Circulating 25-hydroxyvitamin D and physical performance in older

adults: a nationwide study in Taiwan13

Shu-Chun Chuang,4* Hui-Ling Chen,4 Wei-Ting Tseng,4 I-Chien Wu,4 Chih-Cheng Hsu,4 Hsing-Yi Chang,4
Yii-Der Ida Chen,5 Marion M Lee,6 Kiang Liu, 7 and Chao Agnes Hsiung4*

ABSTRACT
Background: A previous study indicated that 31% of the elderly in
Taiwan have a vitamin D deficiency. Vitamin D adequacy has been
associated with physical performance in the elderly.
Objectives: The first aim of the current study was to identify the
determinants for vitamin D deficiency. The second aim was to evaluate
the association between vitamin D status and physical performance.
Design: A total of 5664 community-dwelling participants aged $55 y
were recruited since 2008. Plasma total 25-hydroxyvitamin D
concentrations were originally determined by ELISA and calibrated to
a chemiluminescence measurement. Physical performance was
assessed by handgrip strength, the Short Physical Performance
Battery, timed up and go, a 6-min walk test, and single-leg stands.
Multiple linear regression and logistic regression were used to
estimate the cross-sectional associations.
Results: Vitamin D inadequacy (,50 nmol/L) was related to higher
education (P-trend , 0.01), body mass index [(BMI; in kg/m2) ORs
(95% CIs) for $30 compared with 18.5 to ,25 were 1.78 (1.14, 2.78)
for men and 1.53 (1.11, 2.11) for women], and vegetable intake [fourth
compared with first quartile, 1.58 (1.15, 2.18) for men and 2.38 (1.82,
3.12) for women]. Higher intakes of fish [fourth compared with first
quartile, 0.44 (0.33, 0.59) for men and 0.27 (0.21, 0.36) for women]
and milk [fourth compared with first quartile, 0.46 (0.31, 0.69) for men
and 0.69 (0.49, 0.95) for women] were associated with lower risk of
vitamin D inadequacy. Few subjects had 25-hydroxyvitamin D concentrations ,30 nmol/L. Above that concentration, there was no doseeffect relation with physical performance except for single-leg stands.
Conclusions: The factors associated with vitamin D inadequacy in
Taiwan were higher education, higher BMI, and lower fish and milk
intakes. No dose-effect relation existed between vitamin D concentration
and physical performance except for single-leg stands. This study
was registered at www.clinicaltrials.gov as NCT02677831.
Am J
Clin Nutr 2016;104:133444.
Keywords: epidemiology, HALST, physical performance, vitamin D,
lifestyle factors
INTRODUCTION

The elderly are particularly susceptible to vitamin D deficiency

because of low intestinal absorption, inadequate sunlight exposure,

1334

and impaired skin and renal synthesis (1). It has been estimated that
40100% of community-living elderly adults have vitamin D deficiency in the United States and Europe (24). Traditionally, the
focus on vitamin D has been for its role in maintaining bone health,
for example, in improving bone mineral density and preventing
osteoporosis and fractures. In the past decade, however, associations have been found between vitamin D deficiency and cancer,
cardiovascular disease, diabetes, metabolic syndrome, symptoms
of depression, cognitive deficits, frailty, and total or chronic diseasespecific mortality (5). It was estimated that increasing the plasma
25-hydroxyvitamin D [25(OH)D]8 concentrations of all Europeans
to 40 ng/mL (w100 nmol/L) would reduce the total of economic
burden of disease by 16.7%, or V187,000 million/y (6).
Vitamin D deficiency is a major problem even in subtropical
countries where sun exposure is sufficient to prevent vitamin D deficiency (7). The prevalence of vitamin D inadequacy (,50 nmol/L)
in the elderly has been reported to be 1273% in Malaysia, 91%
in India, 36% in China, 518% in Japan, and 6980% in Korea
(7). Risk factors for vitamin D deficiency in Southeast Asian
countries were identified as younger age, being female, living in
an urban area, and being less physically active (8). In Taiwan,
the dietary reference intake of vitamin D for the elderly ($65 y
old) is 10 mg/d, but the actual intake from food and supplements
was found to be 67 mg/d (9); however, the vitamin D status in
this population was usually adequate (3040 ng/mL) (1012). A
1

Supported by the National Health Research Institute grant PH-104-SP-01.

The funder played no role in designing or conducting the study or in the
collection, management, analysis, or interpretation of the data; nor did they
have any input into the preparation, review, or approval of the manuscript.
3
Supplemental Figure 1 and Tables 13 are available from the Online
Supporting Material link in the online posting of the article and from the
same link in the online table of contents at http://ajcn.nutrition.org.
*To whom correspondence should be addressed. E-mail: scchuang@nhri.org.tw
(S-C Chuang), hsiung@nhri.org.tw (CA Hsiung).
8
Abbreviations used: DEQAS, Vitamin D External Quality Assessment
Scheme; HALST, the Healthy Aging Longitudinal Study in Taiwan; SPPB,
Short Physical Performance Battery; TUG, timed up and go; 25(OH)D,
25-hydroxyvitamin D.
Received September 7, 2015. Accepted for publication August 29, 2016.
First published online October 12, 2016; doi: 10.3945/ajcn.115.122804.
2

Am J Clin Nutr 2016;104:133444. Printed in USA. 2016 American Society for Nutrition
Supplemental Material can be found at:
http://ajcn.nutrition.org/content/suppl/2016/10/12/ajcn.115.1
22804.DCSupplemental.html

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

4
Institute of Population Health Sciences, National Health Research Institutes, Zhunan, Miaoli, Taiwan; 5Institute for Translational Genomics and Biomedical
Sciences, Los Angeles Biomedical Research Institute, Harbor-University of California, Los Angeles Medical Center, Torrance, CA; 6Department of Epidemiology and Biostatistics, University of California San Francisco, San Francisco, CA; and 7Department of Preventive Medicine, Northwestern University
Feinberg School of Medicine, Chicago, IL

VITAMIN D AND PHYSICAL PERFORMANCE IN THE ELDERLY

METHODS

The Healthy Aging Longitudinal Study in Taiwan

The Healthy Aging Longitudinal Study in Taiwan (HALST)
(registered at clinicaltrials.gov as NCT02677831) is a prospective study of community-dwelling elderly adults that has
recruited 5664 volunteers across Taiwan since 2008. The cohort
has been described previously (18). In brief, all eligible residents ($55 y old) living within the catchment area of 7 collaborative hospitals were ascertained from household-registry
archives. Among the 22,563 randomly selected subjects, 6968
subjects could not be located, 8610 subjects declined the invitation, and 6985 (31%) subjects agreed to participate. Subjects were excluded if they had any of the following conditions:
highly contagious infectious diseases (e.g., scabies, open pulmonary tuberculosis); diagnosed dementia; severe illness (e.g., cancer under treatment) and/or being bed-ridden (unable to move);
severe mental disorder (unable to communicate); muteness,
hearing impairment, or blindness (unable to complete the interview); or other conditions, such as living in a long-term care
facility or being hospitalized. The sample was selected to be
representative of the source populations in terms of sex and education (primary school or less compared with more than primary
school).
Data were collected at a home visit and a clinical examination for each participant. A face-to-face interview was conducted during the home visit and collected information on
sociodemographic status, personal and family history of diseases,
physical assessments, cognitive functions, geriatric conditions,
sleep quality, mental health, health care utilization, physical
activity and diet, and quality of life (Supplemental Table 1).
During the clinical examination, participants provided fasting
blood and spot urine samples, and physical characteristics, including anthropometric measures and blood pressure, were
taken.
Informed consent forms were completed by each participant.
The study was approved by the Institutional Review Board at
the National Health Research Institutes and the collaborative
hospitals.

Food-frequency questionnaire
The participants were asked to report the consumption frequency and the portion size of 72 food items over the past year. The
original food-frequency questionnaire was developed for Chinese
Americans (19) and then adapted for a validation study in the
Taiwanese population (20). Average consumption for each food
item (in grams per day) was calculated from frequency and portion
size. Food groups (i.e., fruits, vegetables, meat, fish, milk, and eggs)
were created by summing the consumption of the relevant food items.
Assessment of physical performance
Physical performance was measured through handgrip strength
and the Short Physical Performance Battery (SPPB), which included
gait speed, 5-times chair stands, and standing balance during the
home visit, and timed up and go (TUG), the 6-min walk test, and
single-leg stands during the clinical examination at the hospital.
Handgrip strength was measured with a North Coast hand
dynamometer (North Coast Medical Inc.) (21). The best performance (in kilograms) in 3 trials was used for the analysis.
Success in single-leg stands was defined as being able to stand
on one leg for 30 s with eyes open (22). For the 6-min walk test,
participants were encouraged to walk as far as possible in 6 min.
Total distance walked (in meters) was recorded by trained examiners (23). For the TUG test, the participants were asked to stand
up from a chair, walk 3 m at their usual pace, turn around, and walk
back to sit in the chair (24, 25). Time (in hundredths of a second)
was measured by trained examiners using a handheld stopwatch.
The SPPB scores were determined by using the following
criteria (15): For tests of gait speed, the faster of 2 walks was used
to define the scores. Participants received a score of 0 if they were
not able to perform the test, 1 if their speed was #0.43 m/s, 2 if
0.440.60 m/s, 3 if 0.610.77 m/s, and 4 if .0.77 m/s. For
5-times chair stands, participants received a score of 0 if they
were not able to perform the test, 1 if they completed the test in
.16.7 s, 2 if 16.613.7 s, 3 if 13.611.2 s, and 4 if #11.1 s. For
standing balance, the participants were asked to maintain their
balance in 3 different standing positions for $10 s. Participants
received a score of 0 if they were not able to perform the test,
1 if they could hold a side-by-side stand for 10 s but were unable
to hold a semitandem stand for 10 s, 2 if they could hold
a semitandem stand for 10 s but were unable to hold a fulltandem stand for .2 s, 3 if they could hold a full-tandem stand
for 39 s, and 4 if they could hold a full-tandem stand for 10 s. A
summary performance score was the sum of scores calculated
across the 3 tests, with 0 representing the worst and 12 the best.
Vitamin D measurements and calibration
Biological samples were collected according to a standard
protocol. Briefly, fasting blood and morning urine were collected
and analyzed at a certified clinical laboratory, the Union Clinical
Laboratory. After routine standardization and calibration tests, all
the blood samples were centrifuged at 2500 3 g for 15 min at
48C. The plasma was separated into aliquots and stored at
2808C in a freezer at National Health Research Institutes.
The original total 25(OH)D in plasma was measured with an
OCTEIA 25-Hydroxy Vitamin D EIA kit (Immunodiagnostic
Systems Inc.). The inter- and intra-assay CVs of the measurement
were 710% and 412%, respectively.

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

direct association between vitamin D intake and vitamin D

status in the Taiwanese population has not yet been reported.
Nevertheless, a cross-sectional study of older adults living in
a northern Taiwan community revealed that 31% of the study
subjects had vitamin D deficiency (10).
Previous studies have reported that vitamin D has beneficial
effects with a few exceptions on physical performance (13, 14).
Physical performance, particularly lower extremity functions, has
accurately predicted disability, mortality, and nursing home
admission across diverse populations of the elderly (1517). It is
thus appealing to use vitamin D supplementation to prevent
disability through improvement in physical performance. However, the factors related to vitamin D deficiency and the association between vitamin D status and physical performance have
not yet been properly assessed in Taiwan. The first aim of the
study was to explore the factors related to vitamin D deficiency,
and the second was to evaluate the association between vitamin
D status and physical performance in a community-dwelling
elderly population in Taiwan.

1335

1336

CHUANG ET AL.
TABLE 1
Plasma 25(OH)D concentrations by center1
25(OH)D concentration, nmol/L
30 to ,50

50 to ,75

75 to ,125

Men
Taipei
Yangmei
Miaoli
Changhua
Hualien
Chiayi
Kaohsiung
Total
Women
Taipei
Yangmei
Miaoli
Changhua
Hualien
Chiayi
Kaohsiung
Total
1

$125

25806#
24854#
24841#
24804#
23859#
23828#
22862#

1
0
1
0
1
1
0
4

(0.3)
(0.0)
(0.3)
(0.0)
(0.2)
(0.3)
(0.0)
(0.2)

107
121
120
46
63
13
72
542

(27.0)
(32.1)
(35.4)
(13.6)
(15.3)
(3.9)
(20.9)
(21.4)

251
211
167
196
280
103
204
1412

(63.4)
(56.0)
(49.3)
(57.8)
(68.1)
(31.0)
(59.3)
(55.6)

33
44
44
87
64
150
65
487

(8.3)
(11.7)
(13.0)
(25.7)
(15.6)
(45.2)
(18.9)
(19.2)

4
1
7
10
3
65
3
93

(1.0)
(0.3)
(2.1)
(2.9)
(0.7)
(19.6)
(0.9)
(3.7)

25806#
24854#
24841#
24804#
23859#
23828#
22862#

9
6
9
4
2
1
3
34

(1.9)
(1.6)
(2.5)
(1.0)
(0.5)
(0.3)
(0.8)
(1.2)

194
205
167
107
126
35
109
943

(40.2)
(56.3)
(45.9)
(27.2)
(32.7)
(8.9)
(28.5)
(34.1)

252
145
164
242
229
191
227
1450

(52.3)
(39.8)
(45.1)
(61.6)
(59.5)
(48.7)
(59.4)
(52.5)

26
8
21
40
25
151
42
313

(5.4)
(2.2)
(5.8)
(10.2)
(6.5)
(38.5)
(11.0)
(11.3)

1
0
3
0
3
14
1
22

(0.2)
(0.0)
(0.8)
(0.0)
(0.8)
(3.6)
(0.3)
(0.8)

Values are n (%). 25(OH)D, 25-hydroxyvitamin D.

To quantify the measurement error of the original measurements, the Vitamin D External Quality Assessment Scheme
(DEQAS) standards (5 samples at different concentrations) and
a set of randomly selected samples (n = 62) were sent to the Union
Clinical Laboratory, where the vitamin D was measured by
Chemiluminescence, Diasorin, LIAISON. The bias from the
DEQAS method means ranged from 22.6% to 9.2%.
We randomly split the samples into 2 parts and obtained an
equation between the original and the new measurements (y =
0.0089x2 + 0.3401x + 21.742, R2 = 0.7411, where x = original
measurements by ELISA and y = new measurements by
chemiluminescence). We then applied the equation to the other
part. To validate this equation, we sent another set of samples to
the Union Clinical Laboratory (n = 106). We then applied the
equation to this set of samples. Supplemental Figure 1 presents
the Bland-Altman plots of the average and the differences between the measured and the calibrated results.
Statistical analysis
The plasma 25(OH)D concentrations (in nmol/L) were divided
into ,30 (at risk of vitamin D deficiency), 30 to ,50 (at risk of
inadequacy), 50 to ,75 (sufficient), 75 to ,125 (sufficient, but
not consistently associated with increased benefit), and $125
(may be harmful) (26). We used multiple linear regression models
to determine the variables associated with vitamin D concentrations: age at recruitment (,65, 65 to ,75, or $75 y old);
education; season of blood draw (MarchMay, JuneAugust,
SeptemberNovember, or DecemberFebruary); smoking status
(never, former, or current); drinking status (never, former, or
current); physical activity at leisure time (sex-specific quartile);
physical activity at work (sex-specific quartile); BMI (in kg/m2:
,18.5, 18.5 to ,25, 25 to ,30, or $30); number of chronic
diseases (02, 35, or $6); fruit, vegetable, meat, fish, milk, and
egg intakes (in sex-specific quartiles); and multivitamin use (yes

or no). People with missing data for vitamin D measurements (n =

364), physical performance (n = 399711), or covariates (n = 434)
were deleted from the relevant analyses.
The associations between vitamin D status and physical
performance in a continuous scale were analyzed by multiple
linear regression models. The dependent variables (physical
performance) were log-transformed whenever necessary. The
associations between vitamin D status and success in single-leg
stands ($30 s) and success in full-tandem stands ($10 s for
side-by-side, semitandem, and full-tandem stands) were studied
by logistic regression. Because vitamin D concentrations and
physical performance differed between sexes, the analyses
were separated by sex and adjusted for age at recruitment; education; physical activity at leisure time and at work; BMI;
number of chronic diseases; and fruit, vegetable, meat, fish,
milk, and egg intakes. Season of blood draw, smoking, drinking,
and multivitamin intake were removed from the covariate list
because they were not associated with physical performance in
the model selection process (stepwise method, P , 0.1 for entry
and P . 0.1 for removal from the model). The normality assumption was tested by graphic examination of the residual distribution. The adjusted means were calculated by exponentiating
the natural-log transformed means adjusted for the other effects in
the model. Tests for trend were performed by modeling the categories as a continuous variable.
All analyses were performed by using SAS 9.3 (SAS Institute Inc.).
Statistical significance was defined as a , 0.05. To deal with multiple comparisons, the false-discovery rate (q value) was calculated.
A q value of ,0.05 was considered a significant association. P value
plots were also used to estimate the number of true hypotheses (27).

RESULTS

Overall, the mean 6 SD 25(OH)D concentrations were

66.7 6 28.8 nmol/L for men and 57.7 6 18.8 nmol/L for women.

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

,30

North latitude

1337

VITAMIN D AND PHYSICAL PERFORMANCE IN THE ELDERLY

TABLE 2
The means and 95% CIs of the plasma 25(OH)D concentration (nmol/L) by participants characteristics1
Men
n

Mean (95% CI)

759
1058
681

77.30 (73.62, 80.98)

78.10 (74.51, 81.70)
73.51 (69.66, 77.37)

Mean (95% CI)

832
1278
622

63.35 (58.87, 67.83)

63.39 (58.94, 67.85)
59.74 (55.14, 64.34)

455
1327
301
484
165

68.19
62.51
60.29
59.55
60.26

(63.69,
(58.17,
(55.53,
(54.88,
(54.99,

72.70)
66.85)
65.05)
64.23)
65.52)

855
816
549
512

59.92
64.44
66.66
57.63

(55.47,
(59.90,
(62.04,
(53.06,

64.36)
68.97)
71.28)
62.20)

,0.001

0.003

,0.001
75
990
303
769
361

95.84
75.24
73.98
69.94
66.53

(88.99,
(71.93,
(69.81,
(66.28,
(62.25,

102.69)
78.56)
78.15)
73.61)
70.80)

895
721
435
447

70.61
81.58
83.87
69.17

(67.04,
(77.83,
(79.83,
(65.12,

74.18)
85.33)
87.90)
73.22)

,0.001

,0.001

,0.001

0.05
1024
846
628

77.97 (74.13, 81.80)

76.73 (73.10, 80.35)
74.22 (70.50, 77.95)

971
455
1072

77.84 (74.19, 81.50)

75.08 (71.04, 79.13)
75.99 (72.52, 79.46)

1898
464
136

73.21 (70.15, 76.27)

76.63 (72.83, 80.43)
79.08 (73.57, 84.59)

1850
214
216
218

70.38
75.81
79.42
79.61

(67.24,
(71.18,
(74.71,
(74.93,

73.52)
80.45)
84.13)
84.28)

70
1369
927
132

83.14
79.30
73.26
69.53

(76.20,
(76.27,
(70.13,
(64.22,

90.07)
82.32)
76.38)
74.85)

1116
1048
334

0.35
2666
24
42

60.04 (56.04, 64.03)

65.40 (57.81, 72.99)
61.05 (54.81, 67.29)

2190
86
456

63.41 (58.78, 68.03)

60.68 (55.39, 65.97)
62.40 (57.87, 66.93)

2686
19
27

60.97 (57.30, 64.64)

61.02 (52.84, 69.19)
64.50 (57.11, 71.89)

2029
238
235
230

60.93
61.82
61.80
64.09

(56.61,
(56.99,
(56.96,
(59.22,

65.24)
66.65)
66.64)
68.96)

76
1524
923
209

68.23
61.65
60.03
58.74

(62.49,
(57.31,
(55.60,
(53.90,

73.96)
65.98)
64.46)
63.58)

77.23 (73.74, 80.72)

76.13 (72.63, 79.63)
75.56 (71.26, 79.86)

1030
1273
429

61.88 (57.44, 66.33)

62.41 (57.95, 66.87)
62.19 (57.56, 66.82)

1119
837
674
784
537

76.14
76.55
75.19
76.69
74.86

(72.59,
(72.89,
(71.45,
(73.10,
(71.00,

79.69)
80.22)
78.92)
80.28)
78.71)

1208
1195
1018
797
586

61.77
62.90
60.51
62.48
60.87

(57.33,
(58.44,
(56.01,
(57.98,
(56.38,

66.21)
67.36)
65.00)
66.99)
65.36)

605
620
634
639

77.33
75.26
75.57
77.06

(73.45,
(71.46,
(71.77,
(73.16,

81.21)
79.07)
79.37)
80.97)

682
681
683
686

63.76
61.47
61.01
62.40

(59.20,
(56.90,
(56.46,
(57.90,

68.32)
66.04)
65.56)
66.91)

611
621
632
634

78.11
76.95
76.53
73.64

(74.24,
(73.14,
(72.67,
(69.80,

81.99)
80.76)
80.38)
77.47)

669
681
693
689

65.05
61.82
61.28
60.50

(60.52,
(57.29,
(56.72,
(55.95,

69.57)
66.35)
65.84)
65.05)

0.16

0.27

0.02

0.60

,0.001

0.09

,0.001

0.00

0.52

0.79

0.43

0.03

,0.001

0.05

(Continued)

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

Age, y
55 to ,65
65 to ,75
$75
Education
Illiterate
Primary school
Middle school
High school
.High school
Season
Spring
Summer
Autumn
Winter
Smoking
Never
Former
Current
Drinking
Never
Former
Current
Betel chewing
Never
Former
Current
Work-related physical activity, sex-specific quartile
Q1
Q2
Q3
Q4
BMI, kg/m2
,18.5
18.5 to ,25
25 to ,30
$30
No. of chronic diseases2
02
35
$6
The 5 most prevalent diseases3
Hypertension
Cataract
Hyperlipidemia
Gastric ulcer or gastric diseases
Heart diseases
Fruit intake, g/d
Q1 (M: ,94.4; F: ,90.5)
Q2 (M: 94.4 to ,192.0; F: 90.5 to ,176.3)
Q3 (M: 192.0 to ,343.7; F: 176.3 to ,319.0)
Q4 (M: $343.7; F: $319.0)
Vegetables, g/d
Q1 (M: ,317.3; F: ,325.4)
Q2 (M: 317.3 to ,524.0; F: 325.4 to ,526.6)
Q3 (M: 524.0 to ,835.9; F: 526.6 to ,836.4)
Q4 (M: $835.9; F: $836.4)

Women

1338

CHUANG ET AL.

TABLE 2 (Continued )
Men
n

Mean (95% CI)

Mean (95% CI)

0.23
866
602
597
433

77.60
77.09
76.25
74.29

(73.95,
(73.21,
(72.39,
(70.28,

81.25)
80.97)
80.11)
78.29)

714
599
605
580

69.99
74.72
77.34
83.17

(66.22,
(70.82,
(73.52,
(79.36,

73.76)
78.62)
81.17)
86.99)

1415
369
474
240

74.73
75.20
79.12
76.17

(71.41,
(71.02,
(75.16,
(71.59,

78.06)
79.38)
83.08)
80.74)

509
814
749
426

78.43
76.29
75.72
74.78

(74.46,
(72.57,
(71.99,
(70.79,

82.41)
80.01)
79.46)
78.77)

0.61
677
788
638
629

62.12
62.91
61.76
61.86

(57.52,
(58.36,
(57.26,
(57.33,

66.72)
67.46)
66.25)
66.39)

730
836
704
462

55.82
61.05
65.53
66.24

(51.33,
(56.55,
(60.96,
(61.61,

60.32)
65.56)
70.10)
70.86)

1376
469
663
224

61.01
62.51
63.00
62.13

(56.64,
(57.87,
(58.44,
(57.27,

65.37)
67.15)
67.55)
66.99)

808
958
692
274

63.78
62.01
61.49
61.35

(59.23,
(57.53,
(56.96,
(56.62,

68.33)
66.50)
66.02)
66.09)

,0.001

,0.001

0.03

0.09

0.21

0.06

,0.001

0.26
1738
760

75.62 (72.25, 78.98)

76.99 (73.27, 80.72)

1839
893

60.53 (56.14, 64.91)

63.80 (59.29, 68.30)

The means were adjusted for the other variables in the table. Q, quartile; 25(OH)D, 25-hydroxyvitamin D.
The chronic diseases included self-reported hypertension, diabetes, heart disease, stroke, hyperlipidemia, asthma, chronic respiratory tract disease,
cancer, gastric disease, liver and gallbladder diseases, cataract, gout, anemia, kidney disease, arthritis, spurs, osteoporosis, fracture, and mental disease.
3
The means and 95% CIs were adjusted for age at recruitment (,65, 6575, or $75 y old); education (illiterate, primary school, middle school, high
school, or more than high school); season (spring, summer, autumn, or winter); smoking, drinking, and betel nut chewing status (never, former, or current);
physical activity at work (sex-specific quartile), BMI (in kg/m2: ,18.5, 18.5 to ,25, 25 to ,30, or $30); fruit, vegetable, meat, fish, milk, and egg intakes (in
sex-specific quartiles); and multivitamin use (yes or no).
2

A total of 22% of men and 35% of women were at risk of vitamin

D inadequacy. Table 1 shows the plasma 25(OH)D concentrations
for men and women by center. The frequency of vitamin D inadequacy was higher in participants from the centers in the north
(e.g., Taipei, Yangmei, and Miaoli) than in those from the centers
in the south (e.g., Chiayi).
Table 2 presents vitamin D status by study factors. This model
explains 17% of the variation of the 25(OH)D concentrations for men
and 14% for women. Table 3 shows the factors associated with vitamin D inadequacy. People who had higher 25(OH)D concentrations
were younger and had lower education, higher work-related physical
activity, lower BMI, low vegetable intake, and high fish intake (Table
2). In men, the factors associated with vitamin D inadequacy were
higher education, season of blood draw, no betel nut chewing, lower
work-related physical activity, higher BMI, higher intake of vegetables, lower intakes of fish and milk, and no multivitamin intake (Table
3). In women, factors related to vitamin D inadequacy were older age,
higher education, season of blood draw, higher BMI, higher intake of
vegetables, lower intake of fish, and no multivitamin intake.
Table 4 shows the results of multiple linear regression analyses and the logistic regression of the association between plasma
25(OH)D concentrations and physical performance. Very few subjects had 25(OH)D concentrations ,30 nmol/L (4 men, 34 women).
Above that concentration, no dose-effect relation with physical
performance was apparent, except for single-leg stands. Specifically,
a higher OR of success in single-leg stands was associated with

higher 25(OH)D concentrations (P-trend , 0.01 for both men and

women). However, handgrip strength showed little variation across
different 25(OH)D concentrations. Potential U-shaped associations
were seen for vitamin D concentrations with TUG, the 6-min walk
test, and the SPPB score, whereas there were statistically significant
differences (P , 0.05) between the inadequate (30 to ,50 nmol/L)
and sufficient (50 to ,75 nmol/L, reference) groups. The best
performance was seen in participants whose 25(OH)D concentrations were 75 to ,125 nmol/L; but they were not significantly
different from the vitamin Dsufficient group (50 to ,75 nmol/L).
DISCUSSION

In this elderly Taiwanese population, the prevalence of vitamin

D inadequacy (,50 nmol/L) was 22% in men and 35% in
women. Similar to what has been observed in Western countries,
the factors associated with vitamin D status were behavior related. However, the direction of the associations suggests that
the major source of vitamin D is sun exposure. Also, education
was inversely associated with plasma 25(OH)D concentrations.
Vitamin D measurements are known to vary by methods and by
laboratories using the same methods (28). Based on the original
measurements by ELISA, the mean concentration (49 nmol/L)
was much lower than previously reported [30.7 ng/mL by a commercial radioimmunoassay kit (Incstar Corp.) (11), 36.21 ng/mL
(12), and 39.9 ng/mL (10) by the DiaSorin 25-Hydroxyvitamin

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

Meat intake, g/d

Q1 (M: ,25.7; F: 0 or rarely)
Q2 (M: 25.7 to #60.0; F:.0 to #34.3)
Q3 (M: .60.0 to #120.0; F: .34.3 to #68.6)
Q4 (M: .120; F: .68.6)
Fish intake, g/d
Q1 (M: ,12.9; F: 0 or rarely)
Q2 (M: 12.9 to #47.1; F: .0 to #34.3)
Q3 (M: .47.1 to #68.6; F: .34.3 to #68.6)
Q4 (M: .68.6; F: .68.6)
Milk and dairy products intake, g/d
Q1 (M: 0; F: 0 or rarely)
Q2 (M: .0 to #164.6; F: .0 to #164.6)
Q3 (M: .164.6 to #288.0; F: .164.6 to #288.0)
Q4 (M: .288.0; F: .288.0)
Egg intake, g/d
Q1 (M: 0; F: 0 or rarely)
Q2 (M: .0 to #12.9; F: .0 to #12.9)
Q3 (M: .12.9 to #34.3; F: .12.9 to #34.3)
Q4 (M: .34.3; F: .34.3)
Multivitamin use
No
Yes

Women

1339

VITAMIN D AND PHYSICAL PERFORMANCE IN THE ELDERLY

TABLE 3
The ORs and 95% CIs between the participants characteristics and vitamin D inadequacy (,50 nmol/L)1
Men
$50

,50

OR (95% CI)

589
849
526

170
209
155

1.00
0.94 (0.73, 1.21)
1.15 (0.86, 1.53)

65
809
237
591
262

10
181
66
178
99

0.60 (0.29,
1.00
1.15 (0.82,
1.35 (1.03,
1.72 (1.24,

665
621
364
314

230
100
71
133

2.20 (1.68, 2.88)

1.00
1.24 (0.88, 1.75)
2.77 (2.04, 3.76)

799
672
493

225
174
135

1.00
0.95 (0.74, 1.22)
1.10 (0.83, 1.46)

772
347
845

199
108
227

1.00
1.18 (0.88, 1.59)
1.00 (0.79, 1.27)

1467
383
114

431
81
22

1.00
0.68 (0.50, 0.92)
0.55 (0.33, 0.93)

1421
178
179
186

429
36
37
32

1.00
0.61 (0.41, 0.91)
0.66 (0.44, 0.97)
0.58 (0.38, 0.88)

57
1119
691
97

13
250
236
35

1.01 (0.53, 1.95)

1.00
1.61 (1.30, 2.00)
1.76 (1.13, 2.74)

882
824
258

234
224
76

1.00
0.95 (0.76, 1.19)
0.93 (0.67, 1.29)

476
472
512
504

129
148
122
135

1.00
1.11 (0.82, 1.49)
0.78 (0.57, 1.07)
0.90 (0.65, 1.24)

490
504
485
485

121
117
147
149

1.00
1.03 (0.75, 1.39)
1.47 (1.08, 1.99)
1.54 (1.12, 2.12)

686
464
484
330

180
138
113
103

1.00
1.15 (0.88, 1.50)
0.89 (0.67, 1.18)
1.11 (0.82, 1.51)

$50

,50

OR (95% CI)

523
852
389

309
426
233

1.00
0.97 (0.79, 1.20)
1.39 (1.06, 1.81)

344
818
193
305
104

111
509
108
179
61

0.52 (0.40,
1.00
0.92 (0.70,
1.07 (0.83,
1.02 (0.69,

490
587
418
269

365
229
131
243

1.91 (1.54, 2.37)

1.00
0.82 (0.63, 1.06)
2.40 (1.87, 3.07)

1718
16
30

948
8
12

1.00
0.77 (0.30, 1.99)
0.79 (0.38, 1.64)

1423
52
289

767
34
167

1.00
1.09 (0.66, 1.78)
1.02 (0.81, 1.30)

1732
11
21

954
8
6

1.00
1.29 (0.46, 3.66)
0.31 (0.12, 0.82)

1306
155
147
156

723
83
88
74

1.00
1.02 (0.75, 1.38)
0.96 (0.71, 1.30)
0.84 (0.61, 1.16)

51
1013
578
122

25
511
345
87

1.06 (0.62, 1.80)

1.00
1.31 (1.09, 1.58)
1.52 (1.10, 2.09)

652
839
273

378
434
156

1.00
0.92 (0.76, 1.11)
1.00 (0.77, 1.30)

458
436
423
447

224
245
260
239

1.00
1.08 (0.85, 1.39)
1.17 (0.91, 1.51)
0.98 (0.75, 1.28)

494
445
422
403

175
236
271
286

1.00
1.68 (1.30, 2.16)
2.06 (1.59, 2.66)
2.35 (1.80, 3.07)

410
543
414
397

267
245
224
232

1.00
0.77 (0.60, 0.98)
0.91 (0.70, 1.18)
0.93 (0.71, 1.21)

0.32

0.01

,0.001

0.01
1.23)
1.62)
1.76)
2.38)

0.68)
1.22)
1.36)
1.49)

,0.001

,0.001

0.58

0.72

0.46

0.94

0.01

0.05

0.003

0.76

,0.001

0.01

0.88

0.62

0.11

0.45

,0.001

0.01

0.35

0.17

(Continued)

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

Age, y
55 to ,65
65 to ,75
$75
Education
Illiterate
Primary school
Middle school
High school
.High school
Season
Spring
Summer
Autumn
Winter
Smoking
Never
Former
Current
Drinking
Never
Former
Current
Betel chewing
Never
Former
Current
Work-related physical activity, sex-specific quartile
Q1
Q2
Q3
Q4
BMI, kg/m2
,18.5
18.5 to ,25
25 to ,30
$30
No. of chronic diseases2
02
35
$6
Fruit intake, g/d
Q1 (M: ,94.4; F: ,90.5)
Q2 (M: 94.4 to ,192.0; F: 90.5 to ,176.3)
Q3 (M: 192.0 to ,343.7; F: 176.3 to ,319.0)
Q4 (M: $343.7; F: $319.0)
Vegetables, g/d
Q1 (M: ,317.3; F: ,325.4)
Q2 (M: 317.3 to ,524.0; F: 325.4 to ,526.6)
Q3 (M: 524.0 to ,835.9; F: 526.6 to ,836.4)
Q4 (M: $835.9; F: $836.4)
Meat intake, g/d
Q1 (M: ,25.7; F: 0 or rarely)
Q2 (M: 25.7 to #60.0; F: .0 to #34.3)
Q3 (M: .60.0 to #120.0; F: .34.3 to #68.6)
Q4 (M: .120; F: .68.6)

Women

1340

CHUANG ET AL.

TABLE 3 (Continued )
Men
,50

OR (95% CI)

504
484
495
481

210
115
110
99

1.00
0.56 (0.42, 0.73)
0.49 (0.37, 0.65)
0.44 (0.33, 0.59)

1061
295
401
207

354
74
73
33

1.00
0.80 (0.59, 1.08)
0.51 (0.38, 0.69)
0.45 (0.30, 0.68)

412
635
590
327

97
179
159
99

1.00
1.14 (0.84, 1.53)
1.15 (0.85, 1.55)
1.30 (0.92, 1.83)

1346
618

392
142

1.00
0.71 (0.56, 0.90)

$50

,50

OR (95% CI)

362
562
495
345

368
274
209
117

1.00
0.47 (0.38, 0.59)
0.36 (0.29, 0.46)
0.27 (0.21, 0.36)

853
307
448
156

523
162
215
68

1.00
0.87 (0.68, 1.10)
0.81 (0.65, 1.00)
0.68 (0.49, 0.95)

533
617
441
173

275
341
251
101

1.00
1.16 (0.93, 1.44)
1.07 (0.84, 1.36)
0.99 (0.72, 1.36)

1144
620

695
273

1.00
0.65 (0.54, 0.79)

,0.001

P
,0.001

,0.001

0.05

0.53

0.55

,0.001

0.005

The ORs were adjusted for the other variables in the table. An OR .1 indicates greater odds of vitamin D inadequacy (,50 nmol/L). Q, quartile.
The chronic diseases included self-reported hypertension, diabetes, heart disease, stroke, hyperlipidemia, asthma, chronic respiratory tract disease,
cancer, gastric disease, liver and gallbladder diseases, cataract, gout, anemia, kidney disease, arthritis, spurs, osteoporosis, fracture, and mental disease.
1
2

D 125I RIA kit]. The mean concentration was 62 nmol/L

(w25 ng/mL) after calibration. According to the DEQAS report, at
the same 25(OH)D concentration the method means were highest
for Immunodiagnostic Systems EIA, followed by DiaSorin RIA
and DiaSorin Liaison Total. Although the calibrated concentrations
were still lower than those previously reported, calibration with the
DEQAS improved comparability with other studies.
Physical activity has previously been associated with better
vitamin D status (2935). We observed an association of vitamin
D status with work-related physical activity in men but not with
recreational physical activity (data not shown). In general,
physical activity could be an index of sun exposure, which is
a major source of vitamin D. The lack of association between
recreational physical activity and plasma 25(OH)D concentrations may indicate something about the exercise habits of
Taiwanese (e.g., indoor gym exercises and the covering of skin
during outdoor activities). In addition, sun-avoiding behaviors
may be neglected when performing work-related physical activities, particularly among men.
An inverse association between obesity and vitamin D status
has also been commonly reported (3136). Although those who
were overweight or obese might tend to be less active, it has
been reported that vitamin D is stored in adipose tissue and that
vitamin D released from the skin into the circulatory system is
decreased in obese people. Thus, people with obesity may be at
greater risk of vitamin D deficiency (37).
Education was associated with a higher prevalence of vitamin
D inadequacy in both men and women. A similar result was
observed in Japan (38). In Western countries, low socioeconomic
status or education has been associated with a higher risk of
vitamin D deficiency (3941). Studies on knowledge and perceptions of Asian women revealed a gap between knowing how
to prevent vitamin D deficiency and taking the recommended
actions (29, 4245). Because Taiwan is a sunny, subtropical

country, sun exposure should be sufficient for vitamin D synthesis. However, the culture favors fair skin, and both men and
women with higher educational levels may have adapted lifestyles to limit sun exposure (e.g., sunscreen use).
Dietary vitamin D information is not available in the current
National Taiwan Food Composition Database. Nevertheless,
a previous study showed that the major food sources of vitamin D
in Taiwan are fish and its products along with milk and mushrooms (9). Our study confirms that fish and milk intakes were
associated with higher plasma 25(OH)D concentrations. Although vitamin D fortification of milk is not mandatory in
Taiwan, imported dried milk powder may be fortified, and the
level of fortification depends on the brand of the dairy product.
The inverse association found between vegetable intake and
vitamin D status was unexpected. Vegetable intake was positively
associated with higher education and recreational physical activity (data not shown) in our study. The inverse association
between vegetable intake and vitamin D status could be a result of
residual confounding by education or a chance finding. However,
its false discovery rates were low in both men (q = 0.01) and
women (q = 0.04). The European Prospective Investigation into
Cancer and NutritionOxford study reported that vegetarians
and vegans have lower plasma 25(OH)D concentrations than meat
and fish eaters do (46), suggesting that the differences between
vegan and meat and fish eaters could be caused by the intake of
animal food. However, sun exposure still had a greater influence on
plasma 25(OH)D concentrations than did vitamin D intake (46).
Betel nut chewing was associated with a lower frequency of
vitamin D inadequacy in men in this study. Prior studies from
Western countries have been inconsistent in showing that
smoking and drinking can be risk factors for vitamin D deficiency
(31, 34, 36, 47, 48). The association between vitamin D status and
smoking and drinking could arise from smokers and drinkers
having different levels of vitamin D intake compared with

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

Fish intake, g/d

Q1 (M: ,12.9; F: 0 or rarely)
Q2 (M: 12.9 to #47.1; F: .0 to #34.3)
Q3 (M: .47.1 to #68.6; F: .34.3 to #68.6)
Q4 (M: .68.6; F: .68.6)
Milk and dairy products intake, g/d
Q1 (M: 0; F: 0 or rarely)
Q2 (M: .0 to #164.6; F: .0 to #164.6)
Q3 (M: .164.6 to #288.0; F: .164.6 to #288.0)
Q4 (M: .288.0; F: .288.0)
Egg intake, g/d
Q1 (M: 0; F: 0 or rarely)
Q2 (M: .0 to #12.9; F: .0 to #12.9)
Q3 (M: .12.9 to #34.3; F: .12.9 to #34.3)
Q4 (M: .34.3; F: .34.3)
Multivitamin use
No
Yes

$50

Women

0.03 529
0.74 912

0.74 495
0.21 844

361.47 (289.20, 433.75)

357.38 (333.94, 380.82)

27.05 (19.68, 34.42)

21.58 (19.81, 23.36)

0.66 [0.06, 7.46]4

1.73 [0.73, 4.13]

9.92 [7.97, 11.87]

10.08 [9.33, 10.83]

1.01 (0.76, 1.25)

0.84 (0.76, 0.92)

9.40 (6.61, 13.38)

10.50 (9.26, 11.90)

0.28 [0.03, 2.88]

0.45 [0.19, 1.08]

4
31

4
34

3
29

4
33

4
33

3
30

4
34

0.29 534
0.07 937

0.89 512
0.32 871

0.52 534
0.49 939

0.53 530
0.42 924

0.48 496
0.43 887

11.07 (8.48, 14.46)3

12.90 (11.73, 14.18)
0.50 527
0.41 928

P2

4
32

Value

0.51 [0.37, 0.70]

0.73 [0.57, 0.92]

9.40 (9.00, 9.81)

10.21 (9.83, 10.61)

0.90 (0.86, 0.93)

0.84 (0.82, 0.87)

10.27 [9.99, 10.54]

10.11 [9.88, 10.34]

1.03 [0.81, 1.31]

0.84 [0.68, 1.05]

34.00 (32.97, 35.02)

21.62 (21.06, 22.18)

378.63 (368.01, 389.24)

355.47 (348.27, 362.68)

12.73 (12.26, 13.21)

12.90 (12.53, 13.28)

Value

30 to ,50 nmol/L
n

1332
1325

1348
1372

1405
1440

,0.001 1402
0.01 1437

0.16
0.02

0.03
0.01

0.01 1391
0.002 1423

0.80
0.12

0.002 1375
0.23 1418

0.02 1331
,0.001 1349

,0.001 1389
,0.001 1419

P2

1.00
1.00

9.18 (8.85, 9.52)

9.87 (9.53, 10.22)

0.93 (0.90, 0.95)

0.87 (0.84, 0.89)

10.54 [10.31, 10.77]

10.38 [10.16, 10.59]

1.00
1.00

35.20 (34.34, 36.06)

21.87 (21.35, 22.39)

387.50 (378.52, 396.48)

366.59 (359.91, 373.27)

12.13 (11.75, 12.51)

12.40 (12.07, 12.74)

Value

50 to ,75 nmol/L
n

487
310

Ref 475
Ref 292

Ref 484
Ref 311

Ref 486
Ref 305

462
293

Ref 480
Ref 310

Ref 459
Ref 291

Ref 480
Ref 306

P2

383.30 (366.08, 400.52)

353.78 (326.04, 381.53)

0.67
0.81
0.10
0.32

0.79
0.61

0.52
0.36

10.59 [10.32, 10.85]

10.41 [10.12, 10.71]
0.90 (0.87, 0.94)
0.85 (0.82, 0.88

9.22 (8.85, 9.62)

9.98 (9.51, 10.47)

0.89 [0.62, 1.27]

1.20 [0.82, 1.77]

93
22

91
20

93
22

91
22

0.97 [0.44, 2.15]

0.41 [0.14, 1.18]

9.70 (9.03, 10.41)

9.17 (7.89, 10.66)

0.89 (0.84, 0.95)

0.85 (0.75, 0.95)

10.50 [10.04, 10.95]

10.10 [9.21, 10.99]

4.10 [2.45, 6.89]

3.22 [1.20, 8.63]

88
21

12.43 (11.69, 13.21)

12.64 (11.30, 14.14)

,0.001 90
,0.001 20

0.84
0.31

35.12 (34.13, 36.10)

21.55 (20.85, 22.26)

1.60 [1.26, 2.04]

2.10 [1.56, 2.84]

0.59
0.27

389.68 (379.46, 399.90)

362.14 (353.00, 371.27)

93
22

Value

34.14 (32.41, 35.86)

22.24 (19.98, 24.51)

0.34
1.00

11.96 (11.53, 12.39)

12.40 (11.95, 12.87)

$125 nmol/L

89
20

P2

Value

75 to ,125 nmol/L

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

0.93
0.10

0.38
0.12

0.24
0.72

0.84
0.53

,0.001
0.02

0.03
0.74

0.61
0.36

0.41
0.73

P2

The means and ORs were adjusted for age at recruitment (,65, 65 to ,75, or $75 y old); education (illiterate, primary school, middle school, high school, or more than high school); physical activity at
leisure time and at work (sex-specific quartile); BMI (in kg/m2; ,18.5, 18.5 to ,25, 25 to ,30, or $30); number of chronic diseases (02, 35, or $6); and fruit, vegetable, meat, fish, milk, and egg intakes (in
sex-specific quartiles). Most of the dependent variables were normally distributed except for the timed up and go test and the 5-times chair stands. The 2 variables were log-transformed before fitting a multiple
linear regression. The adjusted (least square) means were estimated by fitting a linear regression model. The adjusted means for timed up and go and 5-times chair stands were then calculated by exponentiating
the adjusted means. Ref, reference; SPPB, Short Physical Performance Battery.
2
The P values compare the adjusted (least squares) means or ORs to the ref group (50 to ,75 noml/L) with Students t test.
3
Mean; 95% CI in parentheses (all such values).
4
OR; 95% CI in brackets (all such values).

Timed up and go, s

Men
Women
6-min walk test, m
Men
Women
Grip strength, kg
Men
Women
Succeed in
single-leg
stands, $30 s
Men
Women
SPPB
Men
Women
Gait speed, m/s
Men
Women
5-times chair
stands, s
Men
Women
Succeed in full
tandem stands
Men
Women

,30 nmol/L

TABLE 4
The adjusted means or ORs and 95% CIs of the participants physical performance by 25-hydroxyvitamin D concentrations1

VITAMIN D AND PHYSICAL PERFORMANCE IN THE ELDERLY

1341

1342

CHUANG ET AL.

between several factors and vitamin D inadequacy in a dataset,

we cannot exclude the possibility of chance findings because
of multiple comparisons. We adopted statistical methods, false
discovery rate (q value), and P value plots to address the issue.
Finally, because only 31% of the invited subjects agreed to participate, there may have been selection bias. Compared with
subjects who agreed to participate, women (60% in nonrespondents compared with 53% in respondents), those who were older
($75 y old: 34% in nonrespondents compared with 26% in respondents), and those who had less than primary school education
(64% in nonrespondents compared with 55% in respondents)
tended to refuse to participate in the study. A total of 41% nonresponders did not give us a reason. The common reasons for
decline included did not want to participate in the clinical examination (15%), had taken physical checkup regularly or recently (14%), did not have time (13%), and others. Generally
speaking, the nonresponders were more likely to be older and
female and to have lower education. In the HALST, all measurements were conducted by 15 centrally trained fieldworkers.
Data collection, management, validation, and processing strictly
followed the study protocols. The data quality was guaranteed
through rigid quality-control procedures (e.g., consistency check,
logic check, or range check). The quality control of the laboratory
performance included repeat measurement of random subsamples
that was built into the schedule to assess both intra- and interassay
variation in an ongoing manner. Furthermore, the vitamin D
measurements were calibrated with the DEQAS standards, which
improve its international comparability. Finally, HALST is a relatively large study in Asia (n . 5000), with a comprehensive list of
covariates, thus giving us power to investigate the independent
association between vitamin D and physical performance.
Because a decline in physical performance is a predictor of
adverse geriatric outcomes (1517), improving or maintaining
physical performance is a target for intervention. We aim to
investigate whether baseline vitamin D status is associated with
declines in physical performance during the follow-up. Given
that vitamin D status is associated with many modifiable lifestyle factors, identifying risk populations and implementing
individualized intervention strategies should be possible.
The authors responsibilities were as followsS-CC, I-CW, C-CH, H-YC,
Y-DIC, MML, KL, and CAH: participated in the data analysis, manuscript
writing, and interpretation of the results; and H-LC and W-TT: were involved
in data collection, laboratory measurements, and interpretation of the results.
None of the authors reported a conflict of interest related to this study.

REFERENCES
1. Kinyamu HK, Gallagher JC, Rafferty KA, Balhorn KE. Dietary calcium
and vitamin D intake in elderly women: effect on serum parathyroid
hormone and vitamin D metabolites. Am J Clin Nutr 1998;67:3428.
2. Holick MF. Vitamin D deficiency. N Engl J Med 2007;357:26681.
3. Barnard K, Colon-Emeric C. Extraskeletal effects of vitamin D in older
adults: cardiovascular disease, mortality, mood, and cognition. Am J
Geriatr Pharmacother 2010;8:433.
4. Holick MF. High prevalence of vitamin D inadequacy and implications
for health. Mayo Clin Proc 2006;81:35373.
5. Theodoratou E, Tzoulaki I, Zgaga L, Ioannidis JP. Vitamin D and multiple
health outcomes: umbrella review of systematic reviews and meta-analyses
of observational studies and randomised trials. BMJ 2014;348:g2035.
6. Grant WB, Cross HS, Garland CF, Gorham ED, Moan J, Peterlik M,
Porojnicu AC, Reichrath J, Zittermann A. Estimated benefit of increased vitamin D status in reducing the economic burden of disease in
western Europe. Prog Biophys Mol Biol 2009;99:10413.

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

nonconsumers (49). In our study, betel nut chewers had lower

fruit and vegetable intake, higher smoking and drinking rates,
lower education, and higher work-related physical activity. To
summarize these associations, it is reasonable to speculate that
higher plasma 25(OH)D is a marker for a population that has
lower education and less health-conscious behavior (lower fruit
and vegetable intakes, and higher rates of smoking, drinking, and
betel nut chewing), as well as a lifestyle that involves more sun
exposure, such as work-related physical activity.
In terms of the association of vitamin D with physical performance, we observed a nonlinear association between vitamin D
status and physical performance (Table 4). A threshold between
vitamin D concentrations and physical performance has been suggested (14), but null associations have been reported in populations
with a very high or very low prevalence of vitamin D deficiency (50,
51). Results from studies on populations with 25(OH)D concentrations similar to those in Taiwan have been inconsistent (5254).
Additionally, vitamin D has been suggested to benefit global
physical performance rather than specific performance (55). Despite
only a small proportion of subjects having 25(OH)D ,30 nmol/L
in this study, we observed a U-shaped association between vitamin
D status and SPPB score, with the highest score belonging to the
group that had 25(OH)D concentrations of 75 to ,125 nmol/L.
Nevertheless, age explained the greatest variations of physical
performance, whereas plasma 25(OH)D concentrations explained
only w0.4% of the variation in physical performance (Supplemental Tables 2 and 3). Physical performance is known to decline
with aging, and participants with multiple comorbidities may have
had restricted physical activity, thus leading to decreased muscle
functions and increased BMI. The null-association between vitamin
D status and physical performance in our study may imply an overall
well-being of the study population due to the cross-sectional study
design. Further investigation to assess the relation between vitamin D
status and the declining rate of physical performance is warranted.
Our study has several limitations. First, because of the crosssectional analysis, it is difficult to infer causality. However, vitamin D status could be a marker for lifestyle patterns. Second,
the model explains only 17% and 14% of the variation in 25(OH)D
concentrations for men and women, respectively. Because our
study had a limited number of subjects with vitamin D deficiency,
this could limit our ability to find explanatory variables to predict
vitamin D variation. Previous studies have reported model R2
from 0.13 to 0.42 (32, 35, 39, 56). The unexplained variation
could be due to measurement error in the self-reported lifestyle
data (e.g., physical activity). Theoretically, only outdoor activity
is related to sun exposure, but we did not ask where the activity
was performed. Many activities can be performed either indoors
or outdoors, which could explain why we did not observe an
association between recreational physical activity and 25(OH)D
concentrations. Furthermore, we do not know the ambient exposure during the activity (e.g., tree cover). Other unexplained
variability may be due to unmeasured confounding factors, such
as genetics involved in vitamin D metabolism or other unmeasured factors related to sun exposure. Third, a single measurement of plasma 25(OH)D concentration cannot be viewed as
being representative of long-term vitamin D status, and the
differences in physical performance between groups do not
provide clinically meaningful information. With longer followup of the participants, we expect to be able to address these
issues in the future. Fourth, because we were testing associations

VITAMIN D AND PHYSICAL PERFORMANCE IN THE ELDERLY

30. Scragg R, Camargo CA Jr. Frequency of leisure-time physical activity

and serum 25-hydroxyvitamin D levels in the US population: results
from the Third National Health and Nutrition Examination Survey. Am
J Epidemiol 2008;168:57786.
31. Pasco JA, Henry MJ, Nicholson GC, Brennan SL, Kotowicz MA.
Behavioural and physical characteristics associated with vitamin D
status in women. Bone 2009;44:108591.
32. Millen AE, Wactawski-Wende J, Pettinger M, Melamed ML, Tylavsky
FA, Liu S, Robbins J, LaCroix AZ, LeBoff MS, Jackson RD. Predictors
of serum 25-hydroxyvitamin D concentrations among postmenopausal
women: the Womens Health Initiative Calcium plus Vitamin D clinical trial. Am J Clin Nutr 2010;91:132435.
33. Brock K, Huang WY, Fraser DR, Ke L, Tseng M, Stolzenberg-Solomon
R, Peters U, Ahn J, Purdue M, Mason RS, et al. Low vitamin D status is
associated with physical inactivity, obesity and low vitamin D intake in
a large US sample of healthy middle-aged men and women. J Steroid
Biochem Mol Biol 2010;121:4626.
34. Shea MK, Houston DK, Tooze JA, Davis CC, Johnson MA, Hausman
DB, Cauley JA, Bauer DC, Tylavsky F, Harris TB, et al. Correlates and
prevalence of insufficient 25-hydroxyvitamin D status in black and
white older adults: the health, aging and body composition study. J Am
Geriatr Soc 2011;59:116574.
35. Bertrand KA, Giovannucci E, Liu Y, Malspeis S, Eliassen AH, Wu K,
Holmes MD, Laden F, Feskanich D. Determinants of plasma 25hydroxyvitamin D and development of prediction models in three US
cohorts. Br J Nutr 2012;108:188996.
36. Shirazi L, Almquist M, Malm J, Wirfalt E, Manjer J. Determinants of
serum levels of vitamin D: a study of life-style, menopausal status,
dietary intake, serum calcium, and PTH. BMC Womens Health 2013;
13:33.
37. Wortsman J, Matsuoka LY, Chen TC, Lu Z, Holick MF. Decreased
bioavailability of vitamin D in obesity. Am J Clin Nutr 2000;72:6903.
38. Nakamura K, Kitamura K, Takachi R, Saito T, Kobayashi R, Oshiki R,
Watanabe Y, Tsugane S, Sasaki A, Yamazaki O. Impact of demographic, environmental, and lifestyle factors on vitamin D sufficiency
in 9084 Japanese adults. Bone 2015;74:107.
39. Giovannucci E, Liu Y, Rimm EB, Hollis BW, Fuchs CS, Stampfer MJ,
Willett WC. Prospective study of predictors of vitamin D status and
cancer incidence and mortality in men. J Natl Cancer Inst 2006;98:
4519.
40. Hintzpeter B, Mensink GB, Thierfelder W, Muller MJ, Scheidt-Nave
C. Vitamin D status and health correlates among German adults. Eur J
Clin Nutr 2008;62:107989.
41. Hirani V, Mosdol A, Mishra G. Predictors of 25-hydroxyvitamin D
status among adults in two British national surveys. Br J Nutr 2009;
101:7604.
42. Ashwell M, Stone EM, Stolte H, Cashman KD, Macdonald H,
Lanham-New S, Hiom S, Webb A, Fraser D. UK Food Standards Agency
Workshop Report: an investigation of the relative contributions of diet
and sunlight to vitamin D status. Br J Nutr 2010;104:60311.
43. Brock KE, Ke L, Tseng M, Clemson L, Koo FK, Jang H, Seibel MJ,
Mpofu E, Fraser DR, Mason RS. Vitamin D status is associated with
sun exposure, vitamin D and calcium intake, acculturation and attitudes
in immigrant East Asian women living in Sydney. J Steroid Biochem
Mol Biol 2013;136:2147.
44. Jang H, Koo FK, Ke L, Clemson L, Cant R, Fraser DR, Seibel MJ,
Tseng M, Mpofu E, Mason RS, et al. Culture and sun exposure in
immigrant East Asian women living in Australia. Women Health 2013;
53:50418.
45. Kung AW, Lee KK. Knowledge of vitamin D and perceptions and
attitudes toward sunlight among Chinese middle-aged and elderly
women: a population survey in Hong Kong. BMC Public Health 2006;
6:226.
46. Crowe FL, Steur M, Allen NE, Appleby PN, Travis RC, Key TJ.
Plasma concentrations of 25-hydroxyvitamin D in meat eaters, fish
eaters, vegetarians and vegans: results from the EPIC-Oxford study.
Public Health Nutr 2011;14:3406.
47. Bjrneboe GE, Johnsen J, Bjrneboe A, Rousseau B, Pedersen JI,
Norum KR, Mrland J, Drevon CA. Effect of alcohol consumption on
serum concentration of 25-hydroxyvitamin D3, retinol, and retinolbinding protein. Am J Clin Nutr 1986;44:67882.
48. Brot C, Jorgensen NR, Sorensen OH. The influence of smoking on
vitamin D status and calcium metabolism. Eur J Clin Nutr 1999;53:
9206.

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016

7. Palacios C, Gonzalez L. Is vitamin D deficiency a major global public

health problem? J Steroid Biochem Mol Biol 2014;144:13845.
8. Nimitphong H, Holick MF. Vitamin D status and sun exposure in
southeast Asia. Dermatoendocrinol 2013;5:347.
9. Lee MS, Li HL, Hung TH, Chang HY, Yang FL, Wahlqvist ML. Vitamin D intake and its food sources in Taiwanese. Asia Pac J Clin Nutr
2008;17:397407.
10. Chang CI, Chan DC, Kuo KN, Hsiung CA, Chen CY. Vitamin D insufficiency and frailty syndrome in older adults living in a Northern
Taiwan community. Arch Gerontol Geriatr 2010;50(Suppl 1):S1721.
11. Tsai KS, Hsu SH, Cheng JP, Yang RS. Vitamin D stores of urban
women in Taipei: effect on bone density and bone turnover, and seasonal variation. Bone 1997;20:3714.
12. Lee WP, Lin LW, Yeh SH, Liu RH, Tseng CF. Correlations among
serum calcium, vitamin D and parathyroid hormone levels in the elderly in southern Taiwan. J Nurs Res 2002;10:6572.
13. Annweiler C, Schott AM, Berrut G, Fantino B, Beauchet O. Vitamin
D-related changes in physical performance: a systematic review. J Nutr
Health Aging 2009;13:8938.
14. Halfon M, Phan O, Teta D. Vitamin D: a review on its effects on
muscle strength, the risk of fall, and frailty. Biomed Res Int 2015;2015:
953241.
15. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB.
Lower-extremity function in persons over the age of 70 years as
a predictor of subsequent disability. N Engl J Med 1995;332:55661.
16. Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF,
Blazer DG, Scherr PA, Wallace RB. A short physical performance
battery assessing lower extremity function: association with selfreported disability and prediction of mortality and nursing home admission. J Gerontol 1994;49:M8594.
17. Guralnik JM, Ferrucci L, Pieper CF, Leveille SG, Markides KS, Ostir
GV, Studenski S, Berkman LF, Wallace RB. Lower extremity function
and subsequent disability: consistency across studies, predictive models,
and value of gait speed alone compared with the short physical performance battery. J Gerontol A Biol Sci Med Sci 2000;55:M22131.
18. Wu IC, Chang HY, Hsu CC, Chiu YF, Yu SH, Tsai YF, Shen SC, Kuo
KN, Chen CY, Liu K, et al. Association between dietary fiber intake
and physical performance in older adults: a nationwide study in Taiwan.
PLoS One 2013;8:e80209.
19. Lee MM, Lee F, Ladenla SW, Miike R. A semiquantitative dietary
history questionnaire for Chinese Americans. Ann Epidemiol 1994;4:
18897.
20. Lee MM, Chang IY, Horng CF, Chang JS, Cheng SH, Huang A. Breast
cancer and dietary factors in Taiwanese women. Cancer Causes Control
2005;16:92937.
21. Rantanen T, Guralnik JM, Foley D, Masaki K, Leveille S, Curb JD,
White L. Midlife hand grip strength as a predictor of old age disability.
JAMA 1999;281:55860.
22. Michikawa T, Nishiwaki Y, Takebayashi T, Toyama Y. One-leg
standing test for elderly populations. J Orthop Sci 2009;14:67585.
23. ATS Committee on Proficiency Standards for Clinical Pulmonary
Function Laboratories. ATS statement: guidelines for the six-minute
walk test. Am J Respir Crit Care Med 2002;166:1117.
24. Bischoff HA, Stahelin HB, Monsch AU, Iversen MD, Weyh A, von
Dechend M, Akos R, Conzelmann M, Dick W, Theiler R. Identifying
a cut-off point for normal mobility: a comparison of the timed up and
go test in community-dwelling and institutionalised elderly women.
Age Ageing 2003;32:31520.
25. Podsiadlo D, Richardson S. The timed Up & Go: a test of basic
functional mobility for frail elderly persons. J Am Geriatr Soc 1991;39:
1428.
26. Institute of Medicine, Food and Nutrition Board. Dietary reference
intakes for calcium and vitamin D. Washington (DC): National
Academy Press; 2011.
27. Schweder T, Spjtvoll E. Plots of P-values to evaluate many tests simultaneously. Biometrika 1982;69:493502.
28. Sempos CT, Vesper HW, Phinney KW, Thienpont LM, Coates PM.
Vitamin D status as an international issue: national surveys and the
problem of standardization. Scand J Clin Lab Invest Suppl 2012;243:
3240.
29. Brock K, Cant R, Clemson L, Mason RS, Fraser DR. Effects of diet
and exercise on plasma vitamin D (25(OH)D) levels in Vietnamese
immigrant elderly in Sydney, Australia. J Steroid Biochem Mol Biol
2007;103:78692.

1343

1344

CHUANG ET AL.

49. Fawehinmi TO, Ilomaki J, Voutilainen S, Kauhanen J. Alcohol consumption and dietary patterns: the FinDrink study. PLoS One 2012;7:
e38607.
50. Mathe C, Van PG, Vaes B, Adriaensen W, Gruson D, Degryse JM. No
relation between vitamin D status and physical performance in
the oldest old: results from the Belfrail study. Age Ageing 2013;42:
18690.
51. Pramyothin P, Techasurungkul S, Lin J, Wang H, Shah A, Ross PD,
Puapong R, Wasnich RD. Vitamin D status and falls, frailty, and
fractures among postmenopausal Japanese women living in Hawaii.
Osteoporos Int 2009;20:195562.
52. Bischoff-Ferrari HA, Dietrich T, Orav EJ, Hu FB, Zhang Y, Karlson
EW, Dawson-Hughes B. Higher 25-hydroxyvitamin D concentrations
are associated with better lower-extremity function in both active and
inactive persons aged . or =60 y. Am J Clin Nutr 2004;80:7528.

53. Boxer RS, Dauser DA, Walsh SJ, Hager WD, Kenny AM. The association between vitamin D and inflammation with the 6-minute walk
and frailty in patients with heart failure. J Am Geriatr Soc 2008;56:
45461.
54. Sohl E, de Jongh RT, Heijboer AC, Swart KM, Brouwer-Brolsma EM,
Enneman AW, de Groot CP, van der Velde N, Dhonukshe-Rutten RA,
Lips P, et al. Vitamin D status is associated with physical performance:
the results of three independent cohorts. Osteoporos Int 2013;24:
18796.
55. Annweiler C, Bridenbaugh S, Schott AM, Berrut G, Kressig RW,
Beauchet O. Vitamin D and muscle function: new prospects?
Biofactors 2009;35:34.
56. Peiris AN, Bailey BA, Guha BN, Copeland R, Manning T. Can a model
predictive of vitamin D status be developed from common laboratory
tests and demographic parameters? South Med J 2011;104:6369.

Downloaded from ajcn.nutrition.org at Viet Nam: ASNA Sponsored on November 6, 2016