummings

, Chertkow

(2005) The Montreal CognitiveAssessment, MoCA: A brief screening tool for mild cognitiveimpairment. J Am Geriatr Soc 53, 695–699.

37.

Pedraza

, Salazar

, Soler

, Castro

, Castillo

, Piñeros

(2016) Confiabilidad, validez de criterio y discriminante delMontreal Cognitive Assessment (MoCA) test, en un grupo de adultos deBogotá. Acta Med Colomb 41, 221–28.

38.

Bahia

, Cecchini

, Cassimiro

, Viana

, Lima-Silva

, De Souza

, Carvalho

, Guimarães

, Caramelli

, Balthazar

MLF

, Damasceno

, Brucki

SMD

, Nitrini

, Yassuda

(2018) The accuracyof INECO frontal screening in the diagnosis of executive dysfunctionin frontotemporal dementia and Alzheimer disease. Alzheimer Dis Assoc Disord 32, 314–319.

39.

Moreira

, Costa

, MacHado

, Castro

, Lima

, Vicente

(2019) Distinguishing mild cognitive impairment from healthy agingand Alzheimer’s disease: The contribution of the INECO Frontal Screening (IFS). PLoS One 14, e0221873.

40.

Torralva

, Roca

, Gleichgerrcht

, López

, Manes

(2009) INECO Frontal Screening (IFS): A brief, sensitive, and specific toolto assess executive functions in dementia. J Int Neuropsychol Soc 15, 777–786.

41.

Khan

(2016) Clinical diagnosis of Alzheimer’s disease. In Biomarkers in Alzheimer’s Disease, Academic Press, pp. 27–48.

42.

Pedrosa

, Juarros-Basterretxea

, Robles-Fernández

, Basteiro

, García-Cueto

(2015) Pruebas de bondad de ajuste endistribuciones simétricas, ¿qué estadístico utilizar? Univ Psychol 14, 245–254.

43.

Uriza Carraszo

, Rodríguez Malagón

(2015) Estudiossobre correlación y concordancia. In Epidemiología clínica: Investigación clínica aplicada Editorial Médica Panamericana, Bogotá D.C., pp. 268–280.

44.

R Core Team (2020) R: A Language and Environment for Statistical Computing.

45.

Geda

, Roberts

, Mielke

, Knopman

, Christianson

TJH

, Pankratz

, Boeve

, Sochor

, Tangalos

, Petersen

, Rocca

(2014) Baseline neuropsychiatric symptoms and the risk of incidentmild cognitive impairment: A population-based study. Am JPsychiatry 171, 572–581.

46.

Amanzio

, Palermo

, Zucca

, Rosato

, Rubino

, Leotta

, Bartoli

, Rainero

(2018) Neuropsychological correlates of instrumentalactivities of daily living in neurocognitive disorders: A possiblerole for executive dysfunction and mood changes. IntPsychogeriatr 30, 1871–1881.

47.

Feldman

, Scheltens

, Scarpini

, Hermann

, Mesenbrink

, Mancione

, Tekin

, Lane

, Ferris

(2004) Behavioral symptoms inmild cognitive impairment. Neurology 62, 1199–201.

48.

Creese

, Brooker

, Ismail

, Wesnes

, Hampshire

, Khan

, Megalogeni

, Corbett

, Aarsland

, Ballard

(2019) Mildbehavioral impairment as a marker of cognitive decline incognitively normal older adults . Am J Geriatr Psychiatry 27, 823–834.

49.

Vik-Mo

, Giil

, Ballard

, Aarsland

(2018) Course ofneuropsychiatric symptoms in dementia: 5-year longitudinal study. Int J Geriatr Psychiatry 33, 1361–1369.

50.

Henao-Arboleda

, Moreno-Carrillo

, Ramos

, Aguirre-Acevedo

, Pineda

, Lopera

(2010) Caracterización de síntomasneuropsiquiátricos en pacientes con DCL de tipo amnésico enuna población colombiana . Rev Chilena Neuropsicol 5, 153–59.

51.

Matsuoka

, Ismail

, Narumoto

(2019) Prevalence of mildbehavioral impairment and risk of dementia in a psychiatricoutpatient clinic. J Alzheimers Dis 70, 503–511.

52.

Cui

, Dai

, Miao

, Zhong

, Liu

, Jing

, Bai

, Kong

, Sun

, Li

, Guo

, Rosa-Neto

, Gauthier

, Wu

(2019) Reliabilityand validity of the Chinese version of the mild behavioralimpairment checklist for screening for Alzheimer’s disease. JAlzheimers Dis 70, 747–756.