Alzheimer’s & Dementia - (2014) 1–10

Featured Article

Prevalence of mild cognitive impairment in an urban community in China: A cross-sectional analysis of the Shanghai Aging Study Ding Dinga, Qianhua Zhaoa, Qihao Guoa, Haijiao Menga, Bei Wanga, Jianfeng Luob, James A. Mortimerc, Amy R. Borensteinc, Zhen Honga,* a Institute of Neurology, Huashan Hospital, Fudan University, Shanghai, China Department of Health Statistics, School of Public Health, Fudan University, Shanghai, China c Department of Epidemiology and Biostatistics, College of Public Health, University of South Florida, Tampa, FL, USA b

Abstract

Background: Substantial variations in the prevalence of mild cognitive impairment (MCI) have been reported, although mostly in Western countries. Less is known about MCI in the Chinese population. Methods: We clinically and neuropsychologically evaluated 3141 community residents
60 years of age. Diagnoses of MCI and its subtypes were made using standard criteria via consensus diagnosis. Results: Among 2985 nondemented individuals, 601 were diagnosed with MCI, resulting in a prevalence of 20.1% for total MCI, 13.2% for amnestic MCI (aMCI), and 7.0% for non-amnestic MCI (naMCI). The proportions of MCI subtypes were: aMCI single domain (SD), 38.9%; aMCI multiple domains (MD), 26.5%; naMCI-SD, 25.0%; and naMCI-MD, 9.6%. The prevalence of aMCI-MD increased rapidly with age in women APOE ε4 carriers (from 60 to 69 years to
80 years, 3.1%–33.3%, P , .001). Conclusions: Our findings suggest that 20% of Chinese elderly are affected by MCI. Prospective studies in China are needed to examine progression to dementia and related risk factors. Ó 2014 The Alzheimer’s Association. All rights reserved.

Keywords:

Mild cognitive impairment; Prevalence; Cross-sectional; Population-based; Aging

1. Introduction Rapid demographic changes with increases in the percent of individuals living to older ages, particularly in developing countries, will lead to larger numbers of elderly being at risk for dementia. The frequency of mild cognitive impairment (MCI), an intermediate state between normal cognitive aging and dementia [1,2], could provide valuable information about the population at risk for becoming demented. Petersen and colleagues initially characterized MCI as amnestic impairment that was presumed to represent an early manifestation of Alzheimer’s disease (AD) [1]. The definition was later expanded to include other cognitive domains, with the expectation that the initial pattern of impairment predicted various diagnostic outcomes [3].

*Corresponding author. Tel.: 86-21-52888158; Fax: 86-21-62481930. E-mail address: [email protected]

Over the last 20 years, a series of population-based epidemiologic studies, mostly in the developed world, have been conducted to explore the prevalence of MCI. A systematic review reported that the prevalence of MCI or amnestic MCI (aMCI) ranges from 0.5% to 42% [4]. More recent studies reported that MCI prevalence ranges from 6.5% to 39.1% in populations of American, Australian, Bulgarian, Mexican, and Japanese elderly [5–11]. The substantial variation in the prevalence of MCI has been considered to be due to differences in detection procedures, implementation of MCI diagnostic criteria, and the demographic characteristics of the source populations. Currently, there are 185 million Chinese .60 years of age, which represents the largest number of elderly people in any country in the world [12]. Identifying the prevalence of MCI in China is crucial for assessments of potential disease burden and therefore the need for interventions to prevent or slow progression of decline to dementia. Since 2001, several studies have shown that the prevalence of MCI in

1552-5260/$ - see front matter Ó 2014 The Alzheimer’s Association. All rights reserved. http://dx.doi.org/10.1016/j.jalz.2013.11.002

2

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

China ranged from 2.4% to 35.9% for the population
60 years of age [13–15]. These studies used the Mini-Mental State Examination (MMSE) or the Montreal Cognitive Assessment to screen for cognitive impairment and lacked standard definitions for failing the tests. The 10/66 dementia research group population-based survey reported the prevalence of MCI in developing countries, including China [16]. However, in that study the investigators surveyed only the prevalence of aMCI and not other subtypes. In the present population-based study, we conducted in-person clinical and neuropsychological evaluations of all individuals aged
60 years of age and residing in a geographically defined urban community of Shanghai to detect MCI and its subtypes using standard definitions and consensus diagnosis, comparable to the methodology used in most Western studies. We report the associations of prevalent MCI and its subtypes with age, gender, education, and apolipoprotein E (APOE) genotype. 2. Methods 2.1. Ethics statement This study was approved by the medical ethics committee of Huashan Hospital, Fudan University, Shanghai, China. Written informed consent was obtained from all participants or their legally acceptable representative. 2.2. Study site and target population The Jingansi community is located in the southwest part of downtown Shanghai. It contains 11 neighborhoods and 14,000 families. The Shanghai government maintains an annual census of registered residents and, from this list, 5138 permanent residents (excluding those living in a nursing home or other institution)
60 years of age were found to live in this community at the end of 2009. Inclusion criteria included being on the list of permanent residents and being
60 years old. Potential participants were excluded if they: (1) were deceased; (2) showed severe schizophrenia or mental retardation on their medical record; or (3) had severe problems of vision, hearing, or speaking, and were not able to participate actively in the neuropsychological evaluation. 2.3. Subject recruitment Potential subjects were identified using the governmentmaintained “residents list,” which includes the name, gender, age, address, and telephone number of every registered resident. Study coordinators went to each home in each neighborhood to introduce the study. People who met inclusion and exclusion criteria and who were willing to participate were consecutively enrolled. At that time, an appointment for a clinical interview (either at Huashan Hospital, or at subjects’ homes) was made. Participants were reminded of the evaluation by a telephone call 1 day before it

was scheduled. Those not agreeing to participate in the study were logged and administered a refusal questionnaire. 2.4. Medical and neurologic examination Study neurologists from Huashan Hospital conducted medical and neurologic examinations on all participants. Demographic data (birthdate, gender, and education) and data on lifestyle habits (cigarette smoking and alcohol drinking) were collected. In Shanghai, individuals keep their medical records with them and have all information entered at each encounter with the healthcare system. We asked participants about their medical histories, including physiciandiagnosed hypertension, diabetes, stroke, and heart disease (including coronary artery disease, valvular heart disease, cardiomyopathy, heart failure, and arrhythmias), and confirmed these from the participants’ medical records. The neurologic examination assessed the sensory neurons, motor responses, and reflexes of each participant. The Zung Self-Rating Anxiety Scale [17] and Hamilton Depression Rating Scale [18] were administered to assess psychiatric status. This information was helpful to identify comorbid neurologic and psychologic conditions that could influence cognitive performance, and to define subgroups with neuropsychiatric symptoms at higher risk of developing dementia in the future [19]. In addition to obtaining information directly from participants, neurologists administered the Clinical Dementia Rating (CDR) scale to proxies to obtain information on cognitive complaints and activities of daily living [20,21]. Participants were considered to have cognitive complaints if they, their proxy, or a nurse or physician indicated that they had problems with memory or thinking. Items from the Lawton and Brody Activity of Daily Living (ADL) scale were used to elicit physical self-maintenance and instrumental activities of daily living, including eating, using the telephone, preparing meals, handling money, and completing chores. Participants were considered to be functionally intact if their ADL score was .16 [22]. 2.5. APOE genotype assessment DNA was extracted from blood or saliva and collected from the study participants. APOE genotyping was conducted by the TaqMan single-nucleotide polymorphism (SNP) method [23]. The presence of at least one ε4 allele was treated as being APOE ε41. 2.6. Neuropsychological measurements Based on the Chinese culture, we translated, adapted, and normed neuropsychological tests from Western countries. Because some tests require vocabulary, writing, or reading skills, we designed the neuropsychological battery according to the education level of each participant. A pilot validation study was conducted in healthy elderly living in the same community. The validation sample size for the group with

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10


6 years of education was 956, and the average age was 71.6 years; the sample size for the group with ,6 years of education was 314, and the average age was 78.0 years. We established a 1.5-standard deviation (SD) cutoff for each cognitive measure using corrections for gender, age, and years of education. The distributions of the neuropsychological test scores (except the MMSE, as it was validated in Chinese population in the 1990s [24]) are presented as the normative data in Table E1 (online material). For participants with
6 years of formal education, the battery comprised the MMSE, Conflicting Instructions Task, Stick Test, Modified Common Objects Sorting Test, Auditory Verbal Learning Test, and Trail Making Test. For participants with ,6 years of education, the battery comprised the MMSE, Conflicting Instructions Task, Stick Test, Modified Common Objects Sorting Test, Modified Fuld Object Memory Evaluation, and Renminbi Test. The battery was administered in Chinese by certified study psychometrists within 90 minutes. Neuropsychological tests and domains are as follows: 1. MMSE was used for screening for global cognition [24]. 2. The Conflicting Instructions Task (Go/No-Go Task), adapted from part of the Frontal Assessment Battery [25], was used to measure executive function. First, subjects were asked to tap fingers following the conflicting instructions (sensitivity to interference): “tap twice when I tap once,” and “tap once when I tap twice.” Then subjects tapped fingers according to the series of “1-1-2-1-2-2-2-1-1-2/-1-2-2-1-1-2-1-2-2-1,” when performed by the examiner. In addition, subjects were asked to do “Go/No-Go” (inhibitory control): “tap once when I tap once,” and “do not tap when I tap twice.” Subjects tapped fingers according to the series of “2-1-2-1-1-2-2-1-1-2/-1-2-1-2-2-1-1-2-2-1,” performed by the examiner. 3. The Stick Test, adapted from the Stick Construction Test [26], was used as a measure of spatial construction function and memory. This 10-item test was first administered as a copying task. Subjects were given 4 wooden sticks and asked to copy the examiner’s model exactly. Subjects were asked to recall and construct the previous pattern after copying the current one. After the 10 designs were copied, the reversal condition was administered in which subjects were asked to construct the reverse pattern of the examiner’s model. 4. The Modified Common Objects Sorting Test, adapted from the Object Sorting Test [27], was used to measure language and executive function. The test material consists of pictures of 42 common objects familiar to Chinese people. Subjects were first required to name each object in the picture. Then subjects were asked to sort all the objects into 7 different groups. Subjects were then asked, “Why do all these belong together?”

3

Tests 1–4 were used for subjects at all education levels. 5. The Auditory Verbal Learning Test, adapted from the California Verbal Learning Test, was used to measure verbal memory of subjects with
6 years of education. Subjects were presented 12 words over 5 trials, followed by delayed recall and recognition trials [28]. 6. The Modified Fuld Object Memory Evaluation, adapted from the Fuld Object Memory Evaluation [29], was used to measure immediate and delayed memory for subjects with ,6 years of education. The test material consists of 12 common objects familiar to Chinese people. Subjects were presented with 12 words to learn over 5 trials, followed by delayed recall and recognition trials. 7. Trail Making Tests A and B, adapted from a subtest of the Halstead–Reitan neuropsychological battery [30], were used for subjects with
6 years of education. Subjects were required to connect 25 consecutive targets with numbers inside squares or circles on a sheet of paper. There were 2 parts of the test: A, in which the targets were all numbers, and subjects needed to connect them in sequential order (1, 2, 3, etc.); and B, in which subjects were asked to connect numbers in sequential order with the alternation of square and circle (“1” in a square, “2” in a circle, etc.). This test measures attention and executive function, including visual motor skills, fast visual search, and cognitive set shifting. 8. The Renminbi (official currency of China) Test, translated from the EURO Test [31], was used to measure language, attention, and memory in subjects with ,6 years of education. After assessing the subject’s knowledge of different coins and bills, the subject performed 5 arithmetical tasks of increasing difficulty with 11 coins (counting, making change, adding, dividing by 2, and dividing by 3). After a distraction task, the subject was asked to recall the number and type of coins used before, and the total amount of money involved.

2.7. Consensus diagnosis After each clinical assessment, study neurologists and neuropsychologists (D.D., Q.Z., Q.G., H.M., B.W., and Z.H.) reviewed the functional, medical, neurologic, psychiatric, and neuropsychological data, and reached a consensus regarding the presence or absence of dementia using the criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) [32]. Only those who were not diagnosed with dementia were considered for a diagnosis of MCI. MCI was defined according to the following criteria [33]: (1) cognitive concern or complaint by the subject, informant, nurse, or physician, with CDR 5 0.5; (2) objective impairment in at least 1 cognitive domain based on performance

4

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

1.5 SD below the mean using the norms obtained in the pilot study; (3) essentially normal functional activities (determined from the CDR and the ADL evaluation); and (4) absence of dementia (by DSM-IV). Based on cognitive test scores, subjects diagnosed with MCI were placed into 1 of 4 groups characterizing their cognitive deficits: amnestic MCI single domain (aMCI-SD); amnestic MCI multiple domains (aMCI-MD); non-amnestic MCI single domain (naMCI-SD); or nonamnestic MCI multiple domains (naMCI-MD) [3,34]. For the aMCI-SD group, a deficit on at least 1 of the memory tests was required with no deficit in other domains. For the aMCI-MD group, at least 1 deficit in memory plus at least 1 additional deficit in another domain was required. For the naMCI-SD group, a deficit in verbal fluency, language, visuospatial skills, speed of mental processing, or executive function was required without a memory deficit. For the naMCI-MD group, deficits in
2 domains other than memory were required. 2.8. Statistical analysis Continuous variables are expressed as mean and SD, and categorical variables as frequency (%). Student’s t-test or 1way analysis of variance was used for comparisons for continuous variables. The c2 test was used for comparisons of categorical variables. Based on a published estimated MCI prevalence of 15%– 30%, we estimated that approximately 450–900 persons would have prevalent MCI among 3000 nondemented persons. The prevalence of MCI would be estimated with 95% confidence intervals (CIs) of approximately 61.3%– 1.6%. Prevalence of MCI and MCI subtypes were calculated for the entire population, and by age (60–64, 65–69, 70–74, 75– 79, 80–84,
85 years), gender, education (less than primary school, primary school, middle school, high school, college and above), and APOE genotype (carriers and noncarriers of ε4 allele). The 95% CIs of prevalence were calculated using the exact Clopper–Pearson method, which is based on the exact binomial distribution. The c2 test was used to compare differences in prevalence between gender, age, education, and APOE genotype categories. All P-values and CIs were estimated in a 2-tailed fashion. Differences were considered statistically significant at P , .05. Data were analyzed using SPSS version 13.0 (SPSS, Inc., Chicago, IL, USA).

Fig. 1. Flowchart of subject recruitment for the prevalence study of mild cognitive impairment.

brief information on a refusal questionnaire, including age; gender; education; history of hypertension, diabetes mellitus, stroke, and heart disease; and smoking and drinking habits. Forty-six percent of participants were men, the average age of participants was 72.3 years (SD 8.1), and they had an average of 11.6 years (SD 4.4) of education. The number of years of education decreased significantly with increasing age (P , .001). Participants aged
80 years of age had fewer years of education (mean 10.3, SD 5.4) than those 70–79 (mean 11.2, SD 4.9) and 60–69 (mean 12.6, SD 3.0) years of age. There was no statistical difference by gender or educational attainment between the 3141 participants and 1286 nonparticipants. However, the average age of participants was significantly lower than that of non-participants (mean 73.3, SD 8.6, P 5 .012) due to the higher proportion (43.7%) of participants aged 60–69 years, and lower proportion (18.4%) of participants
80 years of age. Compared with participants, nonparticipants were less likely to have hypertension, heart disease, and stroke (Table 1). 3.2. Biospecimen collection

3. Results 3.1. Characteristics of study subjects Figure 1 shows that, among 5138 registered long-term residents, 619 were ineligible. From January 1, 2010, through September 30, 2011, we conducted in-person interviews for 3141 (70%) individuals who participated in this study. Among 1378 (30%) nonparticipants, 1286 provided

Biospecimens were collected from 2757 (92.4%) individuals without dementia. DNA was extracted from either blood (97%) or saliva (3%) samples. 3.3. Prevalence of MCI Among 3141 participants, 156 were diagnosed with dementia. Of the 2985 nondemented individuals with average

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

5

Table 1 Characteristics of study participants and nonparticipants

Number of subjects Gender Men, n (%) Women, n (%) Age in years, mean (SD) [min, max] Age group, in years 60–69, n (%) 70–79, n (%)
80, n (%) Education, in years, mean (SD) [min, max] Education level Less than primary school, n (%) Primary school, n (%) Middle school, n (%) High school, n (%) College and above, n (%) Medical history Hypertension, n (%) Diabetes mellitus, n (%) Stroke, n (%) Heart disease, n (%) Life habits Smoking, n (%) Drinking, n (%)

Participants

Nonparticipants

P-value

3141

1286

1438 (45.8) 1703 (54.2) 72.3 (8.1) [60, 101]

628 (48.8) 658 (51.2) 73.3 (8.6) [60, 100]

1373 (43.7) 1191 (37.9) 577 (18.4) 11.6 (4.4) [0, 21]

464 (36.1) 500 (38.9) 322 (25.0) 11.3 (4.7) [0, 23]

134 (4.3) 301 (9.6) 580 (18.5) 978 (31.1) 1145 (36.5)

74 (5.8) 128 (10.0) 238 (18.5) 370 (28.8) 476 (37.0)

0.186

1771 (56.4) 459 (14.6) 444 (14.1) 1200 (38.2)

588 (45.7) 192 (14.9) 115 (8.9) 269 (20.9)

,0.001 0.787 ,0.001 ,0.001

299 (9.5) 247 (7.9)

118 (9.2) 104 (8.1)

0.064 0.012 ,0.001

0.975

0.685

Abbreviation: SD, standard deviation.

age of 71.8 years (SD 7.8), 601 were diagnosed with MCI, resulting in a prevalence of 20.1% (95% CI 18.7%– 21.5%). As shown in Table 2, MCI prevalence was slightly higher among women (21.1%, 95% CI 19.1%–23.1%) compared with men (19.0%, 95% CI 16.9%–21.1%), but this difference was not significant (P 5 .142). The preva-

lence of MCI increased significantly with age (P , .001), and decreased significantly with increasing level of education (P , .001). The prevalence of MCI in subjects carrying 1 or 2 APOE ε4 alleles (20.2%, 95% CI 16.6%–23.8%) was similar to that of noncarriers (18.8%, 95% CI 17.2%–20.4%) (P 5 .486).

Table 2 Prevalence and 95% CI for MCI, aMCI, and naMCI among nondemented participants, by gender, age, education, and APOE ε4 allele

Total Gender Male Female Age range, in years 60–64 65–69 70–74 75–79 80–84
85 Education Less than primary school Primary school Middle school High school College and above APOE ε4 allele ε4 (1) ε4 (2)

MCI (n)

% (95% CI)

P-value

aMCI (n)

% (95% CI)

P-value

naMCI (n)

% (95% CI)

P-value

601

20.1 (18.7–21.5)

—

393

13.2 (12.0–14.4)

—

208

7.0 (6.1–7.9)

—

262 339

19.0 (16.9–21.1) 21.1 (19.1–23.1)

0.142

196 197

14.2 (12.4–16.0) 12.3 (10.7–13.9)

0.124

66 142

4.8 (3.7–5.9) 8.9 (7.5–10.3)

,0.001

76 84 83 149 113 75

9.7 (7.6–11.8) 14.5 (11.6–17.4) 15.3 (12.3–18.3) 25.0 (21.5–28.5) 35.3 (30.1–40.5) 46.0 (38.4–53.7)

,0.001

60 69 48 89 73 54

7.7 (5.8–9.6) 11.9 (9.3–14.5) 8.8 (6.4–11.2) 14.9 (12.0–17.8) 22.8 (18.2–27.4) 33.1 (25.9–40.3)

,0.001

16 15 35 81 40 21

2.0 (1.0–3.0) 2.6 (1.3–3.9) 6.4 (4.3–8.5) 13.6 (10.9–16.4) 12.5 (8.9–16.1) 12.9 (7.8–18.1)

,0.001

51 112 138 156 144

50.5 (40.8–60.3) 41.8 (35.9–47.7) 24.8 (21.2–28.4) 16.5 (14.1–18.9) 12.9 (10.9–14.9)

,0.001

20 52 89 121 111

19.8 (12.0–27.6) 19.4 (14.7–24.1) 16.0 (13.0–19.1) 12.8 (10.7–14.9) 10.0 (8.2–11.8)

,0.001

31 60 49 35 33

30.7 (21.7–39.7) 22.4 (17.4–27.4) 8.8 (6.5–11.2) 3.7 (2.5–4.9) 3.0 (2.0–4.0)

,0.001

98 428

20.2 (16.6–23.8) 18.8 (17.2–20.4)

0.486

72 267

14.8 (11.6–18.0) 11.8 (10.5–13.1)

0.060

26 161

5.4 (3.4–7.4) 7.1 (6.0–8.2)

0.170

Abbreviations: APOE, apolipoprotein E; MCI, mild cognitive impairment; aMCI, amnestic mild cognitive impairment; naMCI, non-amnestic mild cognitive impairment; CI, confidence interval.

6

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

3.4. Prevalence of MCI subtypes Table 2 shows the prevalence by MCI subtypes. The prevalence of aMCI was 13.2% (95% CI 12.0%–14.4%), which was nearly twice that of naMCI (7.0%, 95% CI 6.1%–7.9%). The prevalence of aMCI among men (14.2%, 95% CI 12.4– 16.0%) was higher than that among women (12.3%, 95% CI 10.7%–13.9%), but was not significantly different (P 5 .124). However, the prevalence of naMCI in women (8.9%, 95% CI 7.5%–10.3%) was significantly higher than that in men (4.8%, 95% CI 3.7%–5.9%, P , .001). The age and education patterns of aMCI and naMCI were similar to those seen in total MCI. There was no significant difference in the prevalence of naMCI between APOE ε4 carriers and noncarriers, whereas there was a nearly significant effect of APOE ε4 in aMCI (P 5 .06). Among 601 individuals with MCI, 393 (65.4%) had aMCI (234 [38.9%] aMCI-SD and 159 [26.5%] aMCI-MD) and 208 (34.6%) had naMCI (150 [25.0%] naMCI-SD and 58 [9.6%] naMCI-MD). In individuals without aMCI-SD, the most impaired domain was executive function (86.8% in aMCI-MD, 83.3% in naMCI-SD, and 100% in naMCIMD), and the least impaired domain was language (1.9% in aMCI-MD, 0.0% in naMCI-SD, and 1.7% in naMCI-MD). Figure 2 shows that men had a significantly higher prevalence of aMCI-SD than women (men: 10.4%, 95% CI 8.8%– 12.0%; women: 5.6%, 95% CI 4.5%–6.7%; P ,.001). However, women had a significantly higher prevalence of the other 3 MCI subtypes. APOE ε4 carriers had a significantly higher prevalence of aMCI-MD than noncarriers (carriers: 8.2%, 95% CI 5.8%–10.6%; noncarriers: 3.8%, 95% CI 3.0%– 4.6%, P ,.001). There were no significant differences for other MCI subtypes between APOE ε4 carriers and noncarriers. Table 3 shows the age-specific prevalence for MCI subtypes by APOE and gender. The prevalence of aMCI-MD increased significantly with age. It increased more rapidly in APOE ε4 carrier women (from 60–69 years to
80 years old, 3.1%– 33.3%, P , .001) compared with APOE ε4 carrier men (1.9%–18.8%, P 5 .001), APOE ε4 noncarrier women (1.4%–14.4%, P , .001), and APOE ε4 noncarrier men (1.0%–6.0%, P , .001). The prevalence of naMCI-SD increased significantly with age in APOE ε4 noncarrier women (from 60 to 69 years to
80 years old, 2.1%–10.7%, P ,.001) and APOE ε4 noncarrier men (1.6%–7.2%, P , .001). 4. Discussion In this study we clinically and neuropsychologically evaluated 2985 Chinese individuals without dementia
60 years of age residing in an urban community of Shanghai, and documented a prevalence of 20.1% for total MCI, 13.2% for aMCI, and 7.0% for naMCI. Prevalence of aMCI and naMCI increased strongly with age and decreased with higher levels of education. The proportions of aMCI-SD, aMCI-MD, naMCI-SD, and naMCI-MD were 38.9%, 26.5%, 25.0%, and 9.6%, respectively. In separate analyses,

Fig. 2. NOTE. (A) Prevalence of MCI subtypes by sex. Men had significantly higher prevalence rates of aMCI-SD than women; however, women had significantly higher prevalence rates of other MCI subtypes than men. (B) Prevalence of MCI subtypes by APOE ε4 allele. APOE ε4 allele carriers had significantly higher prevalence rates of aMCI-MD than non-carriers. Abbreviations: MCI, mild cognitive impairment; aMCI-SD, amnestic mild cognitive impairment single domain; aMCI-MD, amnestic mild cognitive impairment multiple domains; naMCI-SD, non-amnestic mild cognitive impairment single domain; naMCI-MD, non-amnestic mild cognitive impairment multiple domains; APOE, Apolipoprotein.

women and APOE ε4 carriers had a significantly higher prevalence of aMCI-MD. This is the first epidemiologic study conducted in China with a study design, operational procedures and MCI (including 4 subtypes) diagnostic criteria similar to most cohort studies in developed countries. It is also the largest study in which all participants were evaluated by comprehensive in-person assessments and used consensus diagnosing. In the current study, the prevalence of MCI among nondemented individuals
70 years of age was 25.9%, which was somewhat higher than that from the Einstein Aging Study (21.5%) [6], but lower than that from the Sydney Memory and Ageing Study (39.1%) [8]. Our study also indicated that, among nondemented community-dwelling individuals aged
75 years, the MCI prevalence was 31.2%, which was comparable with that from the North Manhattan WHICAP study (32.6%) [35]. However, several studies reported MCI prevalence of ,20% among nondemented individuals aged
75 years. A study in Kolkata, India, reported an MCI prevalence of 13.7% [36]. Four studies in Europe, the Leipzig Longitudinal Study of the Aged in Germany

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

7

Table 3 Age-specific prevalence of mild cognitive impairment subtypes by gender and APOE ε4 allele Men

Women

APOE ε4 (1) Age, in years

% (95% CI)

APOE ε4 (2) P-value

% (95% CI)

APOE ε4 (1) P-value

aMCI-SD

60–69 70–79
80

8.5 (3.2–13.8) 9.0 (3.4–14.6) 9.4 (0–19.5)

0.862

11.0 (8.3–13.7) 8.8 (6.1–11.5) 10.8 (6.1–15.5)

aMCI-MD

60–69 70–79
80

1.9 (0–4.5) 5.0 (0.7–9.3) 18.8 (5.3–32.3)

0.001

1.0 (0.1–1.9) 3.8 (2.0–5.6) 6.0 (2.4–9.6)

,0.001

3.1 (0.1–6.1) 8.2 (2.7–13.7) 33.3 (19.5–47.1)

naMCI-SD

60–69 70–79
80

0.9 (0–2.7) 6.0 (1.4–10.7) 3.1 (0–9.1)

0.199

1.6 (0.5–2.7) 5.5 (3.3–7.7) 7.2 (3.3–11.1)

,0.001

3.1 (0.1–6.1) 6.2 (1.4–11.0) 8.9 (0.6–17.2)

naMCI-MD

60–69 70–79
80

0.018

0.2 (0–0.6) 0.7 (0–1.5) 1.2 (0–2.9)

0 0 6.3 (0–14.7)

0.677

% (95% CI)

0.171

4.7 (1.1–8.4) 3.1 (0–6.6) 4.4 (0–10.4)

0.8 (0–2.3) 1.0 (0–3.0) 0

APOE ε4 (2) P-value

% (95% CI)

P-value

0.841

4.9 (3.1–6.7) 5.6 (3.6–7.6) 7.0 (3.3–10.7)

0.299

,0.001

1.4 (0.4–2.4) 4.2 (2.4–6.0) 14.4 (9.4–19.4)

,0.001

0.137

2.1 (0.9–3.3) 7.7 (5.6–10.1) 10.7 (6.3–15.1)

,0.001

1.000

0.3 (0–0.8) 5.8 (3.7–7.9) 5.9 (2.5–9.3)

,0.001

Abbreviations: APOE, apolipoprotein E; aMCI-SD, amnestic mild cognitive impairment single domain; aMCI-MD, amnestic mild cognitive impairment multiple domains; naMCI-SD, non-amnestic mild cognitive impairment single domain; naMCI-MD, non-amnestic mild cognitive impairment multiple domains; CI, confidence interval.

(LEILA751) [37], the Italian Longitudinal Study on Aging [38], the German Study on Ageing, Cognition and Dementia in Primary Care Patients (AgeCoDe) [39], and the Swedish Kungsholmen Project [40], reported MCI prevalence of 19.3%, 16.0%, 15.4%, and 11.1%, respectively. In the USA, the Cardiovascular Health Study [41] and the Mayo Clinic Study of Aging [5] found MCI prevalence to be 19% and 18.2% among nondemented individuals aged
75 years, respectively. The Women Cognitive Impairment Study of Exceptional Aging determined an MCI prevalence of 23.2% among women aged
85 years of age [7], which was lower than findings from our study (53.3% among women aged
85 years). A few studies reported very low prevalence of MCI. The Eugeria Project in France reported an MCI prevalence of 3.2% among people
60 years of age [42]. A study in eastern Finland reported a prevalence of 5.3% among people aged 60–76 years [43]. With regard to the prevalence for aMCI, the Monongahela Valley Independent Elders Survey reported that 3%–4% of nondemented persons aged
65 years had this condition [44], whereas the LEILA751 reported aMCI frequency to be 3.1% among nondemented persons aged
75 years [45]. Several factors likely underlie the wide range of MCI prevalence. The variability of MCI prevalence in different study populations may be due to different neuropsychological tests being used according to the different culture. For example, the Medical Research Council Cognitive Function and Aging Study used the Geriatric Mental State Automated Geriatric Examination for Computer-Assisted Taxonomy and the Cambridge Cognitive Examination to assess and determine cognitive status. The prevalence of MCI amnestic type was reported to be 2.5% among a nondemented population aged
65 years of age [46]. In China, as well as in other

developing countries, appropriate neuropsychological tests need to be administered to those with low education. We chose to use the Modified Fuld Object Memory Evaluation and the Renminbi Test, in place of the Auditory Verbal Learning and Trial Making Tests, to evaluate domains of memory and executive function in those with ,6 years of education. The performance of those with low education was compared with norms for this group in the population with similar cutpoints (1.5 SDs below the mean). Thus, the neuropsychological battery designed in our study can be used for people with a wide range of education. Differential educational attainment is an important factor, considering the strong dependence of MCI prevalence on education. Compared with our study, most previous studies had more educated samples aged
75 years and
85 years. This may explain their lower MCI prevalence. Previous studies have demonstrated that low education is a contributory factor in the development of dementia. A meta-analyses strongly and consistently showed that those with lower education had a higher risk for dementia; the pooled odds ratio was 2.61 (95% CI 2.21–3.07) for prevalence studies and 1.88 (95% CI 1.51–2.34) for incidence studies [47]. Further studies are needed to determine whether low education is a risk factor for the transition from normal to MCI, or from MCI to dementia, or both of these in China. The prevalence can also be affected by cutoff thresholds selected for neuropsychological measures. In the current study, we utilized a 1.5-SD cutoff. Other studies used a 1.0-SD cutoff score [5,38,44], which, assuming a normal distribution of cognitive scores, should lead to an approximate doubling of the MCI prevalence when compared with a 1.5-SD cutoff [37,48]. By definition, the MCI prevalence in our study would be much higher if we

8

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

used the 1.0-SD cutoff. However, longitudinal data from the LEILA751 study suggest that the use of a 1.0-SD cutoff (vs.1.5-SD) produces higher relative predictive power as a result of a large increase in sensitivity with only a minor concurrent decrease in specificity [37]. Additional longitudinal studies will help to choose an appropriate cutoff point to balance the trade-off between higher sensitivity or higher specificity in identifying MCI in the population [48]. The number of previous epidemiologic studies of MCI and dementia in east Asia is relatively small. Recently, the 10/66 dementia research group carried out a survey in Cuba, the Dominican Republic, Peru, Mexico, Venezuela, Puerto Rico, China, and India among 15,376 individuals aged
65 years without dementia [16]. The crude prevalence for aMCI ranged from 0.8% in China to 4.3% in India. The study used immediate and delayed word recall scores from the modified CERAD 10-word list to evaluate objective memory impairment, and did not describe detailed source and recruitment procedures of the target population. A study in Japan indicated an MCI prevalence of 23.4% from among 900 people aged
65 years of age in the town of Ama-cho [11]. The study used a 2-phase study design with MMSE (cutoff score 27) and/or CDR (0.5) as the screening tool. A full evaluation of cognitive function using a set of neuropsychological tests was conducted for screen-positive subjects. The use of an MMSE cutoff score of 27 in the study, however, may miss some MCI cases who scored higher on this measure. Findings from these 2 studies from Asia are not comparable to those of the current study due to the differing designs and procedures. The reported proportion of aMCI among all types of MCI varies widely, ranging from 30% to 77% [5–7,10,11,36–39]. Our data indicate that aMCI (65%) was almost twice as common as naMCI, similar to the proportion of aMCI cases in the Mayo Clinic Study of Aging (69%) [5]. We also found that aMCI-SD (39%) was the most common subtype among MCI subtypes. However, 2 previous studies conducted in Germany reported that naMCI-SD was the most common subtype (37% in LEILA751, 57% in AgeCoDe). Those findings may partly be driven by the definition of cognitive domains based on the Structured Interview for Diagnosis of Dementia of Alzheimer Type, Multiinfarct Dementia, and Dementia of other Etiology, according to DSM-III-R, DSM-IV and International Classification of Diseases, tenth edition (ICD-10; SIDAM) [37,39]. Unlike the Mayo Clinic Study of Aging and the 2 European studies [5,38,49], we did not observe significant differences in MCI and aMCI prevalence between men and women. An interesting finding was the significantly higher prevalence of aMCI-SD in men, but significantly higher rates of the other subtypes in women. When examining the prevalence of each MCI subtype by gender and presence of APOE ε4, we found that APOE ε4 carrier women had a higher prevalence of aMCI-MD than APOE ε4 carrier men or APOE ε4 noncarriers. This pattern is

consistent with that observed for the prevalence of AD versus other types of dementia [50]. Ethnic variation of the APOE alleles is highly variable among different populations. In the current study, the frequency of the APOE ε4 allele in nondemented individuals was 9.3%, which is in the range of that in Asian (including Chinese) populations (6.3%–9.3%) [51–53], but lower than that in Caucasian and African American populations (11%–27%) [54]. Therefore, the APOE ε4 allele may exhibit a synergistic effect with gender (female) for the cognitive decline in the Chinese population. Further multivariate analyses are needed to clarify this issue. Different MCI subtypes likely differ in etiology and eventual outcome. aMCI has a high likelihood of progressing to AD, whereas non-amnestic subtypes have a higher likelihood of progressing to a non-AD dementia [33]. Studies reporting progression of the 4 MCI subtypes remain rare. Busse et al. observed that participants with aMCI-MD were more likely to progress to all types of dementia (53%), but not significantly to AD [37]. Among community-dwelling residents in northern Manhattan, those with aMCI-MD had a higher relative risk (RR) for incident AD than those with aMCI-SD (RR 5 4.3 vs. 3.2) [55]. A hospital-based study conducted by our group found that a higher proportion of aMCI-MD subjects converted to AD compared with aMCI-SD subjects (43% vs. 26%) within an average of 2 years among 130 aMCI subjects at baseline [56]. Longitudinal approaches will be needed to document incident disease, and to confirm the prognosis of the various subtypes of MCI, especially non-amnestic subtypes, in the Chinese population. The current study has some limitations. First, the 70% participation rate could result in selection bias, although it is comparable to many previous studies (62%–78%) [57]. Comparisons of participants and nonparticipants showed that those who refused to participate were older, but had comparable educational attainment. Because age is a strong risk factor for MCI, prevalence for MCI in our study may be underestimated due to lower participation rate among very old residents. Not including individuals living in nursing homes or institutions also may have led to the underestimation of prevalence. On the other hand, higher rates of hypertension, stroke, and heart disease in participants relative to nonparticipants could have led to increased MCI prevalence, particularly of a non-amnestic type. Second, although the sample size (2985 nondemented participants) in our study was comparable to or even larger than in many previous studies, when doing subgroup analyses, the sample size became relatively small, resulting in insufficient statistical power. Third, the population in Shanghai has a higher proportion of elderly (25% vs. 14%
60 years of age) [58], and higher educational attainment than in the general population in China (46% vs. 26% high school or higher education) [59]. Furthermore, living conditions in urban areas are better than those in rural areas. Thus, our study results cannot be generalized to the whole Chinese population and

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

our findings may underestimate the age-specific prevalence of MCI in China. Currently in China, conservatively estimated based on our study results, there are at least 35,000,000 elderly
60 of age living with MCI, most of whom are at risk for dementia. Population-based prospective studies in China are needed to examine incident MCI, the progression rates and risk factors for normal cognition to MCI and its subtypes, and for MCI to dementia. Acknowledgments This project was funded by the Science and Technology Committee, Shanghai, China (09DZ1950400). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors thank Dr. Minhua Shao for the technical assistance with APOE genotype assays; Zhaolan Ding, Yan Zhou, Lirong Yu, Meihua Jin, Meirong Chen, Zeya Wang, Meizheng Shi, Jingping Ye, Meiping He, Lanfang Yu, Deping Chen, Fusheng Gong, Meili Shi, Wenying Zhou, Shumin Chen, Xiudi Xu, Meiling Huang, Linghua Ding, Wenfan Zhu, Zhi Zhou, Xiaoying Liu, Fuqin Gao, Peng Gong, Lin Lu, Meng Wang, Ting Zhang, Yaru Guo, Xiaoli Jin, Shiqi Li, Qiongyi Xu, and Yiping Wang for their efforts in study coordination; and all the participants for their cooperation.

RESEARCH IN CONTEXT

1. Systematic review: We searched PubMed for literature studies reporting the prevalence of mild cognitive impairment (MCI) in different ethnic populations over the last 20 years. Previous data from China can hardly be compared to those from Western studies due to differences in design and implementation of diagnostic criteria. 2. Interpretation: This is the first epidemiologic study conducted in China that used comparable diagnostic procedures and MCI definitions to most cohort studies in developed countries. We documented the prevalence of MCI (4 subtypes) by clinically and neuropsychologically evaluating 2985 nondemented community elderly. We first validated and administered the neuropsychological tests in the Chinese language, especially those with low education. 3. Future directions: Population-based prospective studies in China are needed to: (a) examine the progression rates for normal cognition to MCI and for MCI to dementia; (b) explore the synergistic effect of various factors, such as apolipoprotein E, gender, education, etc., for cognitive decline.

9

References [1] Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment. Arch Neurol 1999;56:303–8. [2] Petersen RC, Stevens JC, Ganguli M, Tangalos EG, Cummings JL, DeKosky ST. Practice parameter: Early detection of dementia: mild cognitive impairment (an evidence- based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2001;56:1133–42. [3] Winblad B, Palmer K, Kivipelto M, Jelic V, Fratiglioni L, Wahlund LO, et al. Mild cognitive impairment—beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med 2004;256:240–6. [4] Ward A, Arrighi HM, Michels S, Cedarbaum JM. Mild cognitive impairment: Disparity of incidence and prevalence estimates. Alzheimers Dement 2012;8:14–21. [5] Petersen RC, Roberts RO, Knopman DS, Geda YE, Cha RH, Pankratz VS, et al. Prevalence of mild cognitive impairment is higher in men. The Mayo Clinic Study of Aging. Neurology 2010;75:889–97. [6] Katz MJ, Lipton RB, Hall CB, Zimmerman ME, Sanders AE, Verghese J, et al. Age-specific and sex-specific prevalence and incidence of mild cognitive impairment, dementia, and alzheimer dementia in blacks and whites. A report from the Einstein Aging Study. Alzheimers Dis Assoc Disord 2012;26:335–43. [7] Yaffe K, Middleton LE, Lui L, Spira AP, Stone K, Racine C, et al. Mild cognitive impairment, dementia, and their subtypes in oldest old women. Arch Neurol 2011;68:631–6. [8] Sachdev PS, Lipnicki DM, Crawford J, Reppermund S, Kochan NA, Trollor JN, et al. Risk profiles for mild cognitive impairment vary by age and sex: the Sydney Memory and Ageing Study. Am J Geriatr Psychiatry 2012;20:854–65. [9] Dimitrov I, Tzourio C, Milanov I, Deleva N, Traykov L. Prevalence of dementia and mild cognitive impairment in a Bulgarian urban population. Am J Alzheimers Dis Other Dement 2012;27:131–5. [10] Juarez-Cedillo T, Sanchez-Arenas R, Sanchez-Garcia S, GarciaPena C, Hsiung GYR, Sepehry AA, et al. Prevalence of mild cognitive impairment and its subtypes in the Mexican population. Dement Geriatr Cogn Disord 2012;34:271–81. [11] Wada-Isoe K, Uemura Y, Nakashita S, Yamawaki M, Tanaka K, Yamamoto M, et al. Prevalence of dementia and mild cognitive impairment in the rural island town of Ama-cho, Japan. Dement Geriatr Cogn Dis Extra 2012;2:190–9. [12] More measures sought for aiding elderly population, Chinadaily, March 6, 2012. Available at: http://usa.chinadaily.com.cn/china/ 2012-03/06/content_14763044.htm. [13] Nie H, Xu Y, Liu B, Zhang Y, Lei T, Hui X, et al. The prevalence of mild cognitive impairment about elderly population in China: a meta-analysis. Int J Geriatr Psychiatry 2011;26:558–63. [14] Yao YH, Xu RF, Tang HD, Jiang GX, Wang Y, Wang G, et al. Cognitive impairment and associated factors among the elderly in the Shanghai suburb: findings from a low-education population. Neuroepidemiology 2010;34:245–52. [15] Lu J, Li D, Li F, Zhou A, Wang F, Zuo X, et al. Montreal cognitive assessment in detecting cognitive impairment in Chinese elderly individuals: a population-based study. J Geriatr Psychiatry Neurol 2011;24:184–90. [16] Sosa1 AL, Albanese E, Stephan BCM, Dewey M, Acosta D, Ferri CP, et al. Prevalence, distribution, and impact of mild cognitive impairment in Latin America, China, and India: a 10/66 population-based study. PLoS Med 2012;9:e1001170. [17] Zung WW. A rating instrument for anxiety disorders. Psychosomatics 1971;12:371–9. [18] Hamilton M. Rating depressive patients. J Clin Psychiatry 1980; 41:21–4. [19] Richard E, Reitz C, Honig LH, Schupf N, Tang MX, Manly JJ, et al. Late-life depression, mild cognitive impairment, and dementia. JAMA Neurol 2013;70:383–9.

10

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

[20] Morris JC. The clinical dementia rating (CDR): current version and scoring rules. Neurology 1993;43:2412–4. [21] Lim WS, Chong MS, Sahadevan S. Utility of the clinical dementia rating in Asian populations. Clin Med Res 2007;5:61–70. [22] Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist 1969; 9:179–86. [23] Smirnov DA, Morley M, Shin E, Spielman RS, Cheung VG. Genetic analysis of radiation-induced changes in human gene expression. Nature 2009;459:587–91. [24] Zhang M, Katzman R, Salmon D, Jin H, Cai G, Wang Z, et al. The prevalence of dementia and Alzheimer’s disease in Shanghai, China: impact of age, gender, and education. Ann Neurol 1990;27:428–37. [25] Dubois B, Slachevsky A, Litvan I, Pillon B. The FAB: a frontal assessment battery at bedside. Neurology 2000;55:1621–6. [26] Lezak MD, Howieson DB, Loring DW, Howieson DB, Fischer JS, eds. In: Neuropsychological assessment. 4th ed. New York: Oxford University Press; 2004. p. 562–3. [27] Lezak MD, Howieson DB, Loring DW, Howieson DB, Fischer JS, eds. In: Neuropsychological assessment. 4th ed. New York: Oxford University Press; 2004. p . 583. [28] Guo Q, Zhao Q, Chen M, Ding D, Hong Z. A comparison study of mild cognitive impairment with 3 memory tests among Chinese individuals. Alzheimers Dis Assoc Disord 2009;23:253–9. [29] Rideaux T, Beaudreau SA, Fernandez S, O’Hara R. Utility of the abbreviated Fuld Object Memory Evaluation and MMSE for detection of dementia and cognitive impairment not dementia in diverse ethnic groups. J Alzheimers Dis 2012;31:371–86. [30] Lu J, Guo Q, Hong Z, Shi W, Lu C. Trail making test used by Chinese elderly patients with mild cognitive impairment and mild Alzheimer’s dementia. Chinese J Clin Psychol 2006;14:118–21. [31] Carnero-Pardo C, Gurpegui M, Sanchez-Cantalejo E, Frank A, Mola S, Barquero MS, et al. Diagnostic accuracy of the Eurotest for dementia: a naturalistic, multicenter phase II study. BMC Neurol 2006;6:15. [32] Diagnostic and statistical manual of mental disorders 4th ed. Washington, DC: American Psychiatric Association; 1994. p. 143–7. [33] Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med 2004;25:183–94. [34] Petersen RC. Early diagnosis of Alzheimer’s disease: is MCI too late? Curr Alzheimer Res 2009;6:324–30. [35] Manly JJ, Bell-McGinty S, Tang MX, Schupf N, Stern Y, Mayeux R. Implementing diagnostic criteria and estimating frequency of mild cognitive impairment in an urban community. Arch Neurol 2005; 62:1739–46. [36] Das SK, Bose P, Biswas A, Dutt A, Banerjee TK, Hazra A, et al. An epidemiologic study of mild cognitive impairment in Kolkata, India. Neurology 2007;68:2019–26. [37] Busse A, Hensel A, Guhne U, Angermeyer MC, Riedel-Heller SG. Mild cognitive impairment: long-term course of four clinical subtypes. Neurology 2006;67:2176–85. [38] Di Carlo A, Lamassa M, Baldereschi M, Inzitari M, Scafato E, Farchi G, et al. CIND and MCI in the Italian elderly: frequency, vascular risk factors, progression to dementia. Neurology 2007; 68:1909–16. [39] Luck T, Riedel-Heller SG, Kaduszkiewicz H, Bickel H, Jessen F, Pentzek M, et al. Mild cognitive impairment in general practice: age-specific prevalence and correlate results from the German study on ageing, cognition and dementia in primary care patients (AgeCoDe). Dement Geriatr Cogn Disord 2007;24:307–16. [40] Palmer K, Backman L, Winblad B, Fratiglioni L. Mild cognitive impairment in the general population: occurrence and progression to Alzheimer’s disease. Am J Geriatr Psychiatry 2008;16:603–11.

[41] Lopez OL, Jagust WJ, DeKosky ST, Becker JT, Fitzpatrick A, Dulberg C, et al. Prevalence and classification of mild cognitive impairment in the Cardiovascular Health Study Cognition Study: part 1. Arch Neurol 2003;60:1385–9. [42] Ritchie K, Artero S, Touchon J. Classification criteria for mild cognitive impairment: a population-based validation study. Neurology 2001;56:37–42. [43] Hanninen T, Hallikainen M, Tumainen S, Vanhanen M, Soininen H. Prevalence of mild cognitive impairment: a population-based study in elderly subjects. Acta Neurol Scand 2002;106:148–54. [44] Ganguli M, Dodge HH, Shen C, DeKosky ST. Mild cognitive impairment, amnestic type: an epidemiologic study. Neurology 2004; 63:115–21. [45] Busse A, Bischkopf J, Riedel-Heller SG, Angermeyer MC. Mild cognitive impairment: prevalence and predictive validity according to current approaches. Acta Neurol Scand 2003;108:71–81. [46] Stephan BM, Matthews FE, McKeith IG, Bond J, Brayne C. Medical Research Council Cognitive Function and Aging Study. J Am Geriatr Soc 2007;55:1534–40. [47] Meng X, D’Arcy C. Education and dementia in the context of the cognitive reserve hypothesis: a systematic review with metaanalyses and qualitative analyses. PLoS One 2012;7:e38268. [48] Trittschuh EH, Crane PK, Larson EB, Cholerton B, McCormick WC, McCurry SM, et al. Effects of varying diagnostic criteria on prevalence of mild cognitive impairment in a community based sample. J Alzheimers Dis 2011;25:163–73. [49] Larrieu S, Letenneur L, Orgogozo JM, Fabrigoule C, Amieva H, Le Carret N, et al. Incidence and outcome of mild cognitive impairment in a population-based prospective cohort. Neurology 2002;59:1594–9. [50] Damoiseaux JS, Seeley WW, Zhou J, Shirer WR, Coppola G, Karydas A, et al. Gender modulates the APOE ε4 effect in healthy older adults: convergent evidence from functional brain connectivity and spinal fluid tau levels. J Neurosci 2012;32:8254–62. [51] Hu P, Qing YH, Jing CX, Lu L, Hu B, Du PF. Does the geographical gradient of ApoE4 allele exist in China? A systemic comparison among multiple Chinese populations. Mol Biol Rep 2011;38:489–94. [52] Liang S, Pan M, Geng HH, Chen H, Gu LQ, Qin XT, et al. Apolipoprotein E polymorphism in normal Han Chinese population: frequency and effect on lipid parameters. Mol Biol Rep 2009;36:1251–6. [53] Katzman R, Zhang MY, Chen PJ, Gu N, Jiang S, Saitoh T, et al. Effects of apolipoprotein E on dementia and aging in the Shanghai Survey of Dementia. Neurology 1997;49:779–85. [54] Jun G, Naj A, Beecham GW, Wang LS, Buros J, Gallins PJ, et al. Metaanalysis confirms CR1, CLU, and PICALM as Alzheimer’s disease risk loci and reveals interactions with APOE genotypes. Arch Neurol 2010;67:1473–84. [55] Manly JJ, Tang MX, Schupf N, Stern Y, Vonsattel JP, Mayeux R. Frequency and course of mild cognitive impairment in a multiethnic community. Ann Neurol 2008;63:494–506. [56] Chen MR, Guo QH, Zhou Y, Zhao Q, Ding D, Hong Z. Longitudinal study of two subtypes of mild cognitive impairment. Chin J Neurol 2010;5:351–3. [57] Roberts RO, Geda YE, Knopman DS, Cha RH, Pankratz VS, Boeve BF, et al. The Mayo Clinic Study of Aging: design and sampling, participation, baseline measures and sample characteristics. Neuroepidemiology 2008;30:58–69. [58] Shanghai is growing old. People’s daily online, China Society, March 30, 2012. Available at: http://english.peopledaily.com.cn/90882/ 7774017.html. [59] Comparison of population with various education attainments per 100,000 persons by region. China Statistical yearbook, 2011. Available at: http://www.stats.gov.cn/tjsj/ndsj/2011/html/D0310e.htm.

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

10.e1

Table E1 The normative data of neuropsychological test scores from a validation pilot study Education
6 years Neuropsychological tests

Domain designation

Conflicting Instructions Task* Conflicting correct tapping

Executive function

Conflicting performance time (s)

Go/No-Go correct tapping

Go/No-Go performance time (s)

Stick Test* Imitate

Rotate

Category naming

Recognition

0.8 1.1 1.9 4.9 6.7 7.5 1.8 2.7 3.7 5.1 7.2 8.8

1.2 1.7 2.9 7.4 10.1 11.3 2.7 4.1 5.6 7.7 10.8 13.2

18.2 18.3 17.4 39.2 42.6 43.1 16.1 16.3 14.2 35.3 38.7 41.5

2.9 3.0 4.5 10.3 10.5 9.2 5.0 3.9 5.4 11.0 10.1 9.2

4.4 4.5 6.8 15.5 15.8 13.8 7.5 5.9 8.1 16.5 15.2 13.8

10.0 9.9 9.8 3.8 3.0 2.1 4.3 3.3 2.3

0.2 0.4 0.5 1.8 1.7 1.6 2.4 2.2 1.9

0.3 0.6 0.8 2.7 2.6 2.4 3.6 3.3 2.9

9.7 9.7 9.6 2.3 2.2 1.5 2.4 2.1 1.7

0.7 0.8 0.9 1.2 1.5 1.1 2.0 1.7 1.7

1.1 1.2 1.4 1.8 2.3 1.7 3.0 2.6 2.6

60–69 70–79
80 60–69 70–79
80 60–69 70–79
80

40.6 39.3 37.7 40.0 39.0 36.4 5.7 5.3 4.6

2.0 2.9 4.0 2.4 4.4 9.0 1.3 1.4 1.7

3.0 4.4 6.0 3.6 6.6 13.5 2.0 2.1 2.6

38.6 36.2 35.3 36.4 33.4 28.8 3.8 4.0 3.1

2.3 4.0 6.0 8.5 11.3 15.2 1.5 1.9 2.1

3.5 6.0 9.0 12.8 17.0 22.8 2.3 2.9 3.2

60–69 70–79
80 60–69 70–79
80 60–69 70–79
80

6.9 5.9 4.7 6.4 5.4 4.2 21.9 21.0 19.7

2.3 2.4 2.4 2.5 2.6 2.5 2.1 2.6 3.3

3.5 3.6 3.6 3.8 3.9 3.8 3.2 3.9 5.0 9.2 9.2 8.5 9.7 9.2 8.4 17.3 18.1 17.6

2.1 1.6 2.2 2.2 1.7 2.2 6.1 5.9 6.1

3.2 2.4 3.3 3.3 2.6 3.3 9.2 8.9 9.2

60–69 70–79
80 60–69 70–79
80 60–69 70–79
80 60–69 70–79
80

19.8 19.6 19.3 29.2 32.3 34.5 19.0 18.2 17.2 27.3 30.7 34.8

60–69 70–79
80 60–69 70–79
80 60–69 70–79
80

60–69 70–79
80 60–69 70–79
80 60–69 70–79
80

Recognition

Performance time of Test B

1.5 SDs

Memory

Delayed recall

Trail Making Test A/By Performance time of Test A

1.0 SD

1.0 SD

Memory

Delayed recall

Modified Fuld Object Memory Evaluationz Immediate recall

Mean

Mean

Language, executive function

Categorization

Auditory Verbal Learning Testy Immediate recall

1.5 SDs

Age, yearsx

Visuospatial skill, memory

Recall

Modified Common Objects Sorting Test* Item naming

Education ,6 years

Attention, executive function 60–69 70–79
80 60–69 70–79

47.2 58.5 71.4 123.1 147.0

14.3 20.8 29.3 39.3 50.1

21.5 31.2 44.0 59.0 75.2 (Continued )

10.e2

D. Ding et al. / Alzheimer’s & Dementia - (2014) 1–10

Table E1 Continued Education
6 years Neuropsychological tests

Domain designation

Renminbi Testz Naming

Language, attention, memory

Calculation

Recall

Education ,6 years

Age, yearsx

Mean

1.0 SD

1.5 SDs


80

172.0

59.2

88.8

60–69 70–79
80 60–69 70–79
80 60–69 70–79
80

Mean

1.0 SD

1.5 SDs

4.0 4.0 3.9 5.1 5.2 4.9 9.0 9.4 8.8

0.2 0.2 0.4 1.4 1.3 1.6 2.3 1.1 1.8

0.3 0.3 0.6 2.1 2.0 2.4 3.5 1.7 2.7

Abbreviation: SD, standard deviation. *Tests for individuals for any education level. y Tests for individuals with
6 years of education. z Tests for individuals with ,6 years of education. x Numbers of subjects aged 60–69, 70–79, and 80–89 years with education
6 years are 516, 252, and 188, respectively. Numbers of subjects aged 60–69, 70–79, and 80–89 years with education ,6 years are 26, 169, and 119, respectively.