Hong Kong Brief Cognitive Test for identifying symptomatic Alzheimer's disease and other types of dementia

Abstract

Background

The Hong Kong Brief Cognitive Test (HKBC) has demonstrated high discriminative ability for patients with cognitive impairment in both Cantonese- and Mandarin-speaking populations.

Objective

To evaluate the diagnostic efficacy of the HKBC in identifying dementia and mild cognitive impairment (MCI) due to Alzheimer's disease (AD) and other common types of dementia.

Methods

Sixty-one patients with dementia due to AD, 30 patients with MCI due to AD, 47 patients with subcortical ischemic vascular dementia (SIVD), 50 patients with frontotemporal lobar degeneration (FTLD), 17 patients with Lewy body dementia (LBD), and 37 cognitively unimpaired controls (CUCs) were recruited and completed the HKBC, the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA). The diagnostic performance of each test was analyzed via receiver operating characteristic curve analysis. Impairment in cognitive domains on the HKBC was analyzed in patients with symptomatic AD.

Results

Scores of the HKBC, MMSE and MoCA were significantly lower in patients with all types of dementia, AD (dementia and MCI), and non-AD dementia (SIVD, FTLD, and LBD) than in CUCs. The most appropriate cutoff scores of the HKBC were 24 for identifying AD and LBD, 22 for identifying SIVD and FTLD, and 26 for identifying MCI due to AD from CUCs. HKBC memory and language scores were significantly lower in patients with MCI due to AD than in CUCs.

Conclusions

This study demonstrated that the HKBC could efficiently identify patients with common types of dementia and was sensitive in screening early AD.

Keywords

Alzheimer's disease dementia frontotemporal lobar degeneration Hong Kong Brief Cognitive Test Lewy body dementia mild cognitive impairment subcortical ischemic vascular dementia

Introduction

As the population ages,^1,2 the prevalence of cognitive impairment and dementia among elderly individuals has increased significantly.^3,4 Globally, approximately 50 million people are living with dementia, a number that is expected to exceed 150 million by 2050.⁵ Early diagnosis of dementia can facilitate social support, advanced care planning, and potential disease-modifying therapy. However, a large portion of individuals with dementia remain undiagnosed or unevaluated, especially in developing countries.⁶

Alzheimer's disease (AD) is the most common type of dementia in elderly individuals, accounting for 50% to 70% of all cases. Although positron emission tomography (PET) imaging and cerebrospinal fluid biomarker analysis are integral components of the early diagnostic algorithm for AD, the widespread adoption of these modalities is hindered by their high costs, limited accessibility, and technical complexities. Moreover, other common types of dementia, such as vascular dementia, dementia with Lewy bodies (DLB), and frontotemporal lobar degeneration (FTLD), still lack specific biomarkers for diagnosis or screening.

Currently, cognitive testing with neuropsychological scales is still a crucial tool in the early detection of AD and related dementia,⁷ with advantages of convenience, noninvasiveness, and cost-effectiveness. The Mini-Mental State Examination (MMSE)⁸ and the Montreal Cognitive Assessment (MoCA)⁹ are the most utilized cognitive screening tools worldwide, with pooled sensitivities of 81% and 94% and specificities of 89% and 60%, respectively, in distinguishing patients with dementia from cognitively unimpaired controls (CUCs).^10,11

However, the MMSE is not sufficiently sensitive to detect subtle and early cognitive changes, such as mild cognitive impairment (MCI), or for identifying cognitive impairment caused by subcortical injury or frontal dysfunction, because most of its items are relatively easy and are designed to evaluate memory and language functions. For example, in a meta-analysis, the MMSE showed a sensitivity of 80% and 71% and a specificity of 81% and 85% for MCI¹⁰ and vascular dementia,¹² respectively. In contrast, the MoCA can be used to evaluate a broader range of cognitive domains, aiding in the identification of MCI¹³ and cognitive dysfunction across various disorders beyond AD, such as cerebrovascular diseases¹⁴ and Parkinson's disease.¹⁵ The MoCA has sensitivity values of 90% and 91% and specificity values of 78% and 80% for vascular dementia and DLB, respectively, with cutoff scores of 22 in Chinese populations.^16,17 In addition, the MoCA showed high sensitivities of 100% and 83.3% for detecting amnestic MCI (aMCI) and multidomain MCI, respectively, at an optimal cutoff score of 24, although the specificity was low (50.0% to 52.0%).⁹

Furthermore, both the MMSE and the MoCA are influenced by cultural and educational backgrounds, necessitating different cutoff scores for varying educational levels. For example, previous studies established the diagnostic cutoff points of the MMSE and MoCA for dementia as 16 and 13 for illiterate individuals, 19 for those with an elementary school education, and 23 and 24 for individuals with higher education levels in Chinese populations.^18,19

The Hong Kong Brief Cognitive Test (HKBC), which is a culturally tailored neuropsychological test that covers various cognitive domains designed to better suit Chinese populations and elderly individuals with lower educational levels, was introduced by an expert panel from the Department of Psychiatry at the Chinese University of Hong Kong in 2018.^9,20 The HKBC has been found to slightly outperform the MMSE and the MoCA as screening tools for neurocognitive disorders (NCDs) in Cantonese-speaking populations.²¹ In a Mandarin-speaking population, the HKBC was subsequently shown in our previous study to more accurately identify clinically diagnosed patients with aMCI than the MMSE, with a stable cutoff score across various age and education groups.²²

In this study, we aimed to explore the diagnostic value of the HKBC for patients with symptomatic AD at both the dementia and MCI stages and whose diagnosis was supported by amyloid PET and in patients with non-AD dementia, including subcortical ischemic vascular dementia (SIVD), FTLD, and Lewy body dementia (LBD). We further investigated the changes in the cognitive domains of the HKBC in patients with symptomatic AD.

Methods

Participants

Ninety-one patients with symptomatic AD, including 61 with dementia and 30 with MCI, 114 patients with non-AD dementia including 47 with SIVD, 50 with FTLD, and 17 with LBD, and 37 CUCs were consecutively recruited from the Cognitive Impairment Clinic (Memory Clinic) at Tianjin Medical University General Hospital. All patients underwent comprehensive evaluations, including medical history, physical and neurological examinations, clinical laboratory tests (e.g., thyroid function, vitamin B12, folate, and syphilis serology), brain magnetic resonance imaging (MRI), and neuropsychological assessments. The study was approved by the Ethics Committee of Tianjin Medical University General Hospital, and written informed consent was obtained from all participants or their legal guardians.

Diagnoses were made by dementia specialists according to specific diagnostic criteria for each condition. Patients with dementia and MCI met the criteria for major NCD and mild NCD, respectively, according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, with Clinical Dementia Rating (CDR) scores of 1–2 and 0.5, respectively. Patients with symptomatic AD met the International Working Group-2 criteria for typical AD with a positive amyloid PET result.²³ Patients with SIVD were diagnosed on the basis of the Vascular Cognitive Disorders diagnostic criteria²⁴ and presented with multiple subcortical lacunes or moderate-to-severe white matter hyperintensities but without significant focal atrophy (e.g., the medial temporal lobe) on MRI.²⁵ Patients with FTLD were diagnosed according to the revised diagnostic criteria for the behavioral variant of frontotemporal dementia⁹ or primary progressive aphasia for nonfluent variants and semantic variants.¹⁰ For patients with LBD, diagnoses were based on the revised criteria for DLB²⁶ or the clinical criteria for probable Parkinson's disease dementia (PDD) established by the Movement Disorder Society.²⁷ CUCs were recruited from community-dwelling volunteers without memory or cognitive complaints, whose objective neuropsychological assessment results were within normal ranges, with a CDR score of 0.

All the participants were aged between 50 and 90 years, possessed adequate vision and hearing, and were able to communicate verbally. Subjects with a history of other mental or neurological disorders (e.g., major depression, schizophrenia, epilepsy, traumatic brain injury, substance abuse, and alcoholism) or any other conditions that might affect neuropsychological performance were excluded.

Cognitive testing

All participants were assessed with the Chinese versions of the HKBC, MMSE, and MoCA in a random order (approximately one-third of patients were first administered the HKBC, one-third were first administered the MMSE, and one-third were first administered the MoCA), with 10-min intervals between each assessment to ensure consistency in testing conditions. These tests were administered by two experienced investigators who were uniformly trained, had passed consistency tests, and were blinded to the clinical groupings. To prevent any influence on orientation questions related to time and place, scoring information was withheld from participants during the tests and intervals.²²

The HKBC was administered according to previously described procedures for approximately 7–10 min.²¹ Consistent with the MMSE and the MoCA, lower HKBC scores indicate poorer cognitive function, with a maximum score of 30 points. Z scores of five cognitive domains on the HKBC were converted using the mean and standard deviation of the CUC group for further analyses, including learning and memory (immediate recall/attention, delayed recall, recent memory, and general knowledge), orientation, frontal executive functions (frontal lobe function and executive function), visuospatial construction, and language.

Statistical analysis

All analyses were performed via R version 4.3.3, with the significance level set at α = 0.05. Continuous variables are presented as mean ± standard deviation, whereas categorical data are expressed as numbers and percentages. One-way analysis of variance (ANOVA) was used to compare age and years of formal education among different diagnostic groups, including the SIVD, FTLD, LBD, AD dementia, MCI, and CUC groups. Pearson's χ² tests were used to compare sex distributions among the groups. Analysis of covariance (ANCOVA) was performed to compare scores on the HKBC, MMSE, and MoCA between groups, with age, sex, and education as covariates. Post hoc analysis was performed via false discovery rate (FDR) correction,²⁸ and the differences between the CUC group and each patient group are presented. Additionally, differences in the Z scores of various cognitive domains on the HKBC among the AD dementia groups, the MCI group, and the CUC group were analyzed via two-sample t tests.

Receiver operating characteristic (ROC) curves were generated to assess the diagnostic accuracy of the HKBC, MMSE, and MoCA in distinguishing patients with symptomatic AD and other types of dementia from CUCs. Optimal cutoff scores were determined using the Youden Index (the maximum sum of sensitivity and specificity minus 1) derived from the ROC curves. The area under the curve (AUC) values of HKBC, MMSE and MoCA were compared using the nonparametric approach presented by DeLong et al.,²⁹ and post hoc analysis was performed with FDR correction.

Results

Demographic and clinical characteristics of all participants

The demographic and clinical characteristics of all participants are presented in Table 1. There were differences in age, sex distribution, and educational level between the groups. Compared with those in the CUC group, the proportion of males was generally lower in the MCI due to AD group (p < 0.05), and the educational level was lower in all dementia groups (all p < 0.05).

Table 1.

Demographic and clinical characteristics of all participants.

	AD dementia (N = 61)	SIVD (N = 47)	FTLD (N = 50)	LBD (N = 17)	MCI due to AD (N = 30)	CUC (N = 37)	p
Age, y	66.66 (7.69)^d	70.00 (7.19)	66.86 (7.24)^d	72.71 (5.92)^a,c	69.70 (5.61)	68.38 (6.31)	<0.001
Sex, male (%)	23 (37.7)^b	35 (74.5)^a,c,e	22 (44.0)^b	12 (70.6)^e	6 (20.0)^b,d	17 (45.9)^b	0.05
Education, y	11.51 (3.94)	11.40 (2.84)	11.34 (2.99)	10.06 (3.11)^e	12.63 (3.19)^d	14.32 (3.61)^a,b,c,d	<0.001

Variables are presented as the mean (SD) except sex, as specifically indicated. Statistical differences in age and education were analyzed via ANOVA, and sex distributions were analyzed via Pearson's χ2 tests between groups. AD: Alzheimer's disease; MCI: mild cognitive impairment; SIVD: subcortical ischemic vascular dementia; FTLD: frontotemporal lobar degeneration; LBD: Lewy body dementia; CUC: cognitively unimpaired control.

p < 0.05 versus AD dementia.

p < 0.05 versus SIVD.

p < 0.05 versus FTLD.

p < 0.05 versus LBD.

p < 0.05 versus MCI due to AD.

Performance of the three cognitive tests in all groups

The pairwise comparisons of the mean HKBC, MMSE, and MoCA scores among all groups are presented in Table 2. After adjusting for age, sex, and education level, the HKBC, MMSE, and MoCA score were lower in patients with all types of dementia, AD dementia, non-AD dementia, and MCI due to AD than in CUCs (all p < 0.05). The pairwise comparisons of the mean HKBC, MMSE, and MoCA scores among non-AD dementia groups (including SIVD, FTLD and LBD) and the CUC group are presented in Supplemental Table 1.

Table 2.

Scores of the three cognitive tests.

	All dementia (N = 175)	AD dementia (N = 61)	non-AD dementia (N = 114)	MCI due to AD (N = 30)	CUC (N = 37)	p
HKBC	14.63 (5.85)^d	14.26 (5.23)^d	14.82 (6.18)^d	21.57 (4.38)^a,b,c	27.97 (1.82)^a,b,c,d	<0.001
MMSE	18.98 (5.69)^d	18.26 (5.67)^d	19.36 (5.69)^d	25.87 (2.70)^a,b,c	28.41 (1.24)^a,b,c,d	<0.001
MoCA	13.90 (5.75)^d	13.28 (5.65)^d	14.23 (5.80)^d	20.57 (2.99)^a,b,c	27.27 (2.04)^a,b,c,d	<0.001

The variables are presented as the mean (SD). Statistical differences between groups were analyzed via ANCOVA, with age, sex, and education as covariates. Post hoc analysis was performed via false discovery rate correction. AD: Alzheimer's disease; MCI: mild cognitive impairment; CUC: cognitively unimpaired control; HKBC: Hong Kong Brief Cognitive Test; MMSE: Mini-Mental State Examination; MoCA: Montreal Cognitive Assessment Scale.

p < 0.05 versus dementia.

p < 0.05 versus AD dementia.

p < 0.05 versus non-AD dementia.

p < 0.05 versus MCI due to AD.

Discriminative validity of cognitive tests for various types of dementia and MCI

The optimal cutoff scores, sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), Youden Index, and AUC for the HKBC, MMSE, and MoCA in discriminating patients with various types of dementia and patients with MCI due to AD from CUCs are summarized in Table 3. The HKBC had an AUC of 0.98, with a sensitivity of 88% and a specificity of 100% at an optimal cutoff of 22, for identifying patients with all types of dementia (Figure 1A). For patients with AD dementia, the HKBC had an AUC of 1.00 with a sensitivity of 98% and a specificity of 97% at an optimal cutoff score of 24 when compared with CUCs (Figure 1B). For patients with non-AD dementia, the HKBC had an AUC of 0.97 with a sensitivity of 87% and a specificity of 100% at an optimal cutoff score of 22 when compared with patients with CUCs (Figure 1C). For patients with MCI due to AD, the HKBC had an AUC of 0.92, with a sensitivity of 87% and a specificity of 86% at an optimal cutoff of 26 when compared with CUCs (Figure 1D).

Figure 1.

ROC curves of the HKBC, MMSE, and MoCA for diagnosing patients with various types of dementia and MCI due to AD. According to the ROC curve, the HKBC showed similar discriminative validity for distinguishing patients with all types of dementia (A), patients with AD dementia (B), patients with non-AD dementia (C), and patients with MCI due to AD (D) from CUCs compared with the MMSE and the MoCA. HKBC: Hong Kong Brief Cognitive Test; MMSE: Mini-Mental State Examination; MoCA: Montreal Cognitive Assessment; CUCs: cognitively unimpaired controls; AD: Alzheimer's disease; MCI: mild cognitive impairment; ROC: receiver operating characteristic.

Table 3.

Diagnostic performance of the HKBC, MMSE and MoCA for dementia and MCI.

	AUC	Cutoff	SEN	SPE	PPV	NPV	Youden Index
Differentiating people with all types of dementia from CUCs
HKBC	0.98	22	0.88	1.00	1.00	0.64	0.88
MMSE	0.95	26	0.87	1.00	1.00	0.62	0.87
MoCA	0.99	22	0.90	0.97	0.99	0.68	0.88
Differentiating people with AD dementia from CUCs
HKBC	1.00	24	0.98	0.97	0.98	0.97	0.96
MMSE	0.98	26	0.92	1.00	1.00	0.88	0.92
MoCA	1.00	22	0.97	0.97	0.98	0.95	0.94
Differentiating people with MCI due to AD from CUCs
HKBC	0.92	26	0.87	0.86	0.84	0.89	0.73
MMSE	0.78	26	0.43	1.00	1.00	0.69	0.43
MoCA	0.97	26	0.97	0.84	0.83	0.97	0.80
Differentiating people with non-AD dementia from CUCs
HKBC	0.97	22	0.87	1.00	1.00	0.71	0.87
MMSE	0.94	26	0.84	1.00	1.00	0.67	0.84
MoCA	0.98	24	0.92	0.92	0.97	0.79	0.84
Differentiating people with SIVD from CUCs
HKBC	0.93	22	0.72	1.00	1.00	0.74	0.72
MMSE	0.89	26	0.77	1.00	1.00	0.77	0.77
MoCA	0.96	24	0.87	0.92	0.93	0.85	0.79
Differentiating people with FTLD from CUCs
HKBC	1.00	22	0.98	1.00	1.00	0.97	0.98
MMSE	0.97	26	0.88	1.00	1.00	0.86	0.88
MoCA	0.99	24	0.94	0.92	0.94	0.92	0.86
Differentiating people with LBD from CUCs
HKBC	1.00	24	0.94	0.97	0.94	0.98	0.91
MMSE	0.96	24	0.94	1.00	1.00	0.97	0.94
MoCA	1.00	22	1.00	0.97	0.94	1.00	0.97

AUC: area under the curve; SEN: sensitivity; SPE: specificity; NPV: negative predictive value; PPV: positive predictive value; HKBC: Hong Kong Brief Cognitive Test; MMSE: Mini-Mental State Examination; AD: Alzheimer's disease; MCI: mild cognitive impairment; SIVD: subcortical ischemic vascular dementia; FTLD: frontotemporal lobar degeneration; LBD: Lewy body dementia; CUCs: cognitively unimpaired controls.

The optimal cutoff score of the MMSE was 26 for discriminating all types of dementia (sensitivity 87%, specificity 100%, AUC 0.95, Figure 1A), AD dementia (sensitivity 92%, specificity 100%, AUC 0.98, Figure 1B), non-AD dementia (sensitivity 84%, specificity 1.00, AUC 0.94, Figure 1C), and MCI due to AD (sensitivity 0.43, specificity 1.00, AUC 0.78, Figure 1D) from CUCs. The MoCA had an optimal cutoff of 22 for all types of dementia (sensitivity 90%, specificity 97%, AUC 0.99, Figure 1A) and AD dementia (sensitivity 97%, specificity 97%, AUC 1.00, Figure 1B), 24 for non-AD dementia (sensitivity 92%, specificity 92%, AUC 0.98, Figure 1C), and 26 for MCI due to AD (sensitivity 97%, specificity 84%, AUC 0.97, Figure 1D) when compared with CUC. No significant differences in AUCs were observed between each test for any analysis (p > 0.05, Supplemental Table 2).

The discriminative validity values of the three cognitive tests for SIVD, FTLD and LBD are summarized in Supplemental Figure 1.

Cognitive domains of the HKBC across patients with symptomatic AD

Five cognitive domains of the HKBC, including learning and memory (12/30), orientation (5/30), frontal executive functions (6/30), visuospatial function (3/30), and language (4/30), were analyzed (Table 4,Figure 2). Significant reductions were observed in learning and memory (−2.62 versus 0, p < 0.05) and language (−1.97 versus 0, p < 0.05) in the MCI due to AD group compared with the CUC group. The AD dementia group presented lower scores in all domains compared with the CUC group and lower scores in learning and memory (−4.82 versus −2.62, p < 0.05), orientation (−4.92 versus −0.79, p < 0.05), frontal executive functions (−1.93 versus −0.31, p < 0.05), visuospatial construction (−3.84 versus −0.79, p < 0.05), and language (−3.86 versus −1.97, p < 0.05) compared with the MCI due to AD group.

Figure 2.

Comparison of different cognitive domains on the HKBC between patients with symptomatic AD and CUCs. Significantly lower Z-scores were observed in domains of memory and language in the MCI due to AD group than in the CUC group and in domains of memory, orientation, frontal executive function, visuospatial, and language in the AD dementia group than in the MCI due to AD group. Compared with the CUC group, the AD dementia group had lower scores in all domains. MCI: mild cognitive impairment; AD: Alzheimer's disease; CUCs: cognitively unimpaired controls. *p < 0.05; **p < 0.01; ***p < 0.001.

Table 4.

Comparison of cognitive domains on the HKBC between symptomatic AD patients and CUCs.

	CUC(N = 37)	MCI due to AD(N = 30)	AD dementia(N = 61)
Learning and memory	0 (1)^b	−2.62 (1.88)^a	−4.82 (1.42)^a,b
Orientation	0 (1)	−0.79 (1.98)	−4.92 (3.84)^a,b
Frontal executive functions	0 (1)	−0.31 (1.05)	−1.93 (3.61)^a,b
Visuospatial construction	0 (1)	−0.79 (1.98)	−3.84 (3.04)^a,b
Language	0 (1)^b	−1.97 (1.96)^a	−3.86 (1.71)^a,b

Z scores for each cognitive domain are presented as the means (SDs), which were calculated from raw scores in reference to the means and SDs of all the CUCs. Statistical differences between two groups were analyzed via t tests.

AD: Alzheimer's disease; MCI: mild cognitive impairment; CUCs: cognitively unimpaired controls.

p < 0.05 versus CUC.

p < 0.05 versus MCI due to AD.

Discussion

This study demonstrated the diagnostic ability of the HKBC and established cutoffs for symptomatic AD and other common types of dementia in Mandarin-speaking elderly individuals. In particular, the HKBC demonstrated excellent diagnostic efficacy in discriminating patients with dementia due to AD and MCI due to AD from CUCs at a cutoff of 24 and 26, respectively. The early impaired cognitive domains on the HKBC were memory and language in patients with MCI due to AD, and almost all domains were impaired at dementia stages, indicating the disease monitoring value of the HKBC for symptomatic AD.

This is the first study to validate the diagnostic value of the HKBC for symptomatic AD patients with amyloid PET-supported diagnosis and for other common types of dementia. For patients with symptomatic AD, the performance of the HKBC was comparable to those of the MMSE and MoCA (AUC: 1.00 versus 0.98 versus 1.00) for identifying dementia but was similar to that of the MoCA and better than that of the MMSE (AUC: 0.92 versus 0.97 versus 0.78) for identifying MCI. Both the original study and our previous study reported that the HKBC score had an AUC of 0.915 at cutoffs of 21 and 23 for identifying mild NCD and aMCI, respectively.^21,22 The present study demonstrated the diagnostic ability of the HKBC in biologically diagnosed AD patients and suggested its application for screening AD patients at both the dementia and MCI stages in the Chinese population.

Moreover, the HKBC test showed good diagnostic ability for discriminating patients with non-AD dementia, such as SIVD (AUC: 0.93 versus 0.89 versus 0.96), FTLD (AUC: 1.00 versus 0.97 versus 0.99), and LBD (AUC: 1.00 versus 0.96 versus 1.00) from CUCs, comparable to both the MMSE and MoCA. Although both tests could discriminate patients with vascular dementia from cognitively healthy adults in a previous study, the MoCA had a greater AUC than did the MMSE (0.95 versus 0.86) at cutoff values of 17 and 26, respectively.³⁰ Similarly, for identifying FTLD, the diagnostic accuracy of the MoCA was greater than that of the MMSE (AUC: 0.93 versus 0.77).³¹ In our study, the three tests showed similar discriminative validity in various types of dementia, although it seems that the MoCA and the HKBC were slightly better than the MMSE. We included dementia patients at mild to moderate stages, at which symptoms are usually obvious, leading to perfect performance of cognitive testing. In addition, only behavioral variant FTD (bvFTD) patients were included in the above study; however, a large proportion of primary progressive aphasia (PPA) patients (55.6%), e.g., nfvPAA and svPPA patients, were included in our study. Since the MMSE includes several items for testing language, such as naming, repetition, comprehension, reading and writing (8 points), it may be sensitive for identifying patients with PPA.

Our findings suggested that the optimal cutoff value of the HKBC for differentiating dementia patients from control individuals was 22, which was slightly higher than the cutoff of 21 for major neurodegenerative disorders reported in the original study.²¹ Compared with our study, the cohort included in that study had a greater mean age and a lower proportion of participants with seven or more years of education in the dementia group (79.4 versus 68.2 years, 22.7% versus 91.6%) and the control group (75.4 versus 68.8, 43.4% versus 100%).²¹ Additionally, the etiology or cause of cognitive impairment for participants may have differed between the studies, since we specifically included different types of dementia, such as SIVD, FTLD and LBD, in addition to AD. All these factors should contribute to the small variations in cutoff values of the HKBC.

The HKBC test showed excellent diagnostic accuracies for differentiating dementia (AUC of 0.98) and MCI due to AD (AUC of 0.92) from CUCs, consistent with previous studies (AUC of 0.96 for major and mild NCD and 0.92 for mild NCD, and AUC of 0.92 for aMCI).²¹^,²² In contrast, although the MMSE and MoCA also identified patients with dementia, their diagnostic accuracies varied across previous studies (e.g., the AUC of MMSE ranging from 0.85 to 0.99 and the AUC of MoCA ranging from 0.89 to 0.97),^10,11,32 suggesting potential influence of demographics and cultural background on these two tests.

A meta-analysis of a few preliminary studies of the HKBC concluded that the HKBC had the highest validity and reliability in detecting early cognitive decline among various screening tools, such as the MMSE, MoCA and CDT.³² The verbal delayed recall test and verbal fluency test are considered two of the most sensitive cognitive screening tools for early AD; these tools are included in several screening tests, including the HKBC and the MoCA (Beijing version), accounting for 4 points and 3 points, respectively.^21,26 Although free delayed recall is included in all three screening tests, cued delayed recall, which reflects the ability to use semantic learning strategies in learning and memory performance, has been demonstrated to be impaired in aMCI patients²⁷ and has been proposed as an early indicator of AD, is only scored in the HKBC.³³ These explanations are also supported by our cognitive domain analysis for the HKBC. Although all the cognitive domains had lower scores in patients with dementia due to AD than in CUCs, memory and language were the only two domains with significantly lower scores in patients with MCI due to AD.

Several limitations should be considered. First, the participants in this study were recruited from one memory clinic in China, and sample size was small. Consequently, CUCs could not be matched with patients at a 1:1 ratio. The discriminative ability of the HKBC and its cutoff scores need further validation in a larger population with diverse demographic and clinical features and in communities. Furthermore, demographic characteristics were not completely matched between all groups. For example, patients in the LBD group had significantly fewer years of education than those in the other groups, which may have influenced the cut-off values of cognitive tests for the identification of LBD patients. Second, biomarker measurements were not conducted to exclude pathology or comorbidity of AD in patients with other types of dementia. In particular, SIVD and DLB sometimes show mixed pathological changes (such as amyloid-β plaques and tau neurofibrillary tangles) as observed in AD patients in previous studies. Finally, patients with FTLD and LBD usually seek medical services when symptoms are obvious. Therefore, the diagnostic ability was undoubtedly quite good for all screening tests. Moreover, we only identified the diagnostic value of the HKBC in MCI patients caused by AD. Its potential use in screening the prodromal stages (e.g., MCI or mild behavioral impairment) of other types of dementia is worthy of further investigation, given that the HKBC includes comprehensive assessments of cognition, including attention, information processing and executive function.

In conclusion, in the present study, we demonstrated the diagnostic value of the HKBC for various types of dementia in a Mandarin-speaking population. Moreover, the HKBC is sensitive for patients in the early stages of AD, with significantly reduced scores in the memory and language domains.

Supplemental Material

sj-docx-1-alz-10.1177_13872877251405903 - Supplemental material for Hong Kong Brief Cognitive Test for identifying symptomatic Alzheimer's disease and other types of dementia

Supplemental material, sj-docx-1-alz-10.1177_13872877251405903 for Hong Kong Brief Cognitive Test for identifying symptomatic Alzheimer's disease and other types of dementia by Yuetao Hu, Xiuli Lu, Zeming Han, Yudi Shi, Yidi Wang, Qingzheng Lu, Yu Wang, Yong He, Yunyao Lu, Feifan Chen, Huifeng Chen and Nan Zhang in Journal of Alzheimer's Disease

Footnotes

Acknowledgements

We thank all the participants who agreed to participate in this study and all the members involved in the selection and assessment.

ORCID iDs

Xiuli Lu

Zeming Han

Yudi Shi

Yunyao Lu

Nan Zhang

Ethical considerations

The institutional ethics committee rigorously evaluated and authorized all research protocols (Ethics Approval Code: IRB2020-YX-161-01).

Consent to participate

Written informed consent was obtained from each participant following a detailed explanation of the study procedures.

Consent for publication

Not applicable

Author contribution(s)

Yuetao Hu: Conceptualization; Data curation; Funding acquisition; Software; Supervision; Validation; Visualization; Writing – original draft; Writing – review & editing.

Xiuli Lu: Data curation; Writing – review & editing.

Zeming Han: Validation; Visualization; Writing – original draft; Writing – review & editing.

Yudi Shi: Data curation; Writing – original draft; Writing – review & editing.

Yidi Wang: Data curation; Writing – review & editing.

Qingzheng Lu: Data curation; Writing – review & editing.

Yu Wang: Data curation; Writing – review & editing.

Yong He: Data curation; Writing – review & editing.

Yunyao Lu: Data curation; Writing – review & editing.

Feifan Chen: Conceptualization; Data curation; Writing – original draft; Writing – review & editing.

Huifeng Chen: Conceptualization; Data curation; Writing – original draft; Writing – review & editing.

Nan Zhang: Conceptualization; Data curation; Funding acquisition; Methodology; Resources; Supervision; Validation; Writing – original draft; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Science and Technology Innovation 2030—Major Project (grant number 2021ZD0201805), the Tianjin Key Medical Discipline (Specialty) Construction Project (grant number TJYXZDXK-004A), Tianjin Natural Science Foundation (24JCYBJC00650) and Tianjin Public Health Science and Technology Major Project (24ZXGZSY00060).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data is not publicly available due to privacy or ethical restrictions. The data has not been previously presented orally or by poster at scientific meetings.

Supplemental material

Supplemental material for this article is available online.

References

Integrated Care for Older People: Guidelines on Community-Level Interventions to Manage Declines in Intrinsic Capacity. Geneva: World Health Organization, http://www.ncbi.nlm.nih.gov/books/NBK488250/ (2017, accessed 28 July 2024).

Sáez de Asteasu

Martínez-Velilla

Zambom-Ferraresi

, et al. Role of physical exercise on cognitive function in healthy older adults: a systematic review of randomized clinical trials. Ageing Res Rev 2017; 37: 117–134.

Langa

Levine

. The diagnosis and management of mild cognitive impairment: a clinical review. JAMA 2014; 312: 2551–2561.

Patnode

Perdue

Rossom

, et al. Screening for cognitive impairment in older adults: updated evidence report and systematic review for the US Preventive Services Task Force. JAMA 2020; 323: 764–785.

Alzheimer's Disease International. World Alzheimer Report 2021: Journey through the diagnosis of dementia, https://www.alzint.org/resource/world-alzheimer-report-2021/ (2021, accessed 28 July 2024).

Chodosh

Petitti

Elliott

, et al. Physician recognition of cognitive impairment: evaluating the need for improvement. J Am Geriatr Soc 2004; 52: 1051–1059.

Kotagal

Langa

Plassman

, et al. Factors associated with cognitive evaluations in the United States. Neurology 2015; 84: 64–71.

Folstein

McHugh

. Mini-mental state’. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189–198.

Nasreddine

Phillips

Bédirian

, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc 2005; 53: 695–699.

10.

Mitchell

. A meta-analysis of the accuracy of the Mini-Mental State Examination in the detection of dementia and mild cognitive impairment. J Psychiatr Res 2009; 43: 411–431.

11.

Tsoi

KKF

Chan

JYC

Hirai

, et al. Cognitive tests to detect dementia: a systematic review and meta-analysis. JAMA Intern Med 2015; 175: 1450–1458.

12.

Lees

Selvarajah

Fenton

, et al. Test accuracy of cognitive screening tests for diagnosis of dementia and multidomain cognitive impairment in stroke. Stroke 2014; 45: 3008–3018.

13.

Albert

DeKosky

Dickson

, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 2011; 7: 270–279.

14.

Hachinski

Iadecola

Petersen

, et al. National Institute of Neurological Disorders and Stroke-Canadian Stroke Network vascular cognitive impairment harmonization standards. Stroke 2006; 37: 2220–2241.

15.

Dalrymple-Alford

MacAskill

Nakas

, et al. The MoCA: well-suited screen for cognitive impairment in Parkinson disease. Neurology 2010; 75: 1717–1725.

16.

Wei

Liu

, et al. The Montreal cognitive assessment: normative data from a large, population-based sample of Chinese healthy adults and validation for detecting vascular cognitive impairment. Front Neurosci 2024; 18: 1455129.

17.

Wang

CS-M

Pai

M-C

Chen

P-L

, et al. Montreal Cognitive Assessment and Mini-Mental State Examination performance in patients with mild-to-moderate dementia with Lewy bodies, Alzheimer’s disease, and normal participants in Taiwan. Int Psychogeriatr 2013; 25: 1839–1848.

18.

Jia

Yang

. Mini-Mental State Examination in elderly Chinese: a population-based normative study. J Alzheimers Dis 2016; 53: 487–496.

19.

, et al. Montreal Cognitive assessment in detecting cognitive impairment in Chinese elderly individuals: a population-based study. J Geriatr Psychiatry Neurol 2011; 24: 184–190.

20.

Chen

K-L

Chu

A-Q

, et al. Validation of the Chinese version of Montreal Cognitive Assessment Basic for screening mild cognitive impairment. J Am Geriatr Soc 2016; 64: e285–e290.

21.

Chiu

HFK

Zhong

Leung

, et al. Development and validation of a new cognitive screening test: the Hong Kong brief cognitive test (HKBC). Int J Geriat Psychiatry 2018; 33: 994–999.

22.

Sun

Chen

, et al. A validation study of the Hong Kong Brief Cognitive Test for screening patients with mild cognitive impairment and Alzheimer’s disease. J Alzheimers Dis 2022; 88: 1523–1532.

23.

Dubois

Feldman

Jacova

, et al. Advancing research diagnostic criteria for Alzheimer’s disease: the IWG-2 criteria. Lancet Neurol 2014; 13: 614–629.

24.

Sachdev

Kalaria

O’Brien

, et al. Diagnostic criteria for vascular cognitive disorders: a VASCOG statement. Alzheimer Dis Assoc Disord 2014; 28: 206–218.

25.

Tian

Zhang

Liu

, et al. A neuropsychological profile and its correlation with neuroimaging markers in patients with subcortical ischaemic vascular dementia. Int J Geriatr Psychiatry 2023; 38: e5900.

26.

Wen

H-B

Zhang

Z-X

Niu

F-S

, et al. The application of Montreal cognitive assessment in urban Chinese residents of Beijing. Zhonghua Nei Ke Za Zhi 2008; 47: 36–39.

27.

Zhang

Sun

W-H

Xing

, et al. Medial temporal lobe atrophy is related to learning strategy changes in amnestic mild cognitive impairment. J Int Neuropsychol Soc 2019; 25: 706–717.

28.

Benjamini

Hochberg

. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B: Stat Methodol 1995; 57: 289–300. (accessed 26 October 2025).

29.

DeLong

Clarke-Pearson

. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988; 44: 837–845.

30.

Freitas

Simões

Alves

, et al. Montreal Cognitive Assessment (MoCA): validation study for vascular dementia. J Int Neuropsychol Soc 2012; 18: 1031–1040.

31.

Freitas

Simões

Alves

, et al. Montreal Cognitive Assessment (MoCA): validation study for frontotemporal dementia. J Geriatr Psychiatry Neurol 2012; 25: 146–154.

32.

Chun

Seward

Patterson

, et al. Evaluation of available cognitive tools used to measure mild cognitive decline: a scoping review. Nutrients 2021; 13: 3974.

33.

Zhang

. Is semantic learning strategy an early clinical marker for amnestic mild cognitive impairment and Alzheimer’s disease? Int Psychogeriatr 2019; 31: 1695–1697.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.16 MB