Content uploaded by Xin Xu
Author content
All content in this area was uploaded by Xin Xu on May 25, 2016
Content may be subject to copyright.
AUTHOR COPY
Journal of Alzheimer’s Disease 43 (2015) 669–675
DOI 10.3233/JAD-141593
IOS Press
669
Tofu Intake is Associated with Poor
Cognitive Performance among
Community-Dwelling Elderly in China
Xin Xu
a,d,∗
, Shifu Xiao
b
, Tri Budi Rahardjo
c
and Eef Hogervorst
a,c,∗
a
Psychology Division, School of Sport, Exercise and Health Sciences, Loughborough University,
Loughborough, UK
b
Department of Geriatric Psychiatry, Shanghai Mental Health Center, School of Medicine, Shanghai Jiaotong
University, Shanghai, China
c
Centre for Ageing Studies, Universitas Indonesia, Depok, Indonesia
d
Memory, Ageing and Cognition Center, Department of Pharmacology, National University of Singapore, Singapore
Accepted 26 June 2014
Abstract. Tofu is a soy product which is commonly consumed in Asian countries, such as China and Indonesia. Several
studies found negative associations of high tofu consumption with cognitive function in older Asian populations. However,
the effect of tofu on cognitive function remains disputed as it was not found in Western populations. In the present study,
the effect of weekly tofu intake on cognitive performance was investigated in an observational cross sectional study of 517
Chinese elderly from Shanghai. Similar to earlier studies, results showed that a higher weekly intake of tofu was associated
with worse memory performance using the Hopkins Verbal Learning Test ( = −0.10, p = 0.01) after controlling for age, gender,
education, being vegetarian, and weekly intake of fruit/juice, green vegetables, and orange/red vegetables. Furthermore, among
older elderly (≥68 years of age), high tofu intake increased the risk of cognitive impairment indicative of dementia (OR = 1.27,
95% CI = 0.99–1.64, p = 0.04), after adjusting for all covariates. Consumption of meat and green vegetables independently also
reduced risk of dementia. To conclude, high intake of tofu was negatively related to cognitive performance among community-
dwelling elderly in China. Similar findings were reported in Indonesia and in Japanese Americans in the US. These findings
suggest that the effect of tofu on cognition in elderly should be further investigated.
Keywords: China, cognition, dementia, memory, tofu
INTRODUCTION
Soy products containing isoflavones, such as tofu,
are common foods consumed in Asian countries. How-
ever, the effects of soy products on cognition remain
debatable. Several authors suggested that higher soy
consumption is associated with worse cognitive per-
formance in Asian populations over the age of 65 years
[1–3]. For instance, high tofu consumption was asso-
∗
Correspondence to: Xin Xu and Eef Hogervorst, Applied Cog-
nitive Research, Psychology Division, School of Sport, Exercise and
Health, Brockington Building, Ashby Road, Loughborough Univer-
sity, Loughborough LE11 3TU, UK. Tel.: +44 1509 223020; Fax:
+44 1509 223940; E-mail: e.hogervorst@lboro.ac.uk.
ciated with worse memory using the Hopkins Verbal
Learning Test (HVLT) [4] in a community-based study
conducted in Indonesia [3]. This negative association
of tofu and global cognitive function was also reported
in two longitudinal studies in the US among Japanese
Americans [1, 2]. On the other hand, a better perfor-
mance on cognitive tests measuring processing speed
(but worse verbal memory performance) was reported
in a longitudinal study conducted in the US among
Asian females during and after menopausal transition
with high phytoestrogen intake [5]. Similarly, genis-
tein, the most potent isoflavone or phytoestrogen, was
reported to be positively related to cognitive abil-
ity among middle-aged participants, but had negative
ISSN 1387-2877/15/$27.50 © 2015 – IOS Press and the authors. All rights reserved
AUTHOR COPY
670 X. Xu et al. / Tofu Intake is Associated with Poor Cognitive Performance among Community-Dwelling Elderly in China
associations in elderly subjects [6, 7]. In addition, supe-
rior memory performance was associated with high
consumption of another type of soy product, tempe,
in several studies in Indonesia [3, 8]. There are also
some studies reporting no association between soy
consumption and cognition, especially among older
European and American people [6, 8, 9]. Age, gender,
type of test used for assessment, level of consump-
tion, ethnicity, being an equol producer, and/or type of
product consumed may explain some of the differences
found.
Intervention studies also reported different results
of soy isoflavones on cognitive function. Improve-
ment of cognitive function, including attention [10,
11], language [12, 13], executive function [11, 12],
and visual memory [10, 13–15] were reported after
daily soy supplement interventions (ranging from
60–2000 mg/day, 1.5–6 months total duration). Yet
no effect of isoflavone intake on cognitive function
was reported with soy supplementation interventions
ranging from 60–160 mg/day, within a total duration
of intervention period from 4 to 12 months [16–20].
Noticeably, among all the reported intervention stud-
ies mentioned, only one of these was conducted in an
Asian sample [18], whereas all the other studies were
conducted in Western countries.
The relatively low treatment dosage here may
have had no effect because the Asian participants
would have already consumed more daily isoflavones
[21]. Soy product intake is generally relatively higher
in Asian countries which may affect isoflavone
metabolism and/or their subsequent effect on the brain.
Only very few studies looked into the effect of tofu
on cognition among Chinese elderly people in com-
munity settings. In a recent study no association of
tofu and cognitive function was found in Chinese
elderly [22]. However, this study was conducted among
elderly above 90 years of age and survival bias may
have played a role. Hence, the current study fur-
ther explored the association between tofu intake and
cognitive ability among community-dwelling elderly
people in Shanghai, China. We used the same mem-
ory test earlier found to be sensitive to phytoestrogen
intake and that found in saliva samples [3, 7, 8]. This
test was also found to have very good sensitivity and
specificity for dementia, in particular for its early stages
[8, 23–25]. Because inter-rater reliability for dementia
is often moderate at best [26] and many older people
who were thought to have mild cognitive impairment
will reverse to normal function [25], in this study we
used the cut-off for that test which best indicated early
dementia rather than the clinical diagnoses.
METHODS
Participants
The present observational cross-sectional study was
carried out between June 1 and August 31, 2011. All 50
to 95-year old persons born between June 1, 1916, and
August 31, 1961, and registered for census purposes in
Shanghai were invited to take part in the study. A total
of 517 participants were recruited from urban sites in
the North Xin Jing District of Shanghai, China. Ethical
approval was obtained from Shanghai Mental Health
Centre before the study was initiated.
Prior to the study, all community elders and staff
at local community health centers had been informed
of the survey based study. Participants were commu-
nally talked to by the community center supervisor and
told about the study, its aims and procedures, as well as
time and other commitments required for participation.
Any questions were answered. The informed consent
sheet was read and then signed by both participants and
their caregivers when they wanted to take part (100%
consented). Testing was done by the trained and super-
vised research assistants between 8–11 am to avoid the
effects of time of day.
Assessments
The survey consisted of the following elements.
General demographics covered a wide variety of
information (e.g., age, gender, education, (past or
present) occupation, and living arrangements). The
food frequency questionnaire (FFQ) is a standard-
ized questionnaire investigating participant’s dietary
consumption habits and frequency of consumption of
particular foods [27]. Using the FFQ, consumption
of foods, such as bread, rice, fruit/juice, green veg-
etables, orange/red vegetables, meat, tofu, and tempe,
were surveyed using daily, weekly, and monthly fre-
quency intakes. In the current study, food intake
frequencies were calculated on a weekly basis (cal-
culated from daily, weekly and monthly, e.g., food
intake once/day, every day = 7 times/week; food intake
once/month = 0.25 times/week, etc.).
The HVLT [4] is widely used to detect memory func-
tion and is a word learning test consisting of 12 words
from 3 low frequency categories. It has 6 parallel ver-
sions but in our study only version A was used. Words
from these 3 categories (‘human shelter’, ‘animals’,
and ‘precious stones’) were repeated 3 times for the
total immediate recall (IR). Delayed Recall was not
included in analyses, as earlier work found this vari-
AUTHOR COPY
X. Xu et al. / Tofu Intake is Associated with Poor Cognitive Performance among Community-Dwelling Elderly in China 671
able to be less sensitive than the IR to associations with
tofu consumption [7].
After cognitive testing, an extensive medical exami-
nation was conducted by trained and qualified medical
clinicians, which led to a consensus diagnosis of
dementia. Dementia and mild cognitive impairment
(MCI, a precursor of dementia) [28] were diag-
nosed according to standard clinical diagnostic criteria,
the Diagnostic and Statistical Manual of Mental
Disorders, Fourth Edition (DSM-IV) [29] and the
revised Petersen’s diagnostic algorithm for MCI [28],
respectively.
Using a combination of cognitive tests and the
clinical investigations, three groups were established
consisting of no cognitive impairment (NCI), MCI, and
dementia.
Data analysis
To establish whether the cut-off on the HVLT IR
obtained in Indonesia [8], the UK [23], and the US
[4] of 18/19 also had good sensitivity and specificity
for dementia in China, receiver operating characteris-
tic (ROC) analyses was employed including MCI and
dementia (possible dementia or DEM) versus controls
(NCI).
Descriptive analyses were performed for fre-
quencies and percentages for demographic, lifestyle
variables, and cognitive performance in the whole
group and between participants who ate tofu and those
who did not. The analyses were done using Chi Square
tests for percentages (e.g., for gender) and indepen-
dent t-tests for continuous data. The demographics
variables were gender, age, education, occupation, and
living status (whom the participants were living with
and whether this was an institution or in the com-
munity), the sample was also described for cognitive
scores on the HVLT IR. Linear regression analysis was
employed to investigate the effect of tofu on HVLT per-
formance using the HVLT IR scores (continuous data
as these showed a normal distribution) as the depen-
dent variable, adjusting for demographic and other
dietary variables, including age, gender, education,
being vegetarian (yes or no to eating meat), and weekly
intake of fruit/juice, green vegetables, and orange/red
vegetables.
Subsequently, having possible dementia based on
the HVLT cut-off was used in logistic regression analy-
ses as dependent variable, controlling for demographic
and other dietary variables including age, gender, edu-
cation, being vegetarian (yes or no), and weekly intake
of fruit/juice, green vegetables, and orange/red vegeta-
bles. Analyses were also stratified by median age (68
years of age) to investigate previously found effects of
age [3]. All data analyses were performed using SPSS
21.0., using a p value of <0.05 for significance.
RESULTS
Table 1 shows the descriptive analyses of the cohort
split into those with possible dementia (DEM) and
those without (NCI). DEM was categorized using
the HVLT total recall (IR) performance. The optimal
Table 1
Descriptives of demographic and lifestyle variables and HVLT performance in DEM and NCI groups
DEM Group NCI Group p value
115 (22.1%) 406 (77.9%)
Demographic Variables
Age 73.8 ± 9.9 65.7 ± 9.6 <0.001
Gender (Female %) 69 (60.0%) 215 (53.0%) NS
Education <0.001
No education or only primary level 69 (60.0%) 93 (22.9%)
Secondary and above level 46 (40.0%) 313 (77.1%)
Profession NS
No job or manual 81 (70.4%) 267 (65.8%)
Non manual 34 (29.6%) 139 (34.2%)
Dietary Habit Variables
Vegetarian (Yes %, not eating meat) 74 (64.3%) 202 (49.8%) 0.006
Type of Food (times/week)
Tofu (mean weekly intake) 1.6 ± 1.5 1.8 ± 1.6 NS
Fruit/Juice (mean weekly intake) 2.7 ± 1.9 2.7 ± 1.9 NS
Vegetables (mean weekly intake) 8.7 ± 4.4 9.9 ± 3.4 0.007
HVLT Performance
HVLT IR 11.8 ± 6.6 25.4 ± 7.1 <0.001
a
trend level significance. HVLT IR, Hopkins Verbal Learning Test Immediate Recall; DEM, possible dementia;
NCI, no cognitive impairment; NS, not significant.
AUTHOR COPY
672 X. Xu et al. / Tofu Intake is Associated with Poor Cognitive Performance among Community-Dwelling Elderly in China
Table 2
Linear regression analyses in 2 steps: step1, entering weekly tofu
consumption (times per day/month × 7) controlled for age, gender,
and education; step 2, also controlled for other dietary intake
HVLT IR
Step 1 Step 2
 p value  p value
Tofu (weekly) −0.11 0.009 −0.10 0.01
Age −0.34 <0.001 −0.31 <0.001
Gender −0.05 NS −0.03 NS
Education 0.27 <0.001 0.3 <0.001
Not being vegetarian – – 0.16 <0.001
Fruit/Juice (weekly) – – 0.06 NS
Green vegetables (weekly) – – 0.17 <0.001
Orange/Red vegetables – – 0.004 NS
(weekly)
HVLT IR, Hopkins Verbal Learning Test Immediate Recall; NS, not
significant.
Fig. 1. Relationship between mean HVLT IR score and weekly tofu
intake.
cut-off score of the HVLT in discriminating between
NCI and MCI and dementia cases was generated
by applying ROC. A HVLT score of equal or less
than this cut-off score (≤19 in the current study)
was defined as “possible dementia (DEM) (Area
Under Curve (AUC) = 0.92, 95% CI = 0.89–0.94), this
rendered 90.0% sensitivity and 79.6% specificity when
distinguishing DEM cases (dementia and MCI) from
NCI cases. This cut-off score was further adopted in
the current study to define cognitive impairment. This
score is similar to that found in earlier studies inves-
tigating dementia in Oxford and Indonesia and also
predicted diagnoses independent of education, age,
gender, and depression [8].
Participants in the DEM group were more likely to
be older (74 versus 66 years, on average) and less edu-
cated (60% versus 23%) but had equal proportions
of females as the NCI. Reflecting the cut-off score,
there was a 13-point difference between DEM and
NCI groups on the HVLT IR performance (12 ver-
sus 25 word recalled) whereas an 8-point difference
was observed on the HVLT Delayed Recall perfor-
mance between these 2 groups (1 word recalled for
DEM versus 9 for NCI, data not shown) (Table 1).
Linear regression models (see Table 2) demonstrated
that weekly tofu intake was negatively associated
with immediate recall memory on the HVLT IR
after controlling for demographic (age, gender, and
education) and other food intake variables (being veg-
etarian, weekly intake of fruit/juice, green vegetables,
and orange/red vegetables). Eating meat (not being
vegetarian) was independently associated with better
memory function, as was consumption of green veg-
etables (Table 2).
A bar graph showed an overall trend for increasing
weekly tofu intake to be negatively and linearly asso-
ciated with subject’s worse performance on the HVLT
IR test (Fig. 1).
A logistic regression model using DEM as a binary
outcome investigated the predictive value of weekly
tofu intake, controlling for demographic variables and
other types of food consumed weekly. These analy-
ses indicated that there was a trend for weekly tofu
intake to increase the risk for cognitive impairment by
20% (OR = 1.20, p = 0.08) after adjusting for age, gen-
der, education, vegetarian habits, and weekly intake of
fruit/juice and green/orange/red vegetables. In our pre-
vious study [3], tofu intake was mainly negatively asso-
ciated with dementia risk and worse memory perfor-
mance among ‘older’ elderly (>68 years of age). There-
fore, in the current study stratification using a median
age split (68 years of age) was applied (see Table 3).
From Table 3 we can see that among younger par-
ticipants, there was no significant association of tofu
consumption with DEM, whereas an increased risk of
almost 30% was seen among older elderly (≥68 years
of age) (OR = 1.27, 95% CI = 0.99–1.64, p = 0.04) after
adjusting for all the other covariates. Independently,
not being vegetarian (eating meat) decreased risk for
cognitive impairment almost 4-fold and eating green
vegetables reduced the risk by almost 20%. Education,
but not gender or age, also reduced the risk indepen-
dently in this older group.
DISCUSSION
In the present study, higher intake of tofu was neg-
atively associated with learning ability and immediate
memory performance on the HVLT. Furthermore,
AUTHOR COPY
X. Xu et al. / Tofu Intake is Associated with Poor Cognitive Performance among Community-Dwelling Elderly in China 673
Table 3
Logistic regression analyses stratified for age using the median split (68 years of age), controlled for age, gender, and education in step 1, and
dietary habits in step 2
<68 years of age ≥68 years of age
Step 1 Step 2 Step 1 Step 2
Odd Ratio (95% CI), p value
Weekly Tofu intake NS NS 1.24 (0.97–1.57), p = 0.08
a
1.27 (0.99–1.64), p = 0.04
Age NS NS 1.10 (1.01–1.18), p = 0.02 NS
Education 0.90 (0.84–0.96), 0.87 (0.81–0.94), p < 0.001 0.85 0.83 (0.74–0.93),
p = 0.002 (0.76–0.94), p = 0.001 p = 0.001
Gender (Male) NS NS 0.49 (0.27–0.90), p = 0.02 0.54 (0.28–1.04), p = 0.06
a
Being vegetarian – NS – 3.80 (1.87–7.70), p < 0.001
Weekly fruit/juice intake – NS – NS
Weekly green vegetables intake – 0.83 (0.75–0.92), p < 0.001 – 0.81 (0.73–0.89), p < 0.001
Weekly orange/red – NS – NS
vegetables intake
a
trend for significance. NS, not significant.
among elderly who were 68 years of age or older,
higher weekly tofu intake was a significant risk factor
which increased the risk of possible dementia, inde-
pendent of the other covariates, including demographic
variables and other dietary habits. Similar results could
be found in another earlier study conducted in Indone-
sia, where high tofu intake was associated with poor
memory using the same test [3]. The authors also
reported that in that cohort tempe intake, a fermented
whole soybean food, was significantly related to better
memory [3, 8]. However, tempe is not a popular type
of food in China and only a very limited number of par-
ticipants reported eating tempe (5 out of 521, less than
1% of the whole sample), hence consumption of tempe
was not included in the current analyses. Tempe, sim-
ilar to green vegetables, contains folate which reduces
homocysteine, a risk factor for dementia and cognitive
decline [30]. Not being vegetarian, e.g., eating meat,
in elderly was associated with a four-fold decrease in
risk of possible dementia. In contrast, one earlier study
noted that meat eaters had a doubled risk of demen-
tia [31]. This may be because meat contains saturated
fats, which is a risk factor for cardiovascular disease,
and risk factors for cardiovascular disease are risk fac-
tors for dementia [32]. However, meat also contains
cobalamin which can further help reduce homocys-
teine levels [30]. This means that a well-balanced diet
with little protein, such as tofu and lean meats, but
plenty of vegetables are probably best suited for elderly
to prevent dementia. On the other hand in Indonesia,
high green vegetable consumption was associated with
increased dementia risk [3], which was possibly due to
pollution and heavy use of pesticides. Hence, moder-
ation of overall food intake is probably best advised,
similar to earlier advice regarding the consumption of
fatty (polluted) fish.
There were several limitations to the current study.
Firstly, the results from the current study may have lim-
ited representativeness. The socioeconomic status was
not different for those with possible dementia and those
without. However, overall socioeconomic status of the
cohort investigated was quite low (the current sam-
ple was drawn from a relatively underdeveloped area
in Shanghai, China). Because the present study was
conducted in a cross-sectional community setting, it is
not possible to examine whether elderly who eat more
tofu actually also deteriorated cognitively. A follow-up
study thus needs to be performed. In addition, the cur-
rent sample was constituted of Chinese elderly only.
Therefore results might not apply to Western countries
as several earlier studies did not find these types of
associations in non-Asian populations. Lastly, it may
well be that lower quantities of tofu consumption do
not lead to cognitive impairment and an optimal dosage
needs to be investigated to maintain optimal health
and cognitive function in the elderly. Our graphs in
China and Indonesia did not suggest optimal intakes
of tofu for elderly, although for those of middle-age
optimal levels of genistein associated with better cog-
nitive function were detected in Indonesia [7]. This
may be associated with its estrogenic effects on brain
function, which may be positive in middle-aged peo-
ple but which may worsen pathology in the old [7]. In
sum, further research needs to investigate whether tofu
really is associated with worse cognitive function and
increased risk for dementia in those over 68 years of age
and whether a balanced diet and exercise particularly
in midlife can affect this risk in later life [33].
AUTHOR COPY
674 X. Xu et al. / Tofu Intake is Associated with Poor Cognitive Performance among Community-Dwelling Elderly in China
ACKNOWLEDGMENTS
The authors thank all study participants for their
participation. They also thank the research team
from Shanghai Mental Health Center for data col-
lection. This study was funded by China Science
and Technology Ministry. National Pillar Program
2009BAI77B03.
Authors’ disclosures available online (http://www.j-
alz.com/disclosures/view.php?id=2413).
REFERENCES
[1] White LR, Petrovitch H, Ross GW, Masaki K, Hardman J,
Nelson J, Davis D, Markesbery W (2000) Brain aging and
midlife tofu consumption. J Am Coll Nutr 19, 242-255.
[2] Rice MM, Graves AB, McCurry SM, Gibbons L, Bowen J,
McCormick W, Larson EB (2000) Postmenopausal estrogen
and estrogen-progestin use and 2-year rate of cognitive change
in a cohort of older Japanese American women: The Kame
Project. Arch Intern Med 160, 1641-1649.
[3] Hogervorst E, Sadjimim T, Yesufu A, Kreager P, Rahardjo
TB (2008) High tofu intake is associated with worse memory
in elderly Indonesian men and women. Dement Geriatr Cogn
Disord 26, 56-57.
[4] Brandt J (1991) The Hopkins Verbal Learning Test: Develop-
ment of a new memory test with six equivalent forms. Clin
Neuropsychologist 5, 125-142.
[5] Greendale G, Huang MH, Leung K, Crawford SL, Gold EB,
Wight R, Waetjen E, Larlamangla AS (2012) Dietary phytoe-
strogen intakes and cognitive function during the menopause
transition: Results from the SWAN phytoestrogen study.
Menopause 19, 894-903.
[6] Soni M, Rahardjo TBW, Soekardi R, Sulistyowati YL,
Yesufu-Udechuku A, Irsan A, Hogervorst E (2014) Phy-
toestrogens and cognitive function: A review. Maturitas 77,
209-220.
[7] Hogervorst E, Kushandy L, Angrianni W (2009) Different
forms of soy processing may determine the positive or neg-
ative impact on cognitive function of Indonesian elderly. In
Hormones, Cognition and Dementia, Hogervorst E, Hender-
son W, Gibbs B, Brinton RD, eds. Cambridge University
Press, Edinburgh, pp. 121-132.
[8] Hogervorst E, Mursjid F, Priandini D (2011) Borobudur revis-
ited: Soy consumption may be associated with better recall in
younger, but not in older, rural Indonesian elderly. Brain Res
1379, 206-212.
[9] Franco OH, Burger H, Lebrun CE, Peeters PH, Lamberts SW,
Grobbee DE, Van Der Schouw YT (2005) Higher dietary
intake of lignans is associated with better cognitive per-
formance in postmenopausal women. J Nutr 135, 1190-
1195.
[10] Duffy R, Wiseman H, File SE (2003) Improved cognitive
function in postmenopausal women after 12 weeks of con-
sumption of a soya extract containing isoflavones. Pharmacol
Biochem Behav 75, 721-729.
[11] Casini ML, Marelli G, Papaleo E, Ferrari A, D’Ambrosio F,
Unfer V (2006) Psychological assessment of the effects of
treatment with phytoestrogens on postmenopausal women:
A randomized, double-blind, crossover, placebo-controlled
study. Fertil Steril 85, 972-978.
[12] Kritz-Silverstein D, Von Muhlen D, Barrett-Connor E, Bres-
sel MAB (2003) Isoflavonesand cognitive function in older
women: The Soy and Postmenopausal Health in Aging
(SOPHIA) Study. Menopause 10(3), 196-202.
[13] Gleason CE, Carlsson CM, Barnet JH, Meade SA, Setchell
KD, Atwood CS, Johnson SC, Ries ML, Asthana S (2009) A
preliminary study of the safety: Feasibility and cognitive effi-
cacy of soy isoflavone supplements in older men and women.
Age Ageing 38, 86-93.
[14] Thorp A, Sinn N, Buckley JD, Coates AM, Howe PR (2009)
Soya isoflavone supplementation enhances spatial working
memory in men. Br J Nutr 102, 1348-1354.
[15] Henderson VW, St John JA, Hodis HN, Kono N, McCleary
CA, Franke AA, Mack WJ (2012) Long-term soy isoflavone
supplementation and cognition in women: A randomized,
controlled trial. Neurology 78, 1841-1848.
[16] Kreijkamp-Kaspers S, Kok L, Grobbee DE, De Haan EH,
Aleman A, Van Der Schouw YT (2007) Dietary phytoestrogen
intake and cognitive function in older women. J Gerontol A
Biol Sci Med Sci 62, 556-562.
[17] Fournier LR, Ryan Borchers TA, Robison LM (2007) The
effects of soy milk and isoflavone supplements on cogni-
tive performance in healthy, postmenopausal women. J Nutr
Health Ageing 11, 155-164.
[18] Ho SC, Chan AS, Ho YP, So EK, Sham A, Zee B, Woo JL
(2007) Effects of soy isoflavone supplementation on cognitive
function in Chinese postmenopausal women: A double-blind,
randomized, controlled trial. Menopause 14, 489-499.
[19] Basaria S, Wisniewski A, Dupree K, Bruno T, Song MY,
Yao F, Ojumu A, John M, Dobs AS (2009) Effect of high-
dose isoflavones on cognition, quality of life, androgens, and
lipoprotein in post-menopausal women. J Endocrinol Invest
32, 150-155.
[20] Maki PM, Dennerstein L, ClarkM. Guthrie J, Lamontagne
P, Fornelli D, Little D, Henderson VW, Resnick SM (2011)
Perimenopausal use of hormone therapy is associated with
enhanced memory and hippocampal function later in life.
Brain Res 1379, 232-243.
[21] Yesufu, A, Rahardjo TB, Hogervorst E (2011) Soy, tofu and
brain function in the elderly. In International Handbook of
Behavior, Diet and Nutrition, Preedy VR, Watson RR, Martin
CR, eds. Springer-Verlag, London, pp. 2783-2815.
[22] Gao L, Dong B, Qiu KH, Xiang D (2013) Association between
cognitive impairment and eating habits in elderly Chinese
subjects over 90 years of age. J Int Med Res 41, 1362-
1369.
[23] Hogervorst E, Combrinck M, Lapuerta P, Rue J, Swales K,
Budge M (2002) The Hopkins Verbal Learning Test and
screening for dementia. Dement Geriatr Cogn Disord 13,
13-20.
[24] De Jager CA, Hogervorst E, Combrinck M, Budge MM
(2003) Sensitivity and specificity of neuropsychological tests
for mild cognitive impairment, vascular cognitive impair-
ment and Alzheimer’s disease. Psychol Med 33, 1039-
1050.
[25] Schrijnemaekers AMC, De Jager CA, Hogervorst E, Budge
MM (2006) Cases with mild cognitive impairment and
Alzheimer’s disease fail to benefit from repeated exposure
to episodic memory tests as compared with controls. J Clin
Exp Neuropsychol 28, 438-455.
[26] Hogervorst E, Barnetson L, Jobst KJ, Combrinck M, Nagy ZS,
Smith AD (2000) Diagnosing dementia: Interrater-reliability
assessment and accuracy of the NINCDS/ARDA criteria
versus CERAD histopathological criteria for Alzheimer’s
Disease. Dement Geriatr Cogn Disord 11, 107-113.
AUTHOR COPY
X. Xu et al. / Tofu Intake is Associated with Poor Cognitive Performance among Community-Dwelling Elderly in China 675
[27] Frankenfeld CL, Lampe JW, Shannon J, Gao DL, Ray RM,
Prunty J, Kalhom TF, Wahala K, Patterson RE, Thomas
DB (2004) Frequency of soy food consumption and serum
isoflavone concentrations among Chinese women in Shang-
hai. Public Health Nutr 7, 765-772.
[28] Petersen RC (2004) Mild cognitive impairment as a diagnostic
entity. J Intern Med 256, 183-194.
[29] American Psychiatric Association (2000) Diagnostic and
Statistical Manual of Mental Disorders, Revised 4th ed.
Washington, DC.
[30] Smith AD (2008) The worldwide challenge of the dementias:
A role for B vitamins and homocysteine? Food Nutr Bull 29,
143-172.
[31] Giem P, Beeson WL, Fraser GE (1993) The incidence of
dementia and intake of animal products: Preliminary find-
ings from the adventist health study. Neuroepidemiology 12,
28-36.
[32] Hogervorst E, Clifford A, Stock J, Xu X, Bandelow S (2012)
Exercise to prevent cognitive decline and Alzheimer’sdisease:
For whom, when, what and (most importantly) how much?
J Alzheimers Dis Parkinsonism 2, 117.
[33] Clifford A, Yesufu UA, Edwards A, Bandelow S, Hogervorst
E (2009) Maintaining cognitive health in elderly women: An
invited review. Aging Health 5, 655-670.