The prevalence of dementia in the People s Republic of China: a systematic analysis of studies

Age and Ageing 2007; 36: 619 624 The Author 2007. Published by Oxford University Press on behalf of the British Geriatrics Society. doi:10.1093/ageing/afm128 All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org Published electronically 25 October 2007 SYSTEMATIC REVIEW The prevalence of dementia in the People s Republic of China: a systematic analysis of 1980 2004 studies MENG-JIE DONG 1,BIN PENG 2, XIANG-TONG LIN 1,JUN ZHAO 1,YAN-RONG ZHOU 2,RUN-HUA WANG 2 1 Department of Nuclear Medicine, Huashan Hospital, Fudan University, Shanghai 200040, China 2 Department of Medical Statistics, Chongqing Medical University, Chongqing 400016, China Address correspondence to: Xiang-Tong Lin. Fax : +86-21-62497591. Email: xtlin@81890.net Abstract Background dementia has been a major public health problem. However, there has not yet been a nationwide investigation or systematic analysis of the prevalence of dementia in China from 1980 to 2004. Objectives the aim of this study was to analyse the prevalence of dementia and its major subtypes [Alzheimer disease (AD), vascular dementia (VD)] among the population aged 60 years and older in China from 1980 to 2004. Methods epidemiological investigations on dementia in China published in journals and covering the period from 1980 to 2004 were identified manually and on-line by using CBMDISK, Chongqing VIP database and CNKI database. Those reported in English journals were identified using MEDLINE. Selected studies had to describe an original study defined by strict screening and diagnostic criteria. The fixed effects model or random effects model was employed according to statistical test for homogeneity. Results twenty-five studies were selected, the statistical information of which was collected for systematic analysis. Our results showed that AD and VD were the two major subtypes of dementia in China, and the pooled prevalence of AD and VD for the population aged 60 years and older was 1.6 and 0.8%, respectively. There was a higher prevalence of AD in the illiterate elderly population (3.2%) than in those who received years of education. The chronological prevalence of AD increased significantly from 1980 to 2004. In southern and northern China, the prevalence of AD was 2.0 and 1.2%, respectively, while VD was 0.6 and 1.1%, respectively. Conclusions in the last 24 years, AD and VD were the two major subtypes of dementia in China. The prevalence of AD may be affected by sex, education, occupation or age. The prevalence of VD, which was higher in northern than in southern China, seems not to be affected by age, sex or education. Keywords: dementia, Alzheimer disease, vascular dementia, prevalence, systematic analysis, China, elderly Introduction For decades, dementia has been a major public health problem in developed countries, and the prevalence rate of dementia among the elderly (aged 65 years and over) in western developed countries has been reported to be around 5 6% [1, 2, 3, 4]. A recent study has shown that agestandardised prevalence is 6.4% for dementia of all types, 4.4% for Alzheimer disease (AD) and 1.6% for vascular dementia (VD); the prevalence of dementia increases with age, which is 0.8% in the 65 69-year-old groups, and 28.5% in those 90 years and older [5]. What is the characteristic of prevalence of dementia in China? Although there are several dementia-related literature offerings, there has not been a nationwide investigation or systematic analysis of the prevalence of dementia in China since 1980. Moreover, China is the largest developing country in the world, with its population group of over 60 years old reaching more than one hundred million, which accounts for 10% of the total population. Therefore, an analysis of the prevalence of dementia in China is of great significance. The aim of the present study was to perform a systematic analysis of published epidemiological studies in the elderly Chinese population aged 60 years and older from 1980 to 2004. 619

M. J. Dong et al. Methods Search strategy Studies were identified in the following electronic databases: CBMDISK, Chongqing VIP, CNKI and MEDLINE, from 1980 to 2004, using the terms dementia, Alzheimer disease, vascular dementia and prevalence. The study did not include epidemiological studies in the areas of Hong Kong, Macao and Taiwan because these data were not easily accessible. As a result, a bibliography of references on dementia in the population in question was established, which covered a total of 58 published studies. Study inclusion criteria In order to meet the analysis requirements and reduce deviation, selected studies fulfilled the following criteria: (i) case collection based on field survey; (ii) the study based on population samples rather than volunteers; (iii) the population studied in most investigations was 60 years and older, and there were two studies with a younger inclusion age of 50 and 55 years old respectively (in our study, data from participants below 60 years of age were omitted); (iv) case collection was conducted in two stages: the screening phase and the diagnostic phase. Trained interviewers, who were mainly health workers and nurses, performed the screening survey. All samples were screened with brief cognitive tests (Mini Mental State Examination (MMSE), Hasegawa dementia scale (HDS) or Blessed Dementia Scale (BDS). Screen positives were scored below a given cutoff point on one or two screening tests, or were clinically suspected of dementia. Suspected cases of dementia were further examined by senior physicians with the aid of laboratory tests, clinical features and neuropsychological tests. Of 58 studies, 25 studies were selected in this investigation while 34 were excluded mainly because they were not original descriptions of population prevalence. Clinical diagnostic criteria employed in the 25 studies included DSM-III-R [10 23, 25 30, 32 33] and ICD-10 [24, 31, 34]. NINCDS-ADRDA [10 34] was used to diagnose possible and probable AD, and the Hachinski Ischaemia Score [15, 16, 18 20, 23, 25, 26, 28, 30, 33] and NINDS-AIREN [24, 29, 31, 34] were used to differentiate AD and VD. For diagnostic criteria of dementia with Lewy bodies (DLB): dementia was first diagnosed by the screening phase, followed by the diagnostic phase previously introduced; the clinical feature included fluctuating cognitive impairment, episodes of confusion and Parkinsonism with some of the visual signs and symptoms of Parkinson s disease. Quality of the studies We accessed the quality of studies using the framework suggested by the Cochrane Collaboration [6]. For the inclusion decision, quality assessment was carried out independently by at least two reviewers. In the case of disagreement, data were reviewed by two other authors. The data from all included studies were clearly tabulated, and deviations were taken into account and identified during the quality assessment stage. Data analysis We used a published systematic analysis technique to calculate the pooled prevalence of dementia from all eligible studies [7 9]. (Full details are in supplementary data available at http://www.ageing.oxfordjournals.org/.) Results The 25 studies covered 12 provinces and municipalities including the area of Shanghai (6), Fujian (1), Sichuan (2), Hunan (1), Jiangsu (1), Anhui (1), Chongqing (1), Guangdong (1), Beijing (8), Inner Mongolia (1), Shanxi (1) and Xinjiang (1), which published in 16 different journals. The total population of 87,761 aged 60 years and older was investigated, and 2,437 cases of dementia were selected. Among these, AD numbered 1,525; VD, 663; mixed dementia, 58; alcohol dementia, 19; Parkinson, 32; Trauma, 10; Lewy Bodies, 4; others, 126. The prevalence of dementia, AD and VD was calculated for each study (Table 1). Prevalence of different pathologic subtypes of dementia The combined result for the pooled prevalence of dementia in the elderly population aged 60 years and older was 2.8%, 95% CI = 2.5 3.1%. AD and VD were the two major subtypes of dementia in China, as the proportions of AD and VD were 54.3 and 25.5%, respectively. The odds ratio of AD to VD was 2.2, 95% CI = 1.6 3.6. The prevalence of different pathologic subtypes of dementia was also analysed (Table 2). Relationship between prevalence of dementia and sex, education, occupation and residence The prevalence of dementia as related to sex, education, occupation and residence was analysed. (Please see Appendix 2 and Appendix 3 in the supplementary data on the Journal website http://www.ageing.oxfordjournals.org.) Chronological prevalence of dementia According to the study years selected for our investigation, data were classified into four groups and then the homogeneity test was performed. Results showed that the data did not meet homogeneity (Table 3). Compared with VD, the prevalence of AD increased significantly from 1980 to 2004. The odds ratios of AD to VD for years 1985 90, 1991 95, 1996 2000 and 2001 04 were 1.1 (95% CI = 0.9 1.4), 1.8 (95% CI = 1.4 2.2), 3.2 (95% CI = 2.8 3.7) and 3.0 (95% CI = 2.3 3.7), respectively, demonstrating an increased tendency as time progressed. 620

A study of dementia in the People s Republic of China Table 1. Twenty-five studies characteristic of the prevalence of dementia included in the studies First Study design Prevalence (%) author & year Subjects no. Urban Alzheimer Vascular Dementia published Location Survey date (male/female) /rural disease dementia... Chen et al., Beijing 1986 8,740 (4,055/4,685) C 0.4 (0.3 0.5) 0.4 (0.3 0.6) 1.3 (1.1 1.5) 1987 [10] Gao et al., Beijing Apr 1986 Jun 1986 906 (409/497) C 1.0 (0.5 1.7) 2.7 (1.7 3.7) 3.9 (2.7 5.1) 1989 [11] Li et al., Beijing 1986 1,090 (505/585) C 0.2 (0.1 0.7) 0.7 (0.3 1.3) 1.3 (0.7 2.0) 1989 [12] Zhang et al., Shanghai 1989 5,055 C 2.0 (1.7 2.4) 0.8 (0.6 1.1) 3.2 (2.7 3.7) 1989 [13] Chen et al., Beijing Jan 1986 May 1986 5,172 (2,385/2,787) C 0.2 (0.1 0.3) 0.5 (0.3 0.7) 0.8 (0.5 1.0) 1992 [14] Gao et al., Shanghai Oct 1990 Dec 1990 3,779 (1,499/2,280) C + R 3.1 (2.6 3.7) 0.9 (0.6 1.4) 4.2 (3.6 4.8) 1993 [15] Mao et al., Fujian Jun 1992 Sep 1992 1,982 (787/1,195) C + R 1.5 (1.1 2.0) 1.1 (0.7 1.3) 2.9 (2.2 3.6) 1993 [16] Wang et al., Shanghai 1991 1,515 (610/905) C 1.8 (1.2 2.5) 0.3 (0.1 0.6) 2.2 (1.5 2.9) 1995 [17] Xue et al., Guangdong 1991 93 3,285 (1,579/1,706) C + R 0.7 (0.4 1.0) 0.6 (0.3 0.8) 1.5 (1.1 1.9) 1997 [18] Li et al., Beijing Nov 1994 May 1995 1,027 (372/655) C 3.7 (2.6 4.9) 2.4 (1.6 3.4) 6.1 (4.7 7.6) 1997 [19] Zhang et al., Beijing Dec 1995 Dec 1996 1,243 (760/483) C 1.4 (0.8 2.1) 1.0 (0.5 1.6) 2.3 (1.6 3.2) 1998 [20] Chen et al., Xinjiang Jan 1995 Mar 1995 2,687 (1,380/1,307) C 0.4 (0.2 0.7) 1.0 (0.6 1.3) 1.8 (1.3 2.3) 1998 [21] Tang et al., Sichuan 1994 5,987 (2,653/3,334) R 1.4 (1.2 1.7) 0.2 (0.1 0.3) 1.7 (1.4 2.0) 1999 [22] Xiao et al., Hu nan Jun 1997 May 1998 3,287 (1,551/1,736) C + R 1.4 (1.1 1.8) 0.8 (0.5 1.1) 2.6 (2.1 3.1) 1999 [23] Li, et al., Beijing 1997 1,593 (697/896) C 0.4 (0.2 2.0) 0.9 (0.5 1.5) 2.5 (1.8 3.3) 1999 [24] Wang et al., Anhui Dec 1997 2,749 (998/1,751) C + R 4.0 (3.3 4.7) 1.0 (0.7 1.4) 5.3 (4.5 6.1) 1999 [25] Fan et al., Jingsu May 1999 Jun 1999 3,268 (1,095/1,044) C + R 1.0 (0.7 1.3) 0.5 (0.3 0.7) 1.5 (1.1 1.9) 2000 [26] Zhang et al., Shanghai May 1999 Feb 1999 1,186 (597/598) C + R 0.9 (0.5 1.5) 0.4 (0.1 0.9) 2.78 (1.9 3.7) 2000 [27] Tang et al., Sichuan Jun 1997 Apr 1998 5,353 (2,552/2,801) C + R 2.1 (1.7 2.4) 0.4 (0.2 0.5) 2.67 (2.3 3.1) 2001 [28] Zhou et al., Shanghai 1997 15,910 (7,180/8,730) C + R 2.2 (2.0 2.4) 0.7 (0.5 0.8) 3.2 (2.9 3.4) 2001 [29] Qu et al., Shanxi Sep 1997 Feb 1998 4,850 (2,040/2,810) C + R 2.1 (1.7 2.4) 1.1 (0.8 1.4) 3.53 (3.0 4.0) 2001 [30] Zhang et al., Shanghai Feb 1999 May 1999 1,186 (597/589) C + R 0.8 (0.4 1.3) 0.4 (0.1 0.9) 2.78 (1.9 3.7) 2001 [31] Tang et al., Beijing 1996 2,788 (1,356/1,432) C + R 5.0 (4.2 5.8) 1.54 (1.1 2.0) 7.5 (6.5 8.4) 2002 [32] Zou et al., Chongqing 2001.3 2001.5 1,519 (2,107/2,654) C 4.8 (3.8 5.8) 0.6 (0.3 1.0) 5.7 (4.6 6.9) 2002 [33] Li et al., 2003 [34] Inner Mongolia Jul 2002 Aug 2002 2,324 (1,846/478) C 1.3 (0.9 1.8) 1.3 (0.9 1.8) 2.7 (2.1 3.3) Figures in parentheses are 95% CIs; C, urban; R, rural; no, number. Prevalence of dementia in terms of age After age 60, the prevalence of dementia increased progressively with age and peaked in the group aged 90 years and older (Table 4); the prevalence of AD showed an almost identical tendency. At each 5-year increase in age, the prevalence of AD increased by 1.9, 3.6, 1.7, 1.9, 3.2 and 1.6 times respectively. Meanwhile, females had a consistently higher prevalence rate than males in every age group. In 60 80-year-olds, the prevalence of VD showed an increasing tendency, but decreased after age 80. 621

M. J. Dong et al. Table 2. The prevalence of different pathologic subtypes of dementia in northern and southern China (%) Alzheimer Vascular disease dementia Mixed Parkinson Alcohol Hurt Lewy bodies Others... Northern China 1.2 1.1 0.25 0.09 0.03 0.05 0.0 0.20 (1.2 1.4) (0.8 1.1) (0.18 0.41) (0.02 0.19) (0.01 0.07) (0.00 0.11) (0 0.05) (0.22 0.37) Southern China 2.0 0.6 0.11 0.06 0.07 0.02 0.03 0.14 (1.7 2.2) (0.4 0.7) (0.10 0.27) (0.06 0.12) (0.04 0.10) (0.01 0.04) (0.01 0.08) (0.11 0.22) Total 1.6 0.8 0.25 0.07 0.05 0.03 0.03 0.16 (1.0 2.7) (0.7 0.8) (0.18 0.29) (0.06 0.11) (0.03 0.08) (0.01 0.04) (0.01 0.04) (0.18 0.27) Figures in parentheses are 95% CIs. Table 3. The chronological prevalence and 95% CI of dementia in China Dementia Alzheimer disease Vascular dementia Year Case Population Prevalence (%) Case Population Prevalence (%) Case Population Prevalence (%)... 1985 90 170 24,742 2.1 (1.8 2.2) 277 24,742 0.9 (0.9 1.2) 170 24,742 0.8 (0.6 0.9) 1991 95 118 17,726 2.4 (2.1 2.6) 231 17,696 1.4 (1.1 1.5) 118 17,726 0.8 (0.6 0. 8) 1996 2000 181 26,312 3.1 (2.1 3.3) 913 41,450 2.1 (1.8 2.6) 309 41,450 0.7 (0.7 0.8) 2001 04 40 3,843 4.0 (3.3 4.4) 104 3,843 2.7 (2.2 3.2) 40 3,843 1.0 (0.7 1.4) Figures in parentheses are 95% CIs. Prevalence differences between northern and southern China In order to find out the geographical characteristics of prevalence of dementia in China, we only analysed the prevalence differences between northern and southern China due to the limited information. (Full details are given in Appendix 3 and Appendix 6 available at http://www.ageing.oxfordjournals.org/.) Discussion In this systematic analysis, 25 epidemiological studies were selected that covered the areas of 12 provinces and municipalities for examination of dementia prevalence in China from 1980 to 2004. This study of Chinese populationbased subject samples over 60 years of age confirmed that dementia was a common medical condition, particularly in the elderly. The pooled prevalence of AD and VD for the Chinese population aged 60 years and older was 1.6 and 0.8%, respectively, similar to that previously reported by Liu et al.[35], but lower than that reported over the last two decades in Europe [5]. We also confirmed that AD and VD are the two major subtypes of dementia in China, as the proportions of AD and VD were 54.3 and 25.5%, respectively. The proportion of AD and VD in dementia was somewhat different from that in Europe and other Asian Table 4. The prevalence of dementia (%) in terms of age in China from 1980 to 2004 Alzheimer disease Vascular dementia Total dementia Age Male Female Total Male Female Total Male Female Total... 60 64 0.3 0.4 0.3 0.4 0.3 0.4 0.7 0.7 0.7 (0.1 0.5) (0.2 0.7) (0.2 0.4) (0.1 1.2) (0.0 1.1) (0.3 0.7) (0.4 0.9) (0.4 1.0) (0.5 0.8) 65 69 0.3 0.5 0.5 0.9 2.0 0.6 1.2 1.3 1.4 (0.2 0.7) (0.4 1.0) (0.4 0.7) (0.6 1.4) (0.4 1.6) (0.4 0.2) (0.9 1.5) (1.0 1.8) (1.3 1.8) 70 74 0.9 1.8 1.7 0.4 0.0 0.7 2.1 2.9 4.2 (0.6 1.5) (1.4 2.5) (1.6 2.2) (0.1 1.4) (0.0 0.9) (0.5 0.9) (1.4 2.4) (2.2 3.5) (2.8 3.5) 75 79 2.3 4.4 2.8 0.6 1.9 1.3 4.8 6.6 4.9 (1.6 2.9) (3.5 5.2) (2.7 3.5) (0.3 1.2) (1.2 3.1) (0.9 1.7) (3.3 4.8) (5.2 7.2) (4.7 5.7) 80 84 3.8 11.0 5.3 1.8 1.1 2.1 5.0 10.4 8.4 (2.9 5.9) (9.0 12.4) (5.0 6.3) (0.7 2.8) (0.4 1.8) (1.5 2.4) (2.9 4.7) (9.1 12.0) (7.0 8.5) 85 89 10.6 23.4 15.0 0.0 0.4 1.7 14.9 25.2 19.5 (7.6 14.1) (19.3 26.8) (13.8 17.7) (0.0 1.3) (0.0 1.3) (0.4 1.6) (10.5 17.7) (22.4 29.7) (18.9 23.5) 90 21.1 31.8 21.4 0.0 2.7 2.0 23.3 34.0 25.7 (9.2 30.4) (21.9 38.5) (16.1 25.5) (0.0 12.8) (0.1 12.8) (0.1 9.5) (11.2 33.3) (23.6 40.7) (21.3 34.2) Figures in parentheses are 95% CIs. 622

A study of dementia in the People s Republic of China countries. For example, VD tends to be more common than AD in Japan and Russia [36, 37]. Several factors such as environmental interactions, ethnicity and genetics (APOE-e4) may explain these differences [38, 39, 40, 41]. In China, the elderly tend to live with their children and are taken care of by their children. Studies have also revealed the possibility that genetic factors may explain these differences. It has been reported that the proportion of the western individuals possessing the apolipoprotein *E4 allele (9.0 16.5%) is higher than their Chinese counterpart (8.1 11.1%) [42, 43]. The findings presented in this paper confirmed the remarkable increase in prevalence of dementia with age in China, further demonstrating that the age-dependent increase was more prominent for AD than VD, which is identical to findings in the international literature [44, 45]. Consistent with the report of Jorm et al.[4], an almost exponential relationship between age and prevalence of AD, which increased significantly with each 5-year increase in age, was also found in this present study. However, Sulkava et al.[46] reported that the prevalence of AD was similar among the young elderly (aged 65 74) and increased significantly after 75 years of age. Our study also showed that the prevalence of AD was higher in women than men within the same age category, but there was no significant difference for VD. The results of several meta-analyses also showed that the prevalence of AD was higher in women than men among the senior population [35, 44]. However, in the Kungsholmen Project in Sweden, there were no observed sex differences found in AD prevalence among those under the age of 85. Among those over 85, women exhibited a higher prevalence of AD [46]. In the Eurodem pooled analysis of prevalence of dementia, men under 75 years of age had a slightly higher prevalence, while women over the age of 75 had a higher prevalence [47]. Importantly, we found that chronological prevalence of AD increased significantly from 1980 to 2004, and the gap of prevalence of dementia between China and the Western world has been shrinking. Changes in lifestyle, nutrition status and health care along with rapid economic growth in China may have had a great impact on the health of the elderly. Li et al. [48] have also proven the increase in prevalence of dementia during the past 10 years by conducting a similar epidemiological survey in the same district of Beijing. There was a trend found in the present study that the more years of education there were, the lower the prevalence rate of AD was. The prevalence of AD was 3.2% in the illiterate group, which was higher than in those receiving an education, although there was no relationship between the prevalence of VD and the amount of education. The state of not having enough education (deficiency in education) might in some way cause AD or influence the causal factor. Education may provide protection or reserve against clinical manifestation of AD, as supported by two studies of regional cerebral blood flow (rcbf) in AD [49, 50]. In China, occupation, cultural activity and socioeconomic status differ greatly between rural and urban areas as well as between northern and southern China. Although there was a higher prevalence of dementia in rural areas, there was no significant difference in the prevalence of AD or VD (Table Appendix 3). Therefore, we think that the associated effect of cultural activity and socioeconomic status on the prevalence of AD and VD between rural and urban areas is weak. It is interesting that there was a higher prevalence of AD in southern China than in northern China, and necessitates further exploration. The prevalence of VD in northern China was higher than in southern China, which may be attributed to a higher prevalence of cardiovascular disease in northern China. Different occupations had different prevalence of AD. The prevalence of AD in the housewife population was 4.4%, the highest among all occupations examined. The occupational category is concomitantly indicative of exposure to chemicals and/or other environmental factors, a reflection of psychosocial characteristics (particularly cognitive stimulation throughout life) and a socioeconomic indicator of medical care. Some authors also have shown that a relationship exists between occupation and prevalence of dementia, and found that the prevalence of AD was higher in some occupations [51 53]. One possible explanation is that women have less chance of receiving an education than meninchina. We do not believe that methodology problems have seriously biased the main results reported here. Although the studies were not designed with identical objectives or methods, they were all carried out in large, representative population samples; the studies included were performed by screening and diagnostic procedures that have been shown to be highly reliable [5, 54]; suspected cases were confirmed by further clinical observation or laboratory tests. Key points AD and VD were the two major subtypes of dementia in China, as the proportions of AD and VD were 54.3 and 25.5%, respectively, and prevalence was 1.6 and 0.8%, respectively. The great chronological change in prevalence of dementia or AD has taken place in China since 1980. The analysis confirmed the remarkable increase of prevalence of dementia or AD with age in China. Conflicts of interest None Supplementary data Supplementary data for this article are available online at http://ageing.oxfordjournals.org. 623

M. J. Dong et al. References PLEASE NOTE: The extensive list of references supporting this review has meant that only the most important have been listed and are represented by bold type throughout the text. The full list of references is available on the Journal website (www.ageing.oxfordjournals.org) as Appendix 5. 1. Plum F. Dementia: an approaching epidemic. Nature 1979; 279: 372 3. 2. Evans DA, Funkenstein HH, Albert MS et al. Prevalenceof Alzheimer s disease in a community population of older person: higher than previous reported. J Am Med Assoc 1989; 262: 2551 6. 3. Hofman A, Rocca WA, Brayne C et al. The prevalence of dementia in Europe: a collaborative study of 1980 1990 findings. Int J Epidemiol 1991; 20: 736 48. 4. Jorm AF, Korten AE, Handeson AS et al. The prevalence of dementia: a quantitative integration of the literature. Acta Psychiatr Scand 1987; 6: 465 79. 5. Lobo A, Launer LJ, Fratiglioni L et al. Prevalence of dementia and major subtypes in Europe: A collaborative study of population-based cohorts. Neurology 2000; 54: S4 9. 6. Mulrow CD, Oxman AD eds. Cochrane collaboration handbook [1997]. In The Cochrane Library, Issue 1. Oxford Update Software 1999. 7. Patel JK, Kapadia CH, Owen DB. Handbook of Statistical Distribution. New York, Basel Marcle Dekker, Inc 1976. 8. Friedl MC, Draijer N, de Jonge P. Prevalence of dissociative disorders in psychiatric in-patients: the impact of study characteristics. Acta Psychiatr Scand 2000; 102: 423 8. 9. Wang Song. Homogeneity test method between samples with small probability event for Meta analysis. Chin J Health Stat 1996; 13: 13 4. 35. Liu Li, Xiu-E Guo, Yong-qing Zhou et al. Prevalence of dementia in China. Dement Geriatr Cogn Disord 2003; 15: 226 30. 36. Karasawa A, Homma A. Recent changes in the prevalence of dementia in the Tokyo metropolis. In: Hasegawa K, Homma A, eds. Psychogeriatrics: Biomedical and Social Advances, vol. 1, Amsterdam Excerpta Medica 1990; 24 9. 37. Ikeda M, Hokoishi K, Maki N et al. Increased prevalence of vascular dementia in Japan: a community-based epidemiological study. Neurology 2001; 57: 839 44. 41. Evans DA, Beckett LA, Field TS et al. Apolipoprotein E [small element of] 4 and incidence of Alzheimer disease in a community population of older persons. JAMA 1997; 277: 822 4. 42. Wang K. Studies of human apolipoprotein E genetic isoform and their phenotype among the Chinese population. Acta Acad Med Sin 1986; 8: 192 8. 43. Hong CJ, Liu TY, Liu HC et al. Epsilon 4 Allele of apolipoprotein E increases risk of Alzheimer s disease in a Chinese population. Neurology 1996; 46: 1749 51. 44. Rocca WA, Hofman A, Brayne C et al. Frequency and distribution of Alzheimer s disease in Europe: a collaborative study of 1980 1990 prevalence findings. The EURODEM- Prevalence Research Group. Ann Neurol 1991; 30: 381 90. 45. Ritchie K, Kildea D, Robine JM et al. The relationship between age and the prevalence of senile dementia: a meta-analysis of recent data. Int J Epidemiol 1992; 21: 763 9. 46. Sulkava R, Wikstrom J, Aromaa A et al. Prevalence of severe dementia in Finland. Neurology 1985; 35: 1025 9. Received 9 October 2006; accepted in revised form 12 July 2007 624