East Asian Arch Psychiatry 2010;20:116-22


Neuropsychological Performance Predicts Decision-making Abilities in Chinese Older Persons with Mild or Very Mild Dementia
认知测试预测华籍老年极轻度至轻度老人癡呆症患者的精神决 策力
VWC Lui, LCW Lam, DNY Luk, HFK Chiu, PS Appelbaum
雷永昌、林翠华、陆雅欣、赵凤琴、PS Appelbaum

Dr Victor Wing-cheong Lui, MRCPsych, LLB, Department of Psychiatry, Tai Po Hospital, Hong Kong SAR, China.
Prof Linda Chiu-wa Lam, MD, FRCPsych, Department of Psychiatry, Chinese University of Hong Kong, Hong Kong SAR, China.
Dr Daisy Nga-yan Luk, MRCPsych, Department of Psychiatry, Tai Po Hospital, Hong Kong.
Prof Helen Fung-kum Chiu, FRCPsych, Department of Psychiatry, Chinese University of Hong Kong, Hong Kong SAR, China.
Dr Paul S. Appelbaum, MD, Division of Psychiatry, Law, and Ethics, Department of Psychiatry, College of Physicians and Surgeons of Columbia University, United States.

Address for correspondence: Dr Victor Wing-cheong Lui, Department of Psychiatry, Tai Po Hospital, 9 Chuen On Road, Tai Po, Hong Kong SAR, China.
Tel: (852) 2607 6111; Fax: (852) 2662 3568; Email: victorluiwc@hotmail.com

Submitted: 23 February 2010; Accepted: 27 April 2010

pdf Full Paper in PDF


Objective: To explore the relationship of the 4 decision-making abilities (Understanding, Appreciation, Reasoning, and Expressing a Choice) and neuropsychological performance in patients with very mild and mild dementia.

Methods: Chinese subjects were recruited from local social centres and residential hostels for elderly people in Hong Kong. Clinical diagnosis was made by experienced geriatric psychiatrists. A battery of neuropsychological tests that assesses general cognitive abilities, verbal memory, executive function, concept formation, and auditory and visual attention, was administered. Mental capacity to consent to treatment was assessed using the Chinese version of the MacArthur Competence Assessment Tool – Treatment.

Results: Fifty participants with very mild or mild dementia were compared with 42 cognitively intact subjects. After controlling for the effects of age and education, stepwise linear regression analysis demonstrated that the 4 decision-making abilities correlated with different neuropsychological test performances, which predicted 45% of the common variance for Understanding, 39% for Appreciation, 20% for Reasoning, and 30% for Expressing a Choice. The Reasoning score was only predicted by the Category Verbal Fluency Test (β = 0.4, p = 0.01).

Conclusion: Neuropsychological test performance differentially predicted different decision-making abilities in older patients with mild or very mild dementia.

Key words: Mental competency; Neuropsychological tests




结果:五十名极轻度至轻度老人癡呆症患者的测试结果与42名认知正常人士作比较。在控制年龄和教育程度因素後,逐步线性迴归分析显示,上述4项决断力因素与不同的认知功能测试表现呈相关,能预测理解力的公共方差占45%、评价占39%、推理占20%,选择表达力则占30%;而推理比分可经分类口头流畅度测试被预测出来(β = 0.4,p = 0.01)。




Decision-making capacity for consent to treatment embodies 4 abilities: to communicate a choice; to understand the relevant information; to appreciate the medical consequences of the situation; and to reason about treatment choices.1,2 Impairment in decision-making capacity occurs frequently among patients with Alzheimer disease (AD).3-5 Studies of the relationship between specific decision-making abilities and neuropsychological impairment in AD patients showed that treatment decisional abilities might be predicted by cognitive performance.6-8 However, in patients with mild- to-moderate dementia such studies are few and limited.

In a recent epidemiological survey, it was estimated that 8.9% of Hong Kong community-dwelling adults aged over 70 years suffer from mild dementia, and a similar proportion may be suffering from milder forms of cognitive impairment and impaired decision-making capacity.9However, literacy levels in the Chinese community differ from those in western populations (by a mean education level of about 2 years). Along with cultural differences, previous research on performance characteristics of mental capacity measures from western developed countries may not be directly applicable to the Chinese.

We previously reported the psychometric properties of the Chinese version of the MacArthur Competence Assessment Tool – Treatment (MacCAT-T).5 In the present study, we aimed to evaluate the association between neuropsychological test performance and decision-making abilities in Chinese older people with very mild dementia and mild AD. Exploration of the association between decision-making dimensions and specific cognitive domains has the potential to enhance the identification of patients at greater risk of impaired mental capacity. When certain patterns of cognitive impairments are observed, the clinician’s attention to such issues may be raised.



Chinese subjects over the age of 60 years were recruited from local social centres and residential hostels for older people in Hong Kong. The participants were volunteers who responded to announcements and advertisements about research participation at these centres / hostels. All subjects were assessed by a trained geriatric psychiatrist. The Clinical Dementia Rating (CDR) scale was used to assess the severity of dementia.10,11 Subjects with a global CDR of 0 were considered to be not demented. Subjects with a global CDR of 0.5 were categorised as having very mild dementia. Subjects with a global CDR of 1 were further evaluated, and those who satisfied the NINCDS-ADRDA criteria for probable or possible AD were recruited for comparison.12 Subjects with very mild and mild dementia were considered together as a single cognitive deficit (CD) group.

Exclusion criteria were: a CDR of 2 or more, and a known history of other neurodegenerative disorders or a major psychiatric disorder. Participants with profound communication difficulties were also excluded. One of the research team’s psychiatrists explained the details of study and obtained written informed consent from each participant. The study was approved by the institutional ethical review board.


Measurement of Decision-making Abilities

The MacCAT-T manual and record forms were translated into Chinese.13 The Chinese MacCAT-T has established reliability and validity in Chinese older people with dementia.5 It is a semi-structured interview that provides relevant treatment information for the patients and evaluates mental capacity along 4 dimensions: (1) Understanding, the ability to comprehend the information disclosed about the disorder and its proposed treatment; (2) Appreciation, the ability to relate such information to one’s own situation; (3) Reasoning, the ability to process the information in a logical fashion toward a decision; and (4) Expressing a Choice, the ability to communicate a decision about treatment. Each dimension yields respective summary scores. No MacCAT-T total score is calculated because significant enough deficits on 1 dimension may result in mental incompetence, even when performance on other dimensions is intact. In the non- demented (NC) group (CDR = 0), the Appreciation subscale was omitted because the concept of appreciating one’s own condition and need for treatment does not apply.

In the present study, the MacCAT-T disclosures and items were customised to refer to treatment of AD with cholinesterase inhibitors. The order and headings for our MacCAT-T questions were: understanding AD, appreciating AD, understanding cholinesterase inhibitors, understanding benefits and risks of cholinesterase inhibitors, appreciating their benefits and risks, expressing a choice, reasoning about the choice, and its logical consistency. The Chinese MacCAT-T was administered to each subject for assessment of decision-making abilities by a trained research assistant.

To focus on the assessment of mental competence and minimise the effect of verbal recall, the research assistant reminded the subjects up of the disease and treatment information disclosed during the MacCAT-T interview to 3 times. The interview was audiotaped and relevant non-verbal communication was separately recorded. For example, a subject with a language barrier could express his / her choice by gesture.

To determine inter-rater agreement, recordings of the MacCAT-T interviews of 33 subjects in the CD group were assessed independently by 2 trained geriatric psychiatrists. The intraclass correlation coefficients were 0.82 for Understanding, 0.71 for Appreciation, 0.69 for Reasoning, and 0.69 for Expressing a Choice. Intraclass correlation coefficients were accepted as reliability measures for MacCAT-T scores in the literature.3,5,14 These figures demonstrated a substantial level of agreement among the local raters and were deemed comparable to those in published western studies.3,14

Neuropsychological Assessment

The Cantonese version of the Mini-Mental State Examination (CMMSE) is widely used in clinical practice for patients with dementia and served as an index of global cognitive function.

The Chinese version of the AD Assessment Scale– Cognitive subscale (ADAS-Cog) is a standard cognitive instrument that was specifically designed for the evaluation of CD in subjects with AD.15-17 The ADAS-Cog comprises subscales that examine different aspects of cognitive function. The maximum score is 70 with increasing scores indicating greater severity of impairment.

Ten-minute Delayed Recall

A word list of 10 items was read to each subject, who was asked to recall the list 10 minutes later. The score was the number of items correctly recalled by the subject.

Category Verbal Fluency Test

In this study, the subjects were asked to generate exemplars in 3 categories (animals, fruits, and vegetables) for 1 minute per category. Information about the number of exemplars generated in 1 minute was recorded. The Category Verbal Fluency Test (CVFT) score represents the total number of exemplars generated after the three 1-minute trials.18

Digit and Visual Spans

Ascending and descending digit and visual spans were used to assess attention and working memory, which are closely related to the integrity of executive function.

Concept Formation

In order to assess abstract thinking, subjects were presented with 3 pairs of objects and asked to explain what each of pair of items had in common. Each comparison was rated at 2 points if it entailed an abstract generalisation, 1 point if a response entailed a specific ‘concrete’ likeness, and 0 points if no similarity was observed. The total score ranged from 0 to 6.


All subjects were assessed first by a trained geriatric psychiatrist, who administered the CDR. Demographic and clinical data were recorded. The neuropsychological test battery was then administered. Next, the MacCAT-T interviews were conducted and audiotaped by an independent trained research assistant to assess mental capacity.

Data Analyses

Spearman correlation coefficients between the MacCAT-T summary scores and demographic and clinical variables were derived. Associations between cognitive test performance and decision-making abilities were evaluated. Stepwise linear regression analyses were conducted to evaluate whether the correlated variables predicted each of the 4 MacCAT-T summary scores. The MacCAT-T summary scores were entered as dependent variables, with the above correlated variables as independent variables. Data analysis was performed using PASW Statistics 17.0. An alpha level of 0.05 was considered statistically significant.


Demographic and Clinical Characteristics

Fifty participants with very mild / mild dementia (CDR = 0.5 or 1) were recruited as the CD group. In the latter, the mean (standard deviation [SD]) age of patients was 80 (7) years, 28% of whom were male. The mean duration of education was 1.7 years. The mean (SD) CMMSE score was 22 (5).

In the NC group, there were 42 participants. The mean (SD) age was 75 (7) years, of whom 41% were male. The mean duration of education was 4.7 years. Subjects in this group were younger (t = 3.51, p = 0.001) and more educated (t = –3.29, p = 0.002). There was no significant difference in gender ratios. Their mean (SD) CMMSE score was 28 (2), significantly higher than that in the CD group (t = –7.4, p < 0.001).

MacArthur Competence Assessment Tool – Treatment Summary Scores

The distribution of MacCAT-T scores for the CD group is shown in Table 1. In the NC group, summary scores for Understanding, Reasoning, and Expressing a Choice were significantly higher.

Neuropsychological Performance

Neuropsychological performance data are summarised in Table 2. Subjects in the NC group performed significantly better in all the tests.

Correlations of MacArthur Competence Assessment Tool – Treatment Scores with Demographic Variables and Neuropsychological Performance

For the CD group, the correlations between the MacCAT-T summary scores and demographic and cognitive variables are shown in Table 3. Age, CMMSE score, ADAS-Cog total score, 10-minute delayed recall and CVFT scores yielded significant correlations with all MacCAT-T summary scores. Years of education (rs = 0.33, p < 0.05) correlated significantly with the Understanding score. Backward visual span correlated significantly with the Understanding (rs = 0.43, p < 0.01) and Appreciation (rs = 0.42, p < 0.01) summary scores. There was no significant correlation between MacCAT-T summary scores and gender.

Regression Analysis

Since age and education were associated with MacCAT-T summary scores in our sample, the effects of these 2 variables were controlled for in the stepwise linear regression analysis. The analysis, as shown in Table 4, revealed that the Understanding score was significantly predicted by the ADAS-Cog total score (β = –0.4, p = 0.01) and the CVFT score (β = 0.3, p = 0.04), and that the relationship accounted for about 45% of the variance of the Understanding score. The Appreciation score was signifi- cantly predicted by the CMMSE score (β = 0.6, p < 0.001) and the relationship accounted for about 39% of the variance. The Reasoning score was predicted by the CVFT score (β = 0.4, p = 0.01) and about 20% of its variance was explained by this relationship. Expressing a Choice score was significantly predicted by the ADAS-Cog total score (β = –0.6, p < 0.001), that relationship accounting for about 30% of its variance.


In this study, our subjects were recruited from the local community. The differences in years of education and age between the CD and NC groups were consistent with a local epidemiological survey.9

The differences in MacCAT-T summary scores between the CD and NC groups show that those with CD had impairments in decision-making abilities. As the members of the CD groups were participants with milder CD and mild AD, the findings suggest that mental capacity for treatment decisions may be impaired in the early phases of cognitive decline in late life.8,19,20

Even after controlling for the effects of education and age, neuropsychological test performance significantly predicted decision-making ability. Understanding appears to be more strongly related to neuropsychological test performance than other decisional abilities. These findings are consistent with the results of a regression analysis in a western population.8 The overall effect of these neuropsychological predictors in our study was smaller, explaining only 45% of the total variance for Understanding and about 20% of the total variance for Reasoning, compared with 78% and 40%, respectively in the study by Gurrera et al.8 The difference may be due to the use of different neuropsychological tests in the Chinese population. The subjects recruited in the Gurrera et al’s study8 had mild- to-moderate dementia. In our study, the subjects had comparatively milder CD. Our findings suggest a need to further examine factors that may affect decision-making capacity.

Non-cognitive factors including older age, fewer years of education and poor insight might affect performance in the course of competence assessments.2,21,22 These factors may exaggerate the impact of cognitive impairment on mental capacity. In a recent local epidemiological survey, the educational attainment of Hong Kong patients with dementia was found to be very low.9 Furthermore, cultural expectations towards medical conditions may also affect decision-making. For example, some older people may think that forgetfulness is part of normal ageing process, rather than a symptom of dementia, and have difficulty in believing treatment information. Existing studies focusing on the effect of cultural and educational differences are lacking and these areas should be further explored.

Notably, although the 10-minute delayed recall (the verbal retrieval test used in this study) correlated with all the 4 MacCAT-T summary scores, it was not a significant predictor for any of the MacCAT-T summary scores in the multiple regression analysis. Marson et al6,7 found that verbal memory was not a strong predictor for any decision- making abilities. It was suggested that the floor effects on memory tasks undertaken by AD patients minimised the role of verbal memory.7 In our study, the mean score of the 10-minute delayed recall was only 2.7 out of 10 in the CD group. Floor effects may also have operated among our subjects. Moreover, if the subject forgot the information, our research assistant reminded them up to 3 times about their illness or treatment information during the MacCAT-T interview. This practice may minimise the demand on verbal memory, especially recall. Moreover, it is similar to actual clinical practice, where the patient is allowed to retain descriptions of his / her disease and treatment or ask for clarification throughout the process of informed consent.

The distinctiveness of the multivariate predictor profiles for specific decision-making abilities supports the idea that Understanding, Appreciation, Reasoning, and Expressing a Choice are discrete elements of decision- making capacity. In the bivariate analyses, Understanding, Appreciation, and Expressing a Choice all correlated significantly with general cognitive tests, including ADAS- Cog total score and CMMSE score. Multiple cognitive functions are likely to be involved in these 3 dimensions of decision-making capacity. Category Verbal Fluency Test was a significant multivariate predictor for Understanding and Reasoning. This suggests that executive dysfunction may have a unique relationship to these 2 decision-making abilities. This influence should be recognised during attempts to maximise patient decision-making capacity, because the simple presentation of material or reminders to compensate for impaired recall may not suffice to enhance these 2 decision-making abilities.

The findings of this study should be interpreted in the context of its methodological limitations. The sample size may not have been large enough to represent the factor structure of neuropsychological performance. However, our sample size was comparable to those detailed in published western studies.3,4,6-8,19 To obtain a more representative sample of the population, the subjects were recruited from the community rather than from a health care setting. Our findings were restricted to the neuropsychological tests employed. However, the tests utilised included the most commonly used tests for Chinese older people.

The utility and limitations of MMSE as an indirect test for mental capacity and as a means of identifying patients of impaired capacity have been described and discussed.3,23 In our study, Reasoning ability was predicted only by the CVFT, and not the tests of general cognitive abilities. This suggests that the MMSE alone may not be able to screen or identify patients with impairment in Reasoning ability only. The results of this study support the use of direct assessment tools for mental capacity and highlight the importance of proper competence assessment in clinical practice.


  1. Dunn LB, Nowrangi MA, Palmer BW, Jeste DV, Saks ER. Assessing decisional capacity for clinical research or treatment: a review of instruments. Am J Psychiatry 2006;163:1323-34.
  2. Appelbaum PS. Clinical practice. Assessment of patients’ competence to consent to treatment. N Engl J Med 2007;357:1834-40.
  3. Karlawish JH, Casarett DJ, James BD, Xie SX, Kim SY. The ability of persons with Alzheimer disease (AD) to make a decision about taking an AD treatment. Neurology 2005;64:1514-9.
  4. Okonkwo O, Griffith HR, Belue K, Lanza S, Zamrini EY, Harrell LE, et al. Medical decision-making capacity in patients with mild cognitive impairment. Neurology 2007;69:1528-35.
  5. Lui VW, Lam LC, Luk DN, Wong LH, Tam CW, Chiu HF, et al. Capacity to make treatment decisions in Chinese older persons with very mild dementia and mild Alzheimer’s disease. Am J Geriatr Psychiatry 2009;17:428-36.
  6. Marson DC, Cody HA, Ingram KK, Harrell LE. Neuropsychologic predictors of competency in Alzheimer’s disease using a rational reasons legal standard. Arch Neurol 1995;52:955-9.
  7. Marson DC, Chatterjee A, Ingram KK, Harrell LE. Toward a neurologic model of competency: cognitive predictors of capacity to consent in Alzheimer’s disease using three different legal standards. Neurology 1996;46:666-72.
  8. Gurrera RJ, Moye J, Karel MJ, Azar AR, Armesto JC. Cognitive performance predicts treatment decisional abilities in mild to moderate dementia. Neurology 2006;66:1367-72.
  9. Lam LC, Tam CW, Lui VW, Chan WC, Chan SS, Wong S, et al. Prevalence of very mild and mild dementia in community-dwelling older Chinese people in Hong Kong. Int Psychoger 2008;20:135-48.
  10. Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL. A new clin- ical scale for the staging of dementia. Br J Psychiatry 1982;140:556-72.
  11. Morris JC. Clinical dementia rating: a reliable and valid diagnostic and staging measure for dementia of the Alzheimer type. Int Psychoger 1997;9 Suppl 1:S173-6.
  12. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS- ADRDA work group under the auspices of Department of Health and Human Service Task Force on Alzheimer’s disease. Neurology 1984;34:939-44.
  13. Grisso TG, Appelbaum PS. Assessing competence to consent to treatment: a guide for physicians and other health professionals. New York: Oxford University Press; 1998.
  14. Grisso T, Appelbaum PS, Hill-Fotouhi C. The MacCAT-T: a clinical tool to assess patients’ capacities to make treatment decisions. Psychiatr Serv 1997;48:1415-9.
  15. Mohs RC, Rosen WG, Davis KL. The Alzheimer’s disease assessment scale: an instrument for assessing treatment efficacy. Psychopharmacol Bull 1983;19:448-50.
  16. Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s disease. Am J Psychiatry 1984;141:1356-64.
  17. Chu LW, Chiu KC, Hui SL, Yu GK, Tsui WJ, Lee PW. The reliability and validity of the Alzheimer’s Disease Assessment Scale Cognitive Subscale (ADAS-Cog) among the elderly Chinese in Hong Kong. Ann Acad Med Singapore 2000;29:474-85.
  18. Lam LC, Ho P, Lui VW, Tam CW. Reduced semantic fluency as an additional screening tool for subjects with questionable dementia. Dement Geriatr Cogn Disord 2006;22:159-64.
  19. Moye J, Karel MJ, Gurrera RJ, Azarm AR. Neuropsychological predictors of decision-making capacity over 9 months in mild-to- moderate dementia. J Gen Intern Med 2005;21:78-83.
  20. Jefferson AL, Lambe S, Moser DJ, Byerly LK, Ozonoff A, Karlawish JH. Decisional capacity for research participation in individuals with mild cognitive impairment. J Am Geriatr Soc 2008;56:1236-43.
  21. Raymont V, Bingley W, Buchanan A, David AS, Hayward P, Wessely S, et al. Prevalence of mental incapacity in medical inpatients and associated risk factors: cross-sectional study. Lancet 2004;364:1421-7.
  22. Maxmin K, Cooper C, Potter L, Livingston G. Mental capacity to consent to treatment and admission decisions in older adult psychiatric inpatients. Int J Geriatr Psychiatry 2009;24:1367-75.
  23. Kim SY, Caine ED. Utility and limits of the mini mental state examination in evaluating consent capacity in Alzheimer’s disease. Psychiatr Serv 2002;53:1322-4.
View My Stats